GCN/TAN
                INTERNET  SOCKET  PACKET  DEFINITION  DOC
            Scott Barthelmy       05 Dec 94  (revised 22 Jul 17)


MODS:  The material in this document on KONUS_Lightcurve notices
       is not quite settled.                                          31jul04
       The watchdog timeout flag bit in SWIFT_FOM_OBS
       moved from 2^25 to 2^29 (to be like all the others).           06apr06
       The MILAGRO notice has been finalized.                         12may06
       The material for GLAST is nearly stable at this time,
       and should be used for guidance as to what things
       will look like for launch/operations.                          16oct06
       Added the catnum to BAT_POS packet.                            12apr07
       Change the definition of the TRIG_ID 2^11 bit
       in the Swift/BAT_POS (and LC, FOM, SLEW, TRANS  packets
       (from 0.0 to 6.5 sigma).                                       04may07
       Added GLAST_LAT_GRB_POS_TEST.                                  12may07
       Finished the GLAST OBS_REQUEST and REQUEST_REPONSE.            13may07
       More stuff in the GLAST_LAT_INI/UPD/FIN.                       15may07
       Start the AGILE stuff.                                         16may07
       Added 'misc' field to SWIFT_POINTDIR.
       Added the near_brt_star flagbit to Swift, AGILE, and GLAST.    30may07
       Tweaked the GLAST_LAT types.                                   02jun07
       Added the Gnd-generated bit to the Swift_PointDir Misc.        13jun07
       Added the Swift-BAT_SUBTHRESHOLD type.                         21jun07
       Added the galaxy coincidence flag bits to Swift-BAT_POS/_TRANS/_TEST
       and XRT_POS/_IMAGE.                                            27jun07
       Added the galaxy coincidence flag bits to Swift-BAT_LC and _SM
       and FOM_OBS and SC_SLEW.                                       30jun07
       Added the Swift-BAT_SLEW_GRB_POSITION type.                    30sep07
       Added the 2^14 bit to SolnStatus in SWIFT_BAT_GRB_POS and
       SWIFT_XRT_POS.                                                 11jan08
       Move GLAST_GBM_LC from 112 to 113; added POS_UPD as 112.       19jan08
       Added the 2^5, 2^14, 2^15 and 2^30 bit fields to PKT_MISC
       in XRT_POS_CENTROID.                                           26jan05
       Added the 2^29 bit field to the PKT_MISC in XRT_POS_CENTROID.  13feb08
       Several changes to the GLAST_LAT_INI/UPD/FIN/TRANS/TEST.       18feb08
       Added the 3 e_range bits to the MISC field in BAT_SLEW_POS.    24feb08
       Added the 'far_away' bit to MISC in XRT_IMAGE and _POS.        03apr08
       Added 'rmvd_catalog' bit to 'soln_status' in BAT_POS/LC/
       SCALEDMAP/FOM/SLEW.                                            12apr08
       Added XRT-BAT_position distance to XRT_POSITION.
       Added UVOT-XRT_position distance to UVOT_POSITION.             14apr08
       Added AGILE_POINTDIR.                                          20apr08
       Added SIMBAD/NED Results.                                      11jun08
       Added many flagbits to TRIG_ID and MISC for the AGILE and
       GLAST packets.                                                 06jul08
       Added the IPN_RAW definition.                                  18jul08
       About a dozen changes/additions to the GLAST-GBM pkts.         03aug08
       Changed all GLAST to FERMI.                                    27aug08
       Added internal/scratch field to slot 38 in IPN_RAW.            18sep08
       Added hi-precission error to XRT_POS and UVOT_POS.
       Fixed the defintion of "burst_error" in UVOT_POS.              19sep08
       Added Trig_Signif and Lon,Lat to Fermi-GBM_ALERT.              04oct08
       Added Trigger_Duration to Fermi-GBM_FLT_POS.                   04oct08
       Added missing definition for bit 22 in MISC in SWIFT_FOM.      27oct08
       Added the TOO-initiated forms of SWIFT_FOM and _SLEW.          08nov08
       Improved IPN_POS description.                                  11jan09
       Changed XTE-PCA_Alert and XTE_ASM_TRANS to NLA.                21jan09
       Added the spatial and temporal coincidence bits to the IPN_POS,
       KONUS_LC, INTEGRAL_SPIACS/_WAKEUP/_REFINED/_OFFLINE, SWIFT_BAT_GRB_ALERT,
       SWIFT_BAT_GRB_POS, SWIFT_XRT_POS, SWIFT_UVOT_POS, SWIFT_BAT_SUBTHRESHOLD,
       SWIFT_BAT_TRANS, SWIFT_BAT_SLEW_GRB_POS, AGILE_GRB_WAKEUP/_GROUND/_REFINED,
       FERMI_GBM_ALERT, FERMI_GBM_FLT_POS, FERMI_GBM_GND_POS,
       FERMI_LAT_GRB_POS_INI/_UPD/_DIAG, FERMI_LAT_GND_REF/_TRIG.     24jan09
       Improved the definition of the Fermi-LAT "temp_stat" and
       "image_stat" fields.                                           18feb09
       Added some fields to the FERMI_LAT_POS types.                  08apr09
       Added the 2^0 flagbit to the PKT_MISC fields in the BAT_LIGHTCURVE,
       BAT_SCALEDMAP, FOM, SLEW, and XRT_POS types.                   21apr09
       Added the PIOTS_OT_POS definition.                             03jul09
       Added TGF to the Fermi classification table.                   24aug09
       Added Inten1/2/3 to FERMI_LAT_GND_REF.                         19oct09
       Added the spatial and temporal coincidence bits to the COUNTERPART.
       Added the temporal coincidence bit to the IPN_RAW.             07nov09
       Made the doc match reality that there is image_signif
       in type=65 and 66 packets if the MISC 2^24 bit is set.         22nov09
       Fixed the Misc field in SWIFT_BAT_QL.                          31dec09
       Added FERMI_GBM_FINAL_POS (type=115).                          05jan10
       Added SWIFT_BAT_MONITOR (type=133).                            08feb10
       Fixed the mistake in FERMI_GND_POS (=112): signif is in
       deci-sigma (not the usual centi-sigma).                        16feb10
       Changed LAT_GND_REF to LAT_GND; LAT_GND_TRIG to LAT_OFFLINE.   11mar10
       Added Nearby_Galaxy flag to BAT_POS_ACK.                       09apr10
       Fixed a mistake in the documentation for IPN_SEG slots 10-26.  09may10
       Started the MAXI packet definitions.                           19may10
       Per request, Changed FERMI_GBM_FINAL_POS to FERMI_GBM_HITL_POS.27jun10
       Added INTEGRAL_WEAK.                                           18jan11
       Added LOCATION_QUALITY to 120 and 121.                         26feb11
       Changed (slightly) definition of 2^0 bit in slot[18] in pkt 112.
       Added 2^1 bit in slot[18] of pkt 112.                          03mar11
       Finished the MAXI packet definitions; made public.             12mar11
       Converted FERMI_OBS_REQ to FERMI_LAT_MONITOR.                  21apr11
       Replaced SNEWS with AAVSO.                                     06may11
       Added neg flux value flag to MAXI TRIGGER_ID field.            08may11
       Fixed definition of slot[15] in MAXI types.                    09may11
       Added Bright-star and Nearby-galaxy flags to BAT_MONITOR MISC. 28may11
       Added DOW_TOD Test.                                            06jun11
       Changed FERMI_GBM_HITL_POS back to FERMI_GBM_FINAL_POS.        06jun11
       Started the MOA entry.                                         13jun11
       Fixed up (finally) the GBM_TEST documentation.                 15jun11
       Started the OGLE entry.                                        03jul11
       Updated/completed/change the FERMI_LAT_GND information.        04aug11
       The MAXI_Unknown type was updated to reflect MAXI changes.     24sep11
       Added BAT_SUB_SUB and KNOWN_SRC.                               06nov11
       Slightly improved the description of 'def_not_grb'.            15jan12
       Added FERMI_GBM_ALERT_INTERNAL, FERMI_GBM_FLT_INTERNAL, and
       FERMI_SC_SLEW_INTERNAL.                                        17mar12
       Fixed up IPN_RAW (type=31) for the new timestamp-only era.     20jun12
       Added In-Gnd-Catalog flagbit to BAT_QUICKLOOK.                 31jul12
       Moved the 2 'scratch' in IPN_RAW from 37-38 to 20-21.
       Added the url string (starting at 22) to IPN_RAW.                06jan13
       Corrected definition of 2^0 bit in TYPE_SWIFT_BAT_MONITOR,
       and added unknown_src flag bit (2^1).                            22jan13
       Added the hi-precission form of the trigger_sod to KONUS_LC.     03feb13
       Added two more bitflags to TYPE_SWIFT_BAT_MONITOR"flags" field.  23mar13
       Added SNEWS back in (now at 149).                                09jun13
       Fixed a few of the SNEWS bitflag definitions.                    14nov13
       Added Fermi-GBM lightcurve url information.                      13apr14
       Added more Fermi-GBM lightcurve url information.                 23apr14
       Added Long-v-Short_GRB in FERMI_GBM_GND_POS(type=112).           03feb15
       Changed the numbering of the 5 LVC-related types.                13feb15
       Added the actual format/content information for the LVC-types.   04apr15
       Started the AMON entry.                                          25apr15
       Added 'internal' flag to LVC_* misc[].                           03jun15
       Added the theta-less-than-10arcmin flagbit in BAT_POS PKT_MISC
       at 2^12.                                                         09jul15
       Change the base-string portion of the skymap URL's in types 151-152.
       Fixed the definitions of the "fluence" and "FAR" quantities
       stored in slots 9 and 11 in packet types 150-153.                23jul15
       Split AMON into AMON_ICECUBE_COINC and AMON_ICECUBE_HESE.        25jul15
       Started CALET_GBM (2 types).                                     08aug15
       Filled in some/all of the gaps in AMON and CALET.                09aug15
       Added the spatial/temporal flagbits to the LVC, AMON, and CALET. 17aug15
       Finally filled in 4 holes in the item descriptions fo
       SWIFT_XRT_IMAGE and SWIFT_XRT_IMAGE_PROCESSED.                   25aug15
       Fixed the multiplicative values for 'eta' and 'chirpmass' in LVC.08sep15
       Added Rank and DetWarn to LVC_COUNTERPART.                       29dec15
       Added already-looking-at-trigger-source to XRT_POS.              31dec15
       Added FERMI_GBM_FIN_POS_INTERNAL.                                08jan16
       Added the long-missing description of 'iexpo' in COUNTERPART.    10jan16
       In AMON_HESE remove hese_type, change signalness to signal_trackness.
                                                                        27mar16
       More adjustments/cleanup to the AMON_HESE packet.                10apr16
       Added ST_LoL flag (bit10, [MISC]) to SPER notice.                26apr16
       Added TYPE_COINCIDENCE.                                          03jul16
       Converted Signalness in AMON_ICECUBE_EHE to log() usage.         06jul16
       AMON_ICECUBE_EHE converted to public distribution.               15jul16
       Added TYPE_GWHEN_COINC.                                          23aug16
       Added subsec field to AMON_ICECUBE_{HESE,EHE,COINC} types        09sep16
       Added TYPE_SWIFT_ACTUAL_POINTDIR.                                03oct16
       Added the long-missing long/latittude fields into the documentation
       for the SWIFT_POINTDIR notice.                                   03oct16
       Correct the packing order for the hi- and lo-energy boundaries
       (in slot 17) for the two CALET notices.  Remove the e_index
       from the MISC slot.                                              08oct16
       Added HardwareIng,Vetted,&OpenAlert flagbits to the LVC P/I/U/T. 20oct16
       Added to this documentation the (long missing) 2^30 bit definition
       in TRIGGER_ID slot for the FERMI_LAT_* notice types.             31oct16
       Added 'charge' and 'energy' to AMON_ICECUBE_EHE.                 02nov16
       Added ProbHasNS and ProbHasRemnant to the LVC types.             23dec16
       Removed Eta,ChirpMass,MaxDistance from LVC (as CBC no longer
       supports these Params).                                          23dec16
       Added the long-missing documentation on the "trig_id" slot
       to FERMI_SLEW [*_INTERNAL).                                      28dec16
       Added explanation that the URL is present for the 2nd issuance
       of the LVC_PRELIMINARY notice.                                   19jan17
       Started adding FERMI_GBM_SUBTHRESHOLD.                           02apr17
       Corrected the definitions (scaling factor) for the "significance" field
       in the FERMI_GBM_ALERT[_INTERNAL], FERMI_GBM_FLT_POS[_INTERNAL],
       FERMI_GBM_TRANS, and FERMI_GBM_POS_TEST notice types.
       (The FERMI_GBM_GND_POS[_INTERNAL] was correct.)                  25may17
       Remove the 6th and 7th slots from COINCIDENCE.                   28may17
       Fermi corrected their correction: GBM_ALERT is back to deci-sigma
       and GBM_FLT is back to centi-sigma.                              02jun17
       Added AMON_ICECUBE_TEST (type=159).                              30jun17
       Put back the 6th and 7th slots from COINCIDENCE.                 15jul17
       Added the long-missing letter-code (eg G-series, M-series, ...)
       to the LVC_COUNTERPART section.                                  22jul17



TABLE OF CONTENTS:
      INTRODUCTION:
            Scope
            Variations in Packeting Format
            Integer vs Floating Point
            Byte and Word Order
            Obtaining an e-copy
      PACKET NAME vs TYPE NUMBER
      COMPARING THE TYPE=1,2,3,4,11,21,22,24-45,51-55,58-59,60-99,110-115,118-136,139-141,144,146,148,149,150-154,157-159,160-161,166,168,169 CONTENTS/FORMATS
      PACKET ITEM MACROS
      TYPE=1 PACKET CONTENTS (BATSE_ORIGINAL)              [NO LONGER AVAILABLE, GRO de-orbit]
            Packet Item Descriptions
      TYPE=2 PACKET CONTENTS (Test coords)
            Packet Item Descriptions
      TYPE=3 PACKET CONTENTS (Imalive)
            Packet Item Descriptions
      TYPE=4 PACKET CONTENTS (Kill)
      TYPE=11 PACKET CONTENTS (BATSE_MAXBC)                [NO LONGER AVAILABLE, GRO de-orbit]
            Packet Item Descriptions (only those unique to type=11)
      TYPE=21 PACKET CONTENTS/DESCRIPTION (Bradford Test)  [NO LONGER AVAILABLE, GRO de-orbit]
      TYPE=22 PACKET CONTENTS/DESCRIPTION (BATSE_FINAL)    [NO LONGER AVAILABLE, GRO de-orbit]
      TYPE=24 PACKET CONTENTS/DESCRIPTION (BATSE_LOCBURST) [NO LONGER AVAILABLE, GRO de-orbit]
      TYPE=25 PACKET CONTENTS (ALEXIS)                     [NO LONGER AVAILABLE, mission completed]
            Packet Item Descriptions (only those unique to type=25)
      TYPE=26 PACKET CONTENTS (XTE-PCA Alert)              [NO LONGER AVAILABLE, project stopped producing]
            Packet Item Descriptions (only those unique to type=26)
      TYPE=27 PACKET CONTENTS (XTE-PCA Source)             [NO LONGER AVAILABLE, project stopped producing]
            Packet Item Descriptions (only those unique to type=27)
      TYPE=28 PACKET CONTENTS (XTE-ASM Alert)              [NOT YET AVAILABLE, project stopped producing]
      TYPE=29 PACKET CONTENTS (XTE-ASM Source)
            Packet Item Descriptions (only those unique to type=29)
      TYPE=30 PACKET CONTENTS (COMPTEL)                    [NO LONGER AVAILABLE, GRO do-orbit]
      TYPE=31 PACKET CONTENTS (IPN_Raw)
            Packet Item Descriptions
      TYPE=32 PACKET CONTENTS (IPN_Segment)                [MAY BE RE-AVAILABLE]
            Packet Item Descriptions (only those unique to type=32)
      TYPE=33 PACKET CONTENTS (SAX-WFC Alert)              [NEVER PRODUCED by project]
      TYPE=34 PACKET CONTENTS (SAX-WFC Source)             [NO LONGER AVAILABLE, mission completed]
            Packet Item Descriptions (only those unique to type=34)
      TYPE=35 PACKET CONTENTS (SAX-NFI Alert)              [NEVER PRODUCED by project]
      TYPE=36 PACKET CONTENTS (SAX-NFI Source)             [NO LONGER AVAILABLE, mission completed]
            Packet Item Descriptions (only those unique to type=36)
      TYPE=37 PACKET CONTENTS (RXTE-ASM_XTRANS)            [NO LONGER AVAILABLE, project stopped producing]
            Packet Item Descriptions (only those unique to type=37)
      TYPE=38 PACKET CONTENTS (spare for development testing)
      TYPE=39 PACKET CONTENTS (IPN_Position)
            Packet Item Descriptions (only those unique to type=39)
      TYPE=40 PACKET CONTENTS (HETE_S/C_ALERT Source)      [NO LONGER AVAILABLE, mission completed]
            Packet Item Descriptions
      TYPE=41 PACKET CONTENTS (HETE_S/C_UPDATE Source)     [NO LONGER AVAILABLE, mission completed]
            Packet Item Descriptions (only those unique to type=41)
      TYPE=42 PACKET CONTENTS (HETE_S/C_LAST Source)       [NO LONGER AVAILABLE, mission completed]
            Packet Item Descriptions (only those unique to type=42)
      TYPE=43 PACKET CONTENTS (HETE_GROUND_ANALYSIS Source)[NO LONGER AVAILABLE, mission completed]
            Packet Item Descriptions (only those unique to type=43)
      HETE FLAG BITS  VS  NOTICE TYPE
      TYPE=44 PACKET CONTENTS (HETE_Test Source)
      HETE-BASED PACKET ITEM MACROS
      TYPE=45 PACKET CONTENTS (GRB_COUNTERPART Source)
            Packet Item Descriptions
      TYPE=46 PACKET CONTENTS (SWIFT_TOO_FOM_OBSERVE)
            Packet Item Descriptions
      TYPE=47 PACKET CONTENTS (SWIFT_TOO_SC_SLEW)
            Packet Item Descriptions
      TYPE=48 PACKET CONTENTS (DOW_TOD Test)
            Packet Item Descriptions
      TYPE=51 PACKET CONTENTS (INTEGRAL_POINTDIR)
            Packet Item Descriptions
      TYPE=52 PACKET CONTENTS (INTEGRAL_SPIACS)
            Packet Item Descriptions
      TYPE=53 PACKET CONTENTS (INTEGRAL_WAKEUP Source)
            Packet Item Descriptions
      TYPE=54 PACKET CONTENTS (INTEGRAL_REFINED Source)
            Packet Item Descriptions
      TYPE=55 PACKET CONTENTS (INTEGRAL_OFFLINE Source)
            Packet Item Descriptions
      TYPE=56 PACKET CONTENTS (INTEGRAL_WEAK Source)
            Packet Item Descriptions
      TYPE=57 PACKET CONTENTS (AAVSO Event)                [NOT YET AVAILABLE]
            Packet Item Descriptions
      TYPE=58 PACKET CONTENTS (MILAGRO Source)             [NO LONGER AVAILABLE, project completed]
            Packet Item Descriptions
      TYPE=59 PACKET CONTENTS (KONUS_Lightcurve)           [SOON TO BE AVAILABLE]
            Packet Item Descriptions
      TYPE=60 PACKET CONTENTS (SWIFT_BAT_GRB_ALERT)
            Packet Item Descriptions
      TYPE=61 PACKET CONTENTS (SWIFT_BAT_GRB_POSITION Source)
            Packet Item Descriptions
      TYPE=62 PACKET CONTENTS (SWIFT_BAT_GRB_NACK_POSITION)
            Packet Item Descriptions
      TYPE=63 PACKET CONTENTS (SWIFT_BAT_GRB_LIGHTCURVE)
            Packet Item Descriptions
      TYPE=64 PACKET CONTENTS (SWIFT_BAT_SCALED_MAP)       [NOT AVAILABLE TO THE PUBLIC]
            Packet Item Descriptions
      TYPE=65 PACKET CONTENTS (SWIFT_FOM_OBSERVE)
            Packet Item Descriptions
      TYPE=66 PACKET CONTENTS (SWIFT_SC_SLEW)
            Packet Item Descriptions
      TYPE=67 PACKET CONTENTS (SWIFT_XRT_POSITION Source)
            Packet Item Descriptions
      TYPE=68 PACKET CONTENTS (SWIFT_XRT_SPECTRUM)
            Packet Item Descriptions
      TYPE=69 PACKET CONTENTS (SWIFT_XRT_IMAGE)
            Packet Item Descriptions
      TYPE=70 PACKET CONTENTS (SWIFT_XRT_LIGHTCURVE)
            Packet Item Descriptions
      TYPE=71 PACKET CONTENTS (SWIFT_XRT_NACK_POSITION)
            Packet Item Descriptions
      TYPE=72 PACKET CONTENTS (SWIFT_UVOT_IMAGE)
            Packet Item Descriptions
      TYPE=73 PACKET CONTENTS (SWIFT_UVOT_SRC_LIST)
            Packet Item Descriptions
      TYPE=76 PACKET CONTENTS (SWIFT_BAT_GRB_LIGHTCURVE_PROC) [NOT YET AVAILABLE]
            Packet Item Descriptions
      TYPE=77 PACKET CONTENTS (SWIFT_XRT_SPECTRUM_PROC)
            Packet Item Descriptions
      TYPE=78 PACKET CONTENTS (SWIFT_XRT_IMAGE_PROC)
            Packet Item Descriptions
      TYPE=79 PACKET CONTENTS (SWIFT_UVOT_IMAGE_PROC)
            Packet Item Descriptions
      TYPE=80 PACKET CONTENTS (SWIFT_UVOT_SRC_LIST_PROC)
            Packet Item Descriptions
      TYPE=81 PACKET CONTENTS (SWIFT_UVOT_POSITION Source)
            Packet Item Descriptions
      TYPE=82 PACKET CONTENTS (SWIFT_BAT_GRB_POS_TEST Source)
            Packet Item Descriptions
      TYPE=83 PACKET CONTENTS (SWIFT_POINTDIR)
            Packet Item Descriptions
      TYPE=84 PACKET CONTENTS (SWIFT_BAT_TRANS Source)
            Packet Item Descriptions
      TYPE=85 PACKET CONTENTS (SWIFT_XRT_THRESHPIX)           [NOT PUBLIC, SWIFT TEAM ONLY]
            Packet Item Descriptions
      TYPE=86 PACKET CONTENTS (SWIFT_XRT_THRESHPIX_PROC)      [NOT PUBLIC, SWIFT TEAM ONLY]
            Packet Item Descriptions
      TYPE=87 PACKET CONTENTS (SWIFT_XRT_SPER)                [NOT PUBLIC, SWIFT TEAM ONLY]
            Packet Item Descriptions
      TYPE=88 PACKET CONTENTS (SWIFT_XRT_SPER_PROC)           [NOT PUBLIC, SWIFT TEAM ONLY]
            Packet Item Descriptions
      TYPE=89 PACKET CONTENTS (SWIFT_UVOT_NACK_POSITION)
            Packet Item Descriptions
      TYPE=97 PACKET CONTENTS (SWIFT_BAT_QUICKLOOK_POSITION Source)
            Packet Item Descriptions
      TYPE=98 PACKET CONTENTS (SWIFT_BAT_SUBTHRESHOLD_POSITION Source)
            Packet Item Descriptions
      TYPE=99 PACKET CONTENTS (SWIFT_BAT_SLEW_GRB_POSITION Source)
            Packet Item Descriptions
      TYPE=100 PACKET CONTENTS (SuperAGILE_GRB_POS_WAKEUP Source)
            Packet Item Descriptions
      TYPE=101 PACKET CONTENTS (SuperAGILE_GRB_POS_GROUND Source)
            Packet Item Descriptions
      TYPE=102 PACKET CONTENTS (SuperAGILE_GRB_POS_REFINED Source)
            Packet Item Descriptions
      TYPE=103 PACKET CONTENTS (SWIFT_ACTUAL_POINTDIR)
            Packet Item Descriptions
      TYPE=107 PACKET CONTENTS (AGILE_POINTDIR)
            Packet Item Descriptions
      TYPE=109 PACKET CONTENTS (SuperAGILE_GRB_POS_TEST Source)
            Packet Item Descriptions
      TYPE=110 PACKET CONTENTS (FERMI_GBM_ALERT)
      TYPE=116 PACKET CONTENTS (FERMI_GBM_ALERT_INT)          [NOT PUBLIC, FERMI TEAM ONLY]
            Packet Item Descriptions
      TYPE=111 PACKET CONTENTS (FERMI_GBM_FLT_POS Source)
      TYPE=117 PACKET CONTENTS (FERMI_GBM_FLT_INT Source)     [NOT PUBLIC, FERMI TEAM ONLY]
            Packet Item Descriptions
      TYPE=112 PACKET CONTENTS (FERMI_GBM_GND_POS Source)
            Packet Item Descriptions
      TYPE=114 PACKET CONTENTS (FERMI_GBM_GND_INT Source)     [NOT PUBLIC, FERMI TEAM ONLY]
            Packet Item Descriptions
      TYPE=115 PACKET CONTENTS (FERMI_GBM_FINAL_POS Source)
            Packet Item Descriptions
      TYPE=119 PACKET CONTENTS (FERMI_GBM_POS_TEST Source)
            Packet Item Descriptions
      TYPE=120 PACKET CONTENTS (FERMI_LAT_GRB_POS_INI Source) [NOT PUBLIC, FERMI TEAM ONLY]
            Packet Item Descriptions
      TYPE=121 PACKET CONTENTS (FERMI_LAT_GRB_POS_UPD Source)
            Packet Item Descriptions
      TYPE=122 PACKET CONTENTS (FERMI_LAT_GRB_POS_DIAG Source)[NOT PUBLIC, FERMI TEAM ONLY]
            Packet Item Descriptions
      TYPE=123 PACKET CONTENTS (FERMI_LAT_TRANS Source)
            Packet Item Descriptions
      TYPE=124 PACKET CONTENTS (FERMI_LAT_GRB_POS_TEST Source)
            Packet Item Descriptions
      TYPE=125 PACKET CONTENTS (FERMI_LAT_MONITOR)
            Packet Item Descriptions
      TYPE=126 PACKET CONTENTS (FERMI_SC_SLEW)
      TYPE=144 PACKET CONTENTS (FERMI_SC_SLEW_INT)            [NOT PUBLIC, FERMI TEAM ONLY]
            Packet Item Descriptions
      TYPE=127 PACKET CONTENTS (FERMI_LAT_GND Source)
            Packet Item Descriptions
      TYPE=128 PACKET CONTENTS (FERMI_LAT_OFFLINE Source)
            Packet Item Descriptions
      TYPE=129 PACKET CONTENTS (FERMI_POINTDIR)
            Packet Item Descriptions
      TYPE=130 PACKET CONTENTS (SIMBAD/NED Search Results)
            Packet Item Descriptions
      TYPE=131 PACKET CONTENTS (FERMI_GBM_SUBTHRESHOLD transient)
            Packet Item Descriptions
      TYPE=133 PACKET CONTENTS (SWIFT_BAT_MONITOR)
            Packet Item Descriptions
      TYPE=134 PACKET CONTENTS (MAXI_UNKNOWN_SOURCE)
            Packet Item Descriptions
      TYPE=135 PACKET CONTENTS (MAXI_KNOWN_SOURCE)
            Packet Item Descriptions
      TYPE=136 PACKET CONTENTS (MAXI_TEST)
            Packet Item Descriptions
      TYPE=137 PACKET CONTENTS (OGLE)                         [NOT YET AVAILABLE]
            Packet Item Descriptions
      TYPE=139 PACKET CONTENTS (MOA)
            Packet Item Descriptions
      TYPE=140 PACKET CONTENTS (SWIFT_BAT_SUB_SUB_THRESH_POS)
            Packet Item Descriptions
      TYPE=141 PACKET CONTENTS (SWIFT_BAT_KNOWN_SOURCE_POS)
            Packet Item Descriptions
      TYPE=145 PACKET CONTENTS (Coincidence)
            Packet Item Descriptions
      TYPE=146 PACKET CONTENTS (FERMI_GBM_FINAL_POS_INT Source)
            Packet Item Descriptions
      TYPE=148 PACKET CONTENTS (SUZAKU_Lightcurve)
            Packet Item Descriptions
      TYPE=149 PACKET CONTENTS (SNEWS)
            Packet Item Descriptions
      TYPE=150 PACKET CONTENTS (LVC_PRELIMIARY)               [NOT YET PUBLIC, LVC TEAM & MOU PARTNERS ONLY]
            Packet Item Descriptions
      TYPE=151 PACKET CONTENTS (LVC_INITIAL)                  [NOT YET PUBLIC, LVC TEAM & MOU PARTNERS ONLY]
            Packet Item Descriptions
      TYPE=152 PACKET CONTENTS (LVC_UPDATE)                   [NOT YET PUBLIC, LVC TEAM & MOU PARTNERS ONLY]
            Packet Item Descriptions
      TYPE=153 PACKET CONTENTS (LVC_TEST)                     [NOT YET PUBLIC, LVC TEAM & MOU PARTNERS ONLY]
            Packet Item Descriptions
      TYPE=154 PACKET CONTENTS (LVC_COUNTERPART)              [NOT YET PUBLIC, LVC TEAM & MOU PARTNERS ONLY]
            Packet Item Descriptions
      TYPE=157 PACKET CONTENTS (AMON_ICECUBE_COINC)           [NOT YET PUBLIC, AMON TEAM ONLY]
            Packet Item Descriptions
      TYPE=158 PACKET CONTENTS (AMON_ICECUBE_HESE)
            Packet Item Descriptions
      TYPE=159 PACKET CONTENTS (AMON_ICECUBE_TEST)
            Packet Item Descriptions
      TYPE=160 PACKET CONTENTS (CALET_GBM_FLT_LC)
            Packet Item Descriptions
      TYPE=160 PACKET CONTENTS (CALET_GBM_GND_LC)
            Packet Item Descriptions
      TYPE=166 PACKET CONTENTS (AMON_ICECUBE_CLUSTER)         [NOT YET PUBLIC, AMON TEAM ONLY]
            Packet Item Descriptions
      TYPE=168 PACKET CONTENTS (GWHEN_COINC)                  [NOT YET PUBLIC, GWHEN TEAM ONLY]
            Packet Item Descriptions
      TYPE=169 PACKET CONTENTS (AMON_ICECUBE_EHE)
            Packet Item Descriptions
      COORDINATES EPOCH and QUANTIZATION COMPARISON




INTRODUCTION:

Scope:
This is the Socket Packet Format Specification Document.
It contains the complete format and contents of each of the packets that are
sent to GCN/TAN Internet socket sites.  This document lists only those packet types
available to the GCN/TAN Sites -- there are other packet types (not listed
in this doc) that are used for internal communication and processing
purposes within the GCN/TAN system (ie types 5-10, 12-20, and 23, and
90-96 are Swift_Tean internal messages).

Variations in Packeting Format:
Given the intrinsic variation of data types across all the missions,
it is not possible to have the same format for all packet types, nor is it
possible to have the same locations for those items that do appear
in more than one mission (packet type).  Whenever possible, herculean effort
was made to put the same items in standard locations within the packet.

Integer vs Floating Point:
The basic packet is 160 bytes composed of 40 long_words (4-byte integers).
Integers are used even for the intrinsically floating point quantities
so as to make the transfer between platforms and operating systems easier.
All floating point quantities have been scaled up to preserve the decimal
fraction and then integerized.  There are only two levels of scaling and
quantization for angles.  For those coordinates source with relative large uncertainties,
the scaling is 100x [0.01 degrees].  And for those sources with better accuracy,
the scalling is 10,000x [0.0001 degrees].  For the few non-angle quantities
ther can be other scaling factors.  [Fl.pt. were handdled this way because
in the early days of GCN, IEEE Fl.Pt. format were not at all common on computers.]
All that remains for inter-platform compatibility is possibly word and/or byte swapping.

Ranges:
For some quantities the dynamic range of the values is given.  Sometimes
the upper limit of the range is given as "infinity".  In practice,
"infinity" means the largest value that can be expressed the data type
being used (32K or 64K for signed/unsigned 4-byte integer). 

Byte and Word Order:
These socket packets were originally created on a Sun Sparc platform
(which is a big-endian machine). But even though the packets are now
created on a small-endian machine, the byte order is adjusted back 
to the big-endian order just prior to being written out the socket 
connections.  This rotation is done to maintain format continuity
over the years.  If the GCN/TAN recipient site is small-endien machine,
you will have to perform byte and/or word swapping on the 40 long-word
fields to get them into a proper order for its computer.  The location
of the '\n' character within the last 4-byte field in the packet
can be used to diagnose the byte and/or word swapping necessary.

Obtaining an e-copy:
Most of this document is preformatted, so the user can easily get a copy
of this document by using the get_document_source command within
their browser and removing the purposely very small number of html-ism's.



PACKET NAME vs TYPE NUMBER

This table lists all the currently valid notices (by their name) versus
the packet type number.

     NUMBER   NAME               COMMENT

      1       BATSE_ORIGINAL     NO LONGER AVAILABLE
      2       Test
      3       Imalive
      4       Kill
     5-10     [internal use]     n/a
     11       BATSE_MAXBC        NO LONGER AVAILABLE
     12-20    [internal use]     n/a
     21       Bradford_TEST      NO LONGER AVAILABLE
     22       BATSE_FINAL        NO LONGER AVAILABLE
     24       BATSE_LOCBURST     NO LONGER AVAILABLE
     25       ALEXIS             NO LONGER AVAILABLE
     26       RXTE-PCA_ALERT     NO LONGER AVAILABLE
     27       RXTE-PCA           NO LONGER AVAILABLE
     28       RXTE-ASM_ALERT     NO LONGER AVAILABLE
     29       RXTE-ASM           NO LONGER AVAILABLE
     30       COMPTEL            NO LONGER AVAILABLE
     31       IPN_RAW
     32       IPN_SEGMENT        MAY BE RE-AVAILABLE
     33       SAX-WFC_ALERT      NOT AVAILABLE
     34       SAX-WFC            NO LONGER AVAILABLE
     35       SAX-NFI_ALERT      NOT AVAILABLE
     36       SAX-NFI            NO LONGER AVAILABLE
     37       RXTE-ASM_XTRANS    NO LONGER AVAILABLE
     38       spare/testing
     39       IPN_POSITION
     40       HETE_S/C_ALERT     NO LONGER AVAILABLE
     41       HETE_S/C_UPDATE    NO LONGER AVAILABLE
     42       HETE_S/C_LAST      NO LONGER AVAILABLE
     43       HETE_GNDANA        NO LONGER AVAILABLE
     44       HETE_Test
     45       GRB_COUNTERPART
     46       SWIFT_TOO_FOM_OBSERVE
     47       SWIFT_TOO_SC_SLEW
     48       DOW_TOD_TEST
     51       INTEGRAL_POINTDIR
     52       INTEGRAL_SPIACS
     53       INTEGRAL_WAKEUP
     54       INTEGRAL_REFINED
     55       INTEGRAL_OFFLINE
     56       INTEGRAL_WEAK
     57       AAVSO                            NOT YET AVAILABLE
     58       MILAGRO                          NO LONGER AVAILABLE
     59       KONUS_LIGHTCURVE                 SOON TO BE AVAILABLE
     60       SWIFT_BAT_GRB_ALERT
     61       SWIFT_BAT_GRB_POSITION
     62       SWIFT_BAT_GRB_NACK_POSITION
     63       SWIFT_BAT_GRB_LIGHTCURVE
     64       SWIFT_BAT_SCALED_MAP             NOT PUBLIC, SWIFT TEAM ONLY
     65       SWIFT_FOM_OBSERVE
     66       SWIFT_SC_SLEW
     67       SWIFT_XRT_POSITION
     68       SWIFT_XRT_SPECTRUM
     69       SWIFT_XRT_IMAGE
     70       SWIFT_XRT_LIGHTCURVE
     71       SWIFT_XRT_NACK_POSITION
     72       SWIFT_UVOT_IMAGE
     73       SWIFT_UVOT_SRC_LIST
     76       SWIFT_BAT_GRB_PROC_LIGHTCURVE    NOT YET AVAILABLE
     77       SWIFT_XRT_PROC_SPECTRUM
     78       SWIFT_XRT_PROC_IMAGE
     79       SWIFT_UVOT_PROC_IMAGE
     80       SWIFT_UVOT_PROC_SRC_LIST
     81       SWIFT_UVOT_POSITION
     82       SWIFT_BAT_GRB_POS_TEST
     83       SWIFT_POINTDIR
     84       SWIFT_BAT_TRANS
     85       SWIFT_XRT_THRESHPIX              NOT PUBLIC, SWIFT TEAM ONLY
     86       SWIFT_XRT_THRESHPIX_PROC         NOT PUBLIC, SWIFT TEAM ONLY
     87       SWIFT_XRT_SPER                   NOT PUBLIC, SWIFT TEAM ONLY
     88       SWIFT_XRT_SPER_PROC              NOT PUBLIC, SWIFT TEAM ONLY
     89       SWIFT_UVOT_NACK_POSITION
     97       SWIFT_BAT_QUICKLOOK_POSITION
     98       SWIFT_BAT_SUBTHRESHOLD_POSITION
     99       SWIFT_BAT_SLEW_GRB_POSITION
     100      SuperAGILE_GRB_POS_WAKEUP
     101      SuperAGILE_GRB_POS_GROUND
     102      SuperAGILE_GRB_POS_REFINED
     103      SWIFT_ACTUAL_POINTDIR
     107      AGILE_POINTDIR
     109      SuperAGILE_GRB_POS_TEST
     110      FERMI_GBM_ALERT
     111      FERMI_GBM_FLT_POS
     112      FERMI_GBM_GND_POS
     114      FERMI_GBM_GND_INTERNAL           NOT PUBLIC, FERMI TEAM ONLY
     115      FERMI_GBM_FIN_POS
     116      FERMI_GBM_ALERT_INTERNAL         NOT PUBLIC, FERMI TEAM ONLY
     117      FERMI_GBM_FLT_INTERNAL           NOT PUBLIC, FERMI TEAM ONLY
     119      FERMI_GBM_POS_TEST
     120      FERMI_LAT_GRB_POS_INI            NOT PUBLIC, FERMI TEAM ONLY
     121      FERMI_LAT_GRB_POS_UPD
     122      FERMI_LAT_GRB_POS_DIAG           NOT PUBLIC, FERMI TEAM ONLY
     123      FERMI_LAT_TRANS
     124      FERMI_LAT_GRB_POS_TEST
     125      FERMI_LAT_MONITOR
     126      FERMI_SC_SLEW
     127      FERMI_LAT_GND
     128      FERMI_LAT_OFFLINE
     129      FERMI_POINTDIR
     130      SIMBAD/NED_SEARCH_RESULTS
     131      FERMI_GBM_SUBTHRESHOLD
     133      SWIFT_BAT_MONITOR
     134      MAXI_UNKNOWN_SOURCE
     135      MAXI_KNOWN_SOURCE
     136      MAXI_TEST
     137      OGLE                             Soon to be AVAILABLE
     139      MOA
     140      SWIFT_BAT_SUB_SUB_THRESH_POS
     141      SWIFT_BAT_KNOWN_SRC_POS
     144      FERMI_SC_SLEW_INTERNAL           NOT PUBLIC, FERMI TEAM ONLY
     145      COINCIDENCE
     146      FERMI_GBM_FIN_POS_INTERNAL       NOT PUBLIC, FERMI TEAM ONLY
     148      SUZAKU_LIGHTCURVE
     149      SNEWS
     150      LVC_PRELIMINARY                  NOT YET PUBLIC, LVC TEAM & MOU PARTNERS ONLY
     151      LVC_INITIAL                      NOT YET PUBLIC, LVC TEAM & MOU PARTNERS ONLY
     152      LVC_UPDATE                       NOT YET PUBLIC, LVC TEAM & MOU PARTNERS ONLY
     153      LVC_TEST                         NOT YET PUBLIC, LVC TEAM & MOU PARTNERS ONLY
     154      LVC_COUNTERPART                  NOT YET PUBLIC, LVC TEAM & MOU PARTNERS ONLY
     157      AMON_ICECUBE_COINC               NOT YET PUBLIC, AMON TEAM ONLY
     158      AMON_ICECUBE_HESE
     159      AMON_ICECUBE_TEST
     160      CALET_GBM_FLT_LC
     161      CALET_GBM_GND_LC
     166      AMON_ICECUBE_CLUSTER             NOT YET PUBLIC, AMON TEAM ONLY
     168      GWHEN_COINC                      NOT YET PUBLIC, GWHEN TEAM ONLY
     169      AMON_ICECUBE_EHE



COMPARING THE TYPE=1,2,3,4,11,21,22,24-45,51-55,58-59,
60-89,98,99,100-102,107-109,110-113,115,118-137,139-141,144,145,146,148,149,
150-154,157,158,160,161,168,169 CONTENTS/FORMATS:

The contents and formats of all the valid packets for distribution are listed
below.  The descriptions and definitions for each of the items within
the packets are given in the sections following each packet type.
Entries with a "-" or a "spare" in the table are undefined spare slots.

Type:  1             2            3            4              11
Loc:   ORIGINAL      TEST         IMALIVE      KILL           MAXBC

0      pkt_type      pkt_type     pkt_type     pkt_type       pkt_type
1      pkt_sernum    pkt_sernum   pkt_sernum   pkt_sernum     pkt_sernum
2      pkt_hopcnt    pkt_hopcnt   pkt_hopcnt   pkt_hopcnt     pkt_hopcnt
3      pkt_sod       pkt_sod      pkt_sod      pkt_sod        pkt_sod
4      trig_num      trig_num     -            -              trig_num
5      burst_tjd     burst_tjd    burst_tjd    burst_tjd      burst_tjd
6      burst_sod     burst_sod    burst_sod    burst_sod      burst_sod
7      burst_ra      burst_ra     -            -              burst_ra
8      burst_dec     burst_dec    -            -              burst_dec
9      burst_inten   burst_inten  -            -              burst_inten
10     burst_peak    burst_peak   -            -              burst_peak
11     burst_error   burst_error  -            -              burst_error
12     sc_az         sc_az=0      -            -              sc_az
13     sc_el         sc_el=0      -            -              sc_el
14     sc_x_ra       sc_x_ra=0    -            -              sc_x_ra
15     sc_x_dec      sc_x_dec=0   -            -              sc_x_dec
16     sc_z_ra       sc_z_ra=0    -            -              sc_z_ra
17     sc_z_dec      sc_z_dec=0   -            -              sc_z_dec
18     trig_id       trig_id      -            -              trig_id
19     misc          misc         -            -              misc
20     earth_sc_az   earth_sc_az  -            -              earth_sc_az   
21     earth_sc_el   earth_sc_el  -            -              earth_sc_el
22     sc_radius     sc_radius    -            -              sc_radius
23     t_peak        spare        -            -              maxc1[0]
24     sam_used      spare        -            -              maxc1[1]
25     spare         spare        -            -              maxc1[2]
26     spare         spare        -            -              maxc1[3]
27     spare         spare        -            -              maxc1[4]
28     spare         spare        -            -              maxc1[5]
29     spare         spare        -            -              maxc1[6]
30     spare         spare        -            -              maxc1[7]
31     spare         spare        -            -              maxbc[0]
32     spare         spare        -            -              maxbc[1]
33     spare         spare        -            -              maxbc[2]
34     spare         spare        -            -              maxbc[3]
35     spare         spare        -            -              maxbc[4]
36     spare         spare        -            -              maxbc[5]
37     spare         spare        -            -              maxbc[6]
38     spare         spare        -            -              maxbc[7]
39     pkt_term      pkt_term     pkt_term     pkt_term       pkt_term



Type:  21            22           24           25
Loc:   BRAD_COORDS   FINAL        LOCBURST     ALEXIS

0      pkt_type      pkt_type     pkt_type     pkt_type
1      pkt_sernum    pkt_esrnum   pkt_sernum   pkt_sernum
2      pkt_hopcnt    pkt_hopcnt   pkt_hopcnt   pkt_hopcnt
3      pkt_sod       pkt_sod      pkt_sod      pkt_sod
4      trig_num      trig_num     trig_num     src_num
5      burst_tjd     burst_tjd    burst_tjd    trans_tjd
6      burst_sod     burst_sod    burst_sod    trans_sod
7      burst_ra      burst_ra     burst_ra     trans_ra
8      burst_dec     burst_dec    burst_dec    trans_dec
9      burst_inten   burst_inten  burst_inten  -
10     burst_peak    burst_peak   burst_peak   -
11     burst_error   burst_error  burst_error  trans_error
12     sc_az=0       sc_az        sc_az        -
13     sc_el=0       sc_el        sc_el        -
14     sc_x_ra=0     sc_x_ra      sc_x_ra      -
15     sc_x_dec=0    sc_x_dec     sc_x_dec     -
16     sc_z_ra=0     sc_z_ra      sc_z_ra      -
17     sc_z_dec=0    sc_z_dec     sc_z_dec     -
18     -             trig_id      trig_id      -
19     misc          misc         misc         -
20     -             earth_sc_az  -            -
21     -             earth_sc_el  -            -
22     -             sc_radius    -            trans_alpha
23     -             -            -            map_duration
24     -             -            -            tele_id
25     -             -            -            -
26     -             -            -            -
27     -             -            -            -
28     -             -            -            -
29     -             -            -            -
30     -             -            -            -
31     -             -            -            -
32     -             -            -            -
33     -             -            -            -
34     -             -            -            -
35     -             -            -            -
36     -             -            -            -
37     -             -            -            -
38     -             -            -            -
39     pkt_term      pkt_term     pkt_term     pkt_term



Type:  26             27          28            29           30 
Loc:   RXTE-PCA_ALERT RXTE-PCA    XTE-ASM_ALERT XTE-ASM      COMPTEL

0      pkt_type       pkt_type    pkt_type      pkt_type     pkt_type
1      pkt_sernum     pkt_sernum  pkt_sernum    pkt_sernum   pkt_sernum
2      pkt_hopcnt     pkt_hopcnt  pkt_hopcnt    pkt_hopcnt   pkt_hopcnt
3      pkt_sod        pkt_sod     pkt_sod       pkt_sod      pkt_sod
4      trig_num       trig_num    trig_num      trig_num     trig_num
5      burst_tjd      burst_tjd   burst_tjd     burst_tjd    burst_tjd
6      burst_sod      burst_sod   burst_sod     burst_sod    burst_sod
7      locburst_ra    burst_ra    burst_ra      burst_ra     burst_ra
8      locburst_dec   burst_dec   burst_dec     burst_dec    burst_dec
9      -              burst_inten -             burst_inten  burst_inten
10     -              burst_peak  -             burst_peak   burst_peak
11     -              burst_error -             burst_error  -
12     -              -           -             burst_cont   -
13     -              -           -             -            -
14     -              -           -             -            -
15     -              -           -             -            -
16     -              -           -             -            -
17     -              -           -             -            -
18     -              -           trig_id       trig_id      trig_id
19     misc           misc        misc          misc         misc
20     -              -           -             -            -
21     -              -           -             -            -
22     -              -           -             -            -
23     -              -           -             -            -
24     obs_tjd        obs_tjd     -             err_ra1      err_ra1
25     obs_sod        obs_sod     -             err_dec1     err_dec1
26     -              -           -             err_ra2      err_ra2
27     -              -           -             err_dec2     err_dec2
28     -              -           -             err_ra3      err_ra3
29     -              -           -             err_dec3     err_dec3
30     -              -           -             err_ra4      err_ra4
31     -              -           -             err_dec4     err_dec4
32     -              -           -             length       det_signif
33     -              -           -             width        -
34     -              -           -             pos_angle    -
35     -              -           -             -            -
36     -              -           -             -            -
37     -              -           -             -            -
38     -              -           -             -            -
39     pkt_term       pkt_term    pkt_term      pkt_term     pkt_term


Type:  31             32
Loc:   IPN_RAW        IPN_SEG

0      pkt_type       pkt_type
1      pkt_sernum     pkt_sernum
2      pkt_hopcnt     pkt_hopcnt
3      pkt_sod        pkt_sod
4      trig_id        trig_num
5      burst_tjd      burst_tjd
6      burst_sod      burst_sod
7      ipn_center_ra  ipn_center_ra
8      ipn_center_dec ipn_center_dec
9      ipn_radius     ipn_radius
10     ipn_t_window   -
11     ipn_width      ipn_width
12     -              3sig_ra
13     -              3sig_dec
14     evt_duration   2sig_ra
15     -              2sig_dec
16     -              1sig_ra
17     -              1sig_dec
18     trig_id        max_prob_ra
19     misc           max_prob_dec
20     scratch        1sig_ra
21     scratch        1sig_dec
22     url_c0-3       2sig_ra
23     url_c4-7       2sig_dec
24     url_c8-11      3sig_ra
25     url_c12-15     3sig_dec
26     url_c16-19     -
27     url_...        -
28     url_...        -
29     url_...        -
30     url_...        -
31     url_...        -
32     url_...        -
33     url_...        -
34     url_...        -
35     url_...        -
36     url_...        -
37     url_...        -
38     url_c64-67     -
39     pkt_term       pkt_term


Type:  33            34           35             36
Loc:   SAX-WFC_ALERT SAX-WFC      SAX-NFI_ALERT  SAX-NFI

0      pkt_type      pkt_type     pkt_type       pkt_type
1      pkt_sernum    pkt_sernum   pkt_sernum     pkt_sernum
2      pkt_hopcnt    pkt_hopcnt   pkt_hopcnt     pkt_hopcnt
3      pkt_sod       pkt_sod      pkt_sod        pkt_sod  
4      -             -            -              -       
5      burst_tjd     burst_tjd    burst_tjd      burst_tjd
6      burst_sod     burst_sod    burst_sod      burst_sod
7      burst_ra      burst_ra     burst_ra       burst_ra 
8      burst_dec     burst_dec    burst_dec      burst_dec
9      -             burst_inten  -              burst_inten 
10     -             burst_peak   -              burst_peak  
11     -             burst_error  -              burst_error
12     -             burst_cont   -              burst_cont
13     -             -            -              -
14     -             -            -              -
15     -             -            -              -
16     -             -            -              -
17     -             -            -              -
18     trig_id       trig_id      trig_id        trig_id
19     misc          misc         misc           misc  
20     -             -            -              -
21     -             -            -              -
22     -             -            -              -
23     -             -            -              -
24     -             -            -              -
25     -             -            -              -
26     -             -            -              -
27     -             -            -              -
28     -             -            -              -
29     -             -            -              -
30     -             -            -              -
31     -             -            -              -
32     -             -            -              -
33     -             -            -              -
34     -             -            -              -
35     -             -            -              -
36     -             -            -              -
37     -             -            -              -
38     -             -            -              -
39     pkt_term      pkt_term     pkt_term       pkt_term


Type:  37             
Loc:   XTE-ASM_XTRANS 

0      pkt_type         
1      pkt_sernum     
2      pkt_hopcnt     
3      pkt_sod        
4      ref_num        
5      trans_tjd      
6      trans_sod      
7      trans_ra       
8      trans_dec      
9      trans_inten    
10     -              
11     trans_error    
12     -              
13     time_since     
14     chi_sq_1       
15     chi_sq_2       
16     SigNoise1      
17     SigNoise2      
18     trig_id        
19     misc           
20     -
21     -
22     -
23     -
24     ra1
25     dec1
26     ra2
27     dec2
28     ra3
29     dec3
30     ra4
31     dec4
32     line_length
33     line_width
34     pos_angle
35     -
36     -
37     -
38     -
39     pkt_term       


Type:  39
Loc:   IPN_POS

0      pkt_type
1      pkt_sernum
2      pkt_hopcnt
3      pkt_sod
4      ref_num
5      burst_tjd
6      burst_sod
7      burst_ra
8      burst_dec
9      burst_fluence
10     burst_flux
11     burst_error 
12     sample_interval
13     -
14     -
15     -
16     burst2_ra
17     burst2_dec
18     sernum
19     misc
20     -
21     box_area
22     duration
23     ra1
24     dec1
25     ra2
26     dec2
27     ra3
28     dec3
29     ra4
30     dec4
31     ra21
32     dec21
33     ra22
34     dec22
35     ra23
36     dec23
37     ra24
38     dec24
39     pkt_term


Type:  40             41              42            43            44
Loc:   HETE_S/C_ALERT HETE_S/C_UPDATE HETE_S/C_LAST HETE_GNDANA   HETE_Test

0      pkt_type       pkt_type        pkt_type      pkt_type      pkt_type
1      pkt_sernum     pkt_sernum      pkt_sernum    pkt_sernum    pkt_sernum
2      pkt_hopcnt     pkt_hopcnt      pkt_hopcnt    pkt_hopcnt    pkt_hopcnt
3      pkt_sod        pkt_sod         pkt_sod       pkt_sod       pkt_sod
4      trig_seq_num   trig_seq_num    trig_seq_num  trig_seq_num  trig_seq_num
5      burst_tjd      burst_tjd       burst_tjd     burst_tjd     burst_jtd
6      burst_sod      burst_sod       burst_sod     burst_sod     burst_sod
7      [burst_ra]     burst_ra        burst_ra      burst_ra      burst_ra
8      [burst_dec]    burst_dec       burst_dec     burst_dec     burst_dec
9      trig_flags     trig_flags      trig_flags    trig_flags    trig_flags
10     gamma_cnts     gamma_cnts      gamma_cnts    gamma_cnts    gamma_cnts
11     wxm_s2n        wxm_s2n         wxm_s2n       wxm_s2n       wxm_s2n 
12     sxc_cnts       sxc_cnts        sxc_cnts      sxc_cnts      sxc_cnts
13     gamma_time     gamma_time      gamma_time    gamma_time    gamma_time
14     wxm_time       wxm_time        wxm_time      wxm_time      wxm_time
15     sc_point       sc_point        sc_point      sc_point      sc_point
16     -              wxra1           wxra1         wxra1         wxra1
17     -              wxdec1          wxdec1        wxdec1        wxdec1
18     -              wxra2           wxra2         wxra2         wxra1
19     -              wxdec2          wxdec2        wxdec2        wxdec2
20     -              wxra3           wxra3         wxra3         wxra1
21     -              wxdec3          wxdec3        wxdec3        wxdec3
22     -              wxra4           wxra4         wxra4         wxra1
23     -              wxdec4          wxdec4        wxdec4        wxdec4
24     -              wx_errors       wx_errors     wx_errors     wx_errors
25     -              wx_dimsig       wx_dimsig     wx_dimsig     wx_dimsig
26     -              sxra1           sxra1         sxra1         sxra1
27     -              sxdec1          sxdec1        sxdec1        sxdec1
28     -              sxra2           sxra2         sxra2         sxra1
29     -              sxdec2          sxdec2        sxdec2        sxdec2
30     -              sxra3           sxra3         sxra3         sxra1
31     -              sxdec3          sxdec3        sxdec3        sxdec3
32     -              sxra4           sxra4         sxra4         sxra1
33     -              sxdec4          sxdec4        sxdec4        sxdec4
34     -              sx_errors       sx_errors     sx_errors     sx_errors
35     -              sx_simsig       sx_simsig     sx_simsig     sx_simsig
36     -              pos_flags       pos_flags     pos_flags     pos_flags
37     -              validity        validity      validity      validity
38     -              lc_img_sn       lc_img_sn     lc_img_sn     lc_img_sn  
39     pkt_term       pkt_term        pkt_term      pkt_term      pkt_term


Type:  45
Loc:   GRB_COUNTERPART

0      pkt_type
1      pkt_sernum
2      pkt_hopcnt
3      pkt_sod
4      ref_num
5      burst_tjd
6      burst_sod
7      cp_ra
8      cp_dec
9      cp_inten
10     inten_uncert
11     pos_error
12     filter
13     seeing
14     obs_start_date
15     obs_start_time
16     obs_dur
17     id_conf_level
18     trig_id
19     misc
20     tele_name_c0-3
21     tele_name_c4-7
22     tele_name_c8-11
23     tele_name_c11-15
24     name_c0-3
25     name_c4-7
26     name_...
27     name_...
28     name_...
29     name_...
30     name_...
31     name_...
32     name_...
33     name_...
34     name_...
35     name_...
36     name_...
37     name_...
38     name_c56-59
39     pkt_term


Type:  48
Loc:   DOW_TOD

0      pkt_type
1      pkt_sernum
2      pkt_hopcnt
3      pkt_sod
4      -
5      pkt_tjd
6      pkt_sod
7      -
8      -
9      -
10     -
11     -
12     -
13     -
14     -
15     -
16     -
17     -
18     -
19     -
20     -
21     -
22     -
23     -
24     -
25     -
26     -
27     -
28     -
29     -
30     -
31     -
32     -
33     -
34     -
35     -
36     -
37     -
38     -
39     pkt_term



Type:  51            52           53           54             55           56
Loc:   INTEGRAL_     INTEGRAL_    INTEGRAL_    INTEGRAL_      INTEGRAL_    INTEGAL_
       POINTDIR      SPIACS       WAKEUP       REFINED        OFFLINE      WEAK

0      pkt_type      pkt_type     pkt_type     pkt_type       pkt_type     pkt_type
1      pkt_sernum    pkt_sernum   pkt_sernum   pkt_sernum     pkt_sernum   pkt_sernum
2      pkt_hopcnt    pkt_hopcnt   pkt_hopcnt   pkt_hopcnt     pkt_hopcnt   pkt_hopcnt
3      pkt_sod       pkt_sod      pkt_sod      pkt_sod        pkt_sod      pkt_sod
4      trig_sub_num  trig_sub_num trig_sub_num trig_sub_num   trig_sub_num trig_sub_num
5      slew_tjd      burst_tjd    burst_tjd    burst_tjd      burst_tjd    burst_tjd
6      slew_sod      burst_sod    burst_sod    burst_sod      burst_sod    burst_sod
7      -             -            burst_ra     burst_ra       burst_ra     burst_ra
8      -             -            burst_dec    burst_dec      burst_dec    burst_dec
9      -             det_flags    det_flags    det_flags      det_flags    det_flags
10     -             inten_sigma  inten_sigma  inten_sigma    inten_sigma  inten_sigma
11     -             -            burst_error  burst_error    burst_error  burst_error
12     test_mpos     test_mpos    test_mpos    test_mpos      test_mpos    test_mpos
13     -             time_scale   time_scale   time_scale     time_scale   time_scale
14     next_sc_ra    -            sc_ra        sc_ra          sc_ra        sc_ra
15     next_sc_dec   -            sc_dec       sc_dec         sc_dec       sc_dec
16     -             time_error   time_error   time_error     time_error   time_error
17     -             -            -            -              -            -
18     -             -            -            -              -            -
19     misc_att      misc_att     misc_att     misc_att       misc_att     misc_att
20     -             -            -            -              -            -
21     -             -            -            -              -            -
22     -             -            -            -              -            -
23     -             -            -            -              -            -
24     -             -            -            -              -            -
25     -             -            -            -              -            -
26     -             -            -            -              -            -
27     -             -            -            -              -            -
28     -             -            -            -              -            -
29     -             -            -            -              -            -
30     -             -            -            -              -            -
31     -             -            -            -              -            -
32     -             -            -            -              -            -
33     -             -            -            -              -            -
34     -             -            -            -              -            -
35     -             -            -            -              -            -
36     -             -            -            -              -            -
37     -             -            -            -              -            -
38     -             -            -            -              -            -
39     pkt_term      pkt_term     pkt_term     pkt_term       pkt_term     pkt_term


Type:  57            58
Loc:   AAVSO         MILAGRO

0      pkt_type      pkt_type
1      pkt_sernum    pkt_sernum
2      pkt_hopcnt    pkt_hopcnt
3      pkt_sod       pkt_sod
4      trig_num      trig_num
5      event_tjd     burst_tjd
6      event_sod     burst_sod
7      event_ra      burst_ra
8      event_dec     burst_dec
9      -             burst_signal
10     -             background
11     event_error   burst_error 
12     -             significance
13     -             duration
14     -             ann_rate
15     -             zen_angle
16     -             -
17     -             -
18     trig_id       trig_id
19     misc          misc
20     -             -
21     -             -
22     -             -
23     -             -
24     -             -
25     -             -
26     -             -
27     -             -
28     -             -
29     -             -
30     -             -
31     -             -
32     -             -
33     -             -
34     -             -
35     -             -
36     -             -
37     -             -
38     -             -
39     pkt_term      pkt_term



Type:  59
Loc:   KONUS_LC

0      pkt_type
1      pkt_sernum
2      pkt_hopcnt
3      pkt_sod
4      -
5      trigger_tjd
6      trigger_sod
7      -
8      -
9      -
10     -
11     - 
12     -
13     -
14     -
15     E_min
16     E_max
17     hi_prec_sod
18     trig_id
19     -
20     -
21     -
22     url_c0-3
23     url_c4-7
24     url_c8-11 
25     url_... 
26     url_... 
27     url_... 
28     url_... 
29     url_... 
30     url_... 
31     url_... 
32     url_... 
33     url_... 
34     url_... 
35     url_... 
36     url_... 
37     url_... 
38     url_c64-67
39     pkt_term



SWIFT PACKETS:

Type:  60             61              62            63            64
Loc:   BAT_GRB_ALERT  BAT_GRB_POS     BAT_GRB_NACK  BAT_GRB_LC    BAT_SCALEDMAP

0      pkt_type       pkt_type        pkt_type      pkt_type      pkt_type
1      pkt_sernum     pkt_sernum      pkt_sernum    pkt_sernum    pkt_sernum
2      pkt_hopcnt     pkt_hopcnt      pkt_hopcnt    pkt_hopcnt    pkt_hopcnt
3      pkt_sod        pkt_sod         pkt_sod       pkt_sod       pkt_sod
4      trig_obs_num   trig_obs_num    trig_obs_num  trig_obs_num  trig_obs_num
5      trig_tjd       burst_tjd       trig_tjd      burst_tjd     burst_tjd
6      trig_sod       burst_sod       trig_sod      burst_sod     burst_sod
7      -              burst_ra        -             burst_ra      point_ra
8      -              burst_dec       -             burst_dec     point_dec
9      -              burst_flue      -             -             -
10     -              burst_ipeak     -             -             -
11     -              burst_error     -             -             -
12     -              phi             -             phi           -
13     -              theta           -             theta         -
14     -              integ_time      integ_time    delta_time    foregnd_dur
15     -              -               -             integ_time    integ_time
16     lon_lat        lon_lat         lon_lat       lon_lat       lon_lat
17     trig_index     trig_index      trig_index    trig_index    trig_index
18     soln_status    soln_status     soln_status   soln_status   soln_status
19     misc           misc            misc          misc          misc
20     -              image_signif    image_signif  image_signif  -
21     rate_signif    rate_signif     rate_signif   rate_signif   -
22     -              bkg_flue        bkg_flue      url_c0-3      url_c0-3
23     -              bkg_start       bkg_start     url_c4-7      url_c4-7
24     -              bkg_dur         bkg_dur       url_c8-11     url_c8-11 
25     -              cat_num         -             url_...       url_... 
26     -              -               -             url_...       url_... 
27     -              -               -             url_...       url_... 
28     -              -               -             url_...       url_... 
29     -              -               -             url_...       url_... 
30     -              -               -             url_...       url_... 
31     -              -               -             url_...       url_... 
32     -              -               -             url_...       url_... 
33     -              -               -             url_...       url_... 
34     -              -               -             url_...       url_... 
35     -              -               -             url_...       url_... 
36     -              merit_0-3       merit_0-3     url_...       url_... 
37     -              merit_4-7       merit_4-7     url_...       url_... 
38     -              merit_8-9       merit_8-9     url_c64-67    url_c64-67
39     pkt_term       pkt_term        pkt_term      pkt_term      pkt_term


Type:  65 (and 46)    66 (and 47)
Loc:   FOM_OBS        SC_SLEW

0      pkt_type       pkt_type
1      pkt_sernum     pkt_sernum
2      pkt_hopcnt     pkt_hopcnt
3      pkt_sod        pkt_sod
4      trig_obs_num   trig_obs_num
5      burst_tjd      burst_tjd
6      burst_sod      burst_sod
7      burst_ra       burst_ra 
8      burst_dec      burst_dec
9      roll           roll
10     lon_lat        lon_lat
11     -              -
12     -              -
13     -              wait_time
14     -              obs_time
15     integ_time     integ_time
16     soln_status    soln_status
17     trig_index     trig_index
18     at_slew_flags  slew_rtn
19     misc           misc
20     image_signif   image_signif
21     rate_signif    rate_signif
22     -              bat_mode
23     -              xrt_mode
24     -              uvot_mode
25     -              -
26     -              -
27     -              -
28     -              -
29     -              -
30     -              -
31     -              -
32     -              -
33     -              -
34     -              -
35     -              -
36     -              -
37     -              -
38     merit_value    merit_value
39     pkt_term       pkt_term


Type:  67             68              69              70            71
Loc:   XRT_POS        XRT_SPEC        XRT_IMAGE       XRT_LC        XRT_NACK_POS

0      pkt_type       pkt_type        pkt_type        pkt_type      pkt_type
1      pkt_sernum     pkt_sernum      pkt_sernum      pkt_sernum    pkt_sernum
2      pkt_hopcnt     pkt_hopcnt      pkt_hopcnt      pkt_hopcnt    pkt_hopcnt
3      pkt_sod        pkt_sod         pkt_sod         pkt_sod       pkt_sod
4      trig_obs_num   trig_obs_num    trig_obs_num    trig_obs_num  trig_obs_num
5      data_tjd       spec_start_tjd  image_start_tjd start_tjd     data_tjd
6      data_sod       spec_start_sod  image_start_sod start_sod     data_sod
7      burst_ra       bore_ra         burst_ra        bore_ra       point_ra
8      burst_dec      bore_dec        burst_dec       bore_dec      point_dec
9      burst_flue     livetime        centroid_cnt    livetime      centroid_cnts
10     -              spec_stop_tjd   -               stop_tjd      min_count
11     burst_error    spec_stop_sod   centroid_stddev stop_sod      centroid_stddev
12     x_tam_i1       mode            centroid_x      -             sig_max
13     y_tam_i1       waveform        centroid_y      -             -
14     x_tam_i2       bias            raw_x           -             -
15     y_tam_i2       -               raw_y           -             -
16     hi_prec_err    -               roll            -             ph2_iter
17     amp_wave       -               gain/mode/wave  -             max_iter
18     soln_status    -               expo_time       -             err_flag
19     misc           misc            misc            misc          misc
20     xrt-bat        -               grb_in_xrt_y    -             -
21     det_signif     term_cond       grb_in_xrt_z    term_cond     -
22     -              url_c0-3        url_c0-3        url_c0-3      -
23     -              url_c4-7        url_c4-7        url_c4-7      -
24     -              url_...         url_...         url_...       -
25     -              url_...         url_...         url_...       -
26     -              url_...         url_...         url_...       -
27     -              url_...         url_...         url_...       -
28     -              url_...         url_...         url_...       -
29     -              url_...         url_...         url_...       -
30     -              url_...         url_...         url_...       -
31     -              url_...         url_...         url_...       -
32     -              url_...         url_...         url_...       -
33     -              url_...         url_...         url_...       -
34     -              url_...         url_...         url_...       -
35     -              url_...         url_...         url_...       -
36     -              url_...         url_...         url_...       -
37     -              url_...         url_...         url_...       -
38     -              url_c64-67      url_c64-67      url_c64-67    -
39     pkt_term       pkt_term        pkt_term        pkt_term      pkt_term


Type:  72             73
Loc:   UVOT_IMAGE     UVOT_SRCLIST

0      pkt_type       pkt_type
1      pkt_sernum     pkt_sernum
2      pkt_hopcnt     pkt_hopcnt
3      pkt_sod        pkt_sod
4      trig_obs_num   trig_obs_num
5      expo_start_tjd expo_start_tjd
6      expo_start_sod expo_start_sod
7      point_ra       point_ra 
8      point_dec      point_dec
9      image_roll     image_roll
10     filter         filter
11     expo_id        bkg_mean
12     x_offset       x_max
13     y_offset       y_max
14     width          n_stars
15     height         x_offset
16     xy_grb         y_offset
17     n_frames       det_thresh
18     swl_lwl        photo_thresh
19     misc           misc
20     -              -
21     -              -
22     url_c0-3       url_c0-3
23     url_c4 7       url_c4-7
24     url_...        url_...
25     url_...        url_...
26     url_...        url_...
27     url_...        url_...
28     url_...        url_...
29     url_...        url_...
30     url_...        url_...
31     url_...        url_...
32     url_...        url_...
33     url_...        url_...
34     url_...        url_...
35     url_...        url_...
36     url_...        url_...
37     url_...        url_...
38     url_c64-67     url_c64-67
39     pkt_term       pkt_term




Type:  76             77              78              79             80
Loc:   BAT_PROC_LC    XRT_PROC_SPEC   XRT_PROC_IMAGE  UVOT_PROC_IM   UVOT_PROC_SL

0      pkt_type       pkt_type        pkt_type        pkt_type       pkt_type
1      pkt_sernum     pkt_sernum      pkt_sernum      pkt_sernum     pkt_sernum
2      pkt_hopcnt     pkt_hopcnt      pkt_hopcnt      pkt_hopcnt     pkt_hopcnt
3      pkt_sod        pkt_sod         pkt_sod         pkt_sod        pkt_sod
4      trig_obs_num   trig_obs_num    trig_obs_num    trig_obs_num   trig_obs_num
5      burst_tjd      spec_start_tjd  image_start_tjd expo_start_tjd expo_start_tjd
6      burst_sod      spec_start_sod  image_start_sod expo_start_sod expo_start_sod
7      burst_ra       bore_ra         burst_ra        point_ra       point_ra
8      burst_dec      bore_dec        burst_dec       point_dec      point_dec
9      T90_time       burst_flue      Centroid_Cnts   image_roll     image_roll
10     -              SpecStopTJD     -               filter         filter
11     -              SpecStopSOD     CentStdDev      expo_id        bkg_mean
12     phi            mode            centroid_x      x_offset       x_max
13     theta          waveform        centroid_y      y_offset       y_max
14     delta_time     -               raw_x           width          n_stars
15     integ_time     -               raw_y           height         x_offset
16     lon_lat        -               roll            xy_grb         y_offset
17     trig_index     -               gain/mode/wave  n_frames       det_thresh
18     soln_status    -               expo_time       swl_lwl        photo_thresh
19     misc           misc            misc            misc           misc
20     -              -               -               -              -
21     -              -               -               -              -
22     url_c0-3       url_c0-3        url_c0-3        url_c0-3       url_c0-3
23     url_c4-7       url_c4-7        url_c4-7        url_c4-7       url_c4-7
24     url_c8-11      url_c8-11       url_c8-11       url_c8-11      url_c8-11
25     url_...        url_...         url_...         url_...        url_... 
26     url_...        url_...         url_...         url_...        url_... 
27     url_...        url_...         url_...         url_...        url_... 
28     url_...        url_...         url_...         url_...        url_... 
29     url_...        url_...         url_...         url_...        url_... 
30     url_...        url_...         url_...         url_...        url_... 
31     url_...        url_...         url_...         url_...        url_... 
32     url_...        url_...         url_...         url_...        url_... 
33     url_...        url_...         url_...         url_...        url_... 
34     url_...        url_...         url_...         url_...        url_... 
35     url_...        url_...         url_...         url_...        url_... 
36     url_...        url_...         url_...         url_...        url_... 
37     url_...        url_...         url_...         url_...        url_... 
38     url_c64-67     url_c64-67      url_c64-67      url_c64-67     url_c64-67
39     pkt_term       pkt_term        pkt_term        pkt_term       pkt_term


Type:  81             82                83             84             103
Loc:   UVOT_POS       BAT_GRB_POS_TEST  POINTDIR       BAT_TRANS      ACTUAL_POINTDIR

0      pkt_type       pkt_type          pkt_type       pkt_type       pkt_type
1      pkt_sernum     pkt_sernum        pkt_sernum     pkt_sernum     pkt_sernum
2      pkt_hopcnt     pkt_hopcnt        pkt_hopcnt     pkt_hopcnt     pkt_hopcnt
3      pkt_sod        pkt_sod           pkt_sod        pkt_sod        pkt_sod
4      trig_obs_num   trig_obs_num      trig_obs_num   trig_obs_num   trig_obs_num
5      burst_tjd      burst_tdj         point_tjd      trans_tjd      point_tjd
6      burst_sod      burst_sod         point_sod      trans_sod      point_sod
7      burst_ra       burst_ra          point_ra       trans_ra       point_ra
8      burst_dec      burst_dec         point_dec      trans_dec      point_dec
9      burst_mag      burst_flue        point_roll     trans_flue     point_roll
10     filter         burst_ipeak       -              trans_ipeak    -
11     burst_error    burst_error       bat_mode       trans_error    -
12     -              phi               xrt_mode       phi            -
13     -              theta             uvot_mode      theta          -
14     -              integ_time        obs_time       integ_time     -
15     -              -                 merit_value    -              -
16     hi_prec_err    lon_lat           lon_lat        lon_lat        lon_lat
17     -              trig_index        -              trig_index     -
18     soln_status    soln_status       -              soln_status    -
19     misc           misc              misc           misc           misc
20     mag_error      image_signif      -              image_signif   -
21     -              rate_signif       -              rate_signif    -
22     -              bkg_flue          tgt_c0-3       bkg_flue       -
23     -              bkg_start         tgt_c4-7       bkg_start      -
24     -              bkg_dur           tgt_c8-11      bkg_dur        -
25     -              cat_num           tgt_...        cat_num        -
26     -              -                 tgt_...        -              -
27     -              -                 tgt_...        -              -
28     -              -                 tgt_...        -              -
29     -              -                 tgt_...        -              -
30     -              -                 tgt_...        -              -
31     -              -                 tgt_...        -              -
32     -              -                 tgt_...        -              -
33     -              -                 tgt_...        -              -
34     -              -                 tgt_...        -              -
35     -              -                 tgt_...        -              -
36     -              merit_0-3         tgt_...        merit_0-3      -
37     -              merit_4-7         tgt_...        merit_4-7      -
38     -              merit_8-9         tgt_c64-67     merit_8-9      -
39     pkt_term       pkt_term          pkt_term       pkt_term       pkt_term
 

Type:  85             86               87             88
Loc:   XRT_THRESHPIX  XRT_TP_PROC      XRT_SPER       XRT_SPER_PROC

0      pkt_type       pkt_type         pkt_type       pkt_type
1      pkt_sernum     pkt_sernum       pkt_sernum     pkt_sernum
2      pkt_hopcnt     pkt_hopcnt       pkt_hopcnt     pkt_hopcnt
3      pkt_sod        pkt_sod          pkt_sod        pkt_sod
4      trig_obs_num   trig_obs_num     trig_obs_num   trig_obs_num
5      burst_tjd      burst_tdj        burst_tjd      burst_tjd
6      burst_sod      burst_sod        burst_sod      burst_sod
7      point_ra       point_ra         point_ra       point_ra
8      point_dec      point_dec        point_dec      point_dec
9      expo_time      expo_time        expo_time      expo_time
10     stop_date      stop_date        stop_date      stop_date
11     stop_time      stop_time        stop_time      stop_time
12     seq_num        seq_num          num_pkt        num_pkt
13     -              -                -              -
14     -              -                -              -
15     -              -                -              -
16     -              -                -              -
17     -              -                -              -
18     -              -                -              -
19     misc           misc             misc           misc
20     -              -                num_evt        num_evt
21     -              -                -              -
22     url_c0-3       url_c0-3        url_c0-3        url_c0-3
23     url_c4-7       url_c4-7        url_c4-7        url_c4-7
24     url_c8-11      url_c8-11       url_c8-11       url_c8-11
25     url_...        url_...         url_...         url_...
26     url_...        url_...         url_...         url_...
27     url_...        url_...         url_...         url_...
28     url_...        url_...         url_...         url_...
29     url_...        url_...         url_...         url_...
30     url_...        url_...         url_...         url_...
31     url_...        url_...         url_...         url_...
32     url_...        url_...         url_...         url_...
33     url_...        url_...         url_...         url_...
34     url_...        url_...         url_...         url_...
35     url_...        url_...         url_...         url_...
36     url_...        url_...         url_...         url_...
37     url_...        url_...         url_...         url_...
38     url_c64-67     url_c64-67      url_c64-67      url_c64-67
39     pkt_term       pkt_term        pkt_term        pkt_term


Type:  89
Loc:   UVOT_POS_NACK

0      pkt_type
1      pkt_sernum
2      pkt_hopcnt
3      pkt_sod
4      trig_obs_num
5      trig_tjd
6      trig_sod
7      src_ra
8      src_dec
9      mag_lim
10     filter
11     src_error
12     src
13     -
14     -
15     -
16     -
17     -
18     soln_status
19     misc
20     lim_sigma
21     -
22     -
23     -
24     -
25     -
26     -
27     -
28     -
29     -
30     -
31     -
32     -
33     -
34     -
35     -
36     -
37     -
38     -
39     pkt_term


Type:  97                     98                     99
Loc:   BAT_QUICKLOOK_POS      BAT_SUBTHRESHOLD_POS   BAT_SLEW_GRB_POS

0      pkt_type               pkt_type               pkt_type
1      pkt_sernum             pkt_sernum             pkt_sernum
2      pkt_hopcnt             pkt_hopcnt             pkt_hopcnt
3      pkt_sod                pkt_sod                pkt_sod
4      trig_obs_num           trig_obs_num           id_num
5      burst_tjd              burst_tjd              burst_tjd
6      burst_sod              burst_sod              burst_sod
7      burst_ra               burst_ra               burst_ra
8      burst_dec              burst_dec              burst_dec
9      roll                   burst_flue             burst_flue
10     lon_lat                burst_ipeak            burst_ipeak
11     burst_error            burst_error            burst_error
12     -                      phi                    -
13     -                      theta                  -
14     -                      integ_time             integ_time
15     -                      -                      -
16     -                      lon_lat                -
17     trig_index             trig_index             trig_index
18     at_slew_flags          soln_status            soln_status
19     misc                   misc                   misc
20     -                      image_signif           image_signif
21     rate_signif            rate_signif            -
22     -                      bkg_flue               -
23     -                      bkg_start              -
24     -                      bkg_dur                -
25     -                      cat_num                -
26     -                      -                      -
27     -                      -                      -
28     -                      -                      -
29     -                      -                      -
30     -                      -                      -
31     -                      -                      -
32     -                      -                      -
33     -                      -                      -
34     -                      -                      -
35     -                      -                      -
36     -                      merit_0-3              -
37     -                      merit_4-7              -
38     merit_value            merit_8-9              -
39     pkt_term               pkt_term               pkt_term




SuperAGILE PACKETS:

Note: 100=Wakeup, 101=Ground, 102=Refined

Type:  100-102        107               108               109
Loc:   SuperAGILE_POS AGILE_POINTDIR    SuperAGILE_TRANS  SuperAGILE_POS_TEST

0      pkt_type       pkt_type          pkt_type          pkt_type
1      pkt_sernum     pkt_sernum        pkt_sernum        pkt_sernum
2      pkt_hopcnt     pkt_hopcnt        pkt_hopcnt        pkt_hopcnt
3      pkt_sod        pkt_sod           pkt_sod           pkt_sod
4      trig_num       -                 trig_num          trig_num
5      burst_tjd      curr_tjd          trans_tjd         burst_tjd
6      burst_sod      curr_sod          trans_sod         burst_sod
7      burst_ra       curr_ra           trans_ra          burst_ra
8      burst_dec      curr_dec          trans_dec         burst_dec
9      burst_intenX   -                 trans_intenX      burst_intenX
10     burst_intenY   -                 trans_intenY      burst_intenY
11     burst_error    -                 trans_error       burst_error
12     -              next_tjd          -                 -
13     -              next_sod          -                 -
14     -              -                 -                 -
15     -              -                 -                 -
16     -              -                 -                 -
17     -              -                 -                 -
18     trig_id        -                 trig_id           trig_id
19     misc           misc              misc              misc
20     burst_signif   -                 trans_signif      burst_signif
21     -              -                 -                 -
22     -              -                 -                 -
23     -              -                 -                 -
24     -              -                 -                 -
25     -              -                 -                 -
26     -              -                 -                 -
27     -              -                 -                 -
28     -              -                 -                 -
29     -              -                 -                 -
30     -              -                 -                 -
31     -              -                 -                 -
32     -              -                 -                 -
33     -              -                 -                 -
34     -              -                 -                 -
35     -              -                 -                 -
36     -              -                 -                 -
37     -              -                 -                 -
38     -              -                 -                 -
39     pkt_term       pkt_term          pkt_term          pkt_term



FERMI PACKETS:


Type:  110&116        111&117         112            113           114            115&146        118            119           131
Loc:   GBM_ALERT      GBM_FLT_POS     GBM_GND_POS    GBM_LC        GBM_INTERNAL   GBM_FINAL      GBM_TRANS      GBM_POS_TEST  GBM_SUBTHRESH

0      pkt_type       pkt_type        pkt_type       pkt_type      pkt_type       pkt_type       pkt_type       pkt_type      pkt_type
1      pkt_sernum     pkt_sernum      pkt_sernum     pkt_sernum    pkt_sernum     pkt_sernum     pkt_sernum     pkt_sernum    pkt_sernum
2      pkt_hopcnt     pkt_hopcnt      pkt_hopcnt     pkt_hopcnt    pkt_hopcnt     pkt_hopcnt     pkt_hopcnt     pkt_hopcnt    pkt_hopcnt
3      pkt_sod        pkt_sod         pkt_sod        pkt_sod       pkt_sod        pkt_sod        pkt_sod        pkt_sod       pkt_sod
4      trig_num       trig_num        trig_num       trig_num      trig_num       trig_num       trig_num       trig_num      trans_num
5      burst_tjd      burst_tjd       burst_tjd      burst_tjd     burst_tjd      burst_tjd      trans_tjd      burst_tjd     event_tjd
6      burst_sod      burst_sod       burst_sod      burst_sod     burst_sod      burst_sod      trans_sod      burst_sod     event_sod
7      -              burst_ra        burst_ra       burst_ra      burst_ra       burst_ra       trans_ra       burst_ra      event_ra
8      -              burst_dec       burst_dec      burst_dec     burst_dec      burst_dec      trans_dec      burst_dec     event_dec
9      algorithm      burst_flue      -              ???           burst_flue     -              trans_flue     burst_flue    -
10     -              -               -              ???           burst_inten    -              trans_inten    burst_inten   -
11     -              burst_error     burst_error    ???           burst_error    burst_error    trans_error    burst_error   event_error
12     -              phi             phi            -             -              -              -              -             evt_sod_subsec
13     -              theta           theta          -             -              -              -              -             earth_angle
14     trig_dur       integ_time      data_interval  ???           integ_time     -              integ_time     integ_time    duration
15     lo_chan        -               -              -             -              -              -              -             -
16     hi_chan        -               -              ???           -              -              -              -             -
17     ADC_LoChan     trig_dur        -              ???           trig_dur       -              trig_dur       trig_dur      -
18     ADC_HiChan     soln_status     -              ???           soln_status    soln_status    soln_status    soln_status   soln_status
19     misc           misc            misc           ???           misc           misc           misc           misc          misc
20     rec_seq_num    rec_seq_num     rec_seq_num    ???           rec_seq_num    -              rec_seq_num    rec_seq_num   -
21     trig_signif    rate_signif     signif         ???           signif         -              rate_signif    rate_signif   -
22     lon            loc_algor       loc_algor      ???           loc_algor      loc_algor      loc_algor      loc_algor     -
23     lat            most_likely     -              ???           -              -              most_likely    most_likely   reliability
24     -              2most_likely    -              ???           -              -              2most_likely   2most_likely  -
25     -              hard_ratio      -              -             -              -              hard_ratio     hard_ratio    -
26     dets           dets            lo_e           -             lo_e           lo_e           dets           detd          -
27     -              -               hi_e           -             hi_e           hi_e           -              -             url_c0-3
28     -              -               sc_x           -             sc_x           -              -              -             url_c4-7
29     -              -               sc_y           -             sc_y           -              -              -             url_c8-11
30     -              -               sc_z           -             sc_z           -              -              -             url_c12-15
31     lc_yymmdd      lc_yymmdd       lc_yymmdd      -             lc_yymmdd      lc_yymmdd      -              lc_yymmdd     url_c16-19
32     lc_fff         lc_fff          lc_fff         -             lc_fff         lc_fff         -              lc_fff        url_c20-23
33     -              -               -              -             -              -              -              -             url_c24-27
34     -              -               -              -             -              -              -              -             url_c28-31
35     -              -               -              -             -              -              -              -             url_c32-35
36     -              -               -              -             -              -              -              -             url_c36-39
37     -              lon             -              -             -              -              -              -             url_c40-43
38     -              lat             -              -             -              -              -              -             url_c44-47
39     pkt_term       pkt_term        pkt_term       pkt_term      pkt_term       pkt_term       pkt_term       pkt_term      pkt_term


Type:  120            121             122             123            124            125
Loc:   LAT_POS_INI    LAT_POS_UPD     LAT_POS_DIAG    LAT_TRANS      LAT_POS_TEST   LAT_MONITOR

0      pkt_type       pkt_type        pkt_type        pkt_type       pkt_type       pkt_type
1      pkt_sernum     pkt_sernum      pkt_sernum      pkt_sernum     pkt_sernum     pkt_sernum
2      pkt_hopcnt     pkt_hopcnt      pkt_hopcnt      pkt_hopcnt     pkt_hopcnt     pkt_hopcnt
3      pkt_sod        pkt_sod         pkt_sod         pkt_sod        pkt_sod        pkt_sod
4      trig_num       trig_num        trig_num        ref_num        trig_num       ref_num
5      burst_tjd      burst_tjd       burst_tjd       trans_tjd      burst_tjd      trans_tjd
6      burst_sod      burst_sod       burst_sod       trans_sod      burst_sod      trans_sod
7      burst_ra       burst_ra        burst_ra        trans_ra       burst_ra       trans_ra
8      burst_dec      burst_dec       burst_dec       trans_dec      burst_dec      trans_dec
9      tot_loc_inten  tot_loc_inten   tot_loc_inten   curr_flux      tot_loc_inten  curr_flux
10     inten_4        inten_4         inten_4         base_flux      inten_4        base_flux
11     burst_error    burst_error     burst_error     trans_error    burst_error    trans_error
12     phi            phi             phi             cur_flx_err    phi            cur_flx_err
13     theta          theta           theta           bas_flx_err    theta          bas_flx_err
14     int_time       int_time        int_time        time_scale     int_time       time_scale
15     -              -               -               e_band         -              e_band
16     -              -               -               -              -              -
17     trig_index     trig_index      trig_index      -              trig_index     -
18     trig_id        trig_id         trig_id         trig_id        trig_id        trig_id
19     misc           misc            misc            misc           misc           misc
20     rec_seq_num    rec_seq_num     rec_seq_num     -              rec_seq_num    -
21     -              -               -               -              -              -
22     -              -               -               -              -              -
23     -              -               -               -              -              -
24     -              -               -               -              -              -
25     temp_signif    temp_signif     temp_signif     signif         temp_signif    signif
26     image_signif   image_signif    image_signif    -              image_signif   -
27     -              -               -               -              -              -
28     -              -               -               -              -              -
29     -              -               -               -              -              -
30     -              -               -               -              -              -
31     1st_phot_tjd   1st_phot_tjd    1st_phot_tjd    -              -              -
32     1st_phot_sod   1st_phot_sod    1st_phot_sod    src_c0-3       -              src_c0-3
33     last_ph_tjd    last_ph_tjd     last_ph_tjd     src_c4-7       -              src_c4-7
34     last_ph_sod    last_ph_sod     last_ph_sod     src_c8-11      -              src_c8-11
35     -              -               -               src_c12-15     -              src_...
36     -              -               -               src_c16-19     -              src_...
37     burst_id       burst_id        burst_id        src_c20-23     -              src_...
38     loc_qual       loc_qual        -               src_c24-27     -              src_c24-27
39     pkt_term       pkt_term        pkt_term        pkt_term       pkt_term       pkt_term


Type:  126&144
Loc:   FERMI_SC_SLEW

0      pkt_type
1      pkt_sernum
2      pkt_hopcnt
3      pkt_sod
4      trig_num
5      point_tjd
6      point_sod
7      point_ra
8      point_dec
9      -
10     -
11     -
12     -
13     -
14     dwell_time
15     -
16     -
17     -
18     -
19     misc
20     -
21     -
22     -
23     -
24     -
25     -
26     -
27     -
28     -
29     -
30     -
31     -
32     -
33     -
34     -
35     -
36     -
37     -
38     -
39     pkt_term


Type:  127            128
Loc:   LAT_GND        LAT_OFFLINE

0      pkt_type       pkt_type
1      pkt_sernum     pkt_sernum
2      pkt_hopcnt     pkt_hopcnt
3      pkt_sod        pkt_sod
4      trig_num       trig_num
5      burst_tjd      burst_tjd
6      burst_sod      burst_sod
7      burst_ra       burst_ra
8      burst_dec      burst_dec
9      tot_inten      -
10     -              -
11     burst_error    burst_error
12     burst_phi      -
13     burst_theta    -
14     -              -
15     -              -
16     -              -
17     -              -
18     trig_id        trig_id
19     misc           misc
20     -              -
21     -              -
22     -              -
23     -              -
24     -              -
25     -              -
26     signif         -
27     inten2         -
28     inten3         -
29     inten4         -
30     -              -
31     -              -
32     -              -
33     -              -
34     -              -
35     -              -
36     -              -
37     -              -
38     -              -
39     pkt_term       pkt_term


Type:  129
Loc:   FERMI_POINTDIR

0      pkt_type
1      pkt_sernum
2      pkt_hopcnt
3      pkt_sod
4      -
5      start_tjd
6      start_sod
7      start_ra
8      start_dec
9      delta_t
10     ra_dec_1t
11     ra_dec_2t
12     ra_dec_3t
13     ra_dec_4t
14     ra_dec_5t
15     ra_dec_6t
16     ra_dec_7t
17     ra_dec_8t
18     ra_dec_9t
19     ra_dec_10t
20     ra_dec_11t
21     ra_dec_12t
22     ra_dec_13t
23     ra_dec_14t
24     ra_dec_15t
25     ra_dec_16t
26     ra_dec_17t
27     ra_dec_18t
28     ra_dec_19t
29     ra_dec_20t
30     ra_dec_21t
31     ra_dec_22t
32     ra_dec_23t
33     ra_dec_24t
34     ra_dec_25t
35     ra_dec_26t
36     ra_dec_27t
37     ra_dec_28t
38     ra_dec_29t
39     pkt_term


Type:  130
Loc:   SIMBAD/NED

0      pkt_type
1      pkt_sernum
2      pkt_hopcnt
3      pkt_sod
4      trig_obs_num
5      trig_tjd
6      trig_sod
7      ra
8      dec
9      -
10     -
11     error
12     orig_type
13     -
14     -
15     -
16     -
17     -
18     -
19     -
20     -
21     -
22     fname_c0-3
23     fname_c4-7
24     fname_c8-11
25     fname_...
26     fname_...
27     fname_...
28     fname_...
29     fname_...
30     fname_...
31     fname_...
32     fname_...
33     fname_...
34     fname_...
35     fname_...
36     fname_...
37     fname_c60-63
38     fname_c64-67
39     pkt_term


Type:  133 
Loc:   SWIFT_BAT_MONITOR

0      pkt_type
1      pkt_sernum
2      pkt_hopcnt
3      pkt_sod
4      ref_num
5      tjd
6      sod
7      ra
8      dec
9      crab_flux
10     curr_flux
11     base_flux
12     signif
13     sname_c0-3
14     sname_c4-7
15     sname_c8-11
16     sname_c12-15
17     sname_c16-19
18     soln_status
19     misc
20     -
21     -
22     url_c0-3
23     url_c4-7
24     url_...
25     url_...
26     url_...
27     url_...
28     url_...
29     url_...
30     url_...
31     url_...
32     url_...
33     url_...
34     url_...
35     url_...
36     url_...
37     url_...
38     url_c64-67
39     pkt_term



Type:  134                    135                    136
Loc:   MAXI_UNKNOWN_SOURCE    MAXI_KNOWN_SOURCE      MAXI_TEST

0      pkt_type               pkt_type               pkt_type
1      pkt_sernum             pkt_sernum             pkt_sernum
2      pkt_hopcnt             pkt_hopcnt             pkt_hopcnt
3      pkt_sod                pkt_sod                pkt_sod
4      id_num                 id_num                 id_num
5      event_tjd              src_tjd                event_tjd
6      event_sod              src_sod                event_sod
7      event_ra               src_ra                 event_ra
8      event_dec              src_dec                event_dec
9      event_flux             src_flux               event_flux
10     event_flux_err         -                      -
11     event_error            src_error              event_error
12     -                      src_bkg_lo             src_bkg_lo
13     -                      src_bkg_med            src_bkg_med
14     -                      src_bkg_lo             src_bkg_hi
15     tscale_eband           tscale_eband           tscale_eband
16     -                      lon_lat                lon_lat
17     -                      -                      -
18     evt_flags              evt_flags              evt_flags
19     misc                   misc                   misc
20     -                      -                      -
21     -                      -                      -
22     -                      srcname_c0-3           -
23     -                      srcname_c4-7           -
24     -                      srcname_c8-11          -
25     -                      srcname_c12-15         -
26     -                      srcname_c16-19         -
27     -                      srcname_c20-23         -
28     -                      srcname_c24-27         -
29     -                      srcname_c28-31         -
30     -                      -                      -
31     -                      -                      -
32     -                      -                      -
33     -                      -                      -
34     -                      -                      -
35     -                      -                      -
36     -                      -                      -
37     -                      -                      -
38     -                      -                      -
39     pkt_term               pkt_term               pkt_term


Type:  137              139 
Loc:   OGLE             MOA

0      pkt_type         pkt_type
1      pkt_sernum       pkt_sernum
2      pkt_hopcnt       pkt_hopcnt
3      pkt_sod          pkt_sod
4      id_num           id_num
5      det_tjd          det_tjd
6      det_sod          det_sod
7      ra               ra
8      dec              dec
9      max_mag          max_mag
10     curr_mag         curr_mag
11     base_mag I0      base_mag
12     base_m_err       base_m_err
13     max_tjd          max_tjd
14     max_sod Tmax     max_sod
15     max_t_err        max_t_err
16     umin             u0
17     umin_err         u0_error
18     flags            flags
19     misc             misc
20     ampl_n_err       ampl_n_err
21     tau              tE_width
22     tau_err          tE_err
23     fbl              -
24     Ibl              -
25     LeadTime         LeadTime
26     -                -
27     -                -
28     -                -
29     -                -
30     url_c0-3         url_c0-3
31     url_c4-7         url_c4-7
32     url_c8-11        url_c8-11
33     url_...          url_...
34     url_...          url_...
35     url_...          url_...
36     url_...          url_...
37     url_...          url_...
38     url_c32-35       url_c32-35
39     pkt_term         pkt_term


Type:  140                       141 
Loc:   BAT_SUB_SUB_THRESH_POS    BAT_KNOWN_SRC

0      pkt_type                  pkt_type
1      pkt_sernum                pkt_sernum
2      pkt_hopcnt                pkt_hopcnt
3      pkt_sod                   pkt_sod
4      ref_num                   ref_num
5      evt_tjd                   evt_tjd
6      evt_sod                   evt_sod
7      evt_ra                    evt_ra
8      evt_dec                   evt_dec
9      evt_inten                 evt_inten
10     -                         -
11     evt_error                 evt_error
12     phi                       phi
13     theta                     theta
14     int_time                  int_time
15     -                         -
16     -                         -
17     -                         -
18     soln_status               soln_status
19     misc                      misc
20     image_signif              image_signif
21     -                         -
22     -                         -
23     -                         -
24     -                         -
25     -                         cat_num
26     apid                      apid
27     -                         -
28     -                         -
29     -                         -
30     -                         -
31     -                         -
32     -                         -
33     -                         -
34     -                         -
35     -                         -
36     -                         -
37     -                         -
38     -                         -
39     pkt_term                  pkt_term


Type:  145
Loc:   COINCIDENCE

0      pkt_type
1      pkt_sernum
2      pkt_hopcnt
3      pkt_sod
4      contrib1_fields
5      contrib1_trignum
6      contrib1_tjd
7      contrib1_sod
8      contrib1_ra
9      contrib1_dec
10     contrib2_fields
11     contrib2_trignum
12     contrib2_tjd
13     contrib2_sod
14     contrib2_ra
15     contrib2_dec
16     contrib3_fields
17     contrib3_trignum
18     contrib3_tjd
19     contrib3_sod
20     contrib3_ra
21     contrib3_dec
22     contrib4_fields
23     contrib4_trignum
24     contrib4_tjd
25     contrib4_sod
26     contrib4_ra
27     contrib4_dec
28     contrib5_fields
29     contrib5_trignum
30     contrib5_tjd
31     contrib5_sod
32     contrib5_ra
33     contrib5_dec
34     contrib6_fields
35     contrib7_fields
36     total_contribute
37     trigger_id
38     misc
39     pkt_term


Type:  148
Loc:   SUZAKU_LC

0      pkt_type
1      pkt_sernum
2      pkt_hopcnt
3      pkt_sod
4      -
5      trigger_tjd
6      trigger_sod
7      -
8      -
9      -
10     -
11     - 
12     -
13     -
14     -
15     E_min
16     E_max
17     -
18     trig_id
19     -
20     -
21     -
22     url_c0-3
23     url_c4-7
24     url_c8-11 
25     url_... 
26     url_... 
27     url_... 
28     url_... 
29     url_... 
30     url_... 
31     url_... 
32     url_... 
33     url_... 
34     url_... 
35     url_... 
36     url_... 
37     url_... 
38     url_c64-67
39     pkt_term


Type:  149
Loc:   SNEWS

0      pkt_type
1      pkt_sernum
2      pkt_hopcnt
3      pkt_sod
4      trig_num
5      event_tjd
6      event_sod
7      event_ra
8      event_dec
9      event_fluence
10     -
11     event_error
12     containment
13     duration
14     -
15     -
16     -
17     -
18     trig_id
19     misc
20     -
21     -
22     -
23     -
24     -
25     -
26     -
27     -
28     -
29     -
30     -
31     -
32     -
33     -
34     -
35     -
36     -
37     -
38     -
39     pkt_term



Type:  150           151           152           153           154
Loc:   PRELIMINARY   INITIAL       UPDATE        TEST          COUNTERPART

0      pkt_type      pkt_type      pkt_type      pkt_type      pkt_type
1      pkt_sernum    pkt_sernum    pkt_sernum    pkt_sernum    pkt_sernum
2      pkt_hopcnt    pkt_hopcnt    pkt_hopcnt    pkt_hopcnt    pkt_hopcnt
3      pkt_sod       pkt_sod       pkt_sod       pkt_sod       pkt_sod
4      id_num        id_num        id_num        id_num        id_num
5      trigger_tjd   trigger_tjd   trigger_tjd   trigger_tjd   trigger_tjd
6      trigger_sod   trigger_sod   trigger_sod   trigger_sod   trigger_sod
7      -             -             -             -             cp_ra
8      -             -             -             -             cp_dec
9      fluence       fluence       fluence       fluence       cp_inten
10     peak_freq     peak_freq     peak_freq     peak_freq     inten_uncert
11     far           far           far           far           pos_error
12     event_type    event_type    event_type    event_type    filter
13     trig_sod_usec trig_sod_usec trig_sod_usec trig_sod_usec seeing
14     duration      duration      duration      duration      obs_start_date
15     -             -             -             -             obs_start_time
16     -             -             -             -             obs_dur
17     -             -             -             -             id_conf_level
18     trigger_ID    trigger_ID    trigger_ID    trigger_ID    trigger_id
19     misc          misc          misc          misc          misc
20     -             ProbNS,Remnt  ProbNS,Remnt  ProbNS,Remnt  tele_name_c0-3
21     -             -             -             -             tele_name_c4-7
22     -             -             -             -             tele_name_c8-11
23     -             -             -             -             tele_name_c12-15
24     -             -             -             -             name_c0-3
25     -             -             -             -             name_c4-7
26     -             -             -             -             name_...
27     -             -             -             -             name_...
28     -             -             -             -             name_...
29     [url_c0-3]    url_c0-3      url_c0-3      url_c0-3      name_...
30     [url_c4-7]    url_c4-7      url_c4-7      url_c4-7      name_...
31     [url_c8-11]   url_c8-11     url_c8-11     url_c8-11     name_...
32     [url_...]     url_...       url_...       url_...       name_...
33     [url_...]     url_...       url_...       url_...       name_...
34     [url_...]     url_...       url_...       url_...       name_...
35     [url_...]     url_...       url_...       url_...       name_...
36     [url_...]     url_...       url_...       url_...       name_...
37     [url_...]     url_...       url_...       url_...       name_...
38     [url_...]     url_c36-39    url_c36-39    url_c36-39    name_c56-59
39     pkt_term      pkt_term      pkt_term      pkt_term      pkt_term



Type:  157                  158                  158                    166                    169
Loc:   AMON_ICECUBE_COINC   AMON_ICECUBE_HESE    AMON_ICECUBE_TEST      AMON_ICECUBE_CLUSTER   AMON_ICECUBE_EHE

0      pkt_type             pkt_type             pkt_type               pkt_type               pkt_type
1      pkt_sernum           pkt_sernum           pkt_sernum             pkt_sernum             pkt_sernum
2      pkt_hopcnt           pkt_hopcnt           pkt_hopcnt             pkt_hopcnt             pkt_hopcnt
3      pkt_sod              pkt_sod              pkt_type               pkt_sod                pkt_sod
4      event_num            event_num            event_num              event_num              event_num
5      trigger_tjd          trigger_tjd          trigger_tjd            trigger_tjd            trigger_tjd
6      trigger_sod          trigger_sod          trigger_sod            trigger_sod            trigger_sod
7      ra                   ra                   ra                     ra                     ra
8      dec                  dec                  dec                    dec                    dec
9      n_events             n_events             n_events               n_events               n_events
10     false_pos            false_pos            false_pos              -                      -
11     pos_error90          pos_error90          pos_error_90           pos_error_50           pos_error50
12     -                    -                    -                      -                      -
13     deltaT               deltaT               deltaT                 -                      deltaT
14     sigmaT               sigmaT               sigmaT                 -                      sigmaT
15     observation          -                    -                      -                      energy
16     stream               stream               stream                 [stream???]            stream
17     rev                  rev                  rev                    -                      rev
18     trigger_ID           trigger_ID           trigger_ID             trigger_ID             trigger_ID
19     misc                 misc                 misc                   misc                   misc
20     pvalue               pvalue               pvalue                 -                      -
21     substream            charge               charge                 -                      charge
22     -                    signal_trackness     signal_trackness                              signalness
23     -                    run_num              run_num                -                      run_num
24     pos_error50          pos_error50          pos_error50            -                      -
25     -                    -                    -                      -                      -
26     -                    -                    -                      -                      -
27     -                    -                    -                      -                      -
28     -                    -                    -                      -                      -
29     -                    -                    -                      -                      -
30     -                    -                    -                      -                      -
31     -                    -                    -                      -                      -
32     -                    -                    -                      -                      -
33     -                    -                    -                      -                      -
34     -                    -                    -                      -                      -
35     -                    -                    -                      -                      -
36     -                    -                    -                      -                      -
37     -                    -                    -                      -                      -
38     -                    -                    -                      -                      -
39     pkt_term             pkt_term             pkt_term               pkt_term               pkt_term


Type:  160                  161
Loc:   CALET_GBM_FLT_LC     CALET_GBM_GND_LC

0      pkt_type             pkt_type
1      pkt_sernum           pkt_ernum
2      pkt_hopcnt           pkt_hopcnt
3      pkt_sod              pkt_sod
4      id_num               id_num
5      trigger_tjd          trigger_tjd
6      trigger_sod          trigger_sod
7      sc_boresight_ra      sc_boresight_ra 
8      sc_boresight_dec     sc_boresight_dec
9      lc_signif            lc_signif
10     fore_dur             fore_dur
11     bkg_dur1             bkg_dur1
12     bkg_dur2             bkg_dur2
13     -                    -
14     -                    -
15     -                    -
16     lon_lat              lon_lat
17     lo_e_hi_e            lo_e_hi_e
18     trigger_ID           trigger_ID
19     misc                 misc
20     -                    -
21     -                    -
22     -                    -
23     -                    -
24     -                    -
25     -                    -
26     -                    -
27     -                    -
28     -                    -
29     url_c0-3             url_c0-3
30     url_c4-7             url_c4-7
31     url_c8-11            url_c8-11
32     url_...              url_...
33     url_...              url_...
34     url_...              url_...
35     url_...              url_...
36     url_...              url_...
37     url_...              url_...
38     url_c36-39           url_c36-39
39     pkt_term             pkt_term


Type:  168
Loc:   GWHEN

0      pkt_type
1      pkt_sernum
2      pkt_hopcnt
3      pkt_sod
4      id_num
5      trigger_tjd
6      trigger_sod
7      coinc_ra
8      coinc_dec
9      fluence
10     peak_freq
11     lvc_far
12     event_type
13     trig_sod_usec
14     duration
15     eta
16     chirpMass
17     MaxDistance
18     trigger_ID
19     misc
20     hen_ra
21     hen_dec
22     hen_error
23     coinc_far
24     -
25     -
26     -
27     -
28     -
29     url_c0-3
30     url_c4-7
31     url_c8-11
32     url_...
33     url_...
34     url_...
35     url_...
36     url_...
37     url_...
38     url_c36-39
39     pkt_term







PACKET ITEM MACROS:

Below are the C-language define's for picking out the various elements
of the array of longs that make up the packet.  They have names which
reflect the contents of packet types 1, 2, and 3 (for historical reasons).
For the other types, these macros are reusable and for others their names
conote the wrong concept, so straight numerical indices are used in accessing
the packet buffer array.  [For the HETE-based packets, a new set of 40 macros
was define'd (see a section towards the end of this document).]

// Indices into the socket packet array of 40 longs:
#define PKT_TYPE      0   // Packet type number
#define PKT_SERNUM    1   // Packet serial number
#define PKT_HOP_CNT   2   // Packet hop counter
#define PKT_SOD       3   // Packet Sec-Of-Day [centi-sec] (sssss.sss*100)
#define BURST_TRIG    4   // BATSE Trigger number
#define BURST_TJD     5   // Truncated Julian Day
#define BURST_SOD     6   // Sec-of-Day [centi-secs] (sssss.sss*100)
#define BURST_RA      7   // RA  [centi-deg] (0.0 to 359.999 *100)
#define BURST_DEC     8   // Dec [centi-deg] (-90.0 to +90.0 *100)
#define BURST_INTEN   9   // Intensity [cnts]
#define BURST_PEAK   10   // Peak Intensity [cnts/1.024sec]
#define BURST_ERROR  11   // Location uncertainty [centi-deg]
#define SC_AZ        12   // Burst SC Az [centi-deg] (0.0 to 359.999 *100)
#define SC_EL        13   // Burst SC El [centi-deg] (-90.0 to +90.0 *100)
#define SC_X_RA      14   // SC X-axis RA [centi-deg] (0.0 to 359.999 *100)
#define SC_X_DEC     15   // SC X-axis Dec [centi-deg] (-90.0 to +90.0 *100)
#define SC_Z_RA      16   // SC Z-axis RA [centi-deg] (0.0 to 359.999 *100)
#define SC_Z_DEC     17   // SC Z-axis Dec [centi-deg] (-90.0 to +90.0 *100)
#define TRIGGER_ID   18   // Flag bits that identify the trigger type
#define MISC         19   // Misc indicator flag bits
#define E_SC_AZ      20   // Earth's center in SC Az
#define E_SC_EL      21   // Earth's center in SC El
#define SC_RADIUS    22   // Orbital radius of the GRO SC [km]
#define BURST_T_PEAK 23   // Time of Peak intensity [centi-sec] (sssss.ss*100)
#define PKT_SPARE24  24   // Beginning of spare section
#define PKT_SPARE38  38   // End of the spare section
#define PKT_TERM     39   // Packet termination character

#define SIZLBUF      40   // Number of longwords in socket packet




TYPE=1 PACKET CONTENTS:

NOTE: Type=1 packets (BATSE_Original) are NO LONGER AVAILABLE (June 2000). (since CGRO was de-orbitted).

The packet consists of 40 four-byte quantities.  The order and contents
are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=1)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       burst_trig   integer         BATSE trigger number
long         5       burst_tjd    [days]          Truncated Julian Day
long         6       burst_sod    [centi-sec]     (int)(sssss.sss *100)
long         7       burst_ra     [centi-deg]     (int)(0.0 to 359.9999 *100)
long         8       burst_dec    [centi-deg]     (int)(-90.0 to +90.0 *100)
long         9       burst_inten  [cnts]          Intensity, 0 to infinity
long         10      burst_peak   [cnts/1.024s]   Peak Intensity, 0 to infinity
long         11      burst_error  [centi-deg]     (int)(0.0 to 180.0 *100)
long         12      sc_az        [centi-deg]     (int)(0.0 to 359.9999 *100)
long         13      sc_el        [centi-deg]     (int)(-90.0 to +90.0 *100)
long         14      sc_x_ra      [centi-deg]     (int)(0.0 to 359.9999 *100)
long         15      sc_x_dec     [centi-deg]     (int)(-90.0 to +90.0 *100)
long         16      sc_z_ra      [centi-deg]     (int)(0.0 to 359.9999 *100)
long         17      sc_z_dec     [centi-deg]     (int)(-90.0 to +90.0 *100)
long         18      trigger_id   bits            Trigger identification flags
long         19      misc         bits            Misc stuff packed in here
long         20      e_sc_az      [centi-deg]     (int)(0.0 to 359.999 *100)
long         21      e_sc_el      [centi-deg]     (int)(-90.0 to +90.0 *100)
long         22      sc_radius    [km]            SC orbital radius
long         23      t_peak       [centi-sec]     (int)(sssss.sss *100)
long         24      sam_used     integer         Num samples integrated
long         25-38   spare[14]    integer         56 bytes for the future
long         39      pkt_term     integer         Pkt Termination (always = \n)


 
PACKET ITEM DESCRIPTIONS for type=1:

The "pkt_type" is an integer with a value of 1, 2 or 3 that tells
which type of packet this is and therefore what the contents are (they change
between the various types).  The types are:  1) Real BATSE trigger
initiated notices (GRBs, SFs, PEs, etc),  2) Test trigger notices,
3) Imalive packets.  The other types (4, 11, 22, and 24-34) are discussed
in other sections below.  And packet types (4-10, 12-21 and 23) never make it
to the outside world -- they are for internal GCN-use only.

The "pkt_sernum" is the serial number for that packet type.  Each packet
type has its own sequential serial numbering.  They all start from 1 when
the GCN programs start up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the machine of origin.  Do not confuse this with the burst_sod (see below).
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized).

The "burst_trig" is the uniquely identifying serial number assigned by the
on-board BATSE flight software to each trigger.  It started with SN=1 just
after launch and is currently (Jun97) in the 6000's.

The "burst_tjd" is the Truncated Julian Day of the BATSE trigger,
eg. TJD=10281 is 17 Jul 96. 

The "burst_sod" is the UT seconds-of-day of the BATSE trigger.  This is the
spacecraft system clock UT time when the on-board software found a large
enough increase in the countrate (the 5.5-sigma threshold) in two or more
detectors in the selected energy band in any of the 3 sampling time scales
(64, 256, or 1024 msec).  It is derived from the spacecraft clock which has
more accuracy, but then is truncated to 10msec level, multiplied by 100,
and converted to a 32-bit integer for the packet.

The "burst_ra" and "burst_dec" are the RA,Dec coordinates of the trigger/burst
as determined by the GCN program (current epoch).  These intrinsically
floating point quantities have been multiplied by 100 and then integerized
to make an integer quantity with units of centi-degrees.

The "burst_inten" is actually the integrated fluence of the trigger (burst).
For Original notices, it is just the counts in the first 1.024 or 2.048 sec
of the burst after the time when there was enough counts for GCN to
calculate a direction solution.

The "burst_peak" is the maximum count rate intensity seen in the
trigger (burst).  Since there are several time domains in which
the notices are sent, the peak intensity is that so far encountered
in the light curve.  For type=1 it is the brightest 1-sec sample
of the 1or2 rate samples used from that Major Frame of telemetry data.

The "burst_error" is the radius of the circle that will contain on average
68% of the bursts.  This number is generated from an emperical comparison
to the Huntsville locations from an ensemble of past bursts.

The "sc_az" and "sc_el" are the spacecraft coordinates of the trigger (burst)
as determined by the GCN program.  The location in this coordinate system
is useful for the other 3 instruments on the GRO spacecraft (especially
the elevation angle allows the two z-axis instruments to determine if it is
in their FOVs).

The "sc_x_ra", "sc_x_dec", "sc_z_ra", and "sc_z_dec" are the RA,Dec (J2000)
coordinates to which the spacecraft x- and z-axes point.  These are the four
quantities needed to turn the spacecraft az,el directions into celestial 
coordinates.

The "trigger_id" field contains bit_flags which attempt to identify
the type of transient event that triggered the BATSE instrument.
The bit packing of the "trigger_id" entry in the packet is:
Bit_Number     Item_name      Description
----------     ---------      -----------
0              suspected_grb  0=No or 1=Yes, it is a suspected GRB
1              definite_grb   0=No or 1=Yes, it is definitely a GRB
2              near_sun       0=No or 1=Yes, the direction is near the Sun
3              soft_sf        0=No or 1=Yes, the spectrum is SF soft
4              suspected_pe   0=No or 1=Yes, it is a suspected Particle evt
5              definite_pe    0=No or 1=Yes, it is definitely a Particle evt
6              x_x_x_x_xxx    spare
7              definite_unk   0=No or 1=Yes, it is definitely an Unknown
8              earthward      0=No or 1=Yes, the direction is in the Earth cone
9              soft_flag      0=No or 1=Yes, the spectrum is a little soft
10             near_saa       0=No or 1=Yes, the spacecraft location is in SAA
11             definite_saa   0=No or 1=Yes, trigger is definitely SAA related
12             suspected_sf   0=No or 1=Yes, it is a suspected Solar Flare
13             definite_sf    0=No or 1=Yes, it is definitely a Solar Flare
14             op_flag        0=No or 1=Yes, the 2 brightest dets are opposites
15             def_not_grb    0=No or 1=Yes, it is definitely not a GRB
16             iso_pe         0=No or 1=Yes, Datlowe isotropic param is low
17             iso_grb        0=No or 1=Yes, Datlowe isotropic param is high
18             neg_h_ratio    0=No or 1=Yes, the hardness ratio is negative
19             neg_iso_bc     0=No or 1=Yes, Datlowe isotropic param is negative
20             not_soft       0=No or 1=Yes, the hardness ratio is not soft
21             hi_iso_ratio   0=No or 1=Yes, Datlowe iso C3/C2 ratio is high
22             low_inten      0=No or 1=Yes, intensity too small to be a GRB
23-30          spare          spare
31             bow            0=No or 1=Yes, it is a Burst OverWrite

These define's are in C-language programs for masking to get the various items
out of "trigger_id":
#define SUSP_GRB            0x00000001  // Suspected GRB
#define DEF_GRB             0x00000002  // Definitely a GRB
#define NEAR_SUN            0x00000004  // Coords are near the Sun (<15deg)
#define SOFT_SF             0x00000008  // Spectrum is soft, h_ratio > 2.0
#define SUSP_PE             0x00000010  // Suspected Particle Event
#define DEF_PE              0x00000020  // Definitely a Particle Event
#define X_X_X_X_XXX         0x00000040  // spare
#define DEF_UNK             0x00000080  // Definitely an Unknown
#define EARTHWARD           0x00000100  // Location towards Earth center
#define SOFT_FLAG           0x00000200  // Small hardness ratio (>1.5)
#define NEAR_SAA            0x00000400  // It is near/in the SAA region
#define DEF_SAA             0x00000800  // Definitely an SAA region
#define SUSP_SF             0x00001000  // Suspected Solar Flare
#define DEF_SF              0x00002000  // Definitely a Solar Flare
#define OP_FLAG             0x00004000  // Op-dets flag set
#define DEF_NOT_GRB         0x00008000  // Definitely not a GRB event
#define ISO_PE              0x00010000  // Datelowe Iso param is small (PE)
#define ISO_GRB             0x00020000  // D-Iso param is large (GRB/SF)
#define NEG_H_RATIO         0x00040000  // Negative h_ratio
#define NEG_ISO_BC          0x00080000  // Negative iso_part[1] or iso_part[2]
#define NOT_SOFT            0x00100000  // Not soft flag, GRB or PE
#define HI_ISO_RATIO        0x00200000  // Hi C3/C2 D-Iso ratio
#define LOW_INTEN           0x00400000  // Inten too small to be a real GRB
#define X_X_X_X_XX2         0x7F800000  // spare
#define BOW_FLAG            0x80000000  // Burst OverWrite trigger



The "misc" field contains various bit_flags, the algorithm level, the
notice type, and the program source-code version numbers.
The bit packing of the "misc" entry in the packet is:
Bit_Number     Item_name      Description
----------     ---------      -----------
0              tm_indicator   0=No_TM, 1=TM_present
1              bad_calc       Singular matrix (bad calculation)
2              comptel_fov    Position is in COMPTEL FOV
3              egret_fov      Position is in EGRET FOV
4              osse_fov       Position is in OSSE FOV
5-7            level          Program coords calc level [1-4]
8-9            notice_type    0=Original coords notice,
                              1=Updates coords notice,
                              2=Delay coords notice,
                              3=Final coords notice.
10             alex_anti_sun  Position is in the Anti-Solar hemisphere (ALEXIS)
11             ipn_peak       Passed the IPN Peak Inten Threshold criteria
12             ipn_flue       Passed the IPN Fluence Threshold criteria
13             xte_crit       Passed the RXTE follow-up criteria
14             xte_type       RXTE-PCA_ALERT and RXTE-PCA packet sub-types:
                              0=RXTE_ALERT_WILL_NOT_OBSERVE
                              1=RXTE_ALERT_WILL_OBSERVE
                              0=RXTE_OBS_SAW_NOTHING
                              1=RXTE_OBS_SAW_SOMETHING
15             no_trig_ind    BATSE is not in a triggerable mode
16-27          ver_minor      Program version, Minor number [0-2047]
28-31          ver_major      Program version, Major number [1-15]

The define's for masking to get the various items out of "misc":
#define TM_IND_MASK         0x00000001  // TM Indicator mask
#define BAD_CALC_MASK       0x00000002  // Singular matrix (bad calculation)
#define C_FOV_MASK          0x00000004  // COMPTEL FOV indicator mask
#define E_FOV_MASK          0x00000008  // EGRET FOV indicator mask
#define O_FOV_MASK          0x00000010  // OSSE FOV indicator mask
#define PROG_LEV_MASK       0x000000E0  // Program Algorithm Level mask
#define NOTICE_TYPE_MASK    0x00000300  // Notification Type mask
#define A_ASD_MASK          0x00000400  // ALEXIS Anti-Solar Dir ind mask
#define IPN_PEAK_MASK       0x00000800  // Passed the IPN Peak Inten thresh
#define IPN_FLUE_MASK       0x00001000  // Passed the IPN Fluence thresh
#define XTE_CRITERIA        0x00002000  // Meets RXTE follow-up criteria
#define XTE_STATUS          0x00004000  // Won't/will obs and Didn't/did see
#define NO_TRIG_MASK        0x00008000  // BATSE is not in a triggerable mode
#define VER_MINOR_MASK      0x0FFF0000  // Program Minor Version Number mask
#define VER_MAJOR_MASK      0xF0000000  // Program Major Version Number mask


The "e_sc_az" and "e_sc_el" are the spacecraft az,el coordinates of the 
center of the Earth.

The "sc_radius" is the distance of the spacecraft from the center of the Earth.

The "t_peak" item is the UT seconds-of-day of the time of the peak/maximum
count rate [cnts/1.024sec] seen so far (remember there are a series of notices)
in the trigger/burst light curve.  This count rate has been normalized back
to be the true planewave intensity of the burst -- not for any given detector.

The "sam_used" item is the number of 1.024-sec samples of rate data
that went into the accumulation (integration).  For type=1 packets it must
be either a 1 or a 2.  For type=22 packets it can be a value of 1 to 32.

There are 14 "spare" items in the packet for future growth.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.




TYPE=2 PACKET CONTENTS:

The TEST packet consist of 40 four-byte quantities.  The order and contents
are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=2)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       burst_trig   integer         BATSE trigger number
long         5       burst_tjd    [days]          Truncated Julian Day
long         6       burst_sod    [centi-sec]     (int)(sssss.sss *100)
long         7       burst_ra     [centi-deg]     (int)(0.0 to 359.9999 *100)
long         8       burst_dec    [centi-deg]     (int)(-90.0 to +90.0 *100)
long         9       burst_inten  [cnts]          Intensity, 0 to infinity
long         10      burst_peak   [cnts/1.024s]   Peak Intensity, 0 to infinity
long         11      burst_error  [centi-deg]     (int)(0.0 to 180.0 *100)
long         12      sc_az        [centi-deg]     (int)(0.0 to 359.9999 *100)
long         13      sc_el        [centi-deg]     (int)(-90.0 to +90.0 *100)
long         14      sc_x_ra      [centi-deg]     (int)(0.0 to 359.9999 *100)
long         15      sc_x_dec     [centi-deg]     (int)(-90.0 to +90.0 *100)
long         16      sc_z_ra      [centi-deg]     (int)(0.0 to 359.9999 *100)
long         17      sc_z_dec     [centi-deg]     (int)(-90.0 to +90.0 *100)
long         18      trigger_id   bits            Trigger identification flags
long         19      misc         bits            Misc stuff packed in here
long         20      e_sc_az      [centi-deg]     (int)(0.0 to 359.999 *100)
long         21      e_sc_el      [centi-deg]     (int)(-90.0 to +90.0 *100)
long         22      sc_radius    [km]            SC orbital radius
long         23-38   spare[16]    integer         64 bytes for the future
long         39      pkt_term     integer         Pkt Termination (always = \n)


 
PACKET ITEM DESCRIPTIONS for type=2:

The "pkt_type" is an integer with a value of 1, 2 or 3 that tells
which type of packet this is and therefore what the contents are (they change
between the various types).  The types are:  1) Real BATSE trigger
initiated notices (GRBs, SFs, PEs, etc),  2) Test trigger notices,
3) Imalive packets.  The other types (4, 11, 22, and 24-34) are discussed
in other sections below.  And packet types (4-10, 12-21 and 23) never make it
to the outside world -- they are for internal GCN-use only.

The "pkt_sernum" is the serial number for that packet type.  Each packet
type has its own sequential serial numbering.  They all start from 1 when
the GCN programs start up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent.
Do not confuse this with the burst_sod (see below).
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized).

The "burst_trig" is the uniquely identifying serial number assigned by the
on-board BATSE flight software to each trigger.  It started with SN=1 just
after launch and is currently (Jun97) in the 6000's.  This is always -1
for type=2 packets.

The "burst_tjd" is the Truncated Julian Day of the BATSE trigger,
eg. TJD=10281 is 17 Jul 96.

The "burst_sod" is the UT seconds-of-day of the BATSE trigger.  This is the
spacecraft system clock UT time when the on-board software found a large
enough increase in the countrate (the 5.5-sigma threshold) in two or more
detectors in the selected energy band in any of the 3 sampling time scales
(64, 256, or 1024 msec).  It is derived from the spacecraft clock which has
more accuracy, but then is truncated to 10msec level, multiplied by 100,
and converted to a 32-bit integer for the packet.

The "burst_ra" and "burst_dec" are the RA,Dec coordinates of the trigger/burst
as determined by the GCN program (current epoch).  These intrinsically
floating point quantities have been multiplied by 100 and then integerized
to make an integer quantity with units of centi-degrees.

The "burst_inten" is set to 1000 cnts/sec for Test Notices.

The "burst_peak" is set to 1000 cnts/sec for Test Notices.

The "burst_error" is the radius of the circle that will contain on average
68% of the bursts.  This number is generated from an emperical comparison
to the Huntsville locations from an ensemble of past bursts.

The "trigger_id" field contains bit_flags which attempt to identify
the type of transient event that triggered the BATSE instrument.
The bit packing of the "trigger_id" entry in the packet is:
Bit_Number     Item_name      Description
----------     ---------      -----------
0              suspected_grb  0=No or 1=Yes, it is a suspected GRB
1              definite_grb   0=No or 1=Yes, it is definitely a GRB
2              near_sun       0=No or 1=Yes, the direction is near the Sun
3              soft_sf        0=No or 1=Yes, the spectrum is SF soft
4              suspected_pe   0=No or 1=Yes, it is a suspected Particle evt
5              definite_pe    0=No or 1=Yes, it is definitely a Particle evt
6              x_x_x_x_xxx    spare
7              definite_unk   0=No or 1=Yes, it is definitely an Unknown
8              earthward      0=No or 1=Yes, the direction is in the Earth cone
9              soft_flag      0=No or 1=Yes, the spectrum is a little soft
10             near_saa       0=No or 1=Yes, the spacecraft location is in SAA
11             definite_saa   0=No or 1=Yes, trigger is definitely SAA related
12             suspected_sf   0=No or 1=Yes, it is a suspected Solar Flare
13             definite_sf    0=No or 1=Yes, it is definitely a Solar Flare
14             op_flag        0=No or 1=Yes, the 2 brightest dets are opposites
15             def_not_grb    0=No or 1=Yes, it is definitely not a GRB
16             iso_pe         0=No or 1=Yes, Datlowe isotropic param is low
17             iso_grb        0=No or 1=Yes, Datlowe isotropic param is high
18             neg_h_ratio    0=No or 1=Yes, the hardness ratio is negative
19             neg_iso_bc     0=No or 1=Yes, Datlowe isotropic param is negative
20             not_soft       0=No or 1=Yes, the hardness ratio is not soft
21             hi_iso_ratio   0=No or 1=Yes, Datlowe iso C3/C2 ratio is high
22             low_inten      0=No or 1=Yes, intensity too small to be a GRB
23-30          spare          spare
31             bow            0=No or 1=Yes, it is a Burst OverWrite

These define's are in C-language programs for masking to get the various items
out of "trigger_id":
#define SUSP_GRB            0x00000001  // Suspected GRB
#define DEF_GRB             0x00000002  // Definitely a GRB
#define NEAR_SUN            0x00000004  // Coords are near the Sun (<15deg)
#define SOFT_SF             0x00000008  // Spectrum is soft, h_ratio > 2.0
#define SUSP_PE             0x00000010  // Suspected Particle Event
#define DEF_PE              0x00000020  // Definitely a Particle Event
#define X_X_X_X_XXX         0x00000040  // spare
#define DEF_UNK             0x00000080  // Definitely an Unknown
#define EARTHWARD           0x00000100  // Location towards Earth center
#define SOFT_FLAG           0x00000200  // Small hardness ratio (>1.5)
#define NEAR_SAA            0x00000400  // It is near/in the SAA region
#define DEF_SAA             0x00000800  // Definitely an SAA region
#define SUSP_SF             0x00001000  // Suspected Solar Flare
#define DEF_SF              0x00002000  // Definitely a Solar Flare
#define OP_FLAG             0x00004000  // Op-dets flag set
#define DEF_NOT_GRB         0x00008000  // Definitely not a GRB event
#define ISO_PE              0x00010000  // Datelowe Iso param is small (PE)
#define ISO_GRB             0x00020000  // D-Iso param is large (GRB/SF)
#define NEG_H_RATIO         0x00040000  // Negative h_ratio
#define NEG_ISO_BC          0x00080000  // Negative iso_part[1] or iso_part[2]
#define NOT_SOFT            0x00100000  // Not soft flag, GRB or PE
#define HI_ISO_RATIO        0x00200000  // Hi C3/C2 D-Iso ratio
#define LOW_INTEN           0x00400000  // Inten too small to be a real GRB
#define X_X_X_X_XX2         0x7F800000  // spare
#define BOW_FLAG            0x80000000  // Burst OverWrite trigger



The "misc" field contains various bit_flags, the algorithm level, the
notice type, and the program source-code version numbers.
The bit packing of the "misc" entry in the packet is:
Bit_Number     Item_name      Description
----------     ---------      -----------
0              tm_indicator   0=No_TM, 1=TM_present
1              bad_calc       Singular matrix (bad calculation)
2              comptel_fov    Position is in COMPTEL FOV
3              egret_fov      Position is in EGRET FOV
4              osse_fov       Position is in OSSE FOV
5-7            level          Program coords calc level [1-4]
8-9            notice_type    0=Original coords notice,
                              1=Updates coords notice,
                              2=Delay coords notice,
                              3=Final coords notice.
10             alex_anti_sun  Position is in the Anti-Solar hemisphere (ALEXIS)
11             ipn_peak       Passed the IPN Peak Inten Threshold criteria
12             ipn_flue       Passed the IPN Fluence Threshold criteria
13             xte_crit       Passed the RXTE follow-up criteria
14             xte_type       RXTE-PCA_ALERT and RXTE-PCA packet sub-types:
                              0=RXTE_ALERT_WILL_NOT_OBSERVE
                              1=RXTE_ALERT_WILL_OBSERVE
                              0=RXTE_OBS_SAW_NOTHING
                              1=RXTE_OBS_SAW_SOMETHING
15             no_trig_ind    BATSE is not in a triggerable mode
16-27          ver_minor      Program version, Minor number [0-2047]
28-31          ver_major      Program version, Major number [1-15]

The define's for masking to get the various items out of "misc":
#define TM_IND_MASK         0x00000001  // TM Indicator mask
#define BAD_CALC_MASK       0x00000002  // Singular matrix (bad calculation)
#define C_FOV_MASK          0x00000004  // COMPTEL FOV indicator mask
#define E_FOV_MASK          0x00000008  // EGRET FOV indicator mask
#define O_FOV_MASK          0x00000010  // OSSE FOV indicator mask
#define PROG_LEV_MASK       0x000000E0  // Program Algorithm Level mask
#define NOTICE_TYPE_MASK    0x00000300  // Notification Type mask
#define A_ASD_MASK          0x00000400  // ALEXIS Anti-Solar Dir ind mask
#define IPN_PEAK_MASK       0x00000800  // Passed the IPN Peak Inten thresh
#define IPN_FLUE_MASK       0x00001000  // Passed the IPN Fluence thresh
#define XTE_CRITERIA        0x00002000  // Meets RXTE follow-up criteria
#define XTE_STATUS          0x00004000  // Won't/will obs and Didn't/did see
#define NO_TRIG_MASK        0x00008000  // BATSE is not in a triggerable mode
#define VER_MINOR_MASK      0x0FFF0000  // Program Minor Version Number mask
#define VER_MAJOR_MASK      0xF0000000  // Program Major Version Number mask



There are 16 "spare" items in the packet for future growth.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.




TYPE=3 PACKET CONTENTS:

The IMALIVE packet consists of 40 four-byte quantities.  The order and contents
are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=3)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       spare        integer         4 bytes for the future
long         5       burst_tjd    [days]          Truncated Julian Day
long         6       burst_sod    [centi-sec]     (int)(sssss.sss *100)
long         7-38    spare[32]    integer         128 bytes for the future
long         39      pkt_term     integer         Pkt Termination (always = \n)


 
PACKET ITEM DESCRIPTIONS for type=3:

The "pkt_type" is an integer with a value of 1, 2 or 3 that tells
which type of packet this is and therefore what the contents are (they change
between the various types).  The types are:  1) Real BATSE trigger
initiated notices (GRBs, SFs, PEs, etc),  2) Test trigger notices,
3) Imalive packets.  The other types (4, 11, 22, and 24-34) are discussed
in other sections below.  And packet types (4-10, 12-21 and 23) never make it
to the outside world -- they are for internal GCN-use only.

The "pkt_sernum" is the serial number for that packet type.  Each packet
type has its own sequential serial numbering.  They all start from 1 when
the GCN programs start up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized).

The "burst_tjd" is the Truncated Julian Day when this packet was formed.
eg. TJD=10281 is 17 Jul 96.  Since this is an "imalive" packet, the "burst"
aspect of this is meaningless.  It is filled because the filtering code
needs to have this field to work for all packet types.

The "burst_sod" is the UT seconds-of-day when this packet was formed.
It is identical in value to the "pkt_sod" field.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized).  Since this is an "imalive" packet, the "burst" aspect
of this is meaningless.  It is filled because the filtering code needs to have
this field to work for all packet types.

There are 32 "spare" items in the packet for future growth.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.




TYPE=4 PACKET CONTENTS:

The KILL packet consists of 40 four-byte quantities.  It is used to send a
signal the site-end of the socket connection that GCN is going to close
the connection immediately (within milliseconds).  This is the only socket
packet that is never echo-ed back to the GCN system.  The order and contents
are listed in the table below.  They have the same function, meaning, and
content as the items of the same name in the previous packet types (1-3).

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=4)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       spare        integer         4 bytes for the future
long         5       burst_tjd    [days]          Truncated Julian Day
long         6       burst_sod    [centi-sec]     (int)(sssss.sss *100)
long         7-38    spare[32]    integer         128 bytes for the future
long         39      pkt_term     integer         Pkt Termination (always = \n)

The "burst_tjd" and "burst"sod" are in this "KILL" packet
because there are one or two places within the GCN codes
that like to have TJD, so this was a convenient place to put it.
They are not for a "burst" but rather are for the "pkt" formation date/time.




TYPE=11 PACKET CONTENTS:

NOTE: This Type=11 (BATSE_MAXBC) is NO LONGER AVAILABLE (June 2000).

The MAXBC packet consists of 40 four-byte quantities.  The order and contents
are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=11)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       burst_trig   integer         BATSE trigger number
long         5       burst_tjd    [days]          Truncated Julian Day
long         6       burst_sod    [centi-sec]     (int)(sssss.sss *100)
long         7       burst_ra     [centi-deg]     (int)(0.0 to 359.9999 *100)
long         8       burst_dec    [centi-deg]     (int)(-90.0 to +90.0 *100)
long         9       burst_inten  [cnts]          Linear scale, 0 to infinity
long         10      burst_peak   [cnts/1.024s]   Linear scale, 0 to infinity
long         11      burst_error  [centi-deg]     (int)(0.0 to 180.0 *100)
long         12      sc_az        [centi-deg]     (int)(0.0 to 359.9999 *100)
long         13      sc_el        [centi-deg]     (int)(-90.0 to +90.0 *100)
long         14      sc_x_ra      [centi-deg]     (int)(0.0 to 359.9999 *100)
long         15      sc_x_dec     [centi-deg]     (int)(-90.0 to +90.0 *100)
long         16      sc_z_ra      [centi-deg]     (int)(0.0 to 359.9999 *100)
long         17      sc_z_dec     [centi-deg]     (int)(-90.0 to +90.0 *100)
long         18      trigger_id   integer         Trigger identification flags
long         19      misc         integer         Misc stuff packed in here
long         20      e_sc_az      [centi-deg]     (int)(0.0 to 359.999 *100)
long         21      e_sc_el      [centi-deg]     (int)(-90.0 to +90.0 *100)
long         22      sc_radius    [km]            SC orbital radius
long         23      maxc1[0]     [cnts/64mSec]   Linear scale, 0 to infinity
long         24      maxc1[1]     [cnts/64mSec]   Linear scale, 0 to infinity
long         25      maxc1[2]     [cnts/64mSec]   Linear scale, 0 to infinity
long         26      maxc1[3]     [cnts/64mSec]   Linear scale, 0 to infinity
long         27      maxc1[4]     [cnts/64mSec]   Linear scale, 0 to infinity
long         28      maxc1[5]     [cnts/64mSec]   Linear scale, 0 to infinity
long         29      maxc1[6]     [cnts/64mSec]   Linear scale, 0 to infinity
long         30      maxc1[7]     [cnts/64mSec]   Linear scale, 0 to infinity
long         31      maxbc[0]     [cnts/64mSec]   Linear scale, 0 to infinity
long         32      maxbc[1]     [cnts/64mSec]   Linear scale, 0 to infinity
long         33      maxbc[2]     [cnts/64mSec]   Linear scale, 0 to infinity
long         34      maxbc[3]     [cnts/64mSec]   Linear scale, 0 to infinity
long         35      maxbc[4]     [cnts/64mSec]   Linear scale, 0 to infinity
long         36      maxbc[5]     [cnts/64mSec]   Linear scale, 0 to infinity
long         37      maxbc[6]     [cnts/64mSec]   Linear scale, 0 to infinity
long         38      maxbc[7]     [cnts/64mSec]   Linear scale, 0 to infinity
long         39      pkt_term     integer         Pkt Termination (always = \n)

 
PACKET ITEM DESCRIPTIONS (only those unique to type=11):

The 8 "maxc1[i]" and the 8 "maxbc[i]" are the 16 maximum (bkg subtracted) rates
[cnts/64mSec] as determined by the BATSE flight software during the 0-10
minutes after the initial trigger.  The index "i" runs from 0 to 7 for the 8
LADs.  The "c1" refers to the C1 DISCLA channel (~20-50 keV).  The "bc" refers
to the "burst channels" which is most of the time set to the two middle
DISCLA channels (50-100-300 keV).



TYPE=21 PACKET CONTENTS/DESCRIPTION:

NOTE: This Type=11 (Bradford) is NO LONGER AVAILABLE (June 2000)

The format and content of this BRAD_COORDS packet is identical to the
Test Coords packet (type=2).  The only difference is that the values
for the test RA,Dec locations are different than the set used for type=2.
The Bradford telescope system needed a different set of locations than
the generic set, so a separate packet type was defined and this different type
allowed the necessary filtering differences between the Bradford site
and the rest of the GCN sites.



TYPE=22 PACKET CONTENTS/DESCRIPTION:

NOTE: This Type=22 (BATSE_Final) is NO LONGER AVAILABLE (June 2000)

The format and content of this FINAL packet is identical to the ORIGINAL
(type=1) with the one exception:  that the "inten" is now a fluence value
for the number of "sam_used" values that went into the integration
of the burst light curve.



TYPE=24 PACKET CONTENTS/DESCRIPTION:

NOTE: This Type=24 (BATSE_LOCBURST) is NO LONGER AVAILABLE (June 2000).

The format and content of this LOCBURST packet is identical to the 
GRB Coords packet (type=1).  The major difference is that the values
of the RA,Dec location are calculated using the LOCBURST algorithm
instead of the ideal-physics and are, therefore, more accurate.
The "t_peak" value at location index=23 is undefined for this packet type.



TYPE=25 PACKET CONTENTS:

NOTE: This Type=25 (ALEXIS) is NO LONGER AVAILABLE (December 1999).

The ALEXIS packet consists of 40 four-byte quantities.  The order and contents
are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=25)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       src_num      integer         ALEXIS source number
long         5       trans_tjd    [days]          Truncated Julian Day
long         6       trans_sod    [centi-sec]     (int)(sssss.sss *100)
long         7       trans_ra     [centi-deg]     (int)(0.0 to 359.9999 *100)
long         8       trans_dec    [centi-deg]     (int)(-90.0 to +90.0 *100)
long         9-10    spare[2]     integer         Unused at this point
long         11      trans_error  [centi-deg]     (int)(0.0 to 180.0 *100)
long         12-22   spare[11]    integer         Unused at this point
long         22      trans_alpha  integer         (int)(alpha *100)
long         23      map_dur      integer         Always 12, 24, or 48
long         24      tele_id      integer         Always 1, 2, 3, 4, 5, or 6
long         25-38   spare[14]    integer         56 bytes for the future
long         39      pkt_term     integer         Pkt Termination (always = \n)


 
PACKET ITEM DESCRIPTIONS (only those unique to type=25):

Only the packet items that are different than the type=1,2,3 packet contents
are listed below.  The standard packet items are described and defined
in the above sections.

The "src_num" is the uniquely identifying number assigned by the ground
ALEXIS data processing software to each identified source.  It starts with 1
for each processing run and progresses upwards (typically 0-20 sources
are identified each 12-hr run).

The "trans_tjd" is the Truncated Julian Day of the ALEXIS source,
eg. TJD=10281 is 17 Jul 96.  It is the TJD of the ending date/time stamp
of the map accumulation in which the transient was found.

The "trans_sod" is the UT seconds-of-day of the ALEXIS source.  It is the
SOD of the ending date/time stamp of the map accumulation in which the
transient was found.  While the units are centi-seconds, the value is always
a whole second.  The use of centi-seconds maintains the format/precission
that is used throughout all the packets.

The "trans_ra" and "trans_dec" are the RA,Dec coordinates of the transient
as determined by the ALEXIS batch processing analysis program (current epoch).
The units are centi-degrees.

The "trans_error" is the 3-sigma radius of the error circle.  It contains
the sum of the statistical and systematic errors.  This number is derived
from a comparison of the known location of the HZ 43 object and the 
instrument's determination.

The "trans_alpha" is the -log10(probability_of_false_detection).  The alpha
parameter is only the statistical significance of the detection, or put another
way, -log10 of the probability that the source could be a natural fluctuation
above a Poisson background.

The "map_duration" is the amount of data (the last 12, 24, or 48 hours)
that went into the accumulation that the transient was found.  The ALEXIS
processing uses three different time scales as a tradeoff between transient
duration and sensitivity.  This integer has only the 12, 24, or 48 numeric
values -- all other values are invalid.

The "tele_id" is the telescope that the transient was detected in.  This is
an integer with only 1 through 6 as the valid values for the six detectors.
The telescope names and energy bandpasses are: 1="1A 93eV", 2="1B 70eV",
3="2A 93eV", 4="2B 66eV", 5="3A 70eV", 6="3B 66eV".  (Telescopes 1 and 2
are co-aligned on ALEXIS, as are telescopes 3 and 4, and telescopes 5 and 6.)



TYPE=26 PACKET CONTENTS:

NOTE: Type=26 packets (RXTE_PCA_ALERT) are NO LONGER AVAILABLE.

The RXTE-PCA_ALERT packet consist of 40 four-byte quantities.  The order and
contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=26)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       burst_trig   integer         BATSE trigger number
long         5       burst_tjd    [days]          Truncated Julian Day
long         6       burst_sod    [centi-sec]     (int)(sssss.sss *100)
long         7       locburst_ra  [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long         8       locburst_dec [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long         9-18    spare[10]    integer         40 bytes for the future
long         19      misc         integer         Misc stuff packed in here
long         20-23   spare[4]     integer         16 bytes for the future
long         24      obs_tjd      integer         Observation start TJD
long         25      obs_sod      integer         (int)(sssss.ss *100)
long         26-38   spare[19]    integer         76 bytes for the future
long         39      pkt_term     integer         Pkt Termination (always = \n)

 
PACKET ITEM DESCRIPTIONS (only those unique to type=26):

For the type=26 (PCA ALERT) packets, the normal burst_ra/_dec locations
contain the LOCBURST RA/Dec location.  This allows the potential follow-up
observer to make tentative observing plans based approximate LOCBURST location
on the sky.

The "misc" field contains various bit_flags (see the section for types=1,2,3
for the definition of all the normal bits), and for the specialty bits
for type=26 (see the partial table below):

Bit_Number     Item_name      Description
----------     ---------      -----------
13             xte_crit       Passed the RXTE-PCA follow-up criteria
14             xte_type       RXTE-PCA_ALERT packet sub-types:
                              0=RXTE_ALERT_WILL_NOT_OBSERVE
                              1=RXTE_ALERT_WILL_OBSERVE

The define's for masking to get the various items out of "misc":
#define XTE_CRITERIA        0x00002000  // Meets RXTE-PCA follow-up criteria
#define XTE_STATUS          0x00004000  // Won't/will observe

The "obs_tjd" and "obs_sod" fields are the Truncated Julian Day and
Seconds-of_day for the start of the RXTE-PCA observation.  Putting this
timestamp in the ALERT notice allows the follow-up observer to know
when the RXTE-PCA observation is planned and about when the RXTE-PCA coordinates
will become available.



TYPE=27 PACKET CONTENTS:

The RXTE-PCA GRB coordinates packet consist of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=27)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       burst_trig   integer         BATSE trigger number
long         5       burst_tjd    [days]          Truncated Julian Day
long         6       burst_sod    [centi-sec]     (int)(sssss.sss *100)
long         7       burst_ra     [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long         8       burst_dec    [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long         9       burst_inten  [mCrab]         Linear scale, 0 to infinity
long         10      spare        integer         spare
long         11      burst_error  [10^-4-deg]     (int)(0.0 to 180.0 *10000)
long         12-18   spare[7]     integer         28 bytes for the future
long         19      misc         integer         Misc stuff packed in here
long         20-23   spare[4]     integer         16 bytes for the future
long         24      obs_tjd      integer         Observation start TJD
long         25      obs_sod      integer         (int)(sssss.ss *100)
long         26-28   spare[13]    integer         42 bytes for the future
long         39      pkt_term     integer         Pkt Termination (always = \n)


 
PACKET ITEM DESCRIPTIONS (only those unique to type=27):

The "burst_ra" and "burst_dec" are the RA,Dec coordinates of the burst
(current epoch).  These intrinsically floating point quantities have been
multiplied by 10000 and then integerized to make an integer quantity
with units of 0.0001-degrees.

The "burst_inten" is slightly different than the burst_inten used in type=1.
It is units of mCrab (instead of the counts/sec).

The "misc" field contains various bit_flags (see the section for types=1,2,3
for the definition of all the normal bits), and for the specialty bits
for type=27 (see the partial table below):

Bit_Number     Item_name      Description
----------     ---------      -----------
13             xte_crit       Passed the RXTE-PCA follow-up criteria
14             xte_type       RXTE-PCA packet sub-types:
                              0=RXTE_OBS_SAW_NOTHING
                              1=RXTE_OBS_SAW_SOMETHING

The define's for masking to get the various items out of "misc":
#define XTE_CRITERIA        0x00002000  // Meets RXTE-PCA follow-up criteria
#define XTE_STATUS          0x00004000  // Didn't/did see a source

The "obs_tjd" and "obs_sod" fields are the Truncated Julian Day and
Seconds-of_day when the observation started.




TYPE=28 PACKET CONTENTS:

The packet for the RXTE-ASM ALERT notice has not been finalized yet.




TYPE=29 PACKET CONTENTS:

The RXTE-ASM GRB coordinates packet consist of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=29)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       burst_trig   integer         BATSE trigger number
long         5       burst_tjd    [days]          Truncated Julian Day
long         6       burst_sod    [centi-sec]     (int)(sssss.sss *100)
long         7       burst_ra     [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long         8       burst_dec    [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long         9       burst_inten  [mCrab]         Linear scale, 0 to infinity
long         10      spare        integer         spare
long         11      burst_error  [10^-4-deg]     (int)(0.0 to 180.0 *10000)
long         12      burst_cont   [10^-2]         (int)(0.0 to 100.0 *100)
long         13-18   spare[6]     integer         24 bytes for the future
long         19      misc         integer         Misc stuff packed in here
long         20-23   spare[4]     integer         16 bytes for the future
long         24      err_ra1      [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long         25      err_dec1     [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long         26      err_ra2      [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long         27      err_dec2     [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long         28      err_ra3      [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long         29      err_dec3     [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long         30      err_ra4      [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long         31      err_dec4     [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long         32      length       [10^-4-deg]     (int)(0.0 to 180.0 *10000)
long         33      width        [10^-4-deg]     (int)(0.0 to 180.0 *10000)
long         34      pos_angle    [10^-4-deg]     (int)(0.0 to 360.0 *10000)
long         35-38   spare[4]     integer         16 bytes for the future
long         39      pkt_term     integer         Pkt Termination (always = \n)

 
PACKET ITEM DESCRIPTIONS (only those unique to type=29):

The "burst_ra" and "burst_dec" are the RA,Dec coordinates of the burst
(current epoch).  These intrinsically floating point quantities have been
multiplied by 10000 and then integerized to make an integer quantity
with units of 0.0001-degrees.

The "burst_inten" is slightly different than the burst_inten used in type=1.
It is units of mCrab (instead of the counts/sec).

The "misc" field contains various bit_flags (see the section for type=1,2,3
for the definition of all the normal bits), and for the specialty bits
for type=32 (see the partial table below):

Bit_Number     Item_name      Description
----------     ---------      -----------
8-9            notice_type    0=Initial coords notice,
                              1=Updates coords notice,
                              3=Final coords notice.
14             box_v_line     RXTE-ASM error box sub-types:
                              0=line
                              1=box

The define's for masking to get the various items out of "misc":
#define NOTICE_TYPE_MASK    0x00000300  // Notification Type mask

The "burst_error" is the radius of a circle which circumscribes the error
box (diamond-shaped).  It is decimal degrees and is for the "burst_cont"
containment (statistical + systematic).

The "burst_cont" is the confidence containment percentage for the quoted
burst_error value.  It is stored in the packet in centi-percentage
(e.g. 98.5% would be an integer value of 9850).

The "err_ra1" and "err_dec1" are the RA,Dec coordinates for the 1st corner
of the error box.  Ditto for "2", "3", and "4" for the second, third, and
fourth corners, respectively.  (These fields are valid when the error box
is type "box".)  Note that these RA,Dec values are J2000 epoch, whereas the
RA,Dec of the center of the box is in Current epoch.  The use of different
epochs is, admittedly, ugly, but is due to historical/evolutionary aspects
that are way too lengthy and of little practical importance to explain here.

The "length" is the length of the error box when the error box is type "line".
The "width" is the width of the error box when the error box is type "line".
The "pos_angle" is the position angle of the error box when the error box
is type "line".  It is measured eastward from north.





================================================================================

TYPE=30 PACKET CONTENTS:

NOTE: Type=30 (COMPTEL) is NO LONGER AVAILABLE (June 2000). (since CGRO was de-orbitted).

{Note that the RA,Dec of the center are in J2000 epoch, and not
in the rest-of-the-program standard Current epoch.  The corners
of the error box are in the J2000 epoch.}




================================================================================

TYPE=31 PACKET CONTENTS:

The IPN_RAW packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=31)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       trig_num     integer         Mission-specific ID number
long         5       burst_tjd    [days]          Truncated Julian Day
long         6       burst_sod    [centi-sec]     (int)(sssss.sss *100)
long         7       ipn_cntr_ra  [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         8       ipn_cntr_dec [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         9       ipn_radius   [0.0001-deg]    Radius of the annulus
long         10      ipn_t_win    [centi-sec]     Earth-center time range (*100)
long         11      ipn_width    [0.0001-deg]    (int)(0.0 to 180.0 *10000)
long         12-13   spare[2]     integers        8 bytes for the future
long         14      event_dur    [centi-sec]     Duration of the event (*100)
long         15-17   spare[3]     integers        12 bytes for the future
long         18      trig_id      bits            Type of source/trigger found
long         19      misc         bits            Misc stuff packed in here
long         20      scratch      integer         Scratch/internal use
long         21      scratch      integer         Scratch/internal use
quad_char    22-38   url[17]      chars           The URL for the lightcurve
long         39      pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 31
that tells which type of packet this is and therefore what the contents are
(the contents change between the various types).

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start from 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the burst_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "trig_num is some mission-specific identification number.  It is optional;
not all misions-instruments have an identification number.

The "burst_tjd" is the Truncated Julian Day of the burst trigger,
eg. TJD=13370 is 01 Jan 2005.

The "burst_sod" is the UT seconds-of-day (SOD) of the burst trigger.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized).

The "ipn_center_ra" and "ipn_center_dec" are the RA,Dec coordinates of the center
of the IPN annulus (J2000).  These intrinsically floating point quantities
have been multiplied by 10,000 and then integerized to make an integer quantity
with units of 0.0001-degrees.  This field is valid only if the "trigger_id"
2^0 bit is a 0 (ie the 'annulus' form of this notice type).

The "ipn_radius" is the radius to the center of the IPN annulus.
The units are 0.0001-degrees (ie the fl.pt. error was multiplied
by 10000 and then integerized).  This field is valid only if the "trigger_id"
2^0 bit is a 0 (ie the 'annulus' form of this notice type).

The "ipn_t_window" is the amount of time slop that the Timestamp could vary
depending on the direction of the burst wavefront (eg for a LEO s/c, this 
would be +-0.022 sec, and for Mars Observer, it would range from +/-250 sec
to +/-1250 sec depending on the relative positions of Mars and Earth).
The units are 0.01-sec (ie the fl.pt. time duration was multiplied
by 100 and then integerized).

The "ipn_width" is the 3-sigma width of the IPN annulus.
The units are 0.0001-degrees (ie the fl.pt. error was multiplied
by 10000 and then integerized).  This field is valid only if the "trigger_id"
2^0 bit is a 0 (ie the 'annulus' form of this notice type).

The "event_duration" is the duration of the event in the specified instrument
(see Misc flag bits).  (Note that the different mission-instruments have
different sensitivities and energy windows, so for a given GRB, there will be
variations in this value from one instrument to the next.)
The units are 0.01-sec (ie the fl.pt. time duration was multiplied
by 100 and then integerized).

The "trig_id" field contains flag bits on the content and expectation 
of this event being a burst or not.
The bit packing of the "trig_id" entry in the packet is:
BitNum  Item_name       Description
------  ---------       -----------
0       times_only      0=No or 1=Yes, this packet contains only the times (trig,window,duration; no ra,dec,err)
//1       not_operational Instrument not able to take data at that time (PwrOff, Cal_mode, SAA)
//2       grb             0=No or 1=Yes, it is definitely a GRB
//3       prob_grb        0=No or 1=Yes, it is probably a GRB
//4       poss_grb        0=No or 1=Yes, it is possibly a GRB
//5       def_not_grb     0=No or 1=Yes, it is definitely not a GRB (bad label; its really a Retraction)
//6       nothing_found   if 1, it indicates no "blip" was found in the corresponding time_window
//7-8     ecliptic        0=Unknown, 1=North hemisphere, 2=South hemisphere, 3=Middle_band (less than X deg)
9       trig_time_only  0=No or 1=Yes, this packet contains only the trigger_time (no t_window, t_duration))
10-28   spare           spare for future use
29      temporal_coinc  0=No or 1=Yes, there was a temporal coincidence with another event
30      test_submit     0=No or 1=Yes, this is a test submission (internal use only)
31      spare           spare for future use
//Bits with a "//" prefix are no longer available (post Sep 2009).

The "misc" field contains miscellaneous flag bits.
The defined bits are:
//2^0   When this bit is a 1, it indicates that no further information will be available on this event.
//2^1   When this bit is a 1, it indicates that further information will probably be available on this event.
//2^2   When this bit is a 1, it indicates that further information will definitely be available on this event.
//2^3   This is an updated notice.  This is a notice that changes/improves
//      the information specified in the original Notice.
2^4-9 Not assigned.
2^10  The Ulysses mission contributed to this Notice.
2^11  The Mars Odyssey-HEND instrument contributed to this Notice.
2^12  The Mars Odyssey-GRS instrument contributed to this Notice.
2^13  The RHESSI mission contributed to this Notice.
2^14  The KONUS-Wind instrument contributed to this Notice.
2^15  The Swift-BAT instrument contributed to this Notice.
2^16  The Messenger-GRNS instrument contributed to this Notice.
2^17  The Suzaku-WAM instrument contributed to this Notice.
2^18  The INTEGRAL-SPIACS instrument contributed to this Notice.
2^19  The SuperAGILE instrument contributed to this Notice.
2^20  The AGILE-GRID instrument contributed to this Notice.
2^21  The AGILE-MCAL instrument contributed to this Notice.
2^22  The FERMI-GBM instrument contributed to this Notice.
2^23  The FERMI-LAT instrument contributed to this Notice.
2^24-31  Not assigned.

The "scratch"  fields are for use internal to GCN (to communicate
the name of the temporary comments files during importation)
and the mission/instrument ID.
It has no value to the outside world and should be ignored.

The "url" is just the filename portion of the URL -- the leading path
for the full URL is "http://gcn.gsfc.nasa.gov/gcn/XXXXXXX".     //DECIDE ON THIS !!!!
This item in the packet spans 17 longwords (ie 68 bytes) and contains 
the ASCII string of the filename portion of the URL.  This string
can be up to 67 characters long, and will always be null-terminated.

There are a total of 5 "spare" items in the packet (usually filled
with zeros).  They are located at items 12-13 and 15-17.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.






TYPE=32 PACKET CONTENTS:

NOTE: Type=32 (IPN_SEGMENT) is no longer available (since it part of the
input to this type was the BATSE position).

NOTE: It might become available again now that FERMI-GBM error circles are available.

The IPN_SEGMENT packet consists of 40 four-byte quantities.  The order and
contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=32)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       burst_trig   integer         BATSE trigger number
long         5       burst_tjd    [days]          Truncated Julian Day
long         6       burst_sod    [centi-sec]     (int)(sssss.sss *100)
long         7       ipn_c_ra     [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long         8       ipn_c_dec    [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long         9       ipn_radius   [10^-4-deg]     (int)(0.0 to +90.0 *10000)
long         10      spare        integer         4 bytes for the future
long         11      ipn_width    [10^-4-deg]     (int)(0.0 to +90.0 *10000)
long         12      3sig_ra      [10^-4-deg]     (int)(0.0 to 360.0 *10000)
long         13      3sig_dec     [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long         14      2sig_ra      [10^-4-deg]     (int)(0.0 to 360.0 *10000)
long         15      2sig_dec     [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long         16      1sig_ra      [10^-4-deg]     (int)(0.0 to 360.0 *10000)
long         17      1sig_dec     [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long         18      max_prob_ra  [10^-4-deg]     (int)(0.0 to 360.0 *10000)
long         19      max_prob_dec [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long         10      1sig_ra      [10^-4-deg]     (int)(0.0 to 360.0 *10000)
long         21      1sig_dec     [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long         22      2sig_ra      [10^-4-deg]     (int)(0.0 to 360.0 *10000)
long         23      2sig_dec     [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long         24      3sig_ra      [10^-4-deg]     (int)(0.0 to 360.0 *10000)
long         25      3sig_dec     [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long         26-38   spare[13]    integer         52 bytes for the future
long         39      pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS (only those unique to type=32):

The "ipn_c_ra" and "ipn_c_dec" are the RA,Dec coordinates of the center of
the IPN annulus.  They are in J2000 epoch.  The units are 0.0001-degrees.
It was necessary to use a scaling factor (1000x) bigger than the normal 100x
(ie centi-degrees) because the intrinsic accuracy and precision of the
IPN coordinates is much better.

The "ipn_radius" and "ipn_width" are the radius and the width of the IPN
annulus.  Full width which contains all the known statistical and
systematic errors (ie greater than 90% confidence).  The units are
0.0001-degrees.

The "3sig_ra" and "3sig__dec" are the RA,Dec coordinates of the point along
the annulus at which the 3-sigma probability is located.  There are
two such locations (one on each side of the maximum probability location.
They are in J2000 epoch.  The units are 0.0001-degrees.

Similarly for the "2sig_ra", "2sig_dec", "1sig_ra", and "1sig_dec".

The "max_prob_ra" and "max_prob_dec" are the RA,Dec coordinates of the
location on the IPN annulus which the has the maximum probability of being
the location of the GRB.  They are in J2000 epoch.  The units are
0.0001-degrees.



================================================================================

TYPE=33 PACKET CONTENTS:

The packet for the SAX-WFC ALERT notice has not been finalized yet.
[03feb01: It probably never will be finalized.  There seems to be
no interest from the community to have/use this Notice type, and
there seems to be no interest amongst the SAX team to generate it.]




TYPE=34 PACKET CONTENTS:

NOTE: Type=34 (SAX_WFC) is no longer available (since BeppoSAX mission was terminated).

The SAX-WFC GRB Position packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=34)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       spare        integer         4 bytes for the future
long         5       burst_tjd    [days]          Truncated Julian Day
long         6       burst_sod    [centi-sec]     (int)(sssss.sss *100)
long         7       burst_ra     [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long         8       burst_dec    [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long         9       burst_inten  [mCrab]         Linear scale, 0 to infinity
long         10      spare        integer         4 bytes for the future
long         11      burst_error  [10^-4-deg]     (int)(0.0 to 180.0 *10000)
long         12      burst_cont   [centi-%]       (int)(0.0 to 100.0 *100)
long         13-17   spare[5]     integer         20 bytes for the future
long         18      trigger_id   integer         Trigger identification flags
long         19      misc         integer         Misc stuff packed in here
long         20-38   spare[19]    integer         56 bytes for the future
long         39      pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS (only those unique to type=34):

The "burst_ra" and "burst_dec" are the RA,Dec coordinates of the burst
(current epoch).  These intrinsically floating point quantities have been
multiplied by 10000 and then integerized to make an integer quantity
with units of 0.0001-degrees.

The "burst_inten" is slightly different than the burst_inten used in type=1.
It is units of mCrab (instead of the counts/sec).

The "burst_cont" is the confidence containment percentage for the quoted
burst_error value.  It is stored in the packet in centi-percentage
(e.g. 98.5% would be an integer value of 9850).




TYPE=35 PACKET CONTENTS:

The packet for the SAX-NFI ALERT notice has not been finalized yet.
[03feb01: It probably never will be finalized.  There seems to be
no interest from the community to have/use this Notice type, and
there seems to be no interest amongst the SAX team to generate it.]





TYPE=36 PACKET CONTENTS:

NOTE: Type=36 (SAX_NFI) is no longer available (since BeppoSAX mission was terminated).

The SAX-NFI GRB Position packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=36)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       spare        integer         4 bytes for the future
long         5       burst_tjd    [days]          Truncated Julian Day
long         6       burst_sod    [centi-sec]     (int)(sssss.sss *100)
long         7       burst_ra     [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long         8       burst_dec    [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long         9       burst_inten  [mCrab]         Linear scale, 0 to infinity
long         10      spare        integer         4 bytes for the future
long         11      burst_error  [10^-4-deg]     (int)(0.0 to 180.0 *10000)
long         12      burst_cont   [centi-%]       (int)(0.0 to 100.0 *100)
long         13-17   spare[5]     integer         20 bytes for the future
long         18      trigger_id   integer         Trigger identification flags
long         19      misc         integer         Misc stuff packed in here
long         20-38   spare[19]    integer         56 bytes for the future
long         39      pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS (only those unique to type=36):

The "burst_ra" and "burst_dec" are the RA,Dec coordinates of the burst
(current epoch).  These intrinsically floating point quantities have been
multiplied by 10000 and then integerized to make an integer quantity
with units of 0.0001-degrees.

The "burst_inten" is slightly different than the burst_inten used in type=1.
It is units of mCrab (instead of the counts/sec).

The "burst_cont" is the confidence containment percentage for the quoted
burst_error value.  It is stored in the packet in centi-percentage
(e.g. 98.5% would be an integer value of 9850).


================================================================================

TYPE=37 PACKET CONTENTS:

NOTE: Type=37 packets (RXTE_ASM_XTRANS) are NO LONGER AVAILABLE.

The RXTE-ASM_XTRANS Position packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=37)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       ref_num      integer         The serial number of trans
long         5       trans_tjd    [days]          Truncated Julian Day
long         6       trans_sod    [centi-sec]     (int)(sssss.sss *100)
long         7       trans_ra     [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long         8       trans_dec    [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long         9       trans_inten  [mCrab]         Linear scale, 0 to infinity
long         10      spare        integer         4 bytes for the future
long         11      trans_error  [10^-4-deg]     (int)(0.0 to 180.0 *10000)
long         12      spare        integer         4 bytes for the future
long         13      time_since   [centi-sec]     (int)(sssss.sss *100)
long         14-17   spare[4]     integer         16 bytes for the future
long         18      trigger_id   integer         Trigger identification flags
long         19      misc         integer         Misc stuff packed in here
long         24      ra1          [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long         25      dec1         [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long         26      ra2          [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long         27      dec2         [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long         28      ra2          [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long         29      dec3         [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long         30      ra3          [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long         31      dec4         [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long         32      line_length  [10^-4-deg]     (int)(0.0 to 180.0 *10000)
long         33      line_width   [10^-4-deg]     (int)(0.0 to 180.0 *10000)
long         34      pos_angle    [10^-4-deg]     (int)(0.0 to 360.0 *10000)
long         35-38   spare[4]     integer         16 bytes for the future
long         39      pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS (only those unique to type=37):

The "trans_ra" and "trans_dec" are the RA,Dec coordinates of the transient
(current epoch).  These intrinsically floating point quantities have been
multiplied by 10000 and then integerized to make an integer quantity
with units of 0.0001-degrees.

The "trans_inten" is slightly different than the burst_inten used in type=1.
It is units of mCrab (instead of the counts/sec).

The "misc" field contains various bit_flags (see the section for type=1,2,3
for the definition of all the normal bits), and for the specialty bits
for type=37 (see the partial table below):

Bit_Number     Item_name      Description
----------     ---------      -----------
8-9            notice_type    0=Initial coords notice,
                              1=Updates coords notice,
                              3=Final coords notice.
14             box_v_line     RXTE-ASM error box sub-types:
                              0=line
                              1=box

The define's for masking to get the various items out of "misc":
#define NOTICE_TYPE_MASK    0x00000300  // Notification Type mask

The "trans_error" is the radius of a circle which circumscribes the error
box (diamond-shaped).

The "ra1" and "dec1" are the RA,Dec coordinates for the 1st corner of the
error box.  Ditto for "2", "3", and "4" for the second, third, and
fourth corners, respectively.  (These fields are valid when the error box
is type "box".)  Note that these RA,Dec values are J2000 epoch, whereas the
RA,Dec of the center of the box is in Current epoch.  The use of different
epochs is, admittedly, ugly, but is due to historical/evolutionary aspects
that are way too lengthy and of little practical importance to explain here.

The "line_length" is the length of the error box when the error box
is type "line" (this and the next two fields are ONLY assigned when
the type is "line").
The "line_width" is the width of the error box when the error box
is type "line".
The "pos_angle" is the position angle of the error box when the error box
is type "line".  It is measured eastward from north.




================================================================================

TYPE=38 PACKET CONTENTS:

This type=38 is a "spare" packet.  It is used for development testing
within the GCN as each new "source" packet is brought online.  It's contents
and format change with each new source.  (The epoch is Current; not J2000.)
This packet type should never be distributed to the GCN sites.




================================================================================

TYPE=39 PACKET CONTENTS:

The IPN_POSITION packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=39)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       ref_num      integer         The serial number of trans
long         5       burst_tjd    [days]          Truncated Julian Day
long         6       burst_sod    [centi-sec]     (int)(sssss.sss *100)
long         7       burst_ra     [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long         8       burst_dec    [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long         9       burst_flue   [10^-9erg/cm2]  Linear scale, 0 to infinity
long         10      burst_flux   [10^-9erg/cm2/s]Linear scale, 0 to infinity
long         11      burst_error  [10^-4-deg]     (int)(0.0 to 180.0 *10000)
long         12      samp_interval[10^-4-sec]     (int)(0.0 to infinity)
long         13-15   spare[3]     integer         16 bytes for the future
long         16      burst2_ra    [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long         17      burst2_dec   [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long         18      trig_id      bits            Trigger identification flags
long         19      misc         integer         Misc stuff packed in here
long         20      spare        integer         4 bytes for the future
long         21      box_area     [10^-4-sq.arcmin](int)(0.0 to X.X *10000)
long         22      duration     [centi-sec]     (int)(sssss.sss *100)
long         23      ra1          [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long         24      dec1         [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long         25      ra2          [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long         26      dec2         [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long         27      ra2          [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long         28      dec3         [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long         29      ra3          [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long         30      dec4         [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long         31      ra21         [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long         32      dec21        [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long         33      ra22         [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long         34      dec22        [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long         35      ra22         [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long         36      dec23        [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long         37      ra23         [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long         38      dec24        [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long         39      pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 39, which tells that this packet
is of type IPN_POSITION and therefore indicates the contents and format
for the other 39 fields.

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start from 1 when
the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 1.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the burst_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "refnum" is a serial number assigned by K.Hurley (UCBerkeley, IPN).
It is used for referencing/citation purposes.  It is sequential for the
Notices -- not the GRBs -- hence two Notices on a single GRB will have
different serial numbers (eg an initial notice and a refined-calculation notice
would have different refnum's).

The "burst_tjd" is the Truncated Julian Day of the burst,
eg. TJD=12275 is 01 Jan 2002.

The "burst_sod" is the time of earth-crossing for the burst
as determined by the Interplanetary Network of spacecraft.
The SOD is multiplied by 100, and converted to a 32-bit integer for the packet
(i.e the units are centi-sec; 10 msec),  eg. 4689568 = 46895.68 sec-of-day =
13:01:35.68 UT.

The "burst_ra" and "burst_dec" fields specify the RA,Dec location of the calculated
burst position (in Current epoch).  These intrinsically floating point
quantities have been multiplied by 10,000 and then integerized to make
an integer quantity with units of 10^-4-degrees.

The "burst_flue" is the burst fluence in units of 10^-9 erg/cm2.
This is not always known for the burst at the time of the issuance
of the Notice, in which case it will have a value of -1.

The "burst_flux" is the burst fluence in units of 10^-9 erg/cm2/sec.
This is not always known for the burst at the time of the issuance
of the Notice, in which case it will have a value of -1.

The "burst_error" is the radius of a circle which circumscribes
the error box (rectangular-or-diamond-shaped). The units are decimal degrees
times 10,000.0.  This radius represents the worst case error distance 
(ie the distance from the center to the farthest corner).

The "sampling_interval" is the interval of time during which the peak flux
was observed (ie the integration interval while scanning for the maximum
flux value).

The "burst2_ra" and "burst2_dec" values (Current epoch!) is the center of the 2nd IPN box
(if the Notice is a 2-box subtype).

The "trigger_id" field contains bit_flags which attempt to identify
the type of burst event.
The bit packing of the "trigger_id" entry in the packet is:
Bit_Number     Item_name      Description
----------     ---------      -----------
0              x_x_x_x        spare
1              definite_grb   0=No or 1=Yes, it is definitely a GRB
2-27           x_x_x_x        spare
28             spatial_coinc  0=No or 1=Yes, there was a spatial coincidence with another event
29             temporal_coinc 0=No or 1=Yes, there was a temporal coincidence with another event
30-31          x_x_x_x        spare

In the "misc" word, the 2^5-6 bits are used to indicate the Notice subtype:
a 00 is a 1-box subtype, and a 01 is a 2-box subtype.  For subtype=0 (1-box),
the various 2nd-box items in the packet format are filled with zeros (ie the
center and the 4 corners).

The "box_area" is the area of the error box in units of 0.0001 sq.arcminutes.
If the Notice is a 2-box notice, then the area of the 2nd box is always
the same as the first box.

The "duration" is the T90 duration of the burst in units of centi-seconds.
This is not always known for the burst at the time of the issuance
of the Notice, in which case it will have a value of -1.

The "ra1" and "dec1" are the RA,Dec coordinates (J2000!) for the 1st corner of the
error box.  Ditto for "2", "3", and "4" for the second, third, and
fourth corners, respectively.
The "ra21" and "dec21" are the RA,Dec coordinates (J2000!) for the 1st corner of the
second box (if the Notice is a 2-box subtype).  Ditto for the 2nd, 3rd, and 4th
corners.

There are a total of tbd "spare" items in the packet (usually filled
with zeros).  They are located at items 11-12, 17-18, and 20-38.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.



================================================================================

TYPE=40 PACKET CONTENTS:

NOTE: Type=40 packets (HETE_S/C_ALERT) are NO LONGER AVAILABLE.

The HETE_S/C_ALERT packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=40)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       trig_seq_num integers        Trigger num & mesg sernum
long         5       burst_tjd    [days]          Truncated Julian Day
long         6       burst_sod    [centi-sec]     (int)(sssss.sss *100)
long         7-8     spare[2]     integer         Mostly empty in Alert packets
long         9       trig_flags   bits            Trigger type ID flag bits
long        10       gamma_cnts   [cnts/sec]      GammaDet count rate
long        11       wxm_signoise [centi-sig/nois]WXM rate trigger signal/noise
long        12       sxc_cnts     [cnts/sec]      SXC count rate
long        13       gamma_time   [milli-sec]     GammaDet timescale
long        14       wxm_time     [milli-sec]     WXM det timescale
long        15       sc_point     integers        S/c RA,Dec pointing dir
long        16-38    spare[23]    integer         92 bytes not filled (zeros)
long        39       pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 40, 41, 42, 43, or 44
that tells which type of packet this is and therefore what the contents are
(they change slightly between the various types).  The types are:
40 = HETE_S/C_ALERT, 41 = HETE_S/C_UPDATE, 42 = HETE_S/C_LAST,
43 = HETE_GNDANA, and 44 = HETE_Test.

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start from 1 when
the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 1.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the burst_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "trig_seq_num" contains two separate items: a) the "trigger number"
and b) the "message sequence number".  The trigger number is assigned 
by the HETE flight software for each new triggering event.  The message
sequence number is a serial number across all 4 message subtypes (S/C_Alert,
S/C_Update, S/C_Last, and GndAna).  The S/C_Alert always has a sequence number of 1,
then there can be one or more S/C_Update messages for a given trigger (2,3,...,N),
then one S/C_Last (seq_num=N+1), and then one GndAnalysis message (seq_num=N+2).
The two parts of the trig_seq_num word can be extracted using these masks:
#define TRIGNUM_MASK   0x0000FFFF     // Trigger number
#define SEQNUM_MASK    0xFFFF0000     // Mesg Seq Num, then downshift 16 bits

The "burst_tjd" is the Truncated Julian Day of the HETE trigger,
eg. TJD=11910 is 01 Jan 2001.

The "burst_sod" is the UT seconds-of-day of the HETE trigger.  This is the
spacecraft system clock UT time when the on-board software found a large
enough increase in the countrate.  The sod is multiplied by 100,
and converted to a 32-bit integer for the packet (ie  the units are centi-sec; 10 msec).

There are 2 "spare" items next in the packet that are not filled in
for the S/C_Alert packet.  These are the locations of the "burst_ra" and "burst_dec"
which will be assigned values in the later packets when the position information
becomes available.  Please note that on some occasions (~10% of the time),
these two fields can, in fact, be filled with the burst_ra and burst_dec
position values for the burst.  This happens when the s/c can not transmit
the Alert when it first becomes available on the s/c (ie when there is a
gap in the downlinks),  if this gap islong enough (about 10-30 sec), then enough
time will have gone by for the "Update" to come along.  The Update has position
information so it fills in those "slots" in the outgoing packet.  But the outgoing
packet keeps the Alert type indentity instead of the Update indentity even though
it has position information.  The TBD flag is also set in this scenario 
(indicating that there are values to be had in the burst_ra/_dec slots).

The "trig_flags" is a collection of bit flags. The content, bit position,
and meaning is contained in the following list of macros:
#define GAMMA_TRIG     0x00000001     // Did FREGATE trigger? 1=yes, 0=no
#define WXM_TRIG       0x00000002     // Did WXM trigger?
#define SXC_TRIG       0x00000004     // Did SXC trigger?
#define ART_TRIG       0x00000008     // Artificial trigger?
#define POSS_GRB       0x00000010     // Possible GRB
#define DEF_GRB        0x00000020     // Definite GRB
#define DEF_NOT_GRB    0x00000040     // Definitely NOT GRB
#define HETE_NEAR_SAA  0x00000080     // S/c is in or near the SAA
#define POSS_SGR       0x00000100     // Possible SGR
#define DEF_SGR        0x00000200     // Definite SGR
#define POSS_XRB       0x00000400     // Possible XRB
#define DEF_XRB        0x00000800     // Definite XRB
#define GAMMA_DATA     0x00001000     // Gamma data in this message
#define WXM_DATA       0x00002000     // WXM data in this message
#define SXC_DATA       0x00004000     // SXC data in this message
#define OPT_DATA       0x00008000     // Optical (s/c ACS) data in this message
#define WXM_POS        0x00010000     // WXM position in this message
#define SXC_POS        0x00020000     // SXC position in this message
#define H_TRIG_spare1  0x000C0000     // spare1
#define H_USE_TRIG_SN  0x00400000     // Use H_WXM_S_N
#define H_TRIG_spare2  0x00800000     // spare2
#define H_SXC_EN_TRIG  0x01000000     // Triggered on the 1.5-12 keV band
#define H_LOW_EN_TRIG  0x02000000     // Triggered on the 2-30 keV band
#define H_MID_EN_TRIG  0x04000000     // Triggered on the 6-120 keV band
#define H_HI_EN_TRIG   0x08000000     // Triggered on the 25-400 keV band
#define H_PROB_XRB     0x10000000     // Probable XRB
#define H_PROB_SGR     0x20000000     // Probable SGR
#define H_PROB_GRB     0x40000000     // Probable GRB
#define H_TRIG_spare3  0x80000000     // spare3
The 9 "*_POSS_*", "*_PROB_*", and "*_DEF_*" flag bits indicate
if the flight s/w thought the trigger was "possibly", "probably" or "definitely"
due to a GRB, SGR, or an X-Ray Burster.  Given the complexity of the problem and
the limitations of flight s/w algorithms,  more than one of these 9 bits
can be set at the same time.
The "WXM_DATA" and "SXC_DATA" flag bits indicate whether or not this packet
contains data describing the brightness of the SXC or WXM data.
The "WXM_POS" and "SXC_POS" flag bits indicate whether or not this packet
contains valid WSM and SXC position information (ie the 4 corners).
These 2 flag bits are only valid for the S/C_UPDATE, S/C_LAST, and GNDANA (41-43)
packets (ie not valid for the S/C_ALERT (40) packet).

The "gamma_counts" field contains the count rate as detected
by the FREGATE detector.  The units are (approximately) counts per sec.

The "wxm_signoise" field contains the rate-trigger detection signal-to-noise
ratio of the WXC detector.  It is the increase in the countrate
that caused the trigger -- the units are an integer ratio [centi-sigma].

The "sxc_counts" field contains the count rate as detected by the SXC detector.
The units are counts per sec.

The "gamma_time" is the trigger_criteria timescale used in the FREGATE
detection -- this is related to the risetime of the burst.

The "wxm_time" is the trigger timescale, which is more like a burst duration,
but not exactly: it's the "foreground" interval duration which gives
the best S/N in the trigger algorithm.

The "sc_point" field contains the RA,Dec coordinates of the s/c pointing
direction.  The RA is stored in the upper word, and Dec is stored
in the lower word.  The units are degrees.

There are 23 "spare" items in the packet that are not used (usually filled
with zeros).

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.





TYPE=41 PACKET CONTENTS:

NOTE: Type=41 packets (HETE_S/C_UPDATE) are NO LONGER AVAILABLE.

The HETE_S/C_UPDATE packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=41)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       trig_seq_num integers        Trigger num & mesg sernum
long         5       burst_tjd    [days]          Truncated Julian Day
long         6       burst_sod    [centi-sec]     (int)(sssss.sss *100)
long         7       burst_ra     [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long         8       burst_dec    [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long         9       trig_flags   bits            Trigger type ID flag bits
long        10       gamma_cnts   [cnts/sec]      GammaDet count rate
long        11       wxm_signoise [centi-sig/nois]WXM rate trigger signal/noise
long        12       sxc_cnts     [cnts/sec]      SXC det count rate
long        13       gamma_time   [milli-sec]     GammaDet timescale
long        14       wxm_time     [milli-sec]     WXM det timescale
long        15       sc_point     integers        S/c RA,Dec pointing dir
long        16       wxra1        [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long        17       wxdec1       [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long        18       wxra2        [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long        19       wxdec2       [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long        20       wxra3        [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long        21       wxdec3       [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long        22       wxra4        [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long        23       wxdec4       [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long        24       wx_errors    bit fields      Packed with Sys & Stat errors
long        25       wx_dimsig    bit fields      Packed numbers
long        26       sxra1        [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long        27       sxdec1       [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long        28       sxra2        [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long        29       sxdec2       [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long        30       sxra3        [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long        31       sxdec3       [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long        32       sxra4        [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long        33       sxdec4       [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long        34       sx_errors    bit fields      Packed with Sys & Stat errors
long        35       sx_dimsig    bit fields      Packed numbers
long        36       pos_flags    bits            Mixed bits: pos, signif, data
long        37       validity     bits            Trigger source ID type
long        38       lc_img_sn    4 bytes         WXM X&Y LightCurve & Image SN
long        39       pkt_term     integer         Pkt Termination (always = \n)



PACKET ITEM DESCRIPTIONS (only those unique to type=41):

The following items are unique to packet type=41.  Items 0-13 and 39
are the same as defined in packet type=40.

The "burst_ra" and "burst_dec" items are the RA,Dec (Current epoch) location
of the burst.  Since there are two instruments that can produce
position information, these two locations are filled with the best position
of the 1 or 2 possibilities.  If both the SXC and WXM positions are available,
then these two fields are filled with the SXC location.  If only the WXM
position is available, the WXM position is contained in these two items.
If neither position is available, then that very rare (but technically
possible) situation is indicated by filling the items with the values
-999.9999 and -999.9999 (illegal values for RA,Dec).

Note that the 3 countrate fields (gamma_cnt, wxm_cnt, and sxc_cnt) may change
their values from from one Notice to the next.

The "trig_flags" field may also change from one Notice to the next.

The "wxraN", "wxdecN", sxraN", and "sxdecN" fields are the RA,Dec corners
of the WXC and SXC instrument source position error boxes (in J2000 epoch),
respectively.  The value "N" runs from 1 to 4 for the four corners of the box.
Coordinates of corners are given in degrees * 10000 (0.36" precision!).
If all 4 corners are the same RA,Dec location, then the error box is circular
and not rectangular.  And the RA,Dec given is the center of the circle (and
is the same as the "burst_ra","burst_dec" fields.

The"wx_errors" contains the statistical error (upper word) and
the systematic error (lower word) for the WXC position determination.
These should be added in quadrature to get the total uncertainty (radius).
To decode the wx_errors word:
#define STATERR_MASK    0xFFFF0000    // Statistical error [arcsec], radius
#define SYSERR_MASK     0x0000FFFF    // Systematic error [arcsec], radius

The "wx_dimsig" contains two things:
a) the maximum dimension of the WXM error box (the distance between the two
corners of the error box that are the farthest apart [units arcsec]),   and
b) the significance of the detection of the point source in the image plane
(in Sig/Noise units) [note that this is different than the wxm_signoise field
which is based on the rate-trigger detection significance].
To decode the ws_dimsig word:
#define MAXDIM_MASK     0xFFFF0000    // Max dimension of location box
#define NSIGMA_MASK     0x0000FFFF    // Detection significance [Sig/Noise]
Note that the "maximum dimension" of the 'error box' is effectively a diameter
for a circumscribing 'error circle'.

The"sx_errors" contains the statistical error (upper word) and
the systematic error (lower word) for the SXC position determination.
The same decoding listed in wx_errors also applies for sx_errors.

The "sx_dimsig" contains two things:
a) the maximum dimension of the SXC error box (the distance between the two
corners of the error box that are farthest apart [units arcsec],   and
b) the significance of the detection (in Sig/Noise units).
The same decoding macros listed in wx_dimsig also apply for sx_dimsig.

The "pos_flags" is a slight misnomer.  It contains a mixture of flag bits.
There are bits about the status of the various detected positions;
bits about the significance of the detections; and bits about if things
have changed/improved during the ground analysis (GNDANA packets only).
Except the "sc_long" field which is present in all 4 types (Alerts through
GNDANA).  It is contains the longitude of the s/c in its orbit, and is measured
eastward from the prime meridian (ie it goes 0-359deg, however because of
bit-packing constraints, it is prescaled by 2).
The pos_flags bits:
#define WXM_SXC_SAME    0x00000001     // Positions are consistent, 1=yes
                                       // WXM and SXC have overlapping boxes
#define WXM_SXC_DIFF    0x00000002     // Positions are inconsistent
                                       // WXM and SXC do not have overlapping boxes
#define WXM_LOW_SIG     0x00000004     // NSIGMA below a threshold
                                       // The S/N is below a "good" value
#define SXC_LOW_SIG     0x00000008     // NSIGMA below a threshold
                                       // The S/N is below a "good" value
// The following bits are assigned (and therefore valid) only
// in the GNDANA packets:
#define GAMMA_REFINED   0x00000010     // Gamma refined since FINAL
                                       // Gamma data changed as a result
                                       // of ground analysis
#define WXM_REFINED     0x00000020     // WXM refined since FINAL
                                       // WXM data changed as a result
                                       // of ground analysis
#define SXC_REFINED     0x00000040     // SXC refined since FINAL
                                       // SXC data changed as a result
                                       // of ground analysis
#define POS_spare       0x00FFFFB0     // spare
#define SC_LONG         0xFF000000     // S/C Longitude East [units of 2-degrees]

The "validity" is set of bit flags specifying the identity of the type
of trigger (i.e. GRB, x-ray src, etc).
Validity is determined after the trigger has been detected.
Fregate looks at background trends before and after the burst:
it can say "hey, this was due to a rising background" and the burst is invalid.
The on-board code can see that a burst came from a known source,
but only after a position has been detected:  the burst is invalid
because it's a known source.
The on-board code can protect against emersion triggers if it's
been properly configured.  If the ground analysis shows
"oops, it was an emersion trigger", then that bit is set.
If there's no position, it's a useless burst.  This bit is defined,
but it is not currently set by the on-board or the ground processing s/w.
The following macros can be used to extract the various flag bits from validity:
#define BURST_VALID     0x00000001     // Burst declared valid
#define BURST_INVALID   0x00000002     // Burst declared invalid
#define H_VAL0x4_spare  0x00000004     // spare; not used
#define H_VAL0x8_spare  0x00000008     // spare; not used
#define EMERGE_TRIG     0x00000010     // Emersion trigger
#define KNOWN_XRS       0x00000020     // Known X-ray source
#define NO_POSITION     0x00000040     // No WXM or SXC position
#define H_VAL0x80_spare 0x00000080     // spare; not used
#define OPS_ERROR       0x00000100     // Bad burst: s/c and inst ops error
#define PARTICLES       0x00000200     // Bad burst: particles
#define BAD_FLT_LOC     0x00000400     // Bad burst: bad flight location
#define BAD_GND_LOC     0x00000800     // Bad burst: bad ground location
#define RISING_BACK     0x00001000     // Bad burst: rising background
#define POISSON_TRIG    0x00002000     // Bad burst: poisson fluctuation trigger
#define H_VALID_spare   0x1FFFC000     // spare
#define H_EMAIL_METHOD  0x20000000     // email import method(=1, socket=0)
#define H_INTERNAL_TEST 0x40000000     // HETE_OPS GndAna test/debug loop
#define H_NOT_A_BURST   0x80000000     // GndAna shows this trigger to be non-GRB

The "lc_img_sn" contains 4 values (the WXM X and Y LightCurve SN and
Image SN values).
WXM_X_LC_SN     in the most significant byte,
WXM_Y_LC_SN     in the next most significant byte,
WXM_X_IMAGE_SN  in the third most significant byte, and
WXM_Y_IMAGE_SN  in the least significant byte.

WXM_X_LC_SN and WXM_Y_LC_SN describe the signal-to-noise of the excess
over background in the WXM light curve (LC) in each of the two modules (X and Y).
The WXM light curve is obtained by summing the data over all position bins
at 320ms time resolution.  The flight software chooses background and
burst time intervals, and performs a standard background subtraction
in each detector.  The resulting net signals are divided by their respective 
root-variances, to yield the quantities reported here.
Note that WXM_*_LC_SN are not directly related to trigger SNR,
since the trigger operates on 80ms resolution data, and may in fact
come from FREGATE data rather than from WXM data.  The light-curve SNR
is designed measure the overall signal-to-noise in the actual data used
to obtain the location, so as to complement WXM_*_IMAGE_SN.
The two values stored in the packet are in units of deci-sigma (ie they
have been multiplied by 10.0 and integerized prior to storing, and thus
have a dynamic range of 0.0 to 25.5).
 
WXM_X_IMAGE_SN and WXM_Y_IMAGE_SN describe the peak signal-to-noise
in the cross-correlation of the background-subtracted image with the mask
in each of the two modules (X and Y).
The background-subtracted signal across each WXM detector is cross-correlated
with the mask.  The resulting cross-correlation functions are mean-subtracted
and divided by their respective root-variances, yielding SNR as a function
of source projection angle in each detector.  The quantities reported here
are the peak values of those functions.
The two values stored in the packet are in units of deci-sigma (ie they
have been multiplied by 10.0 and integerized prior to storing, and thus
have a dynamic range of 0.0 to 25.5).


TYPE=42 PACKET CONTENTS:

NOTE: Type=42 packets (HETE_S/C_LAST) are NO LONGER AVAILABLE.

The HETE_S/C_LAST packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=42)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       trig_seq_num integers        Trigger num & mesg sernum
long         5       burst_tjd    [days]          Truncated Julian Day
long         6       burst_sod    [centi-sec]     (int)(sssss.sss *100)
long         7       burst_ra     [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long         8       burst_dec    [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long         9       trig_flags   bits            Trigger type ID flag bits
long        10       gamma_cnts   [cnts/sec]      GammaDet count rate
long        11       wxm_signoise [centi-sig/nois]WXM rate trigger signal/noise
long        12       sxc_cnts     [cnts/sec]      SXC det count rate
long        13       gamma_time   [milli-sec]     GammaDet timescale
long        14       wxm_time     [milli-sec]     WXM det timescale
long        15       sc_point     integers        S/c RA,Dec pointing dir
long        16       wxra1        [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long        17       wxdec1       [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long        18       wxra2        [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long        19       wxdec2       [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long        20       wxra3        [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long        21       wxdec3       [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long        22       wxra4        [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long        23       wxdec4       [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long        24       wx_errors    bit fields      Packed with Sys & Stat errors
long        25       wx_dimsig    bit fields      Packed numbers
long        26       sxra1        [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long        27       sxdec1       [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long        28       sxra2        [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long        29       sxdec2       [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long        30       sxra3        [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long        31       sxdec3       [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long        32       sxra4        [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long        33       sxdec4       [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long        34       sx_errors    bit fields      Packed with Sys & Stat errors
long        35       sx_simsig    bit fields      Packed numbers
long        36       pos_flags    bits            Mixed bits: pos, signif, data
long        37       validity     bits            Trigger source ID type
long        38       lc_img_sn    4 bytes         WXM X&Y LightCurve & Image SN
long        39       pkt_term     integer         Pkt Termination (always = \n)



PACKET ITEM DESCRIPTIONS (only those unique to type=42):

The following items are unique to packet type=42.  Items 0-33 and 39
are the same as defined in packet type=41.  The only two new 
bit fields in "validity".

The following macros can be used to extract the new flag bits from validity:
#define BURST_VALID     0x00000001     // Is this a real GRB?  1=yes
#define BURST_INVALID   0x00000002     // It is not a real GRB. 1=not_real



TYPE=43 PACKET CONTENTS:

NOTE: Type=43 packets (HETE_GNDANA) are NO LONGER AVAILABLE.

The HETE_GNDANA packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=43)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       trig_seq_num integers        Trigger num & mesg sernum
long         5       burst_tjd    [days]          Truncated Julian Day
long         6       burst_sod    [centi-sec]     (int)(sssss.sss *100)
long         7       burst_ra     [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long         8       burst_dec    [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long         9       trig_flags   bits            Trigger type ID flag bits
long        10       gamma_cnts   [cnts/sec]      GammaDet count rate
long        11       wxm_signoise [centi-sig/nois]WXM rate trigger signal/noise
long        12       sxc_cnts     [cnts/sec]      SXC det count rate
long        13       gamma_time   [milli-sec]     GammaDet timescale
long        14       wxm_time     [milli-sec]     WXM det timescale
long        15       sc_point     integers        S/c RA,Dec pointing dir
long        16       wxra1        [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long        17       wxdec1       [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long        18       wxra2        [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long        19       wxdec2       [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long        20       wxra3        [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long        21       wxdec3       [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long        22       wxra4        [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long        23       wxdec4       [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long        24       wx_errors    bit fields      Packed with Sys & Stat errors
long        25       wx_dimsig    bit fields      Packed numbers
long        26       sxra1        [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long        27       sxdec1       [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long        28       sxra2        [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long        29       sxdec2       [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long        30       sxra3        [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long        31       sxdec3       [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long        32       sxra4        [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long        33       sxdec4       [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long        34       sx_errors    bit fields      Packed with Sys & Stat errors
long        35       sx_simsig    bit fields      Packed numbers
long        36       pos_flags    bits            Mixed bits: pos, signif, data
long        37       validity     bits            Trigger source ID type
long        38       lc_img_sn    4 bytes         WXM X&Y LightCurve & Image SN
long        39       pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS (only those unique to type=43):

There are no unique (new) fields in packet type=43.  All the fields defined
in type=42 are also present in type=43.  However, there are a few new bits
defined in the "pos_flags" and "validity" fields.  Also, the values contained
in the following fields may have changed from what was distributed
in the last type=42 Notice.




HETE FLAG BITS  VS  NOTICE TYPE:

The table below contains the 3 fields that have flag bits in the 4 types
of HETE Notices (S/C_Alert, S/C_Update, S/C_Last, and GroundAnalysis).  It shows 
which bits are assigned (set to 1 or 0) in each Notice type.
An "X" means that bit is tested and set (1or0) in that Notice type.
A blank (" ") means that bit is not determined/controlled in that Notice type.

Table of HETE Packet Flag Bits  vs  Notice Type

WORD:
BIT             Bit_Mask     S/C_Alert S/C_Update S/C_Last GndAna

TRIG_FLAGS:

H_GAMMA_TRIG    0x00000001       X       X        X        X
H_WXM_TRIG      0x00000002       X       X        X        X
H_SXC_TRIG      0x00000004       X       X        X        X
H_ART_TRIG      0x00000008       X       X        X        X
H_POSS_GRB      0x00000010       X       X        X        X
H_DEF_GRB       0x00000020       X       X        X        X
H_DEF_NOT_GRB   0x00000040       X       X        X        X
H_NEAR_SAA      0x00000080       X       X        X        X
H_POSS_SGR      0x00000100       X       X        X        X
H_DEF_SGR       0x00000200       X       X        X        X
H_POSS_XRB      0x00000400       X       X        X        X
H_DEF_XRB       0x00000800       X       X        X        X
H_GAMMA_DATA    0x00001000       X       X        X        X
H_WXM_DATA      0x00002000       X       X        X        X
H_SXC_DATA      0x00004000       X       X        X        X
H_OPT_DATA      0x00008000       X       X        X        X
H_WXM_POS       0x00010000       X       X        X        X
H_SXC_POS       0x00020000       X       X        X        X
H_TRIG_spare    0xFFFD0000    


POS_FLAGS:

H_WXM_SXC_SAME   0x00000001              X        X        X
H_WXM_SXC_DIFF   0x00000002              X        X        X
H_WXM_LOW_SIG    0x00000004   ???
H_SXC_LOW_SIG    0x00000008   ???
H_GAMMA_REFINED  0x00000010                                X
H_WXM_REFINED    0x00000020                                X
H_SXC_REFINED    0x00000040                                X
H_POS_spare      0x00FFFFB0    
H_SC_LONG        0xFF000000      X       X        X        X


VALIDITY:

H_BURST_VALID   0x00000001                         X       X
H_BURST_INVALID 0x00000002                         X       X
H_VAL0x4_spare  0x00000004    
H_VAL0x8_spare  0x00000008
H_EMERGE_TRIG   0x00000010               X         X       X
H_KNOWN_XRS     0x00000020               X         X       X
H_NO_POSITION   0x00000040               X         X       X
H_VAL0x80_spare 0x00000080    
H_OPS_ERROR     0x00000100                                 X
H_PARTICLES     0x00000200                                 X
H_BAD_FLT_LOC   0x00000400                                 X
H_BAD_GND_LOC   0x00000800                                 X
H_RISING_BACK   0x00001000                                 X
H_POISSON_TRIG  0x00002000                                 X
H_VALID_spare   0x1FFFC000    
H_EMAIL_METHOD  0x20000000                                 X
H_INTERNAL_TEST 0x40000000                                 X
H_NOT_A_BURST   0x80000000                                 X


SLOT 38:

H_X_LC          0xff000000                X        X       X
H_Y_LC          0x00ff0000                X        X       X
H_X_IMAGE       0x0000ff00                X        X       X
H_Y_IMAGE       0x000000ff                X        X       X



TYPE=44 PACKET CONTENTS:

The HETE_Test packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=44)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       trig_seq_num integers        Trigger num & mesg sernum
long         5       burst_tjd    [days]          Truncated Julian Day
long         6       burst_sod    [centi-sec]     (int)(sssss.sss *100)
long         7       burst_ra     [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long         8       burst_dec    [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long         9       trig_flags   bits            Trigger type ID flag bits
long        10       gamma_cnts   [cnts/sec]      GammaDet count rate
long        11       wxm_signoise [centi-sig/nois]WXM rate trigger signal/noise
long        12       sxc_cnts     [cnts/sec]      SXC det count rate
long        13       gamma_time   [milli-sec]     GammaDet timescale
long        14       wxm_time     [milli-sec]     WXM det timescale
long        15       sc_point     integers        S/c RA,Dec pointing dir
long        16       wxra1        [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long        17       wxdec1       [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long        18       wxra2        [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long        19       wxdec2       [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long        20       wxra3        [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long        21       wxdec3       [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long        22       wxra4        [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long        23       wxdec4       [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long        24       wx_errors    bit fields      Packed with Sys & Stat errors
long        25       wx_dimsig    bit fields      Packed numbers
long        26       sxra1        [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long        27       sxdec1       [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long        28       sxra2        [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long        29       sxdec2       [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long        30       sxra3        [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long        31       sxdec3       [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long        32       sxra4        [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long        33       sxdec4       [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long        34       sx_errors    bit fields      Packed with Sys & Stat errors
long        35       sx_simsig    bit fields      Packed numbers
long        36       pos_flags    bits            Mixed bits: pos, signif, data
long        37       validity     bits            Trigger source ID type
long        38       lc_img_sn    4 bytes         WXM X&Y LightCurve & Image SN
long        39       pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS (only those unique to type=44):

There are no unique (new) fields in packet type=44.  All the fields defined
in type=43 are also present in type=44.





HETE-BASED PACKET ITEM MACROS:

Below are the C-language define's for picking out the various elements
of the array of longs that make up the HETE-based packets.  They have names
which reflect the contents of packet types 40, 41, 42, 43, & 44.  A few of the
items have the same names as the original BATSE-based packets, because
they are common to all type-bases.  For the other (non-HETE-based) types,
these macros should not be used because these macro names would conote
the wrong information content.

// Indices into the socket packet array of 40 longs:
#define PKT_TYPE         0   // Packet type
#define PKT_SERNUM       1   // Packet serial number
#define PKT_HOP_CNT      2   // Packet hop count
#define PKT_SOD          3   // Packet Sec-of-Day [centi-sec] (sssss.sss*100)
#define BURST_TRIG       4   // HETE trigger number plus message sequence num
#define BURST_TJD        5   // Truncated Julian Day
#define BURST_SOD        6   // Sec-Of-Day [centi-sec] (sssss.sss*100)
#define BURST_RA         7   // RA  [10^-4-deg] (0.0 to 359.9999 *100)
#define BURST_DEC        8   // Dec [10^-4-deg] (-90.0 to +90.0 *100)
#define TRIG_FLAGS       9   // See flags below
#define GAMMA_CTS_S     10   // cts/s defined by gamma, can be 0
#define WXM_CTS_S       11   // cts/s defined by WXM, can be 0
#define SXC_CTS_S       12   // cts/s defined by SXC, can be 0
#define GAMMA_TIMESCALE 13   // in milliseconds
#define WXM_TIMESCALE   14   // in milliseconds
#define SC_POINTING     15   // S/c pointing direction in degrees
#define WXM_CNR1_RA     16   // First corner of WXM error box
#define WXM_CNR1_DEC    17   // First corner of WXM error box
#define WXM_CNR2_RA     18   // Second corner of WXM error box
#define WXM_CNR2_DEC    19   // Second corner of WXM error box
#define WXM_CNR3_RA     20   // Third corner of WXM error box
#define WXM_CNR3_DEC    21   // Third corner of WXM error box
#define WXM_CNR4_RA     22   // Fourth corner of WXM error box
#define WXM_CNR4_DEC    23   // Fourth corner of WXM error box
#define WXM_ERRORS      24   // Arcsec, stat high, sys low
#define WXM_DIM_NSIGMA  25   // Max box size (arcsec), significance of loc'n
#define SXC_CNR1_RA     26   // First corner of SXC error box
#define SXC_CNR1_DEC    27   // First corner of SXC error box
#define SXC_CNR2_RA     28   // Second corner of SXC error box
#define SXC_CNR2_DEC    29   // Second corner of SXC error box
#define SXC_CNR3_RA     30   // Third corner of SXC error box
#define SXC_CNR3_DEC    31   // Third corner of SXC error box
#define SXC_CNR4_RA     32   // Fourth corner of SXC error box
#define SXC_CNR4_DEC    33   // Fourth corner of SXC error box
#define SXC_ERRORS      34   // Arcsec, stat high, sys low
#define SXC_DIM_NSIGMA  35   // Max box size (arcsec), significance of loc'n
#define POS_FLAGS       36   // Collection of various info bits
#define VALIDITY        37   // Trigger identity, source type, etc
#define H_LC_IMG_SN     38   // LightCurve and Image Sig/Noise ratios
#define PKT_TERM        39   // Packet termination character





================================================================================

TYPE=45 PACKET CONTENTS:

The GRB_COUNTERPART packet contains detections of the burst counterpart
in both the optical, radio, and other (x-ray, TeV, neutrinos, etc ...) bandpasses.
However, to make the field_names shorter, they were given names
with an optical bias.  Only the name has this bias, the contents
cover the optical, the radio, the x-ray, particles, etc.

The GRB_COUNTERPART packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=45)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       ref_num      integer         Some reference number (eg trig_num)
long         5       burst_tjd    [days]          Truncated Julian Day
long         6       burst_sod    [centi-sec]     (int)(sssss.sss *100)
long         7       cp_ra        [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long         8       cp_dec       [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long         9       cp_inten     [centi-mag]     (int)(mm.mm *100)
long        10       inten_err    [centi-mag]     (int)(mm.mm *100)
long        11       cp_error     [10^-4-deg]     (int)(0.0 to 180.0 *10000)
long        12       filter       integer         Opt_fltr, Radio_freq, Oth_??
long        13       seeing       [centi-arcsec]  (int)(ss.ss *100)
long        14       obs_tjd      [days]          Truncated Julian Day
long        15       obs_time     [centi-sec]     (int)(sssss.sss *100)
long        16       obs_dur      [centi-sec]     (int)(sssss.sss *100)
long        17       id_conf      [centi-%]       (int)(0.0 to 100.00 *100)
long        18       flags        bits            Flag bits on source-type
long        19       misc         bits            Misc flag bits and fields
quad_char   20-23    telescope[4] chars           The name of the telescope
quad_char   24-38    name[15]     chars           The name of the submittor
long        39       pkt_term     integer         Pkt Termination (always = \n)



PACKET ITEM DESCRIPTIONS (type=45):


The "pkt_type" is an integer with a value of 45, which tells that this packet
is of type GRB_COUNTERPART and therefore indicates the contents and format
for the other 39 fields.

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start from 1 when
the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 1.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the burst_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "ref_num" is a reference number for that burst.  Typically it is a trigger_number
of the burst for that mission that made the original detection of the burst.

The "burst_tjd" is the Truncated Julian Day of the burst,
eg. TJD=12275 is 01 Jan 2002.

The "burst_sod" is the UT seconds-of-day (SOD) of the burst.  The SOD
is multiplied by 100, and converted to a 32-bit integer for the packet
(i.e the units are centi-sec; 10 msec),  eg. 4689568 = 46895.68 sec-of-day =
13:01:35.68 UT.

The "cp_ra" and "cp_dec" fields specify the RA,Dec location of the detected
GRB counterpart (in J2000 epoch).  These intrinsically floating point
quantities have been multiplied by 10,000 and then integerized to make
an integer quantity with units of 10^-4-degrees.

The "cp_inten" field contains the magnitude of the detected optical transient.
The units are centi-magnitudes (ie the magnitude was multiplied by 100 and
then integerized).  If the packet is for a counterpart detection
in the radio band (see the misc flag bits field), then the "cp_mag" field
contains the flux in centi-uJy (ie 10^-8 Janskies).  If the counterpart
detection in the "other" (x-/g-ray) category, then the units type are specified
by the 'expo' subfield in the 'filter' field and then multiplied by 100 and integerized.

The "inten_error" field contains the uncertainty (statistical plus systematic)
of the cp_inten value.  The units are also centi-magnitudes or centi-uJy or
centi-units for 'other' (eg centi-kHz, centi-MHz, centi-GHz, centi-keV,...,centi-TeV).
(See the 'expo' subfield in the 'filter' value.)

The "cp_error" is the radius of the circle that will contain on average
68% of the counterpart locations.  The units are 10^-4-degrees.

The "filter" field specifies the filter that was used for the observation.
The filter_index comes from a predefined list [see below].
  0 n/a
  1 U
  2 B
  3 V
  4 R
  5 I
  6 J
  7 K
  8 L
  9 M
 10 r
The upper 8 bits (of this 32-bit field) contain the 'exponent' of the 'band'
if this is a 'radio' or 'other' counterpart observation:
RADIO:
  3 kHz
  6 MHz
  9 GHz
OTHER:
  3  keV
  6  MeV
  9  GeV
  12 TeV
  15 PeV

The "seeing" field contains the seeing during the observation.  The units
are centi-arcsec.

The "obs_tjd" is the Truncated Julian Day of the start of the observation.
The "obs_sod" is the Seconds-of-Day of the start of the observation.
The "obs_dur" is the duration in centi-seconds of the observation.  The seconds
is multiplied by 100 and then integerized (i.e. the units are centi-sec, ie 10 msec).

The "id_conf" is the confidence level of the identification of the counterpart to the GRB.
The percentage_value is multiplied by 100 and then integerized (i.e. the units are centi-%).

The "trig_id" field contains flag bit bits relating to the source-type identity.
Bit_Number     Item_name      Description
----------     ---------      -----------
0              n/a            n/a
1              GRB_DEF        This is definitely related to a GRB
2-27           n/a            spare
28             spatial_coinc  0=No or 1=Yes, there was a spatial coincidence with another event
29             temporal_coinc 0=No or 1=Yes, there was a temporal coincidence with another event
30-31          n/a            spare

The "misc" field contains various miscellaneous flags and fields.
The bit packing of the "misc" entry in the packet is:
Bit_Number     Item_name      Description
----------     ---------      -----------
0-4            n/a            spare
5-6            type           optical=1/radio=2/other=3
7-23           n/a            spare
24-31          iexpo          Intensity exponent value

The "telescope" is the name of the telescope that made the detection.
This item in the packet spans 4 longwords (ie 16 bytes) and contains 
the ASCII string of the name of the telescope.  This string can be
up to 15 characters long, and will always be null-terminated.

The "name" is the name of the person submitting this counterpart detection notice.
This item in the packet spans 15 longwords (ie 60 bytes) and contains 
the ASCII string of the name of the submittor.  This string can be
up to 59 characters long, and will always be null-terminated.

There are no "spare" items in this packet.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long in
the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of GCN operation
(called BACODINE in those days).  It has since become a permanent fixture
as multiple sites are now operating, and it is not wise to play
with format/content changes.  And it still does function as a fixed-content
termination value for processing/transmission checking.




================================================================================


TYPE=46 PACKET CONTENTS:
TYPE=47 PACKET CONTENTS:

See packet types 65 and 67 respectively.

 
================================================================================




TYPE=48 PACKET CONTENTS:

The DOW_TOD Test packet contain timestamp information about the 
day-of-week and time-of-day you requested this notice.

The DOW_TOD Test packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=45)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       spare        integer         4 bytes for future use
long         5       pkt_tjd      [days]          Truncated Julian Day
long         6       pkt_sod      [centi-sec]     (int)(sssss.sss *100)
long         7-38    spare[22]    integer         88 bytes for the future
long         39      pkt_term     integer         Pkt Termination (always = \n)



PACKET ITEM DESCRIPTIONS (type=48):


The "pkt_type" is an integer with a value of 48, which tells that this packet
is of type DOW_TOD and therefore indicates the contents and format
for the other 39 fields.

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start from 1 when
the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 1.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the burst_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "notice_tjd" is the Truncated Julian Day when the notice was created,
eg. TJD=12275 is 01 Jan 2002.

The "notice_sod" is the UT seconds-of-day (SOD) when the notice was created.
The SOD is multiplied by 100, and converted to a 32-bit integer for the packet
(i.e the units are centi-sec; 10 msec),  eg. 4689568 = 46895.68 sec-of-day =
13:01:35.68 UT.

There are 23 "spare" items in this packet.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long in
the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of GCN operation
(called BACODINE in those days).  It has since become a permanent fixture
as multiple sites are now operating, and it is not wise to play
with format/content changes.  And it still does function as a fixed-content
termination value for processing/transmission checking.

================================================================================

PACKET ITEM DESCRIPTIONS for type=51:

The INTEGRAL_POINTDIR packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=51)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       trig_sub_num integers        Trigger num & mesg sernum
long         5       slew_tjd     [days]          Truncated Julian Day
long         6       slew_sod     [centi-sec]     (int)(sssss.sss *100)
long         7-11    spare[5]     integer         20 bytes for the future
long        12       test_mpos    flag_bits       Test and Multi-Position flags
long        13       spare[1]     integer         4 bytes for the future
long        14       next_sc_ra   [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long        15       next_sc_dec  [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long        16-18    spare[3]     integer         12 bytes for the future
long        19       misc_att     integer         S/c status & attitude
long        20-38    spare[19]    integer         76 bytes for the future
long        39       pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 51, 52, 53 , 54, or 55
that tells which type of packet this is and therefore what the contents are
(the contents change slightly between the various types).  The types are:
51 = INTEGRAL_POINTDIR, 52 = INTEGRAL_SPIACS, 53 = INTEGRAL_WAKEUP,
54 = INTEGRAL_REFINED, and 55 = INTEGRAL_OFFLINE.

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start from 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created and sent from
the GCN computer.  Do not confuse this with the burst_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "trig_sub_num" are the uniquely identifying serial numbers (main and sub)
assigned by the on-board INTEGRAL flight software to each trigger.
The lower 16 bits contain the trigger ("Alert") number, and the upper 16 bits
contain the Alert sub-number.  The sub-number is always 0 for type=53;
goes from 1 to N for type=54; and N+1, N+2, ... for type=55.

The "slew_tjd" is the TJD when the next s/c slew happens.

The "slew_sod" is the seconds-of-day (SOD) when the next s/c slew happens.

The "test_mpos" field contains the Test_Notice flag bit (2^31) and
the multiple_position flag bit (2^0).  The multi-position flag is meaningless
for the POINTDIR Notice.

The "next_sc_ra" and "next_sc_dec" are the RA,Dec (J2000) coordinates
to which the spacecraft will point after this slew.  This is the pointing
direction of the s/c.  The units are 10^04 deg, ie the values
where multiplied by 10000 and integerized.

The "misc_att" field contains status and attitude information on the s/c.
The exact contents and format are still TBD at the INTEGRAL MOC.

There are a total of 28 "spare" items in the packet (usually filled with zeros).
They are located at items 7-11,13, 16-18, and 20-38.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.



TYPE=52 PACKET CONTENTS:

The INTEGRAL_SPIACS packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=52)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       trig_sub_num integers        Trigger num & mesg sernum
long         5       burst_tjd    [days]          Truncated Julian Day
long         6       burst_sod    [centi-sec]     (int)(sssss.sss *100)
long         7-8     spare[2]     integer         8 bytes for the future
long         9       det_flags    flag_bits       Detector flag bits
long         10      inten_sigma  [centi-sigma]   Burst_intensity_sigma*100
long         11      spare[1]     integer         4 bytes for the future
long         12      test_mpos    flag_bits       Test and Multi-Position flags
long         13      time_scale   [10^-4 sec]     Time sampling or trigger time
long         14-15   spare[2]     integer         8 bytes for the future
long         16      time_error   [10^-4 sec]     Accuracy of GRB time
long         17-18   spare[2]     integer         8 bytes for the future
long         19      misc_att     integer         S/c status & attitude
long         20-38   spare[19]    integer         76 bytes for the future
long         39      pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 51, 52, 53 , 54, or 55
that tells which type of packet this is and therefore what the contents are
(the contents change slightly between the various types).  The types are:
51 = INTEGRAL_POINTDIR, 52 = INTEGRAL_SPIACS, 53 = INTEGRAL_WAKEUP,
54 = INTEGRAL_REFINED, and 55 = INTEGRAL_OFFLINE.

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start from 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the burst_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "trig_sub_num" are the uniquely identifying serial numbers (main and sub)
assigned by the on-board INTEGRAL flight software to each trigger.
The lower 16 bits contain the trigger ("Alert") number, and the upper 16 bits
contain the Alert sub-number.  The sub-number is always 0 for type=53;
goes from 1 to N for type=54; and N+1, N+2, ... for type=55.
The Alert number started with s/n=??? just after launch (Oct 02) and
is currently (???03) in the ???'s.

The "burst_tjd" is the Truncated Julian Day of the INTEGRAL burst trigger,
eg. TJD=12640 is 01 Jan 2003.

The "burst_sod" is the UT seconds-of-day (SOD) of the INTEGRAL burst trigger.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized).

The "det_flags" ("trig_id") field contains bit_flags specifing
the trigger process.  The bit packing of the "det_flags" entry in the packet
is unknown as yet (TBD at the INTEGRAL MOC).

The "intensity_sigma" field contains the intensity of the burst in sigma units,
essentially a signal/noise metric.  The units are 'centi-sigma', ie the fl.pt.
sigma was multiplied by 100 and then integerized and put into the packet.

The "test_mpos" field contains the Test_Notice flag bit (2^31) and
the multiple_position flag bit (2^0).  The multi-position flag is meaningless
for the SPIACS Notice.

The "test_mpos" field contains the following bit_flags:
Bit_Number     Item_name      Description
----------     ---------      -----------
0              multi_position 0=No or 1=Yes, the INTEGRAL Gnd s/w there was more
                                             than one likely position for this burst.
                                             In that case, the RA,Dec fields
                                             ????contain/do_not_contain????
                                             the most likely position (I'm waiting
                                             for info from the Ops team).
1-28           x_x_x_x        spare
29             temporal_coinc 0=No or 1=Yes, there was a temporal coincidence with another event
30             def_not_grb    0=No or 1=Yes, it is definitely not a GRB.
31             test           0=No or 1=Yes, it is a Test_Notice.

The "time_scale" field has the sampling interval used.  For type=52 (SPIACS)
it is the lightcurve binning;  for type=53 and 54 (WAKEUP and REFINED) it is
the SIT and IMITS that triggered;  and for type=55 (OFFLINE) it is the TBD
duration of the burst.

The "time_error" field contains the accuracy of the GRB timing information
(units are 10^-4 seconds).  This is usually the sum of the time bin size
used for the image reconstruction and the propogation delay.  It is expected
(by the INTEGRAL MOC) that type=55 packets (OFFLINE) will be corrected
for propogation delays.

The "sc_ra" and "sc_dec" are the RA,Dec (J2000) coordinates
to which the spacecraft points.  This is the pointing direction of the s/c.
The units are 10^04 deg, ie the values where multiplied by 10000 and 
integerized.

The "misc_att" field contains status and attitude information on the s/c.
The exact contents and format are still TBD at the INTEGRAL MOC.

There are a total of 26 "spare" items in the packet (usually filled
with zeros).  They are located at items 7-8, 11, 14-15, and 20-38.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.



TYPE=53 PACKET CONTENTS:

The INTEGRAL_WAKEUP packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=53)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       trig_sub_num integers        Trigger num & mesg sernum
long         5       burst_tjd    [days]          Truncated Julian Day
long         6       burst_sod    [centi-sec]     (int)(sssss.sss *100)
long         7       burst_ra     [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long         8       burst_dec    [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long         9       det_flags    bits            Detector flag bits
long         10      inten_sigma  [centi-sigma]   Burst_intensity_sigma*100
long         11      burst_error  [arcsec]        Burst location uncertainty 
long         12      test_mpos    flag_bits       Test and Multi-Position flags
long         13      time_scale   [10^-4 sec]     Time sampling or trigger time
long         14      sc_ra        [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long         15      sc_dec       [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long         16      time_error   [10^-4 sec]     Accuracy of GRB time
long         17-18   spare[2]     integer         8 bytes for the future
long         19      misc_att     integer         S/c status & attitude
long         20-38   spare[19]    integer         76 bytes for the future
long         39      pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 51, 52, 53 , 54, or 55
that tells which type of packet this is and therefore what the contents are
(the contents change slightly between the various types).  The types are:
51 = INTEGRAL_POINTDIR, 52 = INTEGRAL_SPIACS, 53 = INTEGRAL_WAKEUP,
54 = INTEGRAL_REFINED, and 55 = INTEGRAL_OFFLINE.

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start from 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the burst_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "trig_sub_num" are the uniquely identifying serial numbers (main and sub)
assigned by the on-board INTEGRAL flight software to each trigger.
The lower 16 bits contain the trigger ("Alert") number, and the upper 16 bits
contain the Alert sub-number.  The sub-number is always 0 for type=53;
goes from 1 to N for type=54; and N+1, N+2, ... for type=55.
The Alert number started with s/n=~1 just after launch (Oct 02) and
is currently (Oct2006) in the 3500s.

The "burst_tjd" is the Truncated Julian Day of the INTEGRAL burst trigger,
eg. TJD=12640 is 01 Jan 2003.

The "burst_sod" is the UT seconds-of-day (SOD) of the INTEGRAL burst trigger.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized).

The "burst_ra" and "burst_dec" are the RA,Dec coordinates of the trigger/burst
(current epoch).  These intrinsically floating point quantities
have been multiplied by 10000 and then integerized to make an integer quantity
with units of 10^-4-degrees.

The "det_flags" ("trig_id") field contains bit_flags specifing
the trigger process.  The bit packing of the "det_flags" entry in the packet
is unknown as yet (TBD at INTEGRAL MOC).

The "intensity_sigma" field contains the intensity of the burst in sigma units,
essentially a signal/noise metric.  The units are 'centi-sigma', ie the fl.pt.
sigma was multiplied by 100 and then integerized and put into the packet.

The "burst_error" is the radius of the circle that will contain on average
TBD% c.l. of the bursts.  The units are arcseconds.

The "test_mpos" field contains the following bit_flags:
Bit_Number     Item_name      Description
----------     ---------      -----------
0              multi_position 0=No or 1=Yes, the INTEGRAL Gnd s/w there was more
                                             than one likely position for this burst.
                                             In that case, the RA,Dec fields
                                             ????contain/do_not_contain????
                                             the most likely position (I'm waiting
                                             for info from the Ops team).
1-27           x_x_x_x        spare
28             spatial_coinc  0=No or 1=Yes, there was a spatial coincidence with another event
29             temporal_coinc 0=No or 1=Yes, there was a temporal coincidence with another event
30             def_not_grb    0=No or 1=Yes, it is definitely not a GRB.
31             test           0=No or 1=Yes, it is a Test_Notice.

The "time_scale" field has the sampling interval used.  For type=52 (SPIACS)
it is the lightcurve binning;  for type=53 and 54 (WAKEUP and REFINED) it is
the SIT and IMITS that triggered;  and for type=55 (OFFLINE) it is the TBD
duration of the burst.

The "sc_ra" and "sc_dec" are the RA,Dec (J2000) coordinates
to which the spacecraft points.  This is the pointing direction of the s/c.
The units are 10^04 deg, ie the values where multiplied by 10000 and 
integerized.

The "time_error" field contains the accuracy of the GRB timing information
(units are 10^-4 seconds).  This is usually the sum of the time bin size
used for the image reconstruction and the propogation delay.  It is expected
(by the INTEGRAL MOC) that type=55 packets (OFFLINE) will be corrected
for propogation delays.

The "misc_att" field contains status and attitude information on the s/c.
The exact contents and format are still TBD at the INTEGRAL MOC.

There are a total of 23 "spare" items in the packet (usually filled
with zeros).  They are located at items 11-12, 17-18, and 20-38.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.



TYPE=54 PACKET CONTENTS:

The INTEGRAL_REFINED packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=54)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       trig_sub_num integers        Trigger num & mesg sernum
long         5       burst_tjd    [days]          Truncated Julian Day
long         6       burst_sod    [centi-sec]     (int)(sssss.sss *100)
long         7       burst_ra     [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long         8       burst_dec    [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long         9       det_flags    bits            Detector flag bits
long         10      inten_sigma  [centi-sigma]   Burst_intensity_sigma*100
long         11      burst_error  [arcsec]        Burst location uncertainty 
long         12      test_mpos    flag_bits       Test and Multi-Position flags
long         13      time_scale   [10^-4 sec]     Time sampling or trigger time
long         14      sc_ra        [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long         15      sc_dec       [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long         16      time_error   [10^-4 sec]     Accuracy of GRB time
long         17-18   spare[2]     integer         8 bytes for the future
long         19      misc_att     integer         S/c status & attitude
long         20-38   spare[19]    integer         76 bytes for the future
long         39      pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 51, 52, 53 , 54, or 55
that tells which type of packet this is and therefore what the contents are
(the contents change slightly between the various types).  The types are:
51 = INTEGRAL_POINTDIR, 52 = INTEGRAL_SPIACS, 53 = INTEGRAL_WAKEUP,
54 = INTEGRAL_REFINED, and 55 = INTEGRAL_OFFLINE.

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start from 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the burst_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "trig_sub_num" are the uniquely identifying serial numbers (main and sub)
assigned by the on-board INTEGRAL flight software to each trigger.
The lower 16 bits contain the trigger ("Alert") number, and the upper 16 bits
contain the Alert sub-number.  The sub-number is always 0 for type=53;
goes from 1 to N for type=54; and N+1, N+2, ... for type=55.
The Alert number started with s/n=~1 just after launch (Oct 02) and
is currently (Oct2006) in the 3500s.

The "burst_tjd" is the Truncated Julian Day of the INTEGRAL burst trigger,
eg. TJD=12640 is 01 Jan 2003.

The "burst_sod" is the UT seconds-of-day (SOD) of the INTEGRAL burst trigger.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized).

The "burst_ra" and "burst_dec" are the RA,Dec coordinates of the trigger/burst
(current epoch).  These intrinsically floating point quantities
have been multiplied by 10000 and then integerized to make an integer quantity
with units of 10^-4-degrees.

The "det_flags" ("trig_id") field contains bit_flags specifing
the trigger process.  The bit packing of the "det_flags entry in the packet
is unknown as yet (TBD at INTEGRAL MOC).

The "intensity_sigma" field contains the intensity of the burst in sigma units,
essentially a signal/noise metric.  The units are 'centi-sigma', ie the fl.pt.
sigma was multiplied by 100 and then integerized and put into the packet.

The "burst_error" is the radius of the circle that will contain on average
TBD% c.l. of the bursts.  The units are arcseconds.

The "test_mpos" field contains the following bit_flags:
Bit_Number     Item_name      Description
----------     ---------      -----------
0              multi_position 0=No or 1=Yes, the INTEGRAL Gnd s/w there was more
                                             than one likely position for this burst.
                                             In that case, the RA,Dec fields
                                             ????contain/do_not_contain????
                                             the most likely position (I'm waiting
                                             for info from the Ops team).
1-27           x_x_x_x        spare
28             spatial_coinc  0=No or 1=Yes, there was a spatial coincidence with another event
29             temporal_coinc 0=No or 1=Yes, there was a temporal coincidence with another event
30             def_not_grb    0=No or 1=Yes, it is definitely not a GRB.
31             test           0=No or 1=Yes, it is a Test_Notice.

The "time_scale" field has the sampling interval used.  For type=52 (SPIACS)
it is the lightcurve binning;  for type=53 and 54 (WAKEUP and REFINED) it is
the SIT and IMITS that triggered;  and for type=55 (OFFLINE) it is the TBD
duration of the burst.

The "sc_ra" and "sc_dec" are the RA,Dec (J2000) coordinates
to which the spacecraft points.  This is the pointing direction of the s/c.
The units are 10^04 deg, ie the values where multiplied by 10000 and 
integerized.

The "time_error" field contains the accuracy of the GRB timing information
(units are 10^-4 seconds).  This is usually the sum of the time bin size
used for the image reconstruction and the propogation delay.  It is expected
(by the INTEGRAL MOC) that type=55 packets (OFFLINE) will be corrected
for propogation delays.

The "misc_att" field contains status and attitude information on the s/c.
The exact contents and format are still TBD at the INTEGRAL MOC.

There are a total of 23 "spare" items in the packet (usually filled
with zeros).  They are located at items 11-12, 17-18, and 20-38.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.



TYPE=55 PACKET CONTENTS:

The INTEGRAL_OFFLINE packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=55)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       trig_sub_num integers        Trigger num & mesg sernum
long         5       burst_tjd    [days]          Truncated Julian Day
long         6       burst_sod    [centi-sec]     (int)(sssss.sss *100)
long         7       burst_ra     [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long         8       burst_dec    [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long         9       det_flags    bits            Detector flag bits
long         10      inten_sigma  [centi-sigma]   Burst_intensity_sigma*100
long         11      burst_error  [arcsec]        Burst location uncertainty 
long         12      test_mpos    flag_bits       Test and Multi-Position flags
long         13      time_scale   [10^-4 sec]     Time sampling or trigger time
long         14      sc_ra        [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long         15      sc_dec       [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long         16      time_error   [10^-4 sec]     Accuracy of GRB time
long         17-18   spare[2]     integer         8 bytes for the future
long         19      misc_att     integer         S/c status & attitude
long         20-38   spare[19]    integer         76 bytes for the future
long         39      pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 51, 52, 53 , 54, or 55
that tells which type of packet this is and therefore what the contents are
(the contents change slightly between the various types).  The types are:
51 = INTEGRAL_POINTDIR, 52 = INTEGRAL_SPIACS, 53 = INTEGRAL_WAKEUP,
54 = INTEGRAL_REFINED, and 55 = INTEGRAL_OFFLINE.

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start from 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the burst_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "trig_sub_num" are the uniquely identifying serial numbers (main and sub)
assigned by the on-board INTEGRAL flight software to each trigger.
The lower 16 bits contain the trigger ("Alert") number, and the upper 16 bits
contain the Alert sub-number.  The sub-number is always 0 for type=53;
goes from 1 to N for type=54; and N+1, N+2, ... for type=55.
The Alert number started with s/n=~1 just after launch (Oct 02) and
is currently (Oct2006) in the 3500s.

The "burst_tjd" is the Truncated Julian Day of the INTEGRAL burst trigger,
eg. TJD=12640 is 01 Jan 2003.

The "burst_sod" is the UT seconds-of-day (SOD) of the INTEGRAL burst trigger.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized).

The "burst_ra" and "burst_dec" are the RA,Dec coordinates of the trigger/burst
(current epoch).  These intrinsically floating point quantities
have been multiplied by 10000 and then integerized to make an integer quantity
with units of 10^-4-degrees.

The "det_flags" ("trig_id") field contains bit_flags specifing
the trigger process.  The bit packing of the "det_flags" entry in the packet
is unknown as yet (TBD at INTEGRAL MOC).

The "intensity_sigma" field contains the intensity of the burst in sigma units,
essentially a signal/noise metric.  The units are 'centi-sigma', ie the fl.pt.
sigma was multiplied by 100 and then integerized and put into the packet.

The "burst_error" is the radius of the circle that will contain on average
TBD% c.l. of the bursts.  The units are arcseconds.

The "test_mpos" field contains the following bit_flags:
Bit_Number     Item_name      Description
----------     ---------      -----------
0              multi_position 0=No or 1=Yes, the INTEGRAL Gnd s/w there was more
                                             than one likely position for this burst.
                                             In that case, the RA,Dec fields
                                             ????contain/do_not_contain????
                                             the most likely position (I'm waiting
                                             for info from the Ops team).
1-27           x_x_x_x        spare
28             spatial_coinc  0=No or 1=Yes, there was a spatial coincidence with another event
29             temporal_coinc 0=No or 1=Yes, there was a temporal coincidence with another event
30             def_not_grb    0=No or 1=Yes, it is definitely not a GRB.
31             test           0=No or 1=Yes, it is a Test_Notice.

The "time_scale" field has the sampling interval used.  For type=52 (SPIACS)
it is the lightcurve binning;  for type=53 and 54 (WAKEUP and REFINED) it is
the SIT and IMITS that triggered;  and for type=55 (OFFLINE) it is the TBD
duration of the burst.

The "sc_ra" and "sc_dec" are the RA,Dec (J2000) coordinates
to which the spacecraft points.  This is the pointing direction of the s/c.
The units are 10^04 deg, ie the values where multiplied by 10000 and 
integerized.

The "time_error" field contains the accuracy of the GRB timing information
(units are 10^-4 seconds).  This is usually the sum of the time bin size
used for the image reconstruction and the propogation delay.  It is expected
(by the INTEGRAL MOC) that type=55 packets (OFFLINE) will be corrected
for propogation delays.

The "misc_att" field contains status and attitude information on the s/c.
The exact contents and format are still TBD at the INTEGRAL MOC.

There are a total of 23 "spare" items in the packet (usually filled
with zeros).  They are located at items 11-12, 17-18, and 20-38.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.


TYPE=56 PACKET CONTENTS:

The INTEGRAL_WEAK packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=54)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       trig_sub_num integers        Trigger num & mesg sernum
long         5       burst_tjd    [days]          Truncated Julian Day
long         6       burst_sod    [centi-sec]     (int)(sssss.sss *100)
long         7       burst_ra     [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long         8       burst_dec    [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long         9       det_flags    bits            Detector flag bits
long         10      inten_sigma  [centi-sigma]   Burst_intensity_sigma*100
long         11      burst_error  [arcsec]        Burst location uncertainty 
long         12      test_mpos    flag_bits       Test and Multi-Position flags
long         13      time_scale   [10^-4 sec]     Time sampling or trigger time
long         14      sc_ra        [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long         15      sc_dec       [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long         16      time_error   [10^-4 sec]     Accuracy of GRB time
long         17-18   spare[2]     integer         8 bytes for the future
long         19      misc_att     integer         S/c status & attitude
long         20-38   spare[19]    integer         76 bytes for the future
long         39      pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 51, 52, 53 , 54, or 55
that tells which type of packet this is and therefore what the contents are
(the contents change slightly between the various types).  The types are:
51 = INTEGRAL_POINTDIR, 52 = INTEGRAL_SPIACS, 53 = INTEGRAL_WAKEUP,
54 = INTEGRAL_REFINED, 55 = INTEGRAL_OFFLINE, and 56 = INTEGRAL_WEAK.

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start from 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the burst_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "trig_sub_num" are the uniquely identifying serial numbers (main and sub)
assigned by the on-board INTEGRAL flight software to each trigger.
The lower 16 bits contain the trigger ("Alert") number, and the upper 16 bits
contain the Alert sub-number.  The sub-number is always 0 for type=53;
goes from 1 to N for type=54; and N+1, N+2, ... for type=55.
The Alert number started with s/n=~1 just after launch (Oct 02) and
is currently (Oct2006) in the 3500s.

The "burst_tjd" is the Truncated Julian Day of the INTEGRAL burst trigger,
eg. TJD=12640 is 01 Jan 2003.

The "burst_sod" is the UT seconds-of-day (SOD) of the INTEGRAL burst trigger.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized).

The "burst_ra" and "burst_dec" are the RA,Dec coordinates of the trigger/burst
(current epoch).  These intrinsically floating point quantities
have been multiplied by 10000 and then integerized to make an integer quantity
with units of 10^-4-degrees.

The "det_flags" ("trig_id") field contains bit_flags specifing
the trigger process.  The bit packing of the "det_flags" entry in the packet
is unknown as yet (TBD at INTEGRAL MOC).

The "intensity_sigma" field contains the intensity of the burst in sigma units,
essentially a signal/noise metric.  The units are 'centi-sigma', ie the fl.pt.
sigma was multiplied by 100 and then integerized and put into the packet.

The "burst_error" is the radius of the circle that will contain on average
TBD% c.l. of the bursts.  The units are arcseconds.

The "test_mpos" field contains the following bit_flags:
Bit_Number     Item_name      Description
----------     ---------      -----------
0              multi_position 0=No or 1=Yes, the INTEGRAL Gnd s/w there was more
                                             than one likely position for this burst.
                                             In that case, the RA,Dec fields
                                             ????contain/do_not_contain????
                                             the most likely position (I'm waiting
                                             for info from the Ops team).
1-27           x_x_x_x        spare
28             spatial_coinc  0=No or 1=Yes, there was a spatial coincidence with another event
29             temporal_coinc 0=No or 1=Yes, there was a temporal coincidence with another event
30             def_not_grb    0=No or 1=Yes, it is definitely not a GRB.
31             test           0=No or 1=Yes, it is a Test_Notice.

The "time_scale" field has the sampling interval used.  For type=52 (SPIACS)
it is the lightcurve binning;  for type=53 and 54 (WAKEUP and REFINED) it is
the SIT and IMITS that triggered;  and for type=55 (OFFLINE) it is the TBD
duration of the burst.

The "sc_ra" and "sc_dec" are the RA,Dec (J2000) coordinates
to which the spacecraft points.  This is the pointing direction of the s/c.
The units are 10^04 deg, ie the values where multiplied by 10000 and 
integerized.

The "time_error" field contains the accuracy of the GRB timing information
(units are 10^-4 seconds).  This is usually the sum of the time bin size
used for the image reconstruction and the propogation delay.  It is expected
(by the INTEGRAL MOC) that type=55 packets (OFFLINE) will be corrected
for propogation delays.

The "misc_att" field contains status and attitude information on the s/c.
The exact contents and format are still TBD at the INTEGRAL MOC.

There are a total of 23 "spare" items in the packet (usually filled
with zeros).  They are located at items 11-12, 17-18, and 20-38.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.




================================================================================

TYPE=57 PACKET CONTENTS:

NOTE: This Type=57 (AAVSO) is NOT YET AVAILABLE.

The AAVSO packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=57)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100)
long         4       trig_num     integer         Trigger num
long         5       event_tjd    [days]          Truncated Julian Day
long         6       event_sod    [centi-sec]     (int)(sssss.sss *100)
long         7       event_ra     [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         8       event_dec    [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         9-10    spare[2]     integer         8 bytes for the future
long         11      event_error  [0.0001-deg]    (int)(0.0 to 180.0 *10000)
long         12-17   spare[6]     integer         24 bytes for the future
long         18      trig_id      bits            Trigger identification flags
long         19      misc         bits            Misc stuff packed in here
long         20-38   spare[19]    integer         76 bytes for the future
long         39      pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 57, which tells that this packet
is of type AAVSO and therefore indicates the contents and format
for the other 39 fields.

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start from 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the event_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "trig_num" is the uniquely identifying serial number for each event.

The "event_tjd" is the Truncated Julian Day of the trigger,
eg. TJD=12640 is 01 Jan 2003.

The "event_sod" is the UT seconds-of-day (SOD) of the trigger.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized).  This is the beginning of the 'duration' search window
of the successful trigger.

The "event_ra" and "event_dec" are the RA,Dec coordinates of the event (J2000).
These intrinsically floating point quantities have been multiplied by 10,000
and then integerized to make an integer quantity with units of 0.0001-degrees.

The "event_error" is the radius of the circle that will contain on average
TBD% of the event locations.  The units are 0.0001-deg (ie the error was multiplied 
by 10000 and then integerized).

The "trigger_id" field contains bit_flags which attempt to identify
the type of transient event.
The bit packing of the "trigger_id" entry in the packet is:
Bit_Number     Item_name      Description
----------     ---------      -----------
//0              Subtype        Notice sub-type: 0=Individual, 1=Coincidence
//1              test_flag      Test-type: 0=Real, 1=Test
//2              radec_undef    RA,Dec knowledge: 0=defined, 1=Undefined (no position known)
3-4                    [not yet defined]
//5              retract        0=No or 1=Yes, it is definitely not a GRB (a Retraction)

The "misc" field contains various bit_flags.
The bit packing is:
Bit_Number     Item_name      Description
----------     ---------      -----------
//?

There are a total of 24 "spare" items in the packet (usually filled
with zeros).  They are located at items 10, 14-17, and 20-38.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.


================================================================================

TYPE=58 PACKET CONTENTS:

The MILAGRO packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=58)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       trig_num     integer         Trigger num
long         5       burst_tjd    [days]          Truncated Julian Day
long         6       burst_sod    [centi-sec]     (int)(sssss.sss *100)
long         7       burst_ra     [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         8       burst_dec    [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         9       burst_sig    [evts]          Events [counts] (0 to infinity *10000)
long         10      bkg          [0.0001-evts]   Background events (0 to infinity *10000)
long         11      burst_error  [0.0001-deg]    (int)(0.0 to 180.0 *10000)
long         12      signif       -               (int)(-log10(signif) *100)
long         13      duration     [centi-sec]     (int)(0.0 to inf *100)
long         14      ann_rate     [0.01*N/yr]     (int)(0.0 to inf *100)
long         15      zenith       [centi-deg]     (int)(0.0 to +90.0 *100)
long         16-17   spare[2]     integer         8 bytes for the future
long         18      trig_id      bits            Trigger identification flags
long         19      misc         bits            Misc stuff packed in here
long         20-38   spare[19]    integer         76 bytes for the future
long         39      pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 58, which tells that this packet
is of type MILAGRO and therefore indicates the contents and format
for the other 39 fields.

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start from 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the burst_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "trig_num" is the uniquely identifying serial number
assigned by the MILAGRO software to each trigger.

The "burst_tjd" is the Truncated Julian Day of the trigger,
eg. TJD=12640 is 01 Jan 2003.

The "burst_sod" is the UT seconds-of-day (SOD) of the trigger.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized).  This is the beginning of the 'duration' search window
of the successful trigger.

The "burst_ra" and "burst_dec" are the RA,Dec coordinates of the trigger/burst
as determined by the MILAGRO software (J2000 epoch).  These intrinsically
floating point quantities have been multiplied by 10,000 and then integerized
to make an integer quantity with units of 0.0001-degrees.

The "signal" is actually the integrated fluence of the burst
over the duration of the trigger.  The units are 'events' (ie counts).
Background events have NOT been subtracted.

The "background" is the counts seen in a 2-hr interval prior to the trigger.
This intrinsically floating point quantity have been multiplied by 10,000 and
then integerized to make an integer quantity with units of 0.0001-events.

The "burst_error" is the radius of the circle that will contain on average
TBD% of the bursts.  The units are 0.0001-deg (ie the error was multiplied 
by 10000 and then integerized).

The "signif" is the -100*log10(probability_of_false_detection).
It is only the statistical significance of the detection, or put another way,
-log10 of the probability that the source could be a natural fluctuation
within the Poisson background.   It is typically a small number (10^-12),
but since there are a large number of trials, this by itself is not an indication
of whether this is real GRB or just a chance fluctuation.  The annual_rate (see below)
will give a better handle on this aspect.  The number stored in this location is actually
the negative base_ten log of the significance multiplied by 100 and then integerized.

The "duration" is the time interval of the search window of the event_times stream
looking for an increase.  The units are centi-seconds, ie fl.pt. seconds time 100,
then integerized.

The "ann_rate" is the number of times per year that a fluctuation of this significance
at this duration interval is expected to be seen.  The units are 0.01/year
(ie the annual rate has been multiplied by 100, then integerized, and poked 
into the packet).

The "zenith" is the MILAGRO instrument zenith angle of the trigger (burst)
as determined by the MILAGRO software.  The units are centi-degrees.

The "trigger_id" field contains bit_flags which attempt to identify
the type of transient event that triggered the MILAGRO instrument.
The bit packing of the "trigger_id" entry in the packet is:
Bit_Number     Item_name      Description
----------     ---------      -----------
0              possible_grb   0=No or 1=Yes, it is a possible GRB
1              definite_grb   0=No or 1=Yes, it is definitely a GRB
2-14           x_x_x_x        spare
15             def_not_grb    0=No or 1=Yes, it is definitely not a GRB
16-31          x_x_x_x        spare

The "misc" field contains various bit_flags and the message sequence number. 
The bit packing of the "trigger_id" entry in the packet is:
Bit_Number     Item_name      Description
----------     ---------      -----------
0-3            short          Subtype: 1=Short,2=medium(TBD),3=long(TBD)
4-23           x_x_x_x        spare
24-31          seq_num        The sequence number of this message
The first 4 bits are the Notice subtype.  Currently, only the 'short' subtype
is defined, but there is provisions for other subtypes, tentatively identified
as 'medium' and 'long' timescales of trigger durations/methods.
The 'seq_num' is a sequentially assigned number for each trigger.
As ground-processing (automated and humans-in-the-loop) progresses,
successive messages may be issued.  This sequence number keeps track
of those updates.

There are a total of 21 "spare" items in the packet (usually filled
with zeros).  They are located at items 16-17, and 20-38.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.


================================================================================

TYPE=59 PACKET CONTENTS:

The KONUS_LIGHTCURVE packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=59)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       spare        integer         4 bytes for the future
long         5       trigger_tjd  [days]          Truncated Julian Day
long         6       trigger_sod  [centi-sec]     (int)(sssss.sss *100)
long         7-14    spare[8]     integer         32 bytes for the future
long         15      E_min        integer         Low end of the energy range
long         16      E_max        integer         Upper end of the energy range
long         17      hi_prec_sod  [milli-sec]     (int)(sssss.sss *1000)
long         18      trigger_id   bits            Trigger identification flags
long         19-21   spare[3]     integer         12 bytes for the future
quad_char    22-38   url[17]      chars           The URL for the lightcurve
long         39      pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 59, which tells that this packet
is of type KONUS_LC and therefore indicates the contents and format
for the other 39 fields.

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start from 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the burst_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

[There is no "trig_num" for KONUS_LC at the time the notice is sent to GCN for distribution.
It is assigned later (1-7 days later) by the KONUS Team.  It eventually appears on the Table page.]

The "trigger_tjd" is the Truncated Julian Day of the trigger,
eg. TJD=12640 is 01 Jan 2003.

The "trigger_sod" is the UT seconds-of-day (SOD) of the trigger.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized).

The "E_min" and "E_max" have the lower and upper ends of the energy range
used in making the countrate lightcurve.

The "hi_prec_sod" is a hi-precission form of the seconds-of-day (SOD) of the trigger.
The units are in milli-seconds (ie the fl.pt. seconds were multiplied by 1000
and then integerized).  (ie it the same value as "trigger_sod", just one more decimal place)

The "trigger_id" field contains bit_flags which attempt to identify
the type of event that triggered the KONUS instrument.
The bit packing of the "trigger_id" entry in the packet is:
Bit_Number     Item_name      Description
----------     ---------      -----------
//0            possible_grb   0=No or 1=Yes, it is a possible GRB
//1            definite_grb   0=No or 1=Yes, it is definitely a GRB
2-14           x_x_x_x        spare
//15           def_not_grb    0=No or 1=Yes, it is definitely not a GRB
16-28          x_x_x_x        spare
29             temporal_coinc 0=No or 1=Yes, there was a temporal coincidence with another event
30-31          x_x_x_x        spare

The "url" is just the filename portion of the URL -- the leading path
for the full URL is "http://gcn.gsfc.nasa.gov/gcn/notices_k".
This item in the packet spans 17 longwords (ie 68 bytes) and contains 
the ASCII string of the filename portion of the URL.  This string
can be up to 67 characters long, and will always be null-terminated.

There are a total of 15 "spare" items in the packet (usually filled
with zeros).  They are located at items 7-21.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.


================================================================================

TYPE=60 PACKET CONTENTS:

The SWIFT_BAT_GRB_ALERT packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=60)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       trig_obs_num integers        Trigger num & Observation num
long         5       trig_tjd     [days]          Truncated Julian Day
long         6       trig_sod     [centi-sec]     (int)(sssss.sss *100)
long         7-15    spare[9]     integer         40 bytes for the future
long         16      lon_lat      2_shorts        (int)(Longitude,Lattitude *100)
long         17      trig_index   integer         Rate_Trigger index
long         18      soln_status  bits            Type of source/trigger found
long         19      misc         bits            Misc stuff packed in here
long         20      spare        integer         4 bytes for the future
long         21      rate_signif  [centi-sigma]   (int)(sig2noise *100)
long         22-38   spare[17]    integer         68 bytes for the future
long         39      pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 60, 61, 62, 63, ...84
that tells which type of packet this is and therefore what the contents are
(the contents change between the various types).  The types are:
60 = SWIFT_BAT_GRB_ALERT, 61 = SWIFT_BAT_GRB_POSITION, 62 = SWIFT_BAT_GRB_NACK_POSITION,
63 = SWIFT_BAT_GRB_LIGHTCURVE. (Only the raw BAT Notices are listed here;
see later for the other Swift-XRT/UVOT Notices type values.)

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start from 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the trig_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "trig_obs_num" is a pair of uniquely identifying serial numbers (Trigger_num and Observation_num)
assigned by the on-board BAT flight software to each trigger.
The lower 24 bits contain the Trigger_number, and the upper 8 bits
contain the Observation_number. 
The Observation_number is always 0 since these are always (by definition)
the first observation of this new trigger.
The Trigger_number started with s/n=100,000 just after launch (Nov 2004).

The "trig_tjd" is the Truncated Julian Day of the Swift-BAT trigger,
eg. TJD=13370 is 01 Jan 2005.  (Note that at the time when this Alert Notice
is issued, it is not yet known if this trigger is due to a real GRB,
or some non-GRB cause (eg. failed or false trigger).)

The "trig_sod" is the UT seconds-of-day (SOD) of the Swift-BAT trigger.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized).  (Since this is the Alert Notice, it is not known yet
if this trigger is a real GRB or just a false trigger.)

The "lon_lat" is the Longitude and Lattitude of the s/c at the time of the BAT trigger.
This 4-byte quantity is composed of two 2-byte integers.  The high-order short
is the latitude multiplied by 100 and integerized; the low-order short
is the longitude multiplied by 100 and integerized; ie both values are in centi-degrees.

The "trig_index" is the index value (row_number in the BAT on-board flight s/w table)
of the trigger criteria that was the highest-significance successful trigger.

The "soln_status" (aka trigger_id) field contains (a) flag bits on the type of solution
that was found in the on-board image and about whether it matches anything in the
on-board source catalog, and (b) flag bits that are assigned in the GCN ground-processing.
The bit packing of the "soln_status" entry in the packet is:
BitNum  F-v-G  Item_name      Description
------  ----   ---------      -----------
0-28     -     spare          spare for future use
29       G     temporal_coinc 0=No or 1=Yes, there was a temporal coincidence with another event
30       G     test_submit    0=No or 1=Yes, this is a test submission (internal use only)
31       -     spare          spare for future use
The columns are:  "BitNum" is the 2^N location within the 4-byte word; "F-v-G" is Flight-assigned vs
Ground-assigned;  "Item_name" is the short-description name of the quantity;  and
"Description" is the longer description of the quantity.

The "misc" field contains miscellaneous flag bits.
The defined bits are:
2^22  When this bit is a 1, it indicates that this Notice was generated as a result
      of an uploaded TOO sequence.  (Swift is commanded to look at this RA,Dec location
      and this Notice was generated as a direct result of that observation.)
2^23  Not assigned.
2^24  Not assigned.
2^25  Not assigned.
2^26  Not assigned.
2^27  Not assigned.
2^28  Not assigned.
2^29  Not assigned.
2^30  When this bit is a 1, then this Notice was ground-generated; else flight-generated.
      A Gnd-generated Notice is issued anytime humans during ground operations
      have determined if this Notice should be issued.  This will typically be used
      during the early phases of the mission before the automatic real-time distribution
      capability is enabled.
2^31  If there is a CRC error in one or more of the packets that went into
      the making of this Notice, then this bit is set to 1, else 0.

The "rate_signif" is the significance level of the initial 'rate trigger'.
This is the signal/noise (ie number of sigma) of the rate-trigger detection
multiplied by 100 and then integerized (ie units are centi-sigma).

There are a total of 28 "spare" items in the packet (usually filled
with zeros).  They are located at items 7-16, 20, and 22-38.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.



TYPE=61 PACKET CONTENTS:

The SWIFT_BAT_GRB_POS_ACK packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=61)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       trig_obs_num integers        Trigger num & Observation num
long         5       burst_tjd    [days]          Truncated Julian Day
long         6       burst_sod    [centi-sec]     (int)(sssss.sss *100)
long         7       burst_ra     [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         8       burst_dec    [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         9       burst_flue   [counts]        Num events during trig window, 0 to inf
long         10      burst_ipeak  [cnts*ff]       Counts in image-plane peak, 0 to infinity
long         11      burst_error  [0.0001-deg]    (int)(0.0 to 180.0 *10000)
long         12      phi          [centi-deg]     (int)(0.0 to 359.9999 *100)
long         13      theta        [centi-deg]     (int)(0.0 to +70.0 *100)
long         14      integ_time   [4mSec]         Duration of the trigger interval, 1 to inf
long         15      spare        integer         4 bytes for the future
long         16      lon_lat      2_shorts        (int)(Longitude,Lattitude *100)
long         17      trig_index   integer         Rate_Trigger index
long         18      soln_status  bits            Type of source/trigger found
long         19      misc         bits            Misc stuff packed in here
long         20      image_signif [centi-sigma]   (int)(sig2noise *100)
long         21      rate_signif  [centi-sigma]   (int)(sig2noise *100)
long         22      bkg_flue     [counts]        Num events during the bkg interval, 0 to inf
long         23      bkg_start    [centi-sec]     (int)(sssss.sss *100)
long         24      bkg_dur      [centi-sec]     (int)(0-80,000 *100)
long         25      cat_num      integer         On-board cat match ID number
long         26-35   spare[10]    integer         40 bytes for the future
long         36      merit_0-3    integers        Merit params 0,1,2,3 (-127 to +127)
long         37      merit_4-7    integers        Merit params 4,5,6,7 (-127 to +127)
long         38      merit_8-9    integers        Merit params 8,9     (-127 to +127)
long         39      pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 60, 61, 62, 63, ...84
that tells which type of packet this is and therefore what the contents are
(the contents change between the various types).  The types are:
60 = SWIFT_BAT_GRB_ALERT, 61 = SWIFT_BAT_GRB_POSITION, 62 = SWIFT_BAT_GRB_NACK_POSITION,
63 = SWIFT_BAT_GRB_LIGHTCURVE. (Only the raw BAT Notices are listed here;
see later for the other Swift-XRT/UVOT Notices type values.)

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start from 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the burst_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "trig_obs_num" is a pair of uniquely identifying serial numbers (Trigger_num and Observation_num)
assigned by the on-board BAT flight software to each trigger.
The lower 24 bits contain the Trigger_number, and the upper 8 bits
contain the Observation_number. 
The Observation_number is always 0 since these are always (by definition)
the first observation of this new trigger (this new burst).
The Trigger_number started with s/n=100,000 just after launch (Nov 04).

The "burst_tjd" is the Truncated Julian Day of the Swift-BAT burst trigger,
eg. TJD=13370 is 01 Jan 2005.

The "burst_sod" is the UT seconds-of-day (SOD) of the Swift-BAT burst trigger.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized).  The trigger time can be off by +/- 320 msec
for events that have a trigger duration less than 64 msec.  This has to do with the way
that the flight software is optimized for execution speed for evaluating trigger criteria.

The "burst_ra" and "burst_dec" are the RA,Dec coordinates of the burst (or transient)
as determined by the BAT flight software (J2000 epoch).  These intrinsically
floating point quantities have been multiplied by 10,000 and then integerized
to make an integer quantity with units of 0.0001-degrees.

The "burst_flue" is the number of events (burst+background events)
during the 'integ_time' time interval of the successful trigger criterion.

The "burst_ipeak" is the height of the peak in the sky-image plane (ie the result
of the FFT,MaskConvolution,InvFFT calculation).  The units are counts
after a multiplicative factor of about 0.5-0.7.

The "burst_error" is the radius of the circle that will contain on average
TBD% of the bursts.  The units are 0.0001-degrees (ie the fl.pt. error was multiplied
by 10000 and then integerized).  [Initially, this value was hardwired 
at a constant value of 4arcmin (0.067deg) radius.  As the mission progressed,
this changed to 3 arcmin.  Eventually, it might become intensity-dependant.]

The "phi" and "theta" are the BAT instrument coordinates of the trigger (burst)
as determined by the BAT flight software program.  Phi is measured phi=0
along the +Y-axis of the s/c and increasing towards the -Z axis,
and theta is measured from the boresight of the BAT instrument (it would have
been better to call this the 'zenith' angle).

The "integ_time" is the duration of the sampling interval
of the successful trigger criteria.  The value is the number of 4msec ticks,
eg a value of 16 indicates a 64-msec trigger criterion.

The "lon_lat" is the Longitude and Lattitude of the s/c at the time of the BAT trigger.
This 4-byte quantity is composed of two 2-byte integers.  The high-order short
is the latitude multiplied by 100 and integerized; the low-order short
is the longitude multiplied by 100 and integerized; ie both values are in centi-degrees.

The "trig_index" is the index value (row_number in the BAT on-board flight s/w table)
of the trigger criteria that was the highest-significance successful trigger.

The "soln_status" (aka trigger_id) field contains (a) flag bits on the type of solution
that was found in the on-board image and about whether it matches anything in the
on-board source catalog, and (b) flag bits that are assigned in the GCN ground-processing.
The bit packing of the "soln_status" entry in the packet is:
BitNum  F-v-G  Item_name      Description
------  ----   ---------      -----------
0        F     point_src      0=No or 1=Yes, a point source was found
1        F     grb            0=No or 1=Yes, it is a GRB
2        F     interesting    0=No or 1=Yes, it is an interesting src (ie a flaring known src)
3        F     flt_cat_src    0=No or 1=Yes, it is in the flight catalog (else unknown src)
4        F     image_trig     0=No or 1=Yes, it is an image trigger (else rate trig)
5        G     def_not_grb    0=No or 1=Yes, it is definitely not a GRB (bad label; its really a Retraction)
6        G     uncert_grb     0=No or 1=Yes, it is probably not a GRBorTransient (hi bkg level, ie near SAA)
7        G     uncert_grb     0=No or 1=Yes, it is probably not a GRBorTransient (low image significance; less then 7.0 sigma)
8        G     gnd_cat_src    0=No or 1=Yes, it is in the ground catalog
9        G     uncert_grb     0=No or 1=Yes, it is probably not a GRBorTransient (negative bkg slope, exit SAA)
10       G     st_loss_lock   0=No or 1=Yes, StarTracker not locked so trigger probably bogus
11       G     uncert_grb     0=No or 1=Yes, it is really probably not a GRBorTransient (VERY low image significance; less than 6.5 sigma)
12       G     blk_cat_src    0=No or 1=Yes, it is in the catalog of sources to be blocked (internal use only)
13       G     near_brt_star  0=No or 1=Yes, there is a nearby bright star (this is a duplicate of the "misc" 2^13).
14       G     was_subthresh  0=No or 1=Yes, this was originally a SubThresh, but it is now converted to a real BAT_POS
15       G     onboard_rmvd   0=No or 1=Yes, this is a source that has purposefully been removed from on-board catalog
16       G     nearby_gal     0=No or 1=Yes, this matched a Nearby_Galaxy in the on-board catalog
17-27    -     spare          spare for future use
28       G     spatial_coinc  0=No or 1=Yes, there was a spatial coincidence with another event
29       G     temporal_coinc 0=No or 1=Yes, there was a temporal coincidence with another event
30       G     test_submit    0=No or 1=Yes, this is a test submission (internal use only)
31       -     spare          spare for future use
The columns are:  "BitNum" is the 2^N location within the 4-byte word; "F-v-G" is Flight-assigned vs
Ground-assigned;  "Item_name" is the short-description name of the quantity;  and
"Description" is the longer description of the quantity.

The "misc" field contains miscellaneous flag bits.
The defined bits are:
2^0-11  Not assigned.
2^11  When this bit is a 1, it indicates that 1 (or more) of the RA,Dec,Roll,Theta,Phi values
      were out of valid range (eg RA was 361 deg).
2^12  When this bit is a 1, it indicated the theta value was less than 10arcmin (ie Swift was already pointed at this source).
2^13  When this bit is a 1, it indicates that this position is near (less than 0.3deg)  a bright star (mag less than 6.5).
2^14  When this bit is a 1, it indicates that the distance between the position and the center of the galaxy
      is less than the sum of the position error radius and the galaxy radius AND that the position error radius
      is smaller than the galaxy radius.  This position is inside the (circularized) NGC galaxy (or nearly in).
2^15  When this bit is a 1, it indicates that the distance between the position and the center of the galaxy
      is less than the sum of the position error radius and the galaxy radius AND that the galaxy radius
      is smaller than the position error radius.  This galaxy is inside the position error circle (or nearly in).
2^16-19  Not assigned.
2^20  When this bit is a 1, it indicates that this was originally a SubThresh, but it is now converted to a real BAT_POS.
2^21  When this bit is a 1, it indicates that this is an Image-trigger AND that it occurred
      during a time when the StarTracker had a Loss-of-Lock.
2^22  When this bit is a 1, it indicates that this Notice was generated as a result
      of an uploaded TOO sequence.  (Swift is commanded to look at this RA,Dec location
      and this Notice was generated as a direct result of that observation.)
2^23  When this bit is a 1, it means the Trig_Time was zero and so the s/c system clock
      in the TDRSS Message SecondaryHeader was used to fill the Burst_TJD and Burst_SOD fields.
      (This 0-Trig_Time scenario occurs during a ground-commanded GRB_TOO.)
2^24  When this bit is a 1, it indicates that the TDRSS Message was received on the ground
      more than 60 sec after the BAT Trigger time.  The most likely scenario for this kind of transmission delay
      is that the BAT Trigger occured during a Malindi downlink pass (when TDRSS messages are blocked
      from immediate transmission; they are buffered onboard until the end of the Malindi Pass).
2^25  This is an updated position.  This is a position notice that changes/improves
      the position specified in the original BAT_Position Notice.
2^26  Not assigned.
2^27  Not assigned.
2^28  When this bit is a 1, then this Notice was derived from a SERS message; else flight-generated.
      If none of the regular position-containing messages are received via the real-time
      TDRSS comm channel (eg BAT_Pos_Ack, BAT_LC, etc), then the SERS messages
      (which come from pipe-line processing of the Malindi playback downlinked data)
      will be used to generate a reconstituted BAT_Pos_Ack Notice.  But since
      the SERS message only contains a subset of the information is a regular TDRSS
      messages, these SERS-reconstituted Notices will have (tbd) fields undefined.
      These undefined fields are: tbd, tbd, tbd....
2^29  When this bit is a 1, then this Notice was reconstituted; else flight-generated.
      If a BAT_LC is received from the s/c before a BAT_Pos is received,
      then bits and pieces from other Notices will be used to generate a reconstituted
      BAT_Pos Notice and it will be issued (with this flag bit set to 1).
      But since the LC message only contains a subset of the information in a regular
      TDRSS BAT_POS messages, these LC-reconstituted Notices will have (tbd) fields undefined.
      These undefined fields are: tbd, tbd, tbd....
2^30  When this bit is a 1, then this Notice was ground-generated; else flight-generated.
      A Gnd-generated Notice is issued anytime humans during ground operations
      have determined if this Notice should be issued.  This will typically be used
      during the early phases of the mission before the automatic real-time distribution
      capability is enabled.  Once the burst position has been validated by humans,
      a BAT_Pos Notice will be issued (with this flag bit set to 1).
      After the Verification phase, there may be rare occaisions when a BAT_POS notice
      is (re)created by human-processing and will be distributed (with this bit set to 1). 
2^31  If there is a CRC error in one or more of the packets that went into
      the making of this Notice, then this bit is set to 1, else 0.

The "image_signif" is the significance level of the peak in the FFT image plane.
This is the signal/noise (ie number of sigma) of the grb peak detection
multiplied by 100 and then integerized (ie units are centi-sigma).

The "rate_signif" is the significance level of the initial 'rate trigger'.
This is the signal/noise (ie number of sigma) of the rate-trigger detection
multiplied by 100 and then integerized (ie units are centi-sigma).

The "bkg_flue" is the total number of events in the time interval used
for the background subtraction of the detector_rate_map (used for the image calc).
The units are counts.

The "bkg_start" is the starting time of the background interval.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized).  Note that not all trigger criteria have a background interval
specified, so this "bkg_start" value may be zero for some bursts.

The "bkg_dur" is the duration of the background interval.
The units are in centi-seconds.

The "cat_num" is the catalog number in the on-board catalog for sources
that match a known source in the on-board catalog.  (Typically it is a number
in the low 10,000s, 12,000s, or 15,000s.)

The "merit_0-3", "merit_4-7", and "merit_8-9" are the 10 merit parameters
for this burst.  Their values range from -127 to +127.  These 'merit params'
are different than the 'merit_value' that is in the SWIFT_FOM_OBS and SWIFT_SC_SLEW
messages.  There are 4, 4, & 2 values packed into these 3 longs.  The packing order
is "0" in the lowest_address byte in the long, "1" is in the next_lowest, etc.
The definition of the 10 parameters is:
Param   Description
0       Flag bit indicating GRB or not (1 or 0, resp).
1       Flag bit indicating Transient or not (1 or 0, resp); unknown src with T_trig>64sec.
2       Flag bit indicating Known_src or not (1 or 0, resp); Known_src matches something in on-board catalog.
3       Trigger duration (log2(t_trig/1024msec)).
4       Trigger energy range (0=15-25,1=15-50,2=25-100,3=50-350).
5       Image significance
6       Is it observable (-1 if w/in 30deg of Moon; -5, 20deg Moon; -100, 45deg Sun).
7       Flag bit indicating in a confused or obscurred region (ie Gal Center or Plane).
8       Sun distance (-100*cos(sun_angular_dist)).
9       An offset param to bias for/against PPTs and TOOs wrt ATs (norm param).

There are a total of 13 "spare" items in the packet (usually filled
with zeros).  They are located at items 15-16 and 25-35.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.



TYPE=62 PACKET CONTENTS:

The SWIFT_BAT_GRB_POS_NACK packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=62)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       trig_obs_num integers        Trigger num & Observation num
long         5       trig_tjd     [days]          Truncated Julian Day
long         6       trig_sod     [centi-sec]     (int)(sssss.sss *100)
long         7-13    spare[7]     integer         28 bytes for the future
long         14      integ_time   [4mSec]         Duration of the trigger interval, 1 to inf
long         15      spare        integer         4 bytes for the future
long         16      lon_lat      2_shorts        (int)(Longitude,Lattitude *100)
long         17      trig_index   integer         Rate_Trigger index
long         18      soln_status  bits            Type of source found
long         19      misc         bits            Misc stuff packed in here
long         20      image_signif [centi-sigma]   (int)(sig2noise *100)
long         21      rate_signif  [centi-sigma]   (int)(sig2noise *100)
long         22      bkg_flue     [counts]        Num events during the bkg interval, 0 to inf
long         23      bkg_start    [centi-sec]     (int)(sssss.sss *100)
long         24      bkg_dur      [centi-sec]     (int)(0-80,000 *100)
long         25-35   spare[11]    integer         44 bytes for the future
long         36      merit_0-3    integers        Merit params 0,1,2,3 (-127 to +127)
long         37      merit_4-7    integers        Merit params 4,5,6,7 (-127 to +127)
long         38      merit_8-9    integers        Merit params 8,9     (-127 to +127)
long         39      pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 60, 61, 62, 63, ...84
that tells which type of packet this is and therefore what the contents are
(the contents change between the various types).  The types are:
60 = SWIFT_BAT_GRB_ALERT, 61 = SWIFT_BAT_GRB_POSITION, 62 = SWIFT_BAT_GRB_NACK_POSITION,
63 = SWIFT_BAT_GRB_LIGHTCURVE. (Only the raw BAT Notices are listed here;
see later for the other Swift-XRT/UVOT Notices type values.)

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start from 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the burst_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "trig_obs_num" is a pair of uniquely identifying serial numbers (Trigger_num and Observation_num)
assigned by the on-board BAT flight software to each trigger.
The lower 24 bits contain the Trigger_number, and the upper 8 bits
contain the Observation_number. 
The Observation_number is always 0 since these are always (by definition)
the first observation of this new trigger.
The Trigger_number started with s/n=100,000 just after launch (Nov 04).

The "trig_tjd" is the Truncated Julian Day of the Swift-BAT trigger,
eg. TJD=13370 is 01 Jan 2005.

The "trig_sod" is the UT seconds-of-day (SOD) of the Swift-BAT trigger.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized).

The "integ_time" is the duration of the sampling interval
of the successful trigger criteria.  The value is the number of 4msec ticks,
eg a value of 16 indicates a 64-msec trigger criterion.

The "lon_lat" is the Longitude and Lattitude of the s/c at the time of the BAT trigger.
This 4-byte quantity is composed of two 2-byte integers.  The high-order short
is the latitude multiplied by 100 and integerized; the low-order short
is the longitude multiplied by 100 and integerized; ie both values are in centi-degrees.

The "trig_index" is the index value (row_number in the BAT on-board flight s/w table)
of the trigger criteria that was the highest-significance successful trigger.

The "soln_status" field contains flag bits on the type of solution that would
found in the FFT sky image and about whether it matches anything in the
on-board source catalog.
The bit packing of the "soln_status" entry in the packet is:
BitNum  F-v-G  Item_name      Description
------  ----   ---------      -----------
0        F     point_src      0=No or 1=Yes, a point source was found
1        F     grb            0=No or 1=Yes, it is a GRB
2        F     interesting    0=No or 1=Yes, it is an interesting src (ie a flaring known src)
3        F     flt_cat_src    0=No or 1=Yes, it is in the flight catalog (else unknown src)
4        F     image_trig     0=No or 1=Yes, it is an image trigger (else rate trig)
The columns are:  "BitNum" is the 2^N location within the 4-byte word; "F-v-G" is Flight-assigned vs
Ground-assigned;  "Item_name" is the short-description name of the quantity;  and
"Description" is the longer description of the quantity.

The "misc" field contains miscellaneous flag bits.
The defined bits are:
2^0-10  Not assigned.
2^11  When this bit is a 1, it indicates that 1 (or more) of the RA,Dec,Roll,Theta,Phi values
      were out of valid range (eg RA was 361 deg).
2^12-21  Not assigned.
2^22  When this bit is a 1, it indicates that this Notice was generated as a result
      of an uploaded TOO sequence.  (Swift is commanded to look at this RA,Dec location
      and this Notice was generated as a direct result of that observation.)
2^23  Not assigned.
2^24  Not assigned.
2^25  Not assigned.
2^26  Not assigned.
2^27  Not assigned.
2^28  When this bit is a 1, then this Notice was derived from a SERS message; else flight-generated.
      If the regular BAT_POS_NACK is not received via the real-time TDRSS comm channel,
      then the SERS message (which comes from pipe-line processing of the Malindi
      playback downlinked data) will be used to generate a reconstituted BAT_Pos_Nack Notice.
      But since the SERS message only contains a subset of the information is a regular TDRSS
      message, these SERS-reconstituted Notices will have (tbd) fields undefined.
      These undefined fields are: tbd, tbd, tbd....
2^29  Not assigned.
2^30  When this bit is a 1, then this Notice was ground-generated; else flight-generated.
      A Gnd-generated Notice is issued anytime humans during ground operations
      have determined if the Notice should be issued.  This will typically be used
      during the early phases of the mission before the automatic real-time distribution
      capability is enabled.  Once the trigger has been determined to be non-GRB by humans,
      a BAT_Pos_Nack Notice will be issued (with this flag bit set to 1).
2^31  If there is a CRC error in one or more of the packets that went into
      the making of this Notice, then this bit is set to 1, else 0.

The "image_signif" is the significance level of the peak in the FFT image plane.
This is the signal/noise (ie number of sigma) of the grb peak detection
multiplied by 100 and then integerized (ie units are centi-sigma).
This field is always zero for BAT_Pos_Nack packets.

The "rate_signif" is the significance level of the initial 'rate trigger'.
This is the signal/noise (ie number of sigma) of the rate-trigger detection
multiplied by 100 and then integerized (ie units are centi-sigma).
This field is always zero for BAT_Pos_Nack packets.

The "bkg_flue" is the total number of events in the time interval used
for the background subtraction of the detector_rate_map (used for the image calc).
The units are counts.
This field is always zero for BAT_Pos_Nack packets.

The "bkg_start" is the starting time of the background interval.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized).
This field is always zero for BAT_Pos_Nack packets.

The "bkg_dur" is the duration of the background interval.
The units are in centi-seconds.
This field is always zero for BAT_Pos_Nack packets.

The "merit_0-3", "merit_4-7", and "merit_8-9" are the 10 merit parameters
for this burst.  Their values range from -127 to +127.  These 'merit params'
are different than the 'merit_value' that is in the SWIFT_FOM_OBS and SWIFT_SC_SLEW
messages.  There are 4, 4, & 2 values packed into these 3 longs.  The packing order
is "0" in the lowest_address byte in the long, "1" is in the next_lowest, etc.
These fields are always zero for BAT_Pos_Nack packets.

There are a total of 20 "spare" items in the packet (usually filled
with zeros).  They are located at items 7-13, 15-16, and 25-35.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.


TYPE=63 PACKET CONTENTS:

The SWIFT_BAT_GRB_LC packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=63)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       trig_obs_num integers        Trigger num & Observation num
long         5       burst_tjd    [days]          Truncated Julian Day
long         6       burst_sod    [centi-sec]     (int)(sssss.sss *100)
long         7       burst_ra     [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         8       burst_dec    [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         9-11    spare[3]     integer         12 bytes for the future
long         12      phi          [centi-deg]     (int)(0.0 to 359.9999 *100)
long         13      theta        [centi-deg]     (int)(0.0 to +70.0 *100)
long         14      delta_time   [centi-sec]     (int)(ssss.sss *100)
long         15      integ_time   [4mSec]         Duration of the trigger interval, 1 to inf
long         16      lon_lat      2_shorts        (int)(Longitude,Lattitude *100)
long         17      trig_index   integer         Rate_Trigger index
long         18      soln_status  bits            Type of source/trigger found
long         19      misc         bits            Misc stuff packed in here
long         20      image_signif [centi-sigma]   (int)(sig2noise *100)
long         21      rate_signif  [centi-sigma]   (int)(sig2noise *100)
quad_char    22-38   url[17]      chars           The URL for the lightcurve
long         39      pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 60, 61, 62, 63, ...84
that tells which type of packet this is and therefore what the contents are
(the contents change between the various types).  The types are:
60 = SWIFT_BAT_GRB_ALERT, 61 = SWIFT_BAT_GRB_POSITION, 62 = SWIFT_BAT_GRB_NACK_POSITION,
63 = SWIFT_BAT_GRB_LIGHTCURVE. (Only the raw BAT Notices are listed here;
see later for the other Swift-XRT/UVOT Notices type values.)

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start from 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the burst_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "trig_obs_num" is a pair of uniquely identifying serial numbers (Trigger_num and Observation_num)
assigned by the on-board BAT flight software to each trigger.
The lower 24 bits contain the Trigger_number, and the upper 8 bits
contain the Observation_number. 
The Observation_number is always 0 since these are always (by definition)
the first observation of this new trigger (this new burst).
The Trigger_number started with s/n=100,000 just after launch (Nov 04).

The "burst_tjd" is the Truncated Julian Day of the Swift-BAT burst trigger,
eg. TJD=13370 is 01 Jan 2005.

The "burst_sod" is the UT seconds-of-day (SOD) of the Swift-BAT burst trigger.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized).

The "burst_ra" and "burst_dec" are the RA,Dec coordinates of the burst (or transient)
as determined by the BAT flight software (J2000 epoch).  These will always be
the same values as in the SWIFT_BAT_POS_ACK packet.  These intrinsically
floating point quantities have been multiplied by 10,000 and then integerized
to make an integer quantity with units of 0.0001-degrees.

The "phi" and "theta" are the BAT instrument coordinates of the trigger (burst)
as determined by the BAT flight software program.  Phi is measured phi=0
along the +Y-axis of the s/c and increasing towards the -Z axis,
and theta is measured from the boresight of the BAT instrument.

The "delta_time" is the difference between the first sample in the lightcurve
and the Trigger_time.  The units are centi-seconds (although they will always be whole sec).
This will almost always be -37.00 sec; only when packets are lost will this be some
other value.

The "integ_time" is the duration of the sampling interval
of the successful trigger criteria. This particular field is copied from the BAT_POS
notice.  The value is the number of 4msec ticks,
eg a value of 16 indicates a 64-msec trigger criterion.
(This 16-th 4-byte field is only defined/valid for packets if the 2^24 bit is set
in the 'misc' field.)

The "lon_lat" is the Longitude and Lattitude of the s/c at the time of the BAT trigger.
This 4-byte quantity is composed of two 2-byte integers.  The high-order short
is the latitude multiplied by 100 and integerized; the low-order short
is the longitude multiplied by 100 and integerized; ie both values are in centi-degrees.

The "trig_index" is the index value (row_number in the BAT on-board flight s/w table)
of the trigger criteria that was the highest-significance successful trigger.

The "soln_status" (aka trigger_id) field contains flag bits copied from the BAT_POS_ACK
ie (a) flag bits on the type of solution that was found in the on-board image and
about whether it matches anything in the on-board source catalog, and
(b) flag bits that are assigned in the GCN ground-processing.
(This 18-th 4-byte field is only defined/valid for packets (a) if the 2^24 bit is set
in the 'misc' field,  and (b) if it was generated after 03 May 05 (the time
when the distribution order of the BAT_POS and FOM_OBS packets was changed
to allow for the filling of this field and to allow for the filtering of FOM_OBS
based on these trigger_id bits).)
The bit packing of the "soln_status" entry in the packet is:
BitNum  F-v-G  Item_name      Description
------  ----   ---------      -----------
0        F     point_src      0=No or 1=Yes, a point source was found
1        F     grb            0=No or 1=Yes, it is a GRB
2        F     interesting    0=No or 1=Yes, it is an interesting src (ie a flaring known src)
3        F     flt_cat_src    0=No or 1=Yes, it is in the flight catalog (else unknown src)
4        F     image_trig     0=No or 1=Yes, it is an image trigger (else rate trig)
5        G     def_not_grb    0=No or 1=Yes, it is definitely not a GRB (bad label; its really a Retraction)
6        G     uncert_grb     0=No or 1=Yes, it is probably not a GRBorTransient (hi bkg level, ie near SAA)
7        G     uncert_grb     0=No or 1=Yes, it is probably not a GRBorTransient (low image significance; less then 7.0 sigma)
8        G     gnd_cat_src    0=No or 1=Yes, it is in the ground catalog
9        G     uncert_grb     0=No or 1=Yes, it is probably not a GRBorTransient (negative bkg slope, exit SAA)
10       G     st_loss_lock   0=No or 1=Yes, StarTracker not locked so trigger probably bogus
11       G     uncert_grb     0=No or 1=Yes, it is really probably not a GRBorTransient (VERY low image significance; less than 6.5 sigma)
12       G     blk_cat_src    0=No or 1=Yes, it is in the catalog of sources to be blocked (internal use only)
13       G     near_brt_star  0=No or 1=Yes, there is a nearby bright star (this is a duplicate of the "misc" 2^13).
14       -     spare          spare for future use
15       G     onboard_rmvd   0=No or 1=Yes, this is a source that has purposefully been removed from on-board catalog
16       G     nearby_gal     0=No or 1=Yes, this matched a Nearby_Galaxy in the on-board catalog
17-31    -     spare          spare for future use
The columns are:  "BitNum" is the 2^N location within the 4-byte word; "F-v-G" is Flight-assigned vs
Ground-assigned;  "Item_name" is the short-description name of the quantity;  and
"Description" is the longer description of the quantity.

The "misc" field contains miscellaneous flag bits.
The defined bits are:
2^0   When this bit is a 1, it indicates that the BAT_Pos TDRSS message was not received for this TrigNum.
2^1-10  Not assigned.
2^11  When this bit is a 1, it indicates that 1 (or more) of the RA,Dec,Roll,Theta,Phi values
      were out of valid range (eg RA was 361 deg).
2^12  Not assigned.
2^13  When this bit is a 1, it indicates that this position is near a bright star (mag less than 6.5).
2^14  When this bit is a 1, it indicates that the distance between the position and the center of the galaxy
      is less than the sum of the position error radius and the galaxy radius AND that the position error radius
      is smaller than the galaxy radius.  This position is inside the (circularized) NGC galaxy (or nearly in).
2^15  When this bit is a 1, it indicates that the distance between the position and the center of the galaxy
      is less than the sum of the position error radius and the galaxy radius AND that the galaxy radius
      is smaller than the position error radius.  This galaxy is inside the position error circle (or nearly in).
2^16-19  Not assigned.
2^20  When this bit is a 1, it indicates that this was originally a SubThresh, but it is now converted to a real BAT_LC.
2^21  When this bit is a 1, it indicates that this is an Image-trigger AND that it occurred
      during a time when the StarTracker had a Loss-of-Lock.
2^22  When this bit is a 1, it indicates that this Notice was generated as a result
      of an uploaded TOO sequence.  (Swift is commanded to look at this RA,Dec location
      and this Notice was generated as a direct result of that observation.)
2^23  This includes "extended" data packets.
2^24  This bit is set if the 'soln_status' field was copied to here from BAT_POS_ACK.
2^25  When this bit is a 1, then this Notice was generated with the 3rd of 3 pkts missing.
2^26  When this bit is a 1, then this Notice was generated with the 2nd of 3 pkts missing.
2^27  When this bit is a 1, then this Notice was generated with the 1st of 3 pkts missing.
2^28  Not assigned.
2^29  This bit is set if this LightCurve Noticed was forced-out via the watchdog timeout.
2^30  When this bit is a 1, then this Notice was ground-generated; else flight-generated.
      This will always 0 for the 'raw' form of this Notice (type=63) and
      1 for the ground-processed form (type=76) or the imported-form of type=63.
2^31  If there is a CRC error in one or more of the packets that went into
      the making of this Notice, then this bit is set to 1, else 0.

The "image_signif" is the significance level of the peak in the FFT image plane.
This is the signal/noise (ie number of sigma) of the grb peak detection
multiplied by 100 and then integerized (ie units are centi-sigma).
This field was copied from the BAT_POS notice.

The "rate_signif" is the significance level of the initial 'rate trigger'.
This is the signal/noise (ie number of sigma) of the rate-trigger detection
multiplied by 100 and then integerized (ie units are centi-sigma).
This field was copied from the BAT_POS notice.

The "url" is just the filename portion of the URL -- the leading path
for the full URL is "http://gcn.gsfc.nasa.gov/gcn/notices_s/".
This item in the packet spans 17 longwords (ie 68 bytes) and contains 
the ASCII string of the filename portion of the URL.  This string
can be up to 67 characters long, and will always be null-terminated.

There are a total of 6 "spare" items in the packet (usually filled
with zeros).  They are located at items 9-11, 15-16 and 18.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.


TYPE=64 PACKET CONTENTS:   NOT AVAILABLE TO THE PUBLIC

The SWIFT_BAT_SCALED_MAP packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=64)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       trig_obs_num integers        Trigger num & Observation num
long         5       map_tjd      [days]          Truncated Julian Day
long         6       map_sod      [centi-sec]     (int)(sssss.sss *100)
long         7       point_ra     [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         8       point_dec    [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         9-13    spare[5]     integer         20 bytes for the future
long         14      foregnd_dur  [mSec]          Duration of the foreground interval, 1 to inf
long         15      integ_time   [4mSec]         Duration of the trigger interval, 1 to inf
long         16      lon_lat      2_shorts        (int)(Longitude,Lattitude *100)
long         17      trig_index   integer         Rate_Trigger index [not assigned]
long         18      soln_status  bits            Type of source/trigger found
long         19      misc         bits            Misc stuff packed in here
long         20      image_signif [centi-sigma]   (int)(sig2noise *100)
long         21      rate_signif  [centi-sigma]   (int)(sig2noise *100)
quad_char    22-38   url[17]      chars           The URL for the ScaledMap
long         39      pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 60, 61, 62, 63, ...84
that tells which type of packet this is and therefore what the contents are
(the contents change between the various types).

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start from 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the burst_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "trig_obs_num" is a pair of uniquely identifying serial numbers (Trigger_num and Observation_num)
assigned by the on-board BAT flight software to each trigger.
The lower 24 bits contain the Trigger_number, and the upper 8 bits
contain the Observation_number. 
The Observation_number is always 0 since these are always (by definition)
the first observation of this new trigger.
The Trigger_number started with s/n=100,000 just after launch (Nov 04).

The "map_tjd" is the Truncated Julian Day when the map was generated.
eg. TJD=13370 is 01 Jan 2005.

The "map_sod" is the UT seconds-of-day (SOD) when the map was generated.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized).

The "point_ra" and "point_dec" are the RA,Dec coordinates of the pointing direction
of the s/c at the start/stop? of the integration interval for this map (J2000).
These intrinsically floating point quantities have been multiplied
by 10,000 and then integerized to make an integer quantity with units of 0.0001-degrees.

The "foregnd_dur" is the duration of the sampling interval
of the successful trigger criteria.  This field is native to the ScaledMap notice
(unlike the next field which is copied from BAT_POS).
The units are milliseconds.

The "integ_time" is the duration of the sampling interval
of the successful trigger criteria.  This particular field is copied from the BAT_POS
notice.  The value is the number of 4msec ticks,
eg a value of 16 indicates a 64-msec trigger criterion.
(This 16-th 4-byte field is only defined/valid if the 2^24 bit is set
in the 'misc' field.)

The "lon_lat" is the Longitude and Lattitude of the s/c at the time of the BAT trigger.
This 4-byte quantity is composed of two 2-byte integers.  The high-order short
is the latitude multiplied by 100 and integerized; the low-order short
is the longitude multiplied by 100 and integerized; ie both values are in centi-degrees.

The "trig_index" is the index value (row_number in the BAT on-board flight s/w table)
of the trigger criteria that was the highest-significance successful trigger.
This field is not yet (if ever) assigned.

The "soln_status" (aka trigger_id) field contains flag bits copied from the BAT_POS_ACK
ie (a) flag bits on the type of solution that was found in the on-board image and
about whether it matches anything in the on-board source catalog, and
(b) flag bits that are assigned in the GCN ground-processing.
(This 18-th 4-byte field is only defined/valid for packets (a) if the 2^24 bit is set
in the 'misc' field,  and (b) if it was generated after 03 May 05 (the time
when the distribution order of the BAT_POS and FOM_OBS packets was changed
to allow for the filling of this field and to allow for the filtering of FOM_OBS
based on these trigger_id bits).)
The bit packing of the "soln_status" entry in the packet is:
BitNum  F-v-G  Item_name      Description
------  ----   ---------      -----------
0        F     point_src      0=No or 1=Yes, a point source was found
1        F     grb            0=No or 1=Yes, it is a GRB
2        F     interesting    0=No or 1=Yes, it is an interesting src (ie a flaring known src)
3        F     flt_cat_src    0=No or 1=Yes, it is in the flight catalog (else unknown src)
4        F     image_trig     0=No or 1=Yes, it is an image trigger (else rate trig)
5        G     def_not_grb    0=No or 1=Yes, it is definitely not a GRB (bad label; its really a Retraction)
6        G     uncert_grb     0=No or 1=Yes, it is probably not a GRBorTransient (hi bkg level, ie near SAA)
7        G     uncert_grb     0=No or 1=Yes, it is probably not a GRBorTransient (low image significance; less then 7.0 sigma)
8        G     gnd_cat_src    0=No or 1=Yes, it is in the ground catalog
9        G     uncert_grb     0=No or 1=Yes, it is probably not a GRBorTransient (negative bkg slope, exit SAA)
10       G     st_loss_lock   0=No or 1=Yes, StarTracker not locked so trigger probably bogus
11       G     uncert_grb     0=No or 1=Yes, it is really probably not a GRBorTransient (VERY low image significance; less than 6.5 sigma)
12       G     blk_cat_src    0=No or 1=Yes, it is in the catalog of sources to be blocked (internal use only)
13       G     near_brt_star  0=No or 1=Yes, there is a nearby bright star (mag less than 6.5).
14       -     spare          spare for future use
15       G     onboard_rmvd   0=No or 1=Yes, this is a source that has purposefully been removed from on-board catalog
16       G     nearby_gal     0=No or 1=Yes, this matched a Nearby_Galaxy in the on-board catalog
17-31    -     spare          spare for future use
The columns are:  "BitNum" is the 2^N location within the 4-byte word; "F-v-G" is Flight-assigned vs
Ground-assigned;  "Item_name" is the short-description name of the quantity;  and
"Description" is the longer description of the quantity.

The "misc" field contains miscellaneous flag bits.
The defined bits are:
2^0   When this bit is a 1, it indicates that the BAT_Pos TDRSS message was not received for this TrigNum.
2^1-10  Not assigned.
2^11  When this bit is a 1, it indicates that 1 (or more) of the RA,Dec,Roll,Theta,Phi values
      were out of valid range (eg RA was 361 deg).
2^12  Not assigned.
2^13  When this bit is a 1, it indicates that this position is near a bright star (mag less than 6.5).
2^14  When this bit is a 1, it indicates that the distance between the BAT_Pos position and the center of the galaxy
      is less than the sum of the position error radius and the galaxy radius AND that the position error radius
      is smaller than the galaxy radius.  This position is inside the (circularized) NGC galaxy (or nearly in).
2^15  When this bit is a 1, it indicates that the distance between the BAT_Pos position and the center of the galaxy
      is less than the sum of the position error radius and the galaxy radius AND that the galaxy radius
      is smaller than the position error radius.  This galaxy is inside the position error circle (or nearly in).
2^16-19  Not assigned.
2^20  When this bit is a 1, it indicates that this was originally a SubThresh, but it is now converted to a real BAT_SM.
2^21  When this bit is a 1, it indicates that this is an Image-trigger AND that it occurred
      during a time when the StarTracker had a Loss-of-Lock.
2^22  When this bit is a 1, it indicates that this Notice was generated as a result
      of an uploaded TOO sequence.  (Swift is commanded to look at this RA,Dec location
      and this Notice was generated as a direct result of that observation.)
2^23  Not assigned.
2^24  This bit is set if the 'soln_status' field was copied to here from BAT_POS_ACK.
2^25  Not assigned.
2^26  Not assigned.
2^27  Not assigned.
2^28  Not assigned.
2^29  This bit is set if this Scaled_Map Noticed was forced-out via the watchdog timeout.
2^30  When this bit is a 1, then this Notice was ground-generated; else flight-generated.
2^31  If there is a CRC error in one or more of the packets that went into
      the making of this Notice, then this bit is set to 1, else 0.

The "image_signif" is the significance level of the peak in the FFT image plane.
This is the signal/noise (ie number of sigma) of the grb peak detection
multiplied by 100 and then integerized (ie units are centi-sigma).
This field is copied from the BAT_Position Notice.

The "rate_signif" is the significance level of the initial 'rate trigger'.
This is the signal/noise (ie number of sigma) of the rate-trigger detection
multiplied by 100 and then integerized (ie units are centi-sigma).
This field is copied from the BAT_Position Notice.

The "url" is just the filename portion of the URL -- the leading path
for the full URL is "http://gcn.gsfc.nasa.gov/gcn/notices_s/".
This item in the packet spans 17 longwords (ie 68 bytes) and contains 
the ASCII string of the filename portion of the URL.  This string
can be up to 67 characters long, and will always be null-terminated.

There are a total of 6 "spare" items in the packet (usually filled
with zeros).  They are located at items 9-13 and 16.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.



TYPE=65 (and 46) PACKET CONTENTS:

The SWIFT_FOM_OBS (and SWIFT_TOO_FOM) packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=65 or =46)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       trig_obs_num integers        Trigger num & Observation num
long         5       burst_tjd    [days]          Truncated Julian Day
long         6       burst_sod    [centi-sec]     (int)(sssss.sss *100)
long         7       burst_ra     [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         8       burst_dec    [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         9       roll         [0.0001-deg]    (int)(0.0 to 360.0 *10000)
long         10      lon_lat      2_shorts        (int)(Longitude,Lattitude *100)
long         11-14   spare[4]     integer         16 bytes for the future
long         15      integ_time   [4mSec]         Duration of the trigger interval, 1 to inf
long         16      soln_status  bits            Type of source found
long         17      trig_index   integer         Rate_Trigger index
long         18      AT_slewFlags bits            Two flag bits packed here
long         19      misc         bits            Misc stuff packed in here
long         20      image_signif [centi-sigma]   (int)(sig2noise *100)
long         21      rate_signif  [centi-sigma]   (int)(sig2noise *100)
long         22-37   spare[16]    integer         64 bytes for the future
long         38      merit        [centi-merit]   (int)(Merit_value *100)
long         39      pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 60, 61, 62, 63, ...84
that tells which type of packet this is and therefore what the contents are
(the contents change between the various types).

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start from 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the burst_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "trig_obs_num" is a pair of uniquely identifying serial numbers (Trigger_num and Observation_num)
assigned by the on-board BAT flight software to each trigger.
The lower 24 bits contain the Trigger_number, and the upper 8 bits
contain the Observation_number. 
The Observation_number is always 0 since these are always (by definition)
the first observation of this new trigger.
The Trigger_number started with s/n=100,000 just after launch (Nov 04).

The "burst_tjd" is the Truncated Julian Day of the Swift-BAT burst trigger,
eg. TJD=13370 is 01 Jan 2005.

The "burst_sod" is the UT seconds-of-day (SOD) of the Swift-BAT burst trigger.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized).

The "burst_ra" and "burst_dec" are the RA,Dec coordinates of the burst (or transient)
as determined by the BAT flight software (J2000 epoch).  These intrinsically
floating point quantities have been multiplied by 10,000 and then integerized
to make an integer quantity with units of 0.0001-degrees.

The "roll" of the spacecraft (after this requested slew).  This intrinsically
floating point quantity have been multiplied by 10,000 and then integerized
to make an integer quantity with units of 0.0001-degrees.

The "lon_lat" is the Longitude and Lattitude of the s/c at the time of the BAT trigger.
This 4-byte quantity is composed of two 2-byte integers.  The high-order short
is the latitude multiplied by 100 and integerized; the low-order short
is the longitude multiplied by 100 and integerized; ie both values are in centi-degrees.
This slot is empty for type=46.

The "integ_time" is the duration of the sampling interval
of the successful trigger criteria. This particular field is copied from the BAT_POS
notice.  The value is the number of 4msec ticks,
eg a value of 16 indicates a 64-msec trigger criterion.
(This 16-th 4-byte field is only defined/valid for type=65 packets and
if the 2^24 bit is set in the 'misc' field.)

The "soln_status" (aka trigger_id) field contains flag bits copied from the BAT_POS_ACK
ie (a) flag bits on the type of solution that was found in the on-board image and
about whether it matches anything in the on-board source catalog, and
(b) flag bits that are assigned in the GCN ground-processing.
(This 16-th 4-byte field is only defined/valid for packets (a) if the 2^24 bit is set
in the 'misc' field,  and (b) if it was generated after 03 May 05 (the time
when the distribution order of the BAT_POS and FOM_OBS packets was changed
to allow for the filling of this field and to allow for the filtering of FOM_OBS
based on these trigger_id bits).  Note that this is a non-standard location
for the "soln_status" field in the Swift series of notices.)
The bit packing of the "soln_status" entry in the packet is:
BitNum  F-v-G  Item_name      Description
------  ----   ---------      -----------
0        F     point_src      0=No or 1=Yes, a point source was found
1        F     grb            0=No or 1=Yes, it is a GRB
2        F     interesting    0=No or 1=Yes, it is an interesting src (ie a flaring known src)
3        F     flt_cat_src    0=No or 1=Yes, it is in the flight catalog (else unknown src)
4        F     image_trig     0=No or 1=Yes, it is an image trigger (else rate trig)
5        G     def_not_grb    0=No or 1=Yes, it is definitely not a GRB (bad label; its really a Retraction)
6        G     uncert_grb     0=No or 1=Yes, it is probably not a GRBorTransient (hi bkg level, ie near SAA)
7        G     uncert_grb     0=No or 1=Yes, it is probably not a GRBorTransient (low image significance; less then 7.0 sigma)
8        G     gnd_cat_src    0=No or 1=Yes, it is in the ground catalog
9        G     uncert_grb     0=No or 1=Yes, it is probably not a GRBorTransient (negative bkg slope, exit SAA)
10       G     st_loss_lock   0=No or 1=Yes, StarTracker not locked so trigger probably bogus
11       G     uncert_grb     0=No or 1=Yes, it is really probably not a GRBorTransient (VERY low image significance; less than 6.5 sigma)
12       G     blk_cat_src    0=No or 1=Yes, it is in the catalog of sources to be blocked (internal use only)
13       G     near_brt_star  0=No or 1=Yes, there is a nearby bright star (this is a duplicate of the "misc" 2^13).
14       -     spare          spare for future use
15       G     onboard_rmvd   0=No or 1=Yes, this is a source that has purposefully been removed from on-board catalog
16       G     nearby_gal     0=No or 1=Yes, this matched a Nearby_Galaxy in the on-board catalog
17-31    -     spare          spare for future use
The columns are:  "BitNum" is the 2^N location within the 4-byte word; "F-v-G" is Flight-assigned vs
Ground-assigned;  "Item_name" is the short-description name of the quantity;  and
"Description" is the longer description of the quantity.

The "trig_index" is the index value (row_number in the BAT on-board flight s/w table)
of the trigger criteria that was the highest-significance successful trigger.

The "at_slew_flags" field contains two flag bits related to the status
of the given burst wrt the Automated Target (AT) and Slewing.
The bits are:
2^0  Is this burst worthy of becoming the new Automated Target: 1=yes
2^1  Is this burst of sufficient merit to request a s/c slew: 1=yes
The AT is a special target that the Swift Observatory will keep coming back to
orbit after orbit without any ground commanding intervention.  It is distinct
from the PrePlanned Targets (PPTs) that are in the uploaded stored observing table.
Whenever the AT is observable (ie no observing constraint is in effect),
then the AT will override any previous PPT.  The override lasts for up to
a commandable amount of time -- typically 60,000 observed-seconds.

The "misc" field contains miscellaneous flag bits.
The defined bits are:
2^0   When this bit is a 1, it indicates that the BAT_Pos TDRSS message was not received for this TrigNum.
2^1-10  Not assigned.
2^11  When this bit is a 1, it indicates that 1 (or more) of the RA,Dec,Roll,Theta,Phi values
      were out of valid range (eg RA was 361 deg).
2^12  Not assigned.
2^13  When this bit is a 1, it indicates that this position is near a bright star (mag less than 6.5).
2^14  When this bit is a 1, it indicates that the distance between the position and the center of the galaxy
      is less than the sum of the position error radius and the galaxy radius AND that the position error radius
      is smaller than the galaxy radius.  This position is inside the (circularized) NGC galaxy (or nearly in).
2^15  When this bit is a 1, it indicates that the distance between the position and the center of the galaxy
      is less than the sum of the position error radius and the galaxy radius AND that the galaxy radius
      is smaller than the position error radius.  This galaxy is inside the position error circle (or nearly in).
2^16-19  Not assigned.
2^20  When this bit is a 1, it indicates that this was originally a SubThresh, but it is now converted to a real FOM.
2^21  When this bit is a 1, it indicates that this is an Image-trigger AND that it occurred
      during a time when the StarTracker had a Loss-of-Lock.
2^22  When this bit is a 1, it indicates that this Notice was generated as a result
      of an uploaded TOO sequence.  (Swift is commanded to look at this RA,Dec location
      and this Notice was generated as a direct result of that observation.)
2^23  This bit is set if the FOM_OBS packet was delayed (to come after
      the BAT_POS_ACK packet) so that the 'soln_status' field
      could be added to this packet (copied from BAT_POS_ACK).
2^24  This bit is set if the 'soln_status' field was copied to here from BAT_POS_ACK.
2^25  Not assigned.
2^26  Not assigned.
2^27  Not assigned.
2^28  Not assigned.
2^29  This bit is set if this FOM_OBS Noticed was forced-out via the watchdog timeout.
2^30  When this bit is a 1, then this Notice was ground-generated; else flight-generated.
      A Gnd-generated Notice is issued anytime humans during ground operations
      have determined if this Notice should be issued.  This will typically be used
      during the early phases of the mission before the automatic real-time distribution
      capability is enabled.  Once the burst position has been validated by humans,
      a BAT_Pos Notice will be issued (with this flag bit set to 1).
      After the Verification phase, there may be rare occaisions when a FOM_OBS notice
      is (re)created by human-processing and will be distributed (with this bit set to 1). 
2^31  If there is a CRC error in one or more of the packets that went into
      the making of this Notice, then this bit is set to 1, else 0.

The "image_signif" is the significance level of the peak in the FFT image plane.
This is the signal/noise (ie number of sigma) of the grb peak detection
multiplied by 100 and then integerized (ie units are centi-sigma).
(This 20-th 4-byte field is only defined/valid for type=65 packets and
if the 2^24 bit is set in the 'misc' field.)

The "rate_signif" is the significance level of the initial 'rate trigger'.
This is the signal/noise (ie number of sigma) of the rate-trigger detection
multiplied by 100 and then integerized (ie units are centi-sigma).

The "merit_value" is the total calculated value for the given burst.
It is a fl.pt. quantity which can theoretically range from -infinity to
+infinity, but given normal operations, we expect it to range from about
-100 to +300.  And for the first 6-12 months of flight-ops, it will be
forced to be a constant value of 100 for all bursts.

There are a total of 23 "spare" items in the packet (usually filled
with zeros).  They are located at items 9-14, 20, and 22-37.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.



TYPE=66 (and 47) PACKET CONTENTS:

The SWIFT_SC_SLEW (and SWIFT_TOO_SC_SLEW) packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=66 or =47)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       trig_obs_num integers        Trigger num & Observation num
long         5       burst_tjd    [days]          Truncated Julian Day
long         6       burst_sod    [centi-sec]     (int)(sssss.sss *100)
long         7       burst_ra     [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         8       burst_dec    [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         9       roll         [0.0001-deg]    (int)(0.0 to 360.0 *10000)
long         10      lon_lat      2_shorts        (int)(Longitude,Lattitude *100)
long         11-12   spare[2]     integer         8 bytes for the future
long         13      wait_time    [centi-sec]     (int)(sssss.sss *100)
long         14      obs_time     [centi-sec]     (int)(sssss.sss *100)
long         15      integ_time   [4mSec]         Duration of the trigger interval, 1 to inf
long         16      soln_status  bits            Type of source/trigger found
long         17      trig_index   integer         Rate_Trigger index
long         18      slew_rtn     integer         0=will_slew, non-0=wont
long         19      misc         bits            Misc stuff packed in here
long         20      image_signif [centi-sigma]   (int)(sig2noise *100)
long         21      rate_signif  [centi-sigma]   (int)(sig2noise *100)
long         22      bat_mode     bits            BAT configuration
long         23      xrt_mode     bits            XRT configuration
long         24      uvot_mode    bits            UVOT configuration
long         25-37   spare[13]    integer         52 bytes for the future
long         38      merit        [centi-merit]   Merit_value *100
long         39      pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 60, 61, 62, 63, ...84
that tells which type of packet this is and therefore what the contents are
(the contents change between the various types).

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start from 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the burst_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "trig_obs_num" is a pair of uniquely identifying serial numbers (Trigger_num and Observation_num)
assigned by the on-board BAT flight software to each trigger.
The lower 24 bits contain the Trigger_number, and the upper 8 bits
contain the Observation_number. 
The Observation_number is always 0 since these are always (by definition)
the first observation of this new trigger.
The Trigger_number started with s/n=100,000 just after launch (Nov 04).

The "burst_tjd" is the Truncated Julian Day of the Swift-BAT burst trigger,
eg. TJD=13370 is 01 Jan 2005.

The "burst_sod" is the UT seconds-of-day (SOD) of the Swift-BAT burst trigger.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized).

The "burst_ra" and "burst_dec" are the RA,Dec coordinates of the burst (or transient)
as determined by the BAT flight software (J2000 epoch).  These intrinsically
floating point quantities have been multiplied by 10,000 and then integerized
to make an integer quantity with units of 0.0001-degrees.

The "roll" of the spacecraft (after this requested slew).  This intrinsically
floating point quantity have been multiplied by 10,000 and then integerized
to make an integer quantity with units of 0.0001-degrees.

The "lon_lat" is the Longitude and Lattitude of the s/c at the time of the BAT trigger.
This 4-byte quantity is composed of two 2-byte integers.  The high-order short
is the latitude multiplied by 100 and integerized; the low-order short
is the longitude multiplied by 100 and integerized; ie both values are in centi-degrees.

The "wait_time" is the number of seconds until the target becomes viewable.
A value of zero means that the spacecraft will slew immediately -- ie no waiting.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized).

The "obs_time" is the number of seconds the target is observable.  (This value
does not reflect any time spent inside the SAA.)
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized).

The "integ_time" is the duration of the sampling interval
of the successful trigger criteria. This particular field is copied from the BAT_POS
notice.  The value is the number of 4msec ticks,
eg a value of 16 indicates a 64-msec trigger criterion.
(This 16-th 4-byte field is only defined/valid for type=66 packets and
if the 2^24 bit is set in the 'misc' field.)

The "soln_status" (aka trigger_id) field contains flag bits copied from the BAT_POS_ACK
ie (a) flag bits on the type of solution that was found in the on-board image and
about whether it matches anything in the on-board source catalog, and
(b) flag bits that are assigned in the GCN ground-processing.
(This 16-th 4-byte field is only defined/valid for packets (a) if the 2^24 bit is set
in the 'misc' field,  and (b) if it was generated after 03 May 05 (the time
when the distribution order of the BAT_POS and FOM_OBS packets was changed
to allow for the filling of this field and to allow for the filtering of FOM_OBS
based on these trigger_id bits).  Note that this is a non-standard location
for the "soln_status" field in the Swift series of notices.)
The bit packing of the "soln_status" entry in the packet is:
BitNum  F-v-G  Item_name      Description
------  ----   ---------      -----------
0        F     point_src      0=No or 1=Yes, a point source was found
1        F     grb            0=No or 1=Yes, it is a GRB
2        F     interesting    0=No or 1=Yes, it is an interesting src (ie a flaring known src)
3        F     flt_cat_src    0=No or 1=Yes, it is in the flight catalog (else unknown src)
4        F     image_trig     0=No or 1=Yes, it is an image trigger (else rate trig)
5        G     def_not_grb    0=No or 1=Yes, it is definitely not a GRB (bad label; its really a Retraction)
6        G     uncert_grb     0=No or 1=Yes, it is probably not a GRBorTransient (hi bkg level, ie near SAA)
7        G     uncert_grb     0=No or 1=Yes, it is probably not a GRBorTransient (low image significance; less then 7.0 sigma)
8        G     gnd_cat_src    0=No or 1=Yes, it is in the ground catalog
9        G     uncert_grb     0=No or 1=Yes, it is probably not a GRBorTransient (negative bkg slope, exit SAA)
10       G     st_loss_lock   0=No or 1=Yes, StarTracker not locked so trigger probably bogus
11       G     uncert_grb     0=No or 1=Yes, it is really probably not a GRBorTransient (VERY low image significance; less than 6.5 sigma)
12       G     blk_cat_src    0=No or 1=Yes, it is in the catalog of sources to be blocked (internal use only)
13       G     near_brt_star  0=No or 1=Yes, there is a nearby bright star (this is a duplicate of the "misc" 2^13).
14       -     spare          spare for future use
15       G     onboard_rmvd   0=No or 1=Yes, this is a source that has purposefully been removed from on-board catalog
16       G     nearby_gal     0=No or 1=Yes, this matched a Nearby_Galaxy in the on-board catalog
17-31    -     spare          spare for future use
The columns are:  "BitNum" is the 2^N location within the 4-byte word; "F-v-G" is Flight-assigned vs
Ground-assigned;  "Item_name" is the short-description name of the quantity;  and
"Description" is the longer description of the quantity.

The "trig_index" is the index value (row_number in the BAT on-board flight s/w table)
of the trigger criteria that was the highest-significance successful trigger.

The "slew_rtn" is the return_code from the s/c for the given slew_request.
If value is zero, then the slew will be executed by the s/c;
if non-zero, then slew will not be executed.  The non-zero values are:
1 Sun constraint,
2 Earth_limb constraint,
3 Moon constraint,
4 Ram_vector constraint,
5 invalid slew_request parameter.

The "misc" field contains miscellaneous flag bits.
The defined bits are:
2^0   When this bit is a 1, it indicates that the BAT_Pos TDRSS message was not received for this TrigNum.
2^1-10  Not assigned.
2^11  When this bit is a 1, it indicates that 1 (or more) of the RA,Dec,Roll,Theta,Phi values
      were out of valid range (eg RA was 361 deg).
2^12  Not assigned.
2^13  When this bit is a 1, it indicates that this position is near a bright star (mag less than 6.5).
2^14  When this bit is a 1, it indicates that the distance between the position and the center of the galaxy
      is less than the sum of the position error radius and the galaxy radius AND that the position error radius
      is smaller than the galaxy radius.  This position is inside the (circularized) NGC galaxy (or nearly in).
2^15  When this bit is a 1, it indicates that the distance between the position and the center of the galaxy
      is less than the sum of the position error radius and the galaxy radius AND that the galaxy radius
      is smaller than the position error radius.  This galaxy is inside the position error circle (or nearly in).
2^16-19  Not assigned.
2^20  When this bit is a 1, it indicates that this was originally a SubThresh, but it is now converted to a real SC_SLEW.
2^21  When this bit is a 1, it indicates that this is an Image-trigger AND that it occurred
      during a time when the StarTracker had a Loss-of-Lock.
2^22  When this bit is a 1, it indicates that this Notice was generated as a result
      of an uploaded TOO sequence.  (Swift is commanded to look at this RA,Dec location
      and this Notice was generated as a direct result of that observation.)
2^23  Not assigned.
2^24  This bit is set if the 'soln_status' field was copied to here from BAT_POS_ACK.
2^25  Not assigned.
2^26  Not assigned.
2^27  Not assigned.
2^28  Not assigned.
2^29  Not assigned.
2^30  When this bit is a 1, then this Notice was ground-generated; else flight-generated.
      A Gnd-generated Notice is issued anytime humans during ground operations
      have determined if this Notice should be issued.  This will typically be used
      during the early phases of the mission before the automatic real-time distribution
      capability is enabled.  Once the burst position has been validated by humans,
      a BAT_Pos Notice will be issued (with this flag bit set to 1).
      After the Verification phase, there may be rare occaisions when a SC_SLEW notice
      is (re)created by human-processing and will be distributed (with this bit set to 1). 
2^31  If there is a CRC error in one or more of the packets that went into
      the making of this Notice, then this bit is set to 1, else 0.

The "image_signif" is the significance level of the peak in the FFT image plane.
This is the signal/noise (ie number of sigma) of the grb peak detection
multiplied by 100 and then integerized (ie units are centi-sigma).
(This 20-th 4-byte field is only defined/valid for type=65 packets.)

The "rate_signif" is the significance level of the initial 'rate trigger'.
This is the signal/noise (ie number of sigma) of the rate-trigger detection
multiplied by 100 and then integerized (ie units are centi-sigma).

The 3 "modes" are configuration fields for each instrument.

The "merit_value" is the total calculated value for the given burst.
It is a fl.pt. quantity which can theoretically range from -infinity to
+infinity, but given normal operations, we expect it to range from about
-100 to +500.  And for the first 6-12 months of flight-ops, it will be
forced to be a constant value of 100 for all bursts.
The units are in centi-merit (ie the fl.pt. merit is multiplied by 100
and then integerized).

There are a total of 19 "spare" items in the packet (usually filled
with zeros).  They are located at items 10-12, 15-16, 20, and 25-37.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.



TYPE=67 PACKET CONTENTS:

The SWIFT_XRT_POS packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=67)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       trig_obs_num integers        Trigger num & Observation num
long         5       data_tjd     [days]          Truncated Julian Day
long         6       data_sod     [centi-sec]     (int)(sssss.sss *100)
long         7       burst_ra     [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         8       burst_dec    [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         9       burst_flux   [flux]          (int)(xrt_flux *100)
long         10      spare        integer         4 bytes for the future
long         11      burst_error  [0.0001-deg]    (int)(0.0 to 180.0 *10000)
long         12      X_TAM_I1     [centi-pixel]   (int)(tam[0]*100)
long         13      Y_TAM_I1     [centi-pixel]   (int)(tam[1]*100)
long         14      X_TAM_I2     [centi-pixel]   (int)(tam[2]*100)
long         15      Y_TAM_I2     [centi-pixel]   (int)(tam[3]*100)
long         16      hi_prec_err  [centi-arcsec]  Hi-precission burst_error
long         17      Amp_Wave     dual_int        AmpNum*256 + WaveformNum
long         18      soln_status  bits            The type of event
long         19      misc         bits            Misc stuff packed in here
long         20      xrt-bat      [0.0001-deg]    Distance between XRT & BAT positions
long         21      det_signif   [centi-sigma]   (int)(det_signif *100)
long         22-38   spare[17]    integer         68 bytes for the future
long         39      pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 60, 61, 62, 63, ...84
that tells which type of packet this is and therefore what the contents are
(the contents change between the various types).  The types are:
67 = SWIFT_XRT_POS, 68 = SWIFT_XRT_SPEC, 69 = SWIFT_XRT_IMAGE,
70 = SWIFT_XRT_LC, and 71 = SWIFT_XRT_NACK_POS. (Only the raw XRT Notices are listed here;
see before and later for the other Swift-BAT/UVOT Notices type values.)

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start from 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the data_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "trig_obs_num" is a pair of uniquely identifying serial numbers (Trigger_num and Observation_num)
assigned by the on-board BAT flight software to each trigger.
The lower 24 bits contain the Trigger_number, and the upper 8 bits
contain the Observation_number. 
The Observation_number is always 0 since these are always (by definition)
the first observation of this new trigger.
The Trigger_number started with s/n=100,000 just after launch (Nov 04).

The "data_tjd" is the Truncated Julian Day of the start of the CCD image
from which this position was extracted,  eg. TJD=13370 is 01 Jan 2005.

The "data_sod" is the UT seconds-of-day (SOD) of the start of the CCD image
from which this position was extracted, which will be very close
to the Swift-BAT burst_sod.  The units are in centi-seconds
(ie the fl.pt. seconds were multiplied by 100 and then integerized).

The "burst_ra" and "burst_dec" are the RA,Dec coordinates of the burst/afterglow
as determined by the XRT flight software (J2000 epoch).  These intrinsically
floating point quantities have been multiplied by 10,000 and then integerized
to make an integer quantity with units of 0.0001-degrees.

The "burst_flux" is the sum of pixel intensities that are between the low-level threshold
and an upper-level threshold  divided by the exposure time and a fudge factor.
Note that the pixels intensities have NOT had the bias subtraction made,
so there is an offset to these flux estimates.  [For now, the fudge factor is set to 1.]
This gives a crude relative brightness estimate of the afterglow source;
and is only intended for comparison of previous XRT detections.
After 14nov05 this field contains the approximate flux in CGS units [erg/cm2/sec]
multiplied by 10^14 and then integerized (eg if the field contains a value of 1230,
then the flux is 1.23x10^-11 erg/cm2/sec).

The "burst_error" is the radius of the circle that will contain on average
90% of the bursts afterglow positions.  It contains statistical plus
systematic contributions.  [The value was hardcoded to be 6arcsec
for all Notices at the beginning of the mission -- as experience built, 
this was replaced with a value that is dependant on the flux and/or significance
of the detection.]  (See also the hi_prec_err below.)

The "tam_x/y_1/2 are the positions of the TAM (Telescope Alignment Monitor)
in the 1st and 2nd images.  The units are centi-tam_pixels; the values typically
range from 200.00 to 400.00.  [These values should remain roughly constant
from Notice to Notice, ie the XRT tube is not flexing over time or temperature.]

The "hi_prec_err" is the radius of the circle that will contain on average
90% of the bursts afterglow positions.  It contains statistical plus
systematic contributions.  This is different than the regular "burst_error"
in that the units are 'centi-arcsec' to remove quantization error of 0.36arcsec
introduced by the regular "burst_error" value.

The "amp_wave" field has two values contained in the 32 bits.
The lower 8 bits (2^0-7) are the 'waveform number' of the readout.
And the next higher 8 bits (2^8-15) is the amplifier used during the readout.
The waveform number is typically 134.  The amplifier number is typically 2.

The "soln_status" (aka "trigger_id")field contains flag bits on the type of solution that was found.
The bit packing of the "soln_status" entry in the packet is:
The bit packing is:
BitNum  F-v-G  Item_name      Description
------  ----   ---------      -----------
0        G     cosmic_ray     0=No or 1=Yes, this is probably a cosmic ray.
1        -     undef          Not yet defined.
2        -     undef          Not yet defined.
3        -     undef          Not yet defined.
4        -     undef          Not yet defined.
5        G     def_not_grb    0=No or 1=Yes, it is definitely not a GRB (bad label; its really a Retraction)
6-7      -     undef          Not yet defined.
8        G     gnd_cat_src    0=No or 1=Yes, BAT_loc is in the BAT ground catalog
9-13     -     undef          Not yet defined.
14       G     was_subthresh  0=No or 1=Yes, this was originally a SubThresh, but it is now converted to a real XRT_POS
15-27    -     undef          Not yet defined.
28       G     spatial_coinc  0=No or 1=Yes, there was a spatial coincidence with another event
29       G     temporal_coinc 0=No or 1=Yes, there was a temporal coincidence with another event
30       G     test_submit    0=No or 1=Yes, this is a test submission (internal use only)
31       -     undef          Not yet defined.
The columns are:  "BitNum" is the 2^N location within the 4-byte word; "F-v-G" is Flight-assigned vs
Ground-assigned;  "Item_name" is the short-description name of the quantity;  and
"Description" is the longer description of the quantity.

The "misc" field contains miscellaneous flag bits.
The defined bits are:
2^0   When this bit is a 1, it indicates that the BAT_Pos TDRSS message was not received for this TrigNum.
2^1-9 Not assigned.
2^10  When this bit is a 1, it indicates that the centroided position was more than 100 rows or columns
      from the center of the FOV (ie likely not a real astrophysical peak).
2^11  When this bit is a 1, it indicates that 1 (or more) of the RA,Dec,Roll,Theta,Phi values
      were out of valid range (eg RA was 361 deg).
2^12  When this bit is 1, it is in the catalog of sources to be blocked (internal use only).
2^13  When this bit is a 1, it indicates that this position is near (less than 0.3deg)  a bright star (mag less than 6.5).
2^14  When this bit is a 1, it indicates that the distance between the position and the center of the galaxy
      is less than the sum of the position error radius and the galaxy radius AND that the position error radius
      is smaller than the galaxy radius.  This position is inside the (circularized) NGC galaxy (or nearly in).
2^15  When this bit is a 1, it indicates that the distance between the position and the center of the galaxy
      is less than the sum of the position error radius and the galaxy radius AND that the galaxy radius
      is smaller than the position error radius.  This galaxy is inside the position error circle (or nearly in).
2^16  When this bit is a 1, it indicated the theta value was less than 10arcmin (ie Swift was already pointed at this source).
2^17-19  Not assigned.
2^20  When this bit is a 1, it indicates that this was originally a SubThresh, but it is now converted to a real XRT_POS.
2^21  When this bit is a 1, it indicates that this is an Image-trigger AND that it occurred
      during a time when the StarTracker had a Loss-of-Lock.
2^22  When this bit is a 1, it indicates that this Notice was generated as a result
      of an uploaded TOO sequence.  (Swift is commanded to look at this RA,Dec location
      and this Notice was generated as a direct result of that observation.)
2^23  This bit is set if the XRT_POS packet was momentarily (few sec) delayed
      so that the XRT_IMAGE message could be interrogated to determine
      the 'gain' of the image that generated the xrt_position.
2^24  Not assigned.
2^25  This is an updated position.  This is a position notice that changes/improves
      the position specified in the original XRT_Position Notice.
2^26  Not assigned.
2^27  Not assigned.
2^28  When this bit is a 1, then this Notice was derived from a SERS message; else flight-generated.
      If the regular xrt_position message was not received via the real-time
      TDRSS comm channel, then the SERS message (which come from pipe-line processing
      of the Malindi playback downlinked data) will be used to generate a reconstituted
      XRT_Pos_Ack Notice.  But since the SERS message only contains a subset of the
      information is a regular TDRSS messages, these SERS-reconstituted Notices
      will have (tbd) fields undefined.  These undefined fields are: tbd, tbd, tbd....
2^29  When this bit is a 1, then this Notice was reconstituted from information
      from the XRT_IMAGE notice.  It was generated on the ground because the XRT_POS notice
      was not received on the ground after a fixed amount of time (~200 sec).  Presumably
      the original XRT_POS was lost due to a TDRSS telemetry gap.  We have reconstituted
      most of the contents of an XRT_POS notice from information from the XRT_IMAGE notice.
2^30  When this bit is a 1, then this Notice was ground-generated; else flight-generated.
2^31  If there is a CRC error in one or more of the packets that went into
      the making of this Notice, then this bit is set to 1, else 0.

The "xrt-bat" is the angular distance between the XRT position and the BAT position.
This intrinsically floating point quantity have been multiplied by 10,000 and
then integerized to make an integer quantity with units of 0.0001-degrees.

The "det_signif" is a measure of the significance of the afterglow detection.
It is the square root of the number of pixels above threshold.  Because of the
limited dynamic range of the pixel intensities in the CCD in this initial image,
the "det_signif" suffers from saturation problems.  Example: a det_signif of 5
means there were 25 pixels above threshold, but if there were 1000 pixels,
then the photon pile-up in those pixels will be significant and the "det_signif"
of 32 will not be an accurate measure of the significance (ie it will be under-valued).

There are a total of 10 "spare" items in the packet (usually filled
with zeros).  They are located at items 10, 18, 20, and 22-28.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.



TYPE=68 PACKET CONTENTS:

The SWIFT_XRT_SPEC packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=68)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       trig_obs_num integers        Trigger num & Observation num
long         5       SpecStart_tjd[days]          Truncated Julian Day
long         6       SpecStart_sod[centi-sec]     (int)(sssss.sss *100)
long         7       bore_ra      [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         8       bore_dec     [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         9       LiveTime     [centi-sec]     (int)(sssss.sss *100)
long         10      SpecStopTJD  [days]          Truncated Julian Day
long         11      SpecStopSOD  [centi-sec]     (int)(sssss.sss *100)
long         12      mode         integer         XRT operational mode
long         13      waveform     integer         Waveform number (50 or 60)
long         14      bias         [ADU]           Bias value
long         15-18   spare[4]     integer         16 bytes for the future
long         19      misc         bits            Misc stuff packed in here
long         20      spare        integer         4 bytes for the future
long         21      term_code    integer         Termination condition code
quad_char    22-38   url[17]      chars           The URL for the spectrum
long         39      pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 60, 61, 62, 63, ...84
that tells which type of packet this is and therefore what the contents are
(the contents change between the various types).  The types are:
67 = SWIFT_XRT_POS, 68 = SWIFT_XRT_SPEC, 69 = SWIFT_XRT_IMAGE,
70 = SWIFT_XRT_LC, and 71 = SWIFT_XRT_NACK_POS. (Only the raw XRT Notices are listed here;
see before and later for the other Swift-BAT/UVOT Notices type values.)

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start from 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the burst_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "trig_obs_num" is a pair of uniquely identifying serial numbers (Trigger_num and Observation_num)
assigned by the on-board BAT flight software to each trigger.
The lower 24 bits contain the Trigger_number, and the upper 8 bits
contain the Observation_number. 
The Observation_number is always 0 since these are always (by definition)
the first observation of this new trigger.
The Trigger_number started with s/n=100,000 just after launch (Nov 04).

The "SpecStart_tjd" is the Truncated Julian Day of the start of the first
CCD integration in the spectrum accumulation, eg. TJD=13370 is 01 Jan 2005.

The "SpecStart_sod" is the UT seconds-of-day (SOD) of the start of the first
CCD integration in the spectrum accumulation.  The units are in centi-seconds
(ie the fl.pt. seconds were multiplied by 100 and then integerized).

The "bore_ra/dec" is the RA,dec of the XRT bore sight (J2000 epoch).
This is the center of the FOV of the XRT image.  These intrinsically floating point
quantities have been multiplied by 10,000 and then integerized to make
an integer quantity with units of 0.0001-degrees.

The "live_time" is the exposure time duration of the spectrum integration.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized).

The "SpecStop_tjd" is the Truncated Julian Day of the last CCD integration
in the spectrum accumulation, eg. TJD=13370 is 01 Jan 2005.

The "SpecStop_sod" is the UT seconds-of-day (SOD) of the last CCD integration
in the spectrum accumulation.  The units are in centi-seconds
(ie the fl.pt. seconds were multiplied by 100 and then integerized).

The "mode" is the CCD Readout mode.  The values are:
1  Null
2  Short image
3  Long image
4  Piled-up Photodiode
5  Low Rate Photodiode
6  Windowed Timing
7  Photo-counting
8  Raw data
9  Bias map
10 Stop

The "waveform" is the CCD waveform ID. The values are either 50 or 60.

The "bias" is the bias value of the last LRPD frame (ignore this for WT spectrum).
The units are ADU.

The "misc" field contains miscellaneous flag bits.
The defined bits are:
2^0-10  Not assigned.
2^11  When this bit is a 1, it indicates that 1 (or more) of the RA,Dec,Roll,Theta,Phi values
      were out of valid range (eg RA was 361 deg).
2^12  When this bit is a 1, it is in the catalog of sources to be blocked (internal use only).
2^13  When this bit is a 1, it indicates that there is nearby bright star (mag less than 6.5).
2^14-19  Not assigned.
2^20  When this bit is a 1, it indicates that this was originally a SubThresh, but it is now converted to a real XRT_SPEC.
2^21  When this bit is a 1, it indicates that this is an Image-trigger AND that it occurred
      during a time when the StarTracker had a Loss-of-Lock.
2^22  When this bit is a 1, it indicates that this Notice was generated as a result
      of an uploaded TOO sequence.  (Swift is commanded to look at this RA,Dec location
      and this Notice was generated as a direct result of that observation.)
2^23  Not assigned.
2^24  Not assigned.
2^25  When this bit is a 1, then this Notice was generated with the 3rd of 3 pkts missing.
2^26  When this bit is a 1, then this Notice was generated with the 2nd of 3 pkts missing.
2^27  When this bit is a 1, then this Notice was generated with the 1st of 3 pkts missing.
2^28  Not assigned.
2^29  This bit is set if this XRT_SPECTRUM Noticed was forced-out via the watchdog timeout.
2^30  When this bit is a 1, then this Notice was ground-generated; else flight-generated.
      This will always 0 for the 'raw' form of this Notice (type=68) and
      1 for the ground-processed form (type=77).
2^31  If there is a CRC error in one or more of the packets that went into
      the making of this Notice, then this bit is set to 1, else 0.

The "term_cond" is the termination condition of the lightcurve collection process.
The values are:
0  Normal,
1  Terminated by time,
2  Terminated by snapshot,
3  Terminated by entering SAA,
4  Spectrum generated at the LRPD-to-WT transition,
5  Spectrum generated at the WT-to-LRorPC transition.

The "url" is just the filename portion of the URL -- the leading path
for the full URL is "http://gcn.gsfc.nasa.gov/gcn/notices_s/".
This item in the packet spans 17 longwords (ie 68 bytes) and contains 
the ASCII string of the filename portion of the URL.  This string
can be up to 67 characters long, and will always be null-terminated.

There are a total of 5 "spare" items in the packet (usually filled
with zeros).  They are located at items 15-18, and 20.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.



TYPE=69 PACKET CONTENTS:

The SWIFT_XRT_IMAGE packet (aka postage-stamp image) consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=69)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       trig_obs_num integers        Trigger num & Observation num
long         5       ImgStartTJD  [days]          Image Start Truncated Julian Day
long         6       ImgStartSOD  [centi-sec]     (int)(sssss.sss *100)
long         7       burst_ra     [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         8       burst_dec    [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         9       n_bright     [pixels]        Num of bright pixels in the image
long         10      spare        integer         4 bytes for the future
long         11      cent_sdev    [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         12      cent_x       [centi-pixels]  (int)(xxx.xx *100)
long         13      cent_y       [centi-pixels]  (int)(yyy.yy *100)
long         14      iraw_x       [pixels]        Center of the 51x51 pixel postage stamp image
long         15      iraw_y       [pixels]        Center of the 51x51 pixel postage stamp image
long         16      roll         [0.0001-deg]    (int)(-180.0 to +180.0 *10000)
long         17      gain/mod/wav mixed           gain, mode, & waveform; ints
long         18      expo_time    [centi-sec]     (int)(sssss.sss *100)
long         19      misc         bits            Misc stuff packed in here
long         20      grb_in_xrt_y [centi-pixels]  Position of GRB afterglow in XRT pixels*100
long         21      grb_in_xrt_z [centi-pixels]  Position of GRB afterglow in XRT pixels*100
quad_char    22-38   url[17]      chars           The URL for the image
long         39      pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 60, 61, 62, 63, ...84
that tells which type of packet this is and therefore what the contents are
(the contents change between the various types).  The types are:
67 = SWIFT_XRT_POS, 68 = SWIFT_XRT_SPEC, 69 = SWIFT_XRT_IMAGE,
70 = SWIFT_XRT_LC, and 71 = SWIFT_XRT_NACK_POS. (Only the raw XRT Notices are listed here;
see before and later for the other Swift-BAT/UVOT Notices type values.)

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start from 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the burst_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "trig_obs_num" is a pair of uniquely identifying serial numbers (Trigger_num and Observation_num)
assigned by the on-board BAT flight software to each trigger.
The lower 24 bits contain the Trigger_number, and the upper 8 bits
contain the Observation_number. 
The Observation_number is always 0 since these are always (by definition)
the first observation of this new trigger.
The Trigger_number started with s/n=100,000 just after launch (Nov 04).

The "ImgStart_tjd" is the Truncated Julian Day of the start of the CCD image integration.
eg. TJD=13370 is 01 Jan 2005.

The "ImgStart_sod" is the UT seconds-of-day (SOD) start of the CCD image integration.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized).

The "burst_ra/dec" is the RA,dec of the centroid of the source found
in the image (J2000 epoch).  These intrinsically floating point quantities
have been multiplied by 10,000 and then integerized to make
an integer quantity with units of 0.0001-degrees.

The "n_bright" is the number of bright pixels in the 51x51 postage-stamp image,
where bright means the pixel intensity is above the LLD and below the ULD thresholds.

The "cent_sdev" is the standard deviation of the centroid peak in the image.
The units are 0.0001-degrees (ie the fl.pt. std_dev was multiplied by 10000
and then integerized).

The "cent_x/y" is the centroid position in raw units.
The units are centi-pixels (ie the fl.pt. position was multiplied by 100
and then integerized).

The "iraw_x/y" are the row/column numbers (in units of CCD pixels) of the center
of the 51x51 pixel image that is this 'postage-stamp' subarray of the whole CCD.

The "roll" angle of the spacecraft.  This intrinsically floating point quantity
has been multiplied by 10,000 and then integerized to make an integer quantity
with units of 0.0001-degrees.

The "gain/mod/wav" is a triplet of 1-byte values in a single 4-byte quantity.
The "gain" is the gain of the amplifier used during the readout of the CCD.
It is packed in bits 2^16-23.  The value will be 1, 2, 4, 8, or 16.  If the gain
is greater than 1, then the "object" in the image is likely due to a cosmic ray,
and not due to a GRB afterglow.
The "mode" is the CCD Readout mode.  It is packed in bits 2^8-15.  The values are:
1  Null
2  Short image
3  Long image
4  Piled-up Photodiode
5  Low Rate Photodiode
6  Windowed Timing
7  Photo-counting
8  Raw data
9  Bias map
10 Stop
The "waveform" is the CCD waveform ID.  It is packed in bits 2^0-7.

The "expo_time" is the nominal exposure time for this image.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized).

The "misc" field contains miscellaneous flag bits.
The defined bits are:
2^0-9  Not assigned.
2^10  When this bit is a 1, it indicates that the centroided position was more than 100 rows or columns
      from the center of the FOV (ie likely not a real astrophysical peak).
2^11  When this bit is a 1, it indicates that 1 (or more) of the RA,Dec,Roll,Theta,Phi values
      were out of valid range (eg RA was 361 deg).
2^12  When this bit is a 1, it is in the catalog of sources to be blocked (internal use only).
2^13  When this bit is a 1, it indicates that this position is near a bright star (mag less than 6.5).
2^14  When this bit is a 1, it indicates that the distance between the position and the center of the galaxy
      is less than the sum of the position error radius and the galaxy radius AND that the position error radius
      is smaller than the galaxy radius.  This position is inside the (circularized) NGC galaxy (or nearly in).
2^15  When this bit is a 1, it indicates that the distance between the position and the center of the galaxy
      is less than the sum of the position error radius and the galaxy radius AND that the galaxy radius
      is smaller than the position error radius.  This galaxy is inside the position error circle (or nearly in).
2^16-19  Not assigned.
2^20  When this bit is a 1, it indicates that this was originally a SubThresh, but it is now converted to a real XRT_IMAGE.
2^21  When this bit is a 1, it indicates that this is an Image-trigger AND that it occurred
      during a time when the StarTracker had a Loss-of-Lock.
2^22  When this bit is a 1, it indicates that this Notice was generated as a result
      of an uploaded TOO sequence.  (Swift is commanded to look at this RA,Dec location
      and this Notice was generated as a direct result of that observation.)
2^23  Not assigned.
2^24  Not assigned.
2^25  When this bit is a 1, then this Notice was generated with the 3rd of 3 pkts missing.
2^26  When this bit is a 1, then this Notice was generated with the 2nd of 3 pkts missing.
2^27  When this bit is a 1, then this Notice was generated with the 1st of 3 pkts missing.
2^28  Not assigned.
2^29  This bit is set if this XRT_IMAGE Noticed was forced-out via the watchdog timeout.
2^30  When this bit is a 1, then this Notice was ground-generated; else flight-generated.
      This will always 0 for the 'raw' form of this Notice (type=69) and
      1 for the ground-processed form (type=78).
2^31  If there is a CRC error in one or more of the packets that went into
      the making of this Notice, then this bit is set to 1, else 0.

The "grb_in_xrt_y/z" is the GRB x-ray afterglow position in XRT coordinates.
The units are centi-pixels (but this was overkill because the intrinsic numbers
are in XRT CCD row,col pixels, so there is no non-zero fractional portion).

The "url" is just the filename portion of the URL -- the leading path
for the full URL is "http://gcn.gsfc.nasa.gov/gcn/notices_s/".
This item in the packet spans 17 longwords (ie 68 bytes) and contains 
the ASCII string of the filename portion of the URL.  This string
can be up to 67 characters long, and will always be null-terminated.

There is a total of 1 "spare" items in the packet (usually filled
with zeros).  It is located at item 10.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.



TYPE=70 PACKET CONTENTS:

The SWIFT_XRT_LC packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=70)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       trig_obs_num integers        Trigger num & Observation num
long         5       start_tjd    [days]          Truncated Julian Day
long         6       start_sod    [centi-sec]     (int)(sssss.sss *100)
long         7       bore_ra      [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         8       bore_dec     [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         9       live_time    [centi-sec]     (int)(sssss.sss *100)
long         10      stop_tjd     [days]          Truncated Julian Day
long         11      stop_sod     [centi-sec]     (int)(sssss.sss *100)
long         12-18   spare[7]     integer         28 bytes for the future
long         19      misc         bits            Misc stuff packed in here
long         20      n_bins       integer         Number of valid bins in lightcurve
long         21      term_cond    integer         Termination condition
long         20-21   spare[2]     integer         8 bytes for the future
quad_char    22-38   url[17]      chars           The URL for the lightcurve
long         39      pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 60, 61, 62, 63, ...84
that tells which type of packet this is and therefore what the contents are
(the contents change between the various types).  The types are:
67 = SWIFT_XRT_POS, 68 = SWIFT_XRT_SPEC, 69 = SWIFT_XRT_IMAGE,
70 = SWIFT_XRT_LC, and 71 = SWIFT_XRT_NACK_POS. (Only the raw XRT Notices are listed here;
see before and later for the other Swift-BAT/UVOT Notices type values.)

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start from 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the burst_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "trig_obs_num" is a pair of uniquely identifying serial numbers (Trigger_num and Observation_num)
assigned by the on-board BAT flight software to each trigger.
The lower 24 bits contain the Trigger_number, and the upper 8 bits
contain the Observation_number. 
The Observation_number is always 0 since these are always (by definition)
the first observation of this new trigger.
The Trigger_number started with s/n=100,000 just after launch (Nov 04).

The "start_tjd" is the Truncated Julian Day of the start of the first
CCD integration in the lightcurve accumulation, eg. TJD=13370 is 01 Jan 2005.

The "start_sod" is the UT seconds-of-day (SOD) of the start of the first
CCD integration in the lightcurve accumulation.  The units are in centi-seconds
(ie the fl.pt. seconds were multiplied by 100 and then integerized).

The "bore_ra/dec" is the RA,dec of the XRT bore sight (J2000 epoch).
This is the center of the FOV of the XRT.  These intrinsically floating point
quantities have been multiplied by 10,000 and then integerized to make
an integer quantity with units of 0.0001-degrees.

The "live_time" is the exposure time duration of the lightcurve integration.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized).

The "SpecStop_tjd" is the Truncated Julian Day of the last CCD integration
in the lightcurve accumulation, eg. TJD=13370 is 01 Jan 2005.

The "SpecStop_sod" is the UT seconds-of-day (SOD) of the last CCD integration
in the lightcurve accumulation.  The units are in centi-seconds
(ie the fl.pt. seconds were multiplied by 100 and then integerized).

The "misc" field contains miscellaneous flag bits.
The defined bits are:
2^0-10  Not assigned.
2^11  When this bit is a 1, it indicates that 1 (or more) of the RA,Dec,Roll,Theta,Phi values
      were out of valid range (eg RA was 361 deg).
2^12  When this bit is a 1, it is in the catalog of sources to be blocked (internal use only).
2^13  When this bit is a 1, there is a nearby bright star (mag less than 6.5).
2^14-19  Not assigned.
2^20  When this bit is a 1, it indicates that this was originally a SubThresh, but it is now converted to a real XRT_LC.
2^21  When this bit is a 1, it indicates that this is an Image-trigger AND that it occurred
      during a time when the StarTracker had a Loss-of-Lock.
2^22  When this bit is a 1, it indicates that this Notice was generated as a result
      of an uploaded TOO sequence.  (Swift is commanded to look at this RA,Dec location
      and this Notice was generated as a direct result of that observation.)
2^23  Not assigned.
2^24  Not assigned.
2^25  Not assigned.
2^26  Not assigned.
2^27  Not assigned.
2^28  Not assigned.
2^29  Not assigned.
2^30  Not assigned.
2^31  If there is a CRC error in one or more of the packets that went into
      the making of this Notice, then this bit is set to 1, else 0.

The "n_bins" is the number of valid bins in the collected lightcurve (0-100).

The "term_cond" is the termination condition of the lightcurve collection process.
The values are:
0  Normal (100 bins),
1  Terminated by time (1-99 bins),
2  Terminated by snapshot (1-99 bins),
3  Terminated by entering SAA (1-99 bins).

The "url" is just the filename portion of the URL -- the leading path
for the full URL is "http://gcn.gsfc.nasa.gov/gcn/notices_s/".
This item in the packet spans 17 longwords (ie 68 bytes) and contains 
the ASCII string of the filename portion of the URL.  This string
can be up to 67 characters long, and will always be null-terminated.

There are a total of 9 "spare" items in the packet (usually filled
with zeros).  They are located at items 12-18 and 20-21.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.



TYPE=71 PACKET CONTENTS:

The SWIFT_XRT_NACK_POS packet (aka centroid error) consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=71)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       trig_obs_num integers        Trigger num & Observation num
long         5       data_tjd     [days]          Truncated Julian Day
long         6       data_sod     [centi-sec]     (int)(sssss.sss *100)
long         7       point_ra     [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         8       point_dec    [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         9       centroid_cnt [cnts]          Integer counts in centroid
long         10      min_cnt      integer         Minimum num cnts needed in centroid
long         11      centroid_std [0.0001-deg]    StdDev of centroid
long         12      sig_max      [0.0001-deg]    Mx_Allowed_StdDev to be good
long         13-15   spare        integer         12 bytes for the future
long         16      ph2_iter     integer         Number of iteration attempts
long         17      max_iter     integer         Max allowed number of attempts
long         18      err_flag     integer         Centroid rtn code: 1,2,3
long         19      misc         bits            Misc stuff packed in here
long         20-38   spare[19]    integer         76 bytes for the future
long         39      pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 60, 61, 62, 63, ...84
that tells which type of packet this is and therefore what the contents are
(the contents change between the various types).  The types are:
67 = SWIFT_XRT_POS, 68 = SWIFT_XRT_SPEC, 69 = SWIFT_XRT_IMAGE,
70 = SWIFT_XRT_LC, and 71 = SWIFT_XRT_NACK_POS. (Only the raw XRT Notices are listed here;
see before and later for the other Swift-BAT/UVOT Notices type values.)

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start from 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the burst_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "trig_obs_num" is a pair of uniquely identifying serial numbers (Trigger_num and Observation_num)
assigned by the on-board BAT flight software to each trigger.
The lower 24 bits contain the Trigger_number, and the upper 8 bits
contain the Observation_number. 
The Observation_number is always 0 since these are always (by definition)
the first observation of this new trigger.
The Trigger_number started with s/n=100,000 just after launch (Nov 04).

The "data_tjd" is the Truncated Julian Day when this data packet was generated on-board.
eg. TJD=13370 is 01 Jan 2005.

The "data_sod" is the UT seconds-of-day (SOD) when this data packet was generated on-board,
which will be very close to the Swift-BAT burst trigger.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized).

The "point_ra" and "point_dec" are the RA,Dec coordinates (J2000)
of the pointing direction of the s/c at the start?/stop???
of the integration interval for the image.  These intrinsically floating point
quantities have been multiplied by 10,000 and then integerized to make
an integer quantity with units of 0.0001-degrees.

The "centroid_cnt" is the "calculated number of events" in the centroid.
The units are counts.

The "min_count" is the minimum number of events needed to calculate
a proper centroid (ie a proper position).
The units are counts.

The "centroid_std" is the standard deviation of any centroid found in the image.
The units are 0.0001-deg (ie the fl.pt. value multiplied by 10000 and
then integerized).

The "sig_max" is the maximum allowable std-deviation for a valid centroid.
The units are 0.0001-deg (ie the fl.pt. value multiplied by 10000 and
then integerized).

The "ph2_iter" is the number of iterations used in Phase 2 of the search
algorithm.

The "max_iter" is the maximum number of iterations allowed in Phase 2
of the search algorithm.

The "err_flag" is the error_code value for why a position was not found.
The values are:  1= no source found in image, 2= algorithm did not converge,
3= standard deviation too large, and 0xFFF= general error.

The "misc" field contains miscellaneous flag bits.
The defined bits are:
2^0-4  Not assigned.
2^5    It is definitely not a GRB (bad label; its really a Retraction).  Note that for this flagbit
       is in a nonstandard field (MISC instead of the usual TRIGGER_ID) for this one pkt type.
2^6-10 Not assigned.
2^11   When this bit is a 1, it indicates that 1 (or more) of the RA,Dec,Roll,Theta,Phi values
       were out of valid range (eg RA was 361 deg).
2^12   When this bit is a 1, it is in the catalog of sources to be blocked (internal use only).
2^13   When this bit is a 1, there is a nearby bright star (mag less than 6.5).
2^14   SPER date was used (by the UL processing) to determine this trigger is a non-burst trigger.
2^15   No SPER date was used (by the UL processing) but it is still determined that this is a non-burst trigger.
2^16-19  Not assigned.
2^20   When this bit is a 1, it indicates that this was originally a SubThresh, but it is now converted to a real XRT_POS_NACK.
2^21   When this bit is a 1, it indicates that this is an Image-trigger AND that it occurred
       during a time when the StarTracker had a Loss-of-Lock.
2^22   When this bit is a 1, it indicates that this Notice was generated as a result
       of an uploaded TOO sequence.  (Swift is commanded to look at this RA,Dec location
       and this Notice was generated as a direct result of that observation.)
2^23   Not assigned.
2^24   Not assigned.
2^25   Not assigned.
2^26   Not assigned.
2^27   Not assigned.
2^28   Not assigned.
2^29   test_submit    0=No or 1=Yes, this is a test submission (internal use only). Non-std location!
2^30   When this bit is a 1, then this Notice was ground-generated; else flight-generated.
2^31   If there is a CRC error in one or more of the packets that went into
       the making of this Notice, then this bit is set to 1, else 0.

There are a total of 22 "spare" items in the packet (usually filled
with zeros).  They are located at items 13-15 and 20-38.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.



TYPE=72 PACKET CONTENTS:

The SWIFT_UVOT_IMAGE packet (aka DarkBurst, aka Genie) consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=72)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       trig_obs_num integers        Trigger num & Observation num
long         5       ExpStart_tjd [days]          Truncated Julian Day
long         6       ExpStart_sod [centi-sec]     (int)(sssss.sss *100)
long         7       point_ra     [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         8       point_dec    [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         9       image_roll   [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         10      filter       integer         Which filter was used
long         11      expo_id      [sec]           Exposure_ID expressed in s/c_sec
long         12      X0           [det_coords]    0-2095
long         13      Y0           [det_coords]    0-2095
long         14      width        [det_coords]    0-2095
long         15      height       [det_coords]    0-2095
long         16      xy_grb       [det_coords]    0-2095; two packed
long         17      n_frames     integer         1-inf
long         18      swl_lwl      integers        2 4-bit integers
long         19      misc         bits            Misc stuff packed in here
long         20-21   spare[2]     integer         8 bytes for the future
quad_char    22-38   url[17]      chars           The URL for the image
long         39      pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 60, 61, 62, 63, ...84
that tells which type of packet this is and therefore what the contents are
(the contents change between the various types).  The types are:
72 = SWIFT_UVOT_IMAGE, and 73 = SWIFT_UVOT_SRCLIST (Only the raw UVOT Notices are listed here;
see before and later for the other Swift-BAT/XRT Notices type values.)

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start from 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the burst_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "trig_obs_num" is a pair of uniquely identifying serial numbers (Trigger_num and Observation_num)
assigned by the on-board BAT flight software to each trigger.
The lower 24 bits contain the Trigger_number, and the upper 8 bits
contain the Observation_number. 
The Observation_number is always 0 since these are always (by definition)
the first observation of this new trigger.
The Trigger_number started with s/n=100,000 just after launch (Nov 04).

The "ExpStart_tjd" is the Truncated Julian Day of the start of the CCD integration,
eg. TJD=13370 is 01 Jan 2005.

The "ExpStart_sod" is the UT seconds-of-day (SOD) of the start of the CCD integration.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100 and then integerized).

The "point_ra" and "point_dec" are the RA,Dec coordinates (J2000)
of the pointing direction of the s/c at the start?/stop???
of the integration interval for the image.  These intrinsically floating point
quantities have been multiplied by 10,000 and then integerized to make
an integer quantity with units of 0.0001-degrees.

The "image_roll" is the Roll of the target.
This intrinsically floating point quantity has been multiplied by 10,000
and then integerized to make an integer quantity with units of 0.0001-degrees.

The "filter" is the filter used for the image.  The values are:
0  Blocked
1  UV_Grism
2  UVW2
3  V
4  UVM2
5  Vis_Grism
6  UVW1
7  U
8  Magnifier
9  B
10 White
11 unknown

The "expo_id" is the exposure_ID expressed in spacecraft seconds.
(???what is range?  ambiguity????)

The "X/Y_offset" are the coordinates of pixel_zero in the Image subarray.
The units are detector coordinates: 0-2095.

The "width" and "height" are the sizes of the Image subarray.
The units are detector coordinates: 0-2095.

The "xy_grb" is the position of the GRB in unbinned detector units.  The source
of this position is determined by the 2^28 bit in the "misc" field.

The "n_frames" is the total number of read-out frames that went into
compiling this Image.

The "swl_lwl" is the 'short word length' and the 'long word length' used
in the compress of the as-transmitted image packet.  There is no compression
in the GCN-version of this Notice (so they are pretty much useless by the time
it gets to GCN distribution).

The "misc" field contains miscellaneous flag bits.
The defined bits are:
2^0-3   Four bits encoding the pixel binning, eg =0 means 1x1 binning (ie none), =1 mean 2x2 binning,
        =2 means 4x4 binning, up to =6 mean 64x64 binning.
2^4-10  The height of the Detector Window used in the "Blue" Processing Electronics.
2^11    When this bit is a 1, it indicates that 1 (or more) of the RA,Dec,Roll,Theta,Phi values
        were out of valid range (eg RA was 361 deg).
2^12    When this bit is a 1, it is in the catalog of sources to be blocked (internal use only).
2^13    When this bit is a 1, there is a nearby bright star (mag less than 6.5).
2^14-19  Not assigned.
2^20    When this bit is a 1, it indicates that this was originally a SubThresh, but it is now converted to a real UVOT_IMAGE.
2^21    When this bit is a 1, it indicates that this is an Image-trigger AND that it occurred
        during a time when the StarTracker had a Loss-of-Lock.
2^22    When this bit is a 1, it indicates that this Notice was generated as a result
        of an uploaded TOO sequence.  (Swift is commanded to look at this RA,Dec location
        and this Notice was generated as a direct result of that observation.)
2^23    Not assigned.
2^24    Not assigned.
2^25    Not assigned.
2^26    Not assigned.
2^27    Not assigned.
2^28    When this bit is a 1, then the source of the GRB Position came from an XRT Position command,
        else it came from the Window Position in the Mode command.
2^29    This bit is set if this UVOT_IMAGE Noticed was forced-out via the watchdog timeout.
2^30    When this bit is a 1, then this Notice was ground-generated; else flight-generated.
        This will always 0 for the 'raw' form of this Notice (type=72) and
        1 for the ground-processed form (type=79).
2^31    If there is a CRC error in one or more of the packets that went into
        the making of this Notice, then this bit is set to 1, else 0.

The "url" is just the filename portion of the URL -- the leading path
for the full URL is "http://gcn.gsfc.nasa.gov/gcn/notices_s/".
This item in the packet spans 17 longwords (ie 68 bytes) and contains 
the ASCII string of the filename portion of the URL.  This string
can be up to 67 characters long, and will always be null-terminated.

There are a total of 2 "spare" items in the packet (usually filled
with zeros).  They are located at items 20-21.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.



TYPE=73 PACKET CONTENTS:

The SWIFT_UVOT_SRCLIST packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=73)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       trig_obs_num integers        Trigger num & Observation num
long         5       start_tjd    [days]          Truncated Julian Day
long         6       start_sod    [centi-sec]     (int)(sssss.sss *100)
long         7       point_ra     [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         8       point_dec    [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         9       image_roll   [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         10      filter       integer         Which filter was used
long         11      bkg_mean     [0.001-dn]      Mean of the background
long         12      x_max        [det_coords]    0-2095
long         13      y_max        [det_coords]    0-2095
long         14      n_stars      integer         1-infinity
long         15      x_offset     [det_coords]    0-2095
long         16      y_offset     [det_coords]    0-2095
long         17      det_thresh   integer         0-????
long         18      photo_thresh integer         0-????
long         19      misc         bits            Misc stuff packed in here
long         20-22   spare[2]     integer         8 bytes for the future
quad_char    22-38   url[17]      chars           The URL for the image
long         39      pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 60, 61, 62, 63, ...84
that tells which type of packet this is and therefore what the contents are
(the contents change between the various types).  The types are:
72 = SWIFT_UVOT_IMAGE, and 73 = SWIFT_UVOT_SRCLIST (Only the raw UVOT Notices are listed here;
see before and later for the other Swift-BAT/XRT Notices type values.)

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start from 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the burst_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "trig_obs_num" is a pair of uniquely identifying serial numbers (Trigger_num and Observation_num)
assigned by the on-board BAT flight software to each trigger.
The lower 24 bits contain the Trigger_number, and the upper 8 bits
contain the Observation_number. 
The Observation_number is always 0 since these are always (by definition)
the first observation of this new trigger.
The Trigger_number started with s/n=100,000 just after launch (Nov 04).

The "Start_tjd" is the Truncated Julian Day of the start of the Finding Chart
CCD integration, eg. TJD=13370 is 01 Jan 2005.

The "Start_sod" is the UT seconds-of-day (SOD) of the start of the Finding Chart
CCD integration.  The units are in centi-seconds (ie the fl.pt. seconds were multiplied
by 100 and then integerized).

The "point_ra" and "point_dec" are the RA,Dec coordinates (J2000)
of the pointing direction of the s/c at the start?/stop???
of the integration interval for the image.  These intrinsically floating point
quantities have been multiplied by 10,000 and then integerized to make
an integer quantity with units of 0.0001-degrees.

The "image_roll" is the Roll of the target.
This intrinsically floating point quantity has been multiplied by 10,000
and then integerized to make an integer quantity with units of 0.0001-degrees.

The "filter" is the filter used for the image.  The values are:
0  Blocked
1  UV_Grism
2  UVW2
3  V
4  UVM2
5  Vis_Grism
6  UVW1
7  U
8  Magnifier
9  B
10 White
11 unknown

The "bkg_mean" is the mean value of the background level in the exposure.
This intrinsically floating point quantity has been multiplied by 1000
and then integerized to make an integer quantity with units of 0.001-degrees.

The "x/y_max"  are the maximum extent of the image window expressed
in detector coordinates (0-2095).

The "n_stars" is the total number of stars in the source list.

The "x/y_offset0" are the X and Y offsets of the image window expressed
in detector coordinates (0-2095).

The "det_thresh" is the detection threshold used in this exposure.
The units are "dn" (digital number), and the range is 0 to ????.

The "photo_thresh" is the photometry threshold used in this exposure.
The units are "dn" (digital number), and the range is 0 to ????.

The "misc" field contains miscellaneous flag bits.
The defined bits are:
2^0-10  Not assigned.
2^11  When this bit is a 1, it indicates that 1 (or more) of the RA,Dec,Roll,Theta,Phi values
      were out of valid range (eg RA was 361 deg).
2^12  When this bit is a 1, it is in the catalog of sources to be blocked (internal use only).
2^13  When this bit is a 1, there is a nearby bright star (mag less than 6.5).
2^14-19  Not assigned.
2^20    When this bit is a 1, it indicates that this was originally a SubThresh, but it is now converted to a real UVOT_SL.
2^21  When this bit is a 1, it indicates that this is an Image-trigger AND that it occurred
      during a time when the StarTracker had a Loss-of-Lock.
2^22  When this bit is a 1, it indicates that this Notice was generated as a result
      of an uploaded TOO sequence.  (Swift is commanded to look at this RA,Dec location
      and this Notice was generated as a direct result of that observation.)
2^23  Not assigned.
2^24  Not assigned.
2^25  Not assigned.
2^26  Not assigned.
2^27  Not assigned.
2^28  Not assigned.
2^29  This bit is set if this UVOT_SrcList Noticed was forced-out via the watchdog timeout.
2^30  When this bit is a 1, then this Notice was ground-generated; else flight-generated.
      This will always 0 for the 'raw' form of this Notice (type=73) and
      1 for the ground-processed form (type=80).
2^31  If there is a CRC error in one or more of the packets that went into
      the making of this Notice, then this bit is set to 1, else 0.

The "url" is just the filename portion of the URL -- the leading path
for the full URL is "http://gcn.gsfc.nasa.gov/gcn/notices_s/".
This item in the packet spans 17 longwords (ie 68 bytes) and contains 
the ASCII string of the filename portion of the URL.  This string
can be up to 67 characters long, and will always be null-terminated.

There are a total of 3 "spare" items in the packet (usually filled
with zeros).  They are located at items 20-22.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.




TYPE=76 PACKET CONTENTS:

The SWIFT_BAT_GRB_LC_PROCESSED packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=76)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       trig_obs_num integers        Trigger num & Observation num
long         5       burst_tjd    [days]          Truncated Julian Day
long         6       burst_sod    [centi-sec]     (int)(sssss.sss *100)
long         7       burst_ra     [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         8       burst_dec    [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         9-11    spare[3]     integer         12 bytes for the future
long         12      phi          [centi-deg]     (int)(0.0 to 359.9999 *100)
long         13      theta        [centi-deg]     (int)(0.0 to +70.0 *100)
long         14      delta_time   [centi-sec]     (int)(ssss.sss *100)
long         15      spare        integer         4 bytes for the future
long         16      lon_lat      2_shorts        (int)(Longitude,Lattitude *100)
long         17      trig_index   integer         Rate_Trigger index
long         18      soln_status  bits            Type of source/trigger found
long         19      misc         bits            Misc stuff packed in here
long         20      image_signif [centi-sigma]   (int)(sig2noise *100)
long         21      rate_signif  [centi-sigma]   (int)(sig2noise *100)
quad_char    22-38   url[17]      chars           The URL for the lightcurve
long         39      pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 60, 61, 62, 63, ...84
that tells which type of packet this is and therefore what the contents are
(the contents change between the various types).  The types are:
76 = SWIFT_BAT_GRB_LIGHTCURVE_PROC, 77 = SWIFT_XRT_SPECTRUM_PROC,
78 = SWIFT_XRT_IMAGE_PROC, 79 = SWIFT_UVOT_IMAGE_PROC, and
80 = SWIFT_UVOT_SRCLIST_PROC. (Only the ground-processed Notices are listed here;
see before for all the other Swift Notices type values.)

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start from 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the burst_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "trig_obs_num" is a pair of uniquely identifying serial numbers (Trigger_num and Observation_num)
assigned by the on-board BAT flight software to each trigger.
The lower 24 bits contain the Trigger_number, and the upper 8 bits
contain the Observation_number. 
The Observation_number is always 0 since these are always (by definition)
the first observation of this new trigger.
The Trigger_number started with s/n=100,000 just after launch (Nov 04).

The "burst_tjd" is the Truncated Julian Day of the Swift-BAT burst trigger,
eg. TJD=13370 is 01 Jan 2005.

The "burst_sod" is the UT seconds-of-day (SOD) of the Swift-BAT burst trigger.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized).

The "burst_ra" and "burst_dec" are the RA,Dec coordinates of the burst (or transient)
as determined by the BAT flight software (J2000 epoch).  These will always be
the same values as in the SWIFT_BAT_POS_ACK packet.  These intrinsically
floating point quantities have been multiplied by 10,000 and then integerized
to make an integer quantity with units of 0.0001-degrees.

The "phi" and "theta" are the BAT instrument coordinates of the trigger (burst)
as determined by the BAT flight software program.  Phi is measured phi=0
along the +Y-axis of the s/c and increasing towards the -Z axis,
and theta is measured from the boresight of the BAT instrument.

The "delta_time" is the difference between the first sample in the lightcurve
and the Trigger_time.  The units are centi-seconds (although they will always be whole sec).

The "lon_lat" is the Longitude and Lattitude of the s/c at the time of the BAT trigger.
This 4-byte quantity is composed of two 2-byte integers.  The high-order short
is the latitude multiplied by 100 and integerized; the low-order short
is the longitude multiplied by 100 and integerized; ie both values are in centi-degrees.

The "trig_index" is the index value (row_number in the BAT on-board flight s/w table)
of the trigger criteria that was the highest-significance successful trigger.

The "soln_status" (aka trigger_id) field contains flag bits copied from the BAT_POS_ACK
ie (a) flag bits on the type of solution that was found in the on-board image and
about whether it matches anything in the on-board source catalog, and
(b) flag bits that are assigned in the GCN ground-processing.
(This 18-th 4-byte field is only defined/valid for packets (a) if the 2^24 bit is set
in the 'misc' field,  and (b) if it was generated after 03 May 05 (the time
when the distribution order of the BAT_POS and FOM_OBS packets was changed
to allow for the filling of this field and to allow for the filtering of FOM_OBS
based on these trigger_id bits).)
The bit packing of the "soln_status" entry in the packet is:
BitNum  F-v-G  Item_name      Description
------  ----   ---------      -----------
0        F     point_src      0=No or 1=Yes, a point source was found
1        F     grb            0=No or 1=Yes, it is a GRB
2        F     interesting    0=No or 1=Yes, it is an interesting src (ie a flaring known src)
3        F     flt_cat_src    0=No or 1=Yes, it is in the flight catalog (else unknown src)
4        F     image_trig     0=No or 1=Yes, it is an image trigger (else rate trig)
5        G     def_not_grb    0=No or 1=Yes, it is definitely not a GRB (bad label; its really a Retraction)
6        G     uncert_grb     0=No or 1=Yes, it is probably not a GRBorTransient (hi bkg level, ie near SAA)
7        G     uncert_grb     0=No or 1=Yes, it is probably not a GRBorTransient (low image significance; less then 7.0 sigma)
8        G     gnd_cat_src    0=No or 1=Yes, it is in the ground catalog
9        G     uncert_grb     0=No or 1=Yes, it is probably not a GRBorTransient (negative bkg slope, exit SAA)
10       G     st_loss_lock   0=No or 1=Yes, StarTracker not locked so trigger probably bogus
11       G     uncert_grb     0=No or 1=Yes, it is really probably not a GRBorTransient (VERY low image significance; less than 6.5 sigma)
12       G     blk_cat_src    0=No or 1=Yes, it is in the catalog of sources to be blocked (internal use only)
13       G     near_brt_star  0=No or 1=Yes, there is a nearby bright star (this is a duplicate of the "misc" 2^13).
14-31    -     spare          spare for future use
The columns are:  "BitNum" is the 2^N location within the 4-byte word; "F-v-G" is Flight-assigned vs
Ground-assigned;  "Item_name" is the short-description name of the quantity;  and
"Description" is the longer description of the quantity.

The "misc" field contains miscellaneous flag bits.
The defined bits are:
2^0-10  Not assigned.
2^11  When this bit is a 1, it indicates that 1 (or more) of the RA,Dec,Roll,Theta,Phi values
      were out of valid range (eg RA was 361 deg).
2^12  Not assigned.
//2^13  When this bit is a 1, it indicates that this position is near a bright star (mag less than 6.5).
//2^14  When this bit is a 1, it indicates that the distance between the position and the center of the galaxy
//      is less than the sum of the position error radius and the galaxy radius AND that the position error radius
//      is smaller than the galaxy radius.  This position is inside the (circularized) NGC galaxy (or nearly in).
//2^15  When this bit is a 1, it indicates that the distance between the position and the center of the galaxy
//      is less than the sum of the position error radius and the galaxy radius AND that the galaxy radius
//      is smaller than the position error radius.  This galaxy is inside the position error circle (or nearly in).
2^16-19  Not assigned.
2^20  When this bit is a 1, it indicates that this was originally a SubThresh, but it is now converted to a real BAT_PROC_LC.
2^21  When this bit is a 1, it indicates that this is an Image-trigger AND that it occurred
      during a time when the StarTracker had a Loss-of-Lock.
2^22  When this bit is a 1, it indicates that this Notice was generated as a result
      of an uploaded TOO sequence.  (Swift is commanded to look at this RA,Dec location
      and this Notice was generated as a direct result of that observation.)
2^23  This includes "extended" data packets.
2^24  This bit is set if the 'soln_status' field was copied to here from BAT_POS_ACK.
2^25  When this bit is a 1, then this Notice was generated with the 3rd of 3 pkts missing.
2^26  When this bit is a 1, then this Notice was generated with the 2nd of 3 pkts missing.
2^27  When this bit is a 1, then this Notice was generated with the 1st of 3 pkts missing.
2^28  When this bit is a 1, then this lightcurve contains GT 50 illegal values.
      This is probably an empty lightcurve for a long image trigger.
2^29  This bit is set if this original 'raw' LightCurve Noticed was forced-out via the watchdog timeout.
2^30  When this bit is a 1, then this Notice was ground-generated; else flight-generated.
      This will always be 0 for the 'raw' form of this Notice (type=63) and
      1 for this ground-processed form (type=76).
2^31  If there is a CRC error in one or more of the packets that went into
      the making of this Notice, then this bit is set to 1, else 0.

The "image_signif" is the significance level of the peak in the FFT image plane.
This is the signal/noise (ie number of sigma) of the grb peak detection
multiplied by 100 and then integerized (ie units are centi-sigma).
This field was copied from the BAT_POS notice.

The "rate_signif" is the significance level of the initial 'rate trigger'.
This is the signal/noise (ie number of sigma) of the rate-trigger detection
multiplied by 100 and then integerized (ie units are centi-sigma).
This field was copied from the BAT_POS notice.

The "url" is just the filename portion of the URL -- the leading path
for the full URL is "http://gcn.gsfc.nasa.gov/gcn/notices_s/".
This item in the packet spans 17 longwords (ie 68 bytes) and contains 
the ASCII string of the filename portion of the URL.  This string
can be up to 67 characters long, and will always be null-terminated.

There are a total of 4 "spare" items in the packet (usually filled
with zeros).  They are located at items 9-11 and 15.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.



TYPE=77 PACKET CONTENTS:

The SWIFT_XRT_SPEC_PROCESSED packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=77)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       trig_obs_num integers        Trigger num & Observation num
long         5       SpecStart_tjd[days]          Truncated Julian Day
long         6       SpecStart_sod[centi-sec]     (int)(sssss.sss *100)
long         7       bore_ra      [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         8       bore_dec     [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         9       LiveTime     [centi-sec]     (int)(sssss.sss *100)
long         10      SpecStopTJD  [days]          Truncated Julian Day
long         11      SpecStopSOD  [centi-sec]     (int)(sssss.sss *100)
long         12      mode         integer         XRT operational mode
long         13      waveform     integer         Waveform number (50 or 60)
long         14      bias         [ADU]           Bias value
long         15-18   spare[4]     integer         16 bytes for the future
long         19      misc         bits            Misc stuff packed in here
long         20      spare        integer         4 bytes for the future
long         21      term_code    integer         Termination condition code
quad_char    22-38   url[17]      chars           The URL for the spectrum
long         39      pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 60, 61, 62, 63, ...84
that tells which type of packet this is and therefore what the contents are
(the contents change between the various types).  The types are:
76 = SWIFT_BAT_GRB_LIGHTCURVE_PROC, 77 = SWIFT_XRT_SPECTRUM_PROC,
78 = SWIFT_XRT_IMAGE_PROC, 79 = SWIFT_UVOT_IMAGE_PROC, and
80 = SWIFT_UVOT_SRCLIST_PROC. (Only the ground-processed Notices are listed here;
see before for all the other Swift Notices type values.)

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start from 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the burst_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "trig_obs_num" is a pair of uniquely identifying serial numbers (Trigger_num and Observation_num)
assigned by the on-board BAT flight software to each trigger.
The lower 24 bits contain the Trigger_number, and the upper 8 bits
contain the Observation_number. 
The Observation_number is always 0 since these are always (by definition)
the first observation of this new trigger.
The Trigger_number started with s/n=100,000 just after launch (Nov 04).

The "SpecStart_tjd" is the Truncated Julian Day of the start of the first
CCD integration in the spectrum accumulation, eg. TJD=13370 is 01 Jan 2005.

The "SpecStart_sod" is the UT seconds-of-day (SOD) of the start of the first
CCD integration in the spectrum accumulation.  The units are in centi-seconds
(ie the fl.pt. seconds were multiplied by 100 and then integerized).

The "bore_ra/dec" is the RA,dec of the XRT bore sight (J2000 epoch).
This is the center of the FOV of the XRT image.  These intrinsically floating point
quantities have been multiplied by 10,000 and then integerized to make
an integer quantity with units of 0.0001-degrees.

The "live_time" is the exposure time duration of the spectrum integration.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized).

The "SpecStop_tjd" is the Truncated Julian Day of the last CCD integration
in the spectrum accumulation, eg. TJD=13370 is 01 Jan 2005.

The "SpecStop_sod" is the UT seconds-of-day (SOD) of the last CCD integration
in the spectrum accumulation.  The units are in centi-seconds
(ie the fl.pt. seconds were multiplied by 100 and then integerized).

The "mode" is the CCD Readout mode.  The values are:
1  Null
2  Short image
3  Long image
4  Piled-up Photodiode
5  Low Rate Photodiode
6  Windowed Timing
7  Photo-counting
8  Raw data
9  Bias map
10 Stop

The "waveform" is the CCD waveform ID. The values are either 50 or 60.

The "bias" is the bias value of the last LRPD frame (ignore this for WT spectrum).
The units are ADU.

The "misc" field contains miscellaneous flag bits.
The defined bits are:
2^0-10  Not assigned.
2^11  When this bit is a 1, it indicates that 1 (or more) of the RA,Dec,Roll,Theta,Phi values
      were out of valid range (eg RA was 361 deg).
2^12  When this bit is a 1, it is in the catalog of sources to be blocked (internal use only).
2^13  When this bit is a 1, there is a nearby bright star (mag less than 6.5).
2^14-19  Not assigned.
2^20  When this bit is a 1, it indicates that this was originally a SubThresh, but it is now converted to a real XRT_PROC_SPEC.
2^21  When this bit is a 1, it indicates that this is an Image-trigger AND that it occurred
      during a time when the StarTracker had a Loss-of-Lock.
2^22  When this bit is a 1, it indicates that this Notice was generated as a result
      of an uploaded TOO sequence.  (Swift is commanded to look at this RA,Dec location
      and this Notice was generated as a direct result of that observation.)
2^23  Not assigned.
2^24  When this bit is a 1, the xrt_proc_sp{1,2}.fits file had an error in processing and does not exist.
2^25  Not assigned.
2^26  Not assigned.
2^27  Not assigned.
2^28  Not assigned.
2^29  This bit is set if this original 'raw' Spectrum Noticed was forced-out via the watchdog timeout.
2^30  When this bit is a 1, then this Notice was ground-generated; else flight-generated.
      This will always be 0 for the 'raw' form of this Notice (type=68) and
      1 for the ground-processed form (type=77).
2^31  If there is a CRC error in one or more of the packets that went into
      the making of this Notice, then this bit is set to 1, else 0.

The "term_cond" is the termination condition of the lightcurve collection process.
The values are:
0  Normal,
1  Terminated by time,
2  Terminated by snapshot,
3  Terminated by entering SAA,
4  Spectrum generated at the LRPD-to-WT transition,
5  Spectrum generated at the WT-to-LRorPC transition.

The "url" is just the filename portion of the URL -- the leading path
for the full URL is "http://gcn.gsfc.nasa.gov/gcn/notices_s/".
This item in the packet spans 17 longwords (ie 68 bytes) and contains 
the ASCII string of the filename portion of the URL.  This string
can be up to 67 characters long, and will always be null-terminated.

There are a total of 5 "spare" items in the packet (usually filled
with zeros).  They are located at items 15-18, and 20.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.



TYPE=78 PACKET CONTENTS:

The SWIFT_XRT_IMAGE_PROCESSED packet (aka poststamp image) consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=78)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       trig_obs_num integers        Trigger num & Observation num
long         5       ImgStartTJD  [days]          Image Start Truncated Julian Day
long         6       ImgStartSOD  [centi-sec]     (int)(sssss.sss *100)
long         7       burst_ra     [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         8       burst_dec    [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         9       n_bright     [pixels]        Num of bright pixels in the image
long         10      spare        integer         4 bytes for the future
long         11      cent_sdev    [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         12      cent_x       [centi-pixels]  (int)(xxx.xx *100)
long         13      cent_y       [centi-pixels]  (int)(yyy.yy *100)
long         14      iraw_x       [pixels]        Center of the 51x51 pixel postage stamp image
long         15      iraw_y       [pixels]        Center of the 51x51 pixel postage stamp image
long         16      roll         [0.0001-deg]    (int)(-180.0 to +180.0 *10000)
long         17      gain/mod/wav mixed           gain, mode, & waveform; ints
long         18      expo_time    [centi-sec]     (int)(sssss.sss *100)
long         19      misc         bits            Misc stuff packed in here
long         20      grb_in_xrt_y [centi-pixels]  Position of GRB afterglow in XRT pixels*100
long         21      grb_in_xrt_z [centi-pixels]  Position of GRB afterglow in XRT pixels*100
quad_char    22-38   url[17]      chars           The URL for the image
long         39      pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 60, 61, 62, 63, ...84
that tells which type of packet this is and therefore what the contents are
(the contents change between the various types).  The types are:
76 = SWIFT_BAT_GRB_LIGHTCURVE_PROC, 77 = SWIFT_XRT_SPECTRUM_PROC,
78 = SWIFT_XRT_IMAGE_PROC, 79 = SWIFT_UVOT_IMAGE_PROC, and
80 = SWIFT_UVOT_SRCLIST_PROC. (Only the ground-processed Notices are listed here;
see before for all the other Swift Notices type values.)

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start from 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the burst_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "trig_obs_num" is a pair of uniquely identifying serial numbers (Trigger_num and Observation_num)
assigned by the on-board BAT flight software to each trigger.
The lower 24 bits contain the Trigger_number, and the upper 8 bits
contain the Observation_number. 
The Observation_number is always 0 since these are always (by definition)
the first observation of this new trigger.
The Trigger_number started with s/n=100,000 just after launch (Nov 04).

The "ImgStart_tjd" is the Truncated Julian Day of the start of the image integration.
eg. TJD=13370 is 01 Jan 2005.

The "ImgStart_sod" is the UT seconds-of-day (SOD) start of the image integration.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized).

The "burst_ra/dec" is the RA,dec of the centroid of the source found
in the image (J2000 epoch).  These intrinsically floating point quantities
have been multiplied by 10,000 and then integerized to make
an integer quantity with units of 0.0001-degrees.

The "n_bright" is the number of bright pixels in the 51x51 postage-stamp image,
where bright means the pixel intensity is above the LLD and below the ULD thresholds.

The "cent_sdev" is the standard deviation of the centroid peak in the image.
The units are 0.0001-degrees (ie the fl.pt. std_dev was multiplied by 10000
and then integerized).

The "cent_x/y" is the centroid position in raw units.
The units are centi-pixels (ie the fl.pt. position was multiplied by 100
and then integerized).

The "iraw_x/y" are the row/column numbers (in units of CCD pixels) of the center
of the 51x51 pixel image that is this 'postage-stamp' subarray of the whole CCD.

The "roll" angle of the spacecraft.  This intrinsically floating point quantity
has been multiplied by 10,000 and then integerized to make an integer quantity
with units of 0.0001-degrees.

The "gain/mod/wav" is a triplet of 1-byte values in a single 4-byte quantity.
The "gain" is the gain of the amplifier used during the readout of the CCD.
It is packed in bits 2^16-23.  The value will be 1, 2, 4, 8, or 16.
The "mode" is the CCD Readout mode.  It is packed in bits 2^8-15.  The values are:
1  Null
2  Short image
3  Long image
4  Piled-up Photodiode
5  Low Rate Photodiode
6  Windowed Timing
7  Photo-counting
8  Raw data
9  Bias map
10 Stop
The "waveform" is the CCD waveform ID.  It is packed in bits 2^0-7.

The "expo_time" is the nominal exposure time for this image.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized).

The "misc" field contains miscellaneous flag bits.
The defined bits are:
2^0-10  Not assigned.
2^11  When this bit is a 1, it indicates that 1 (or more) of the RA,Dec,Roll,Theta,Phi values
      were out of valid range (eg RA was 361 deg).
2^12  When this bit is a 1, it is in the catalog of sources to be blocked (internal use only).
2^13  When this bit is a 1, it indicates that this position is near a bright star (mag less than 6.5).
2^14  When this bit is a 1, it indicates that the distance between the position and the center of the galaxy
      is less than the sum of the position error radius and the galaxy radius AND that the position error radius
      is smaller than the galaxy radius.  This position is inside the (circularized) NGC galaxy (or nearly in).
2^15  When this bit is a 1, it indicates that the distance between the position and the center of the galaxy
      is less than the sum of the position error radius and the galaxy radius AND that the galaxy radius
      is smaller than the position error radius.  This galaxy is inside the position error circle (or nearly in).
2^16-19  Not assigned.
2^20  When this bit is a 1, it indicates that this was originally a SubThresh, but it is now converted to a real XRT_PROC_IMAGE.
2^21  When this bit is a 1, it indicates that this is an Image-trigger AND that it occurred
      during a time when the StarTracker had a Loss-of-Lock.
2^22  When this bit is a 1, it indicates that this Notice was generated as a result
      of an uploaded TOO sequence.  (Swift is commanded to look at this RA,Dec location
      and this Notice was generated as a direct result of that observation.)
2^23  Not assigned.
2^24  When this bit is a 1, the xrt_proc_image.fits file had an error in processing and does not exist.
2^25  Not assigned.
2^26  Not assigned.
2^27  Not assigned.
2^28  Not assigned.
2^29  This bit is set if this original 'raw' Image Noticed was forced-out via the watchdog timeout.
2^30  When this bit is a 1, then this Notice was ground-generated; else flight-generated.
      This will always be 0 for the 'raw' form of this Notice (type=69) and
      1 for the ground-processed form (type=78).
2^31  If there is a CRC error in one or more of the packets that went into
      the making of this Notice, then this bit is set to 1, else 0.

The "grb_in_xrt_y/z" is the GRB x-ray afterglow position in XRT coordinates.
The units are centi-pixels (but this was overkill because the intrinsic numbers
are in XRT CCD row,col pixels, so there is no non-zero fractional portion).

The "url" is just the filename portion of the URL -- the leading path
for the full URL is "http://gcn.gsfc.nasa.gov/gcn/notices_s/".
This item in the packet spans 17 longwords (ie 68 bytes) and contains 
the ASCII string of the filename portion of the URL.  This string
can be up to 67 characters long, and will always be null-terminated.

There is a total of 1 "spare" items in the packet (usually filled
with zeros).  It is located at item 10.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.



TYPE=79 PACKET CONTENTS:

The SWIFT_UVOT_IMAGE_PROCESSED packet (aka DarkBurst, aka Genie) consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=79)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       trig_obs_num integers        Trigger num & Observation num
long         5       ExpStart_tjd [days]          Truncated Julian Day
long         6       ExpStart_sod [centi-sec]     (int)(sssss.sss *100)
long         7       point_ra     [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         8       point_dec    [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         9       image_roll   [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         10      filter       integer         Which filter was used
long         11      expo_id      [sec]           Exposure_ID expressed in s/c_sec
long         12      X0           [det_coords]    0-2095
long         13      Y0           [det_coords]    0-2095
long         14      width        [det_coords]    0-2095
long         15      height       [det_coords]    0-2095
long         16      xy_grb       [det_coords]    0-2095; two packed
long         17      n_frames     integer         1-inf
long         18      swl_lwl      integers        2 4-bit integers
long         19      misc         bits            Misc stuff packed in here
long         20-21   spare[2]     integer         8 bytes for the future
quad_char    22-38   url[17]      chars           The URL for the image
long         39      pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 60, 61, 62, 63, ...84
that tells which type of packet this is and therefore what the contents are
(the contents change between the various types).  The types are:
76 = SWIFT_BAT_GRB_LIGHTCURVE_PROC, 77 = SWIFT_XRT_SPECTRUM_PROC,
78 = SWIFT_XRT_IMAGE_PROC, 79 = SWIFT_UVOT_IMAGE_PROC, and
80 = SWIFT_UVOT_SRCLIST_PROC. (Only the ground-processed Notices are listed here;
see before for all the other Swift Notices type values.)

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start from 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the burst_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "trig_obs_num" is a pair of uniquely identifying serial numbers (Trigger_num and Observation_num)
assigned by the on-board BAT flight software to each trigger.
The lower 24 bits contain the Trigger_number, and the upper 8 bits
contain the Observation_number. 
The Observation_number is always 0 since these are always (by definition)
the first observation of this new trigger.
The Trigger_number started with s/n=100,000 just after launch (Nov 04).

The "ExpStart_tjd" is the Truncated Julian Day of the start of the CCD integration,
eg. TJD=13370 is 01 Jan 2005.

The "ExpStart_sod" is the UT seconds-of-day (SOD) of the start of the CCD integration.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100 and then integerized).

The "point_ra" and "point_dec" are the RA,Dec coordinates (J2000)
of the pointing direction of the s/c at the start?/stop???
of the integration interval for the image.  These intrinsically floating point
quantities have been multiplied by 10,000 and then integerized to make
an integer quantity with units of 0.0001-degrees.

The "image_roll" is the Roll of the target.
This intrinsically floating point quantity has been multiplied by 10,000
and then integerized to make an integer quantity with units of 0.0001-degrees.

The "filter" is the filter used for the image.  The values are:
0  Blocked
1  UV_Grism
2  UVW2
3  V
4  UVM2
5  Vis_Grism
6  UVW1
7  U
8  Magnifier
9  B
10 White
11 unknown

The "expo_id" is the exposure_ID expressed in spacecraft seconds.
(???what is range?  ambiguity????)

The "X/Y_offset" are the coordinates of pixel_zero in the Image subarray.
The units are detector coordinates: 0-2095.

The "width" and "height" are the sizes of the Image subarray.
The units are detector coordinates: 0-2095.

The "xy_grb" is the position of the GRB in unbinned detector units.  The source
of this position is determined by the 2^28 bit in the "misc" field.

The "n_frames" is the total number of read-out frames that went into
compiling this Image.

The "swl_lwl" is the 'short word length' and the 'long word length' used
in the compress of the as-transmitted image packet.  There is no compression
in the GCN-version of this Notice (so they are pretty much useless by the time
it gets to GCN distribution).

The "n_stars" is the total number of stars in the source list.

The "misc" field contains miscellaneous flag bits.
The defined bits are:
2^0-3   Four bits encoding the pixel binning, eg =0 means 1x1 binning (ie none), =1 mean 2x2 binning,
        =2 means 4x4 binning, up to =6 mean 64x64 binning.
2^4-10  The height of the Detector Window used in the "Blue" Processing Electronics.
2^11    When this bit is a 1, it indicates that 1 (or more) of the RA,Dec,Roll,Theta,Phi values
        were out of valid range (eg RA was 361 deg).
2^12    When this bit is a 1, it is in the catalog of sources to be blocked (internal use only).
2^13    When this bit is a 1, there is a nearby bright star (mag less than 6.5).
2^14-19 Not assigned.
2^20    When this bit is a 1, it indicates that this was originally a SubThresh, but it is now converted to a real UVOT_PROC_IMAGE.
2^21    When this bit is a 1, it indicates that this is an Image-trigger AND that it occurred
        during a time when the StarTracker had a Loss-of-Lock.
2^22    When this bit is a 1, it indicates that this Notice was generated as a result
        of an uploaded TOO sequence.  (Swift is commanded to look at this RA,Dec location
        and this Notice was generated as a direct result of that observation.)
2^23    Not assigned.
2^24    When this bit is a 1, the uvot_sky_image.ps file had an error in processing and does not exist.
2^25    When this bit is a 1, the uvot_sources_image.fits file had an error in processing and does not exist.
2^26    When this bit is a 1, the uvot_catalog_image.fits file had an error in processing and does not exist.
2^27    When this bit is a 1, the uvot_field_image.ps file had an error in processing and does not exist.
2^28    When this bit is a 1, then the source of the GRB Position came from an XRT Position command,
        else it came from the Window Position in the Mode command.
2^29    This bit is set if this original 'raw' Image Noticed was forced-out via the watchdog timeout.
2^30    When this bit is a 1, then this Notice was ground-generated; else flight-generated.
        This will always be 0 for the 'raw' form of this Notice (type=72) and
        1 for the ground-processed form (type=79).
2^31    If there is a CRC error in one or more of the packets that went into
        the making of this Notice, then this bit is set to 1, else 0.

The "url" is just the filename portion of the URL -- the leading path
for the full URL is "http://gcn.gsfc.nasa.gov/gcn/notices_s/".
This item in the packet spans 17 longwords (ie 68 bytes) and contains 
the ASCII string of the filename portion of the URL.  This string
can be up to 67 characters long, and will always be null-terminated.

There are a total of 2 "spare" items in the packet (usually filled
with zeros).  They are located at items 20-21.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.



TYPE=80 PACKET CONTENTS:

The SWIFT_UVOT_SRCLIST_PROCESSED packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=80)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       trig_obs_num integers        Trigger num & Observation num
long         5       start_tjd    [days]          Truncated Julian Day
long         6       start_sod    [centi-sec]     (int)(sssss.sss *100)
long         7       point_ra     [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         8       point_dec    [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         9       image_roll   [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         10      filter       integer         Which filter was used
long         11      spare        integer         4 bytes for the future
long         12      x_max        [det_coords]    0-2095
long         13      y_max        [det_coords]    0-2095
long         14      n_stars      integer         1-infinity
long         15      x_offset     [det_coords]    0-2095
long         16      y_offset     [det_coords]    0-2095
long         17-18   spare[2]     integer         8 bytes for the future
long         19      misc         bits            Misc stuff packed in here
long         20-22   spare[2]     integer         8 bytes for the future
quad_char    22-38   url[17]      chars           The URL for the image
long         39      pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 60, 61, 62, 63, ...84
that tells which type of packet this is and therefore what the contents are
(the contents change between the various types).  The types are:
76 = SWIFT_BAT_GRB_LIGHTCURVE_PROC, 77 = SWIFT_XRT_SPECTRUM_PROC,
78 = SWIFT_XRT_IMAGE_PROC, 79 = SWIFT_UVOT_IMAGE_PROC, and
80 = SWIFT_UVOT_SRCLIST_PROC. (Only the ground-processed Notices are listed here;
see before for all the other Swift Notices type values.)

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start from 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the burst_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "trig_obs_num" is a pair of uniquely identifying serial numbers (Trigger_num and Observation_num)
assigned by the on-board BAT flight software to each trigger.
The lower 24 bits contain the Trigger_number, and the upper 8 bits
contain the Observation_number. 
The Observation_number is always 0 since these are always (by definition)
the first observation of this new trigger.
The Trigger_number started with s/n=100,000 just after launch (Nov 04).

The "Start_tjd" is the Truncated Julian Day of the start of the Finding Chart
CCD integration, eg. TJD=13370 is 01 Jan 2005.

The "Start_sod" is the UT seconds-of-day (SOD) of the start of the Finding Chart
CCD integration.  The units are in centi-seconds (ie the fl.pt. seconds were multiplied
by 100 and then integerized).

The "point_ra" and "point_dec" are the RA,Dec coordinates (J2000)
of the pointing direction of the s/c at the start?/stop???
of the integration interval for the image.  These intrinsically floating point
quantities have been multiplied by 10,000 and then integerized to make
an integer quantity with units of 0.0001-degrees.

The "image_roll" is the Roll of the target.
This intrinsically floating point quantity has been multiplied by 10,000
and then integerized to make an integer quantity with units of 0.0001-degrees.

The "filter" is the filter used for the image.  The values are:
0  Blocked
1  UV_Grism
2  UVW2
3  V
4  UVM2
5  Vis_Grism
6  UVW1
7  U
8  Magnifier
9  B
10 White
11 unknown

The "bkg_mean" is the mean value of the background level in the exposure.

The "x/y_max"  are the maximum extent of the image window expressed
in detector coordinates (0-2095).

The "n_stars" is the total number of stars in the source list.

The "x/y_offset0" are the X and Y offsets of the image window expressed
in detector coordinates (0-2095).

The "det_thresh" is the detection threshold used in this exposure.
The units are "dn" (digital number), and the range is 0 to ????.

The "photo_thresh" is the photometry threshold used in this exposure.
The units are "dn" (digital number), and the range is 0 to ????.

The "misc" field contains miscellaneous flag bits.
The defined bits are:
2^0-10  Not assigned.
2^11  When this bit is a 1, it indicates that 1 (or more) of the RA,Dec,Roll,Theta,Phi values
      were out of valid range (eg RA was 361 deg).
2^12  When this bit is a 1, it is in the catalog of sources to be blocked (internal use only).
2^13  When this bit is a 1, there is a nearby bright star (mag less than 6.5).
2^14-19  Not assigned.
2^20  When this bit is a 1, it indicates that this was originally a SubThresh, but it is now converted to a real UVOT_PROC_SL.
2^21  When this bit is a 1, it indicates that this is an Image-trigger AND that it occurred
      during a time when the StarTracker had a Loss-of-Lock.
2^22  When this bit is a 1, it indicates that this Notice was generated as a result
      of an uploaded TOO sequence.  (Swift is commanded to look at this RA,Dec location
      and this Notice was generated as a direct result of that observation.)
2^23  Not assigned.
2^24  When this bit is a 1, the uvot_sky_srclist.ps file had an error in processing and does not exist.
2^25  When this bit is a 1, the uvot_sources_srclist.fits file had an error in processing and does not exist.
2^26  When this bit is a 1, the uvot_catalog_srclist.fits file had an error in processing and does not exist.
2^27  When this bit is a 1, the uvot_field_srclist.ps file had an error in processing and does not exist.
2^28  Not assigned.
2^29  This bit is set if this original 'raw' SrcList Noticed was forced-out via the watchdog timeout.
2^30  When this bit is a 1, then this Notice was ground-generated; else flight-generated.
      This will always be 0 for the 'raw' form of this Notice (type=73) and
      1 for the ground-processed form (type=80).
2^31  If there is a CRC error in one or more of the packets that went into
      the making of this Notice, then this bit is set to 1, else 0.

The "url" is just the filename portion of the URL -- the leading path
for the full URL is "http://gcn.gsfc.nasa.gov/gcn/notices_s/".
This item in the packet spans 17 longwords (ie 68 bytes) and contains 
the ASCII string of the filename portion of the URL.  This string
can be up to 67 characters long, and will always be null-terminated.

There are a total of 3 "spare" items in the packet (usually filled
with zeros).  They are located at items 20-22.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.



TYPE=81 PACKET CONTENTS:

The SWIFT_UVOT_POS packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=81)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       trig_obs_num integers        Trigger num & Observation num
long         5       burst_tjd    [days]          Truncated Julian Day
long         6       burst_sod    [centi-sec]     (int)(sssss.sss *100)
long         7       burst_ra     [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         8       burst_dec    [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         9       burst_mag    [arb]           (int)(uvot_mag *100)
long         10      filter       integer         Filter_value: x-yy
long         11      burst_error  [0.0001-deg]    (int)(0.0 to 180.0 *10000)
long         12-15   spare[4]     integer         16 bytes for the future
long         16      hi_prec_err  [centi-arcsec]  hi precission error
long         17      spare        integer         24 bytes for the future
long         18      soln_status  bits            Type of source found
long         19      misc         bits            Misc stuff packed in here
long         20      mag_error    [arb]           (int)(mag_error *100)
long         21      uvot-xrt     [0.0001-deg]    Distance between XRT & UVOT pos
long         22-38   spare[17]    integer         68 bytes for the future
long         39      pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 60, 61, 62, 63, ...84
that tells which type of packet this is and therefore what the contents are
(the contents change between the various types).  The types are:
76 = SWIFT_BAT_GRB_LIGHTCURVE_PROC, 77 = SWIFT_XRT_SPECTRUM_PROC,
78 = SWIFT_XRT_IMAGE_PROC, 79 = SWIFT_UVOT_IMAGE_PROC, and
80 = SWIFT_UVOT_SRCLIST_PROC. (Only the ground-processed Notices are listed here;
see before for all the other Swift Notices type values.)

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start from 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the burst_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "trig_obs_num" is a pair of uniquely identifying serial numbers (Trigger_num and Observation_num)
assigned by the on-board BAT flight software to each trigger.
The lower 24 bits contain the Trigger_number, and the upper 8 bits
contain the Observation_number. 
The Observation_number is always 0 since these are always (by definition)
the first observation of this new trigger.
The Trigger_number started with s/n=100,000 just after launch (Nov 04).

The "burst_tjd" is the Truncated Julian Day of the Swift-UVOT data,
eg. TJD=13370 is 01 Jan 2005.

The "burst_sod" is the UT seconds-of-day (SOD) of the Swift-UVOT data.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized).

The "burst_ra" and "burst_dec" are the RA,Dec coordinates of the trigger/burst
as determined by the UVOT ground software (J2000 epoch).  These intrinsically
floating point quantities have been multiplied by 10,000 and then integerized
to make an integer quantity with units of 0.0001-degrees.

The "burst_mag" is the magnitude (in the specified filter) of the burst/afterglow.
The units are in centi-mag (ie the fl.pt. magnitude is multiplied by 100
and then integerized).

The "filter" is the filter used for the image.  The values are:
0  Blocked
1  UV_Grism
2  UVW2
3  V
4  UVM2
5  Vis_Grism
6  UVW1
7  U
8  Magnifier
9  B
10 White
11 unknown

The "burst_error" is the radius of the circle that will contain on average
90% of the bursts afterglow positions.  (See the "hi_prec_err" below.)

The "hi_prec_err" is the radius of the circle that will contain on average
90% of the bursts afterglow positions.  It contains statistical plus
systematic contributions.  This is different than the regular "burst_error"
in that the units are 'centi-arcsec' to remove quantization error of 0.36arcsec
introduced by the regular "burst_error" value.

The "soln_status" (aka "trig_id") field contains flag bits on the type of solution that was found.
The bit packing of the "soln_status" entry in the packet is:
Bit_Number     Item_name      Description
----------     ---------      -----------
0              point_src      0=No or 1=Yes, a point source was found, 0=extended src
1              grb            0=No or 1=Yes, it is a GRB
2              undef          Not yet defined.
3              catalog_src    0=No or 1=Yes, it is in the catalog (else unknown src)
4              undef          Not yet defined.
5              def_not_grb    0=No or 1=Yes, it is definitely not a GRB (bad label; its really a Retraction)
6-7            undef          Not yet defined.
8        G     gnd_cat_src    0=No or 1=Yes, BAT_loc is in the BAT ground catalog
9-27           undef          Not yet defined.
28       G     spatial_coinc  0=No or 1=Yes, there was a spatial coincidence with another event
29       G     temporal_coinc 0=No or 1=Yes, there was a temporal coincidence with another event
30             test_submit    0=No or 1=Yes, this is a test submission (internal use only)
31             undef          Not yet defined.

The "misc" field contains miscellaneous flag bits.
The defined bits are:
2^0-11  Not assigned.
2^12  When this bit is a 1, it is in the catalog of sources to be blocked (internal use only).
2^13  When this bit is a 1, it indicates that this position is near a bright star (mag less than 6.5).
//2^14  When this bit is a 1, it indicates that the distance between the position and the center of the galaxy
//      is less than the sum of the position error radius and the galaxy radius AND that the position error radius
//      is smaller than the galaxy radius.  This position is inside the (circularized) NGC galaxy (or nearly in).
//2^15  When this bit is a 1, it indicates that the distance between the position and the center of the galaxy
//      is less than the sum of the position error radius and the galaxy radius AND that the galaxy radius
//      is smaller than the position error radius.  This galaxy is inside the position error circle (or nearly in).
2^16-21  Not assigned.
2^22  When this bit is a 1, it indicates that this Notice was generated as a result
      of an uploaded TOO sequence.  (Swift is commanded to look at this RA,Dec location
      and this Notice was generated as a direct result of that observation.)
2^23  Not assigned.
2^24  Not assigned.
2^25  This is an updated position.  This is a position notice that changes/improves
      the position specified in the original UVOT_Position Notice.
2^26  Not assigned.
2^27  Not assigned.
2^28  Not assigned.
2^29  Not assigned.
2^30  When this bit is a 1, then this Notice was ground-generated; else flight-generated.
2^31  If there is a CRC error in one or more of the packets that went into
      the making of this Notice, then this bit is set to 1, else 0.

The "mag_error" is the uncertainty in the magnitude value (in the specified filter).
The units are in centi-mag (ie the fl.pt. mag_error is multiplied by 100
and then integerized).

The "uvot-xrt" is the angular distance between the UVOT position and the XRT position.
This intrinsically floating point quantity have been multiplied by 10,000 and
then integerized to make an integer quantity with units of 0.0001-degrees.

There are a total of 23 "spare" items in the packet (usually filled
with zeros).  They are located at items 12-15, 17, and 22-38.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.





TYPE=82 PACKET CONTENTS:

The SWIFT_BAT_GRB_POS_TEST packet consists of 40 four-byte quantities.
This is identical in form and content to SWIFT_BAT_GRB_POS_ACK(type=61)
in everything except the first 4-byte quantity: ie the pkt_type value
is 82 instead of 61.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=82)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       trig_obs_num integers        Trigger num & Observation num
long         5       burst_tjd    [days]          Truncated Julian Day
long         6       burst_sod    [centi-sec]     (int)(sssss.sss *100)
long         7       burst_ra     [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         8       burst_dec    [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         9       burst_flue   [counts]        Num events during trig window, 0 to inf
long         10      burst_ipeak  [cnts*ff]       Counts in image-plane peak, 0 to infinity
long         11      burst_error  [0.0001-deg]    (int)(0.0 to 180.0 *10000)
long         12      phi          [centi-deg]     (int)(0.0 to 359.9999 *100)
long         13      theta        [centi-deg]     (int)(0.0 to +70.0 *100)
long         14      integ_time   [4mSec]         Duration of the trigger interval, 1 to inf
long         15      spare        integer         4 bytes for the future
long         16      lon_lat      2_shorts        (int)(Longitude,Lattitude *100)
long         17      trig_index   integer         Rate_Trigger index
long         18      soln_status  bits            Type of source found
long         19      misc         bits            Misc stuff packed in here
long         20      image_signif [centi-sigma]   (int)(sig2noise *100)
long         21      rate_signif  [centi-sigma]   (int)(sig2noise *100)
long         22      bkg_flue     [counts]        Num events during the bkg interval, 0 to inf
long         23      bkg_start    [centi-sec]     (int)(sssss.sss *100)
long         24      bkg_dur      [centi-sec]     (int)(0-80,000 *100)
long         25-35   spare[11]    integer         44 bytes for the future
long         36      merit_0-3    integers        Merit params 0,1,2,3 (-127 to +127)
long         37      merit_4-7    integers        Merit params 4,5,6,7 (-127 to +127)
long         38      merit_8-9    integers        Merit params 8,9     (-127 to +127)
long         39      pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 60, 61, 62, 63, ...84
that tells which type of packet this is and therefore what the contents are
(the contents change between the various types).  The SWIFT_BAT_GRB_POS_TEST is 82.

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start from 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the burst_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "trig_obs_num" is a pair of uniquely identifying serial numbers (Trigger_num and Observation_num)
assigned by the on-board BAT flight software to each trigger.
The lower 24 bits contain the Trigger_number, and the upper 8 bits
contain the Observation_number. 
The Observation_number is always 0 since these are always (by definition)
the first observation of this new trigger (this new burst).
The Trigger_number started with s/n=100,000 just after launch (Nov 04).

The "burst_tjd" is the Truncated Julian Day of the Swift-BAT burst trigger,
eg. TJD=13370 is 01 Jan 2005.

The "burst_sod" is the UT seconds-of-day (SOD) of the Swift-BAT burst trigger.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized).

The "burst_ra" and "burst_dec" are the RA,Dec coordinates of the burst (or transient)
as determined by the BAT flight software (J2000 epoch).  These intrinsically
floating point quantities have been multiplied by 10,000 and then integerized
to make an integer quantity with units of 0.0001-degrees.

The "burst_flue" is the number of events (burst+background events)
during the 'integ_time' time interval of the successful trigger criterion.

The "burst_ipeak" is the height of the peak in the sky-image plane (ie the result
of the FFT,MaskConvolution,InvFFT calculation).  The units are counts
after a multiplicative factor of 0.5-0.7.

The "burst_error" is the radius of the circle that will contain on average
TBD% of the bursts.  The units are centi-degrees (ie the fl.pt. error was multiplied
by 100 and then integerized).  [Initially, this value will be hardwired 
at a constant value of 4arcmin (0.067deg) radius.  As the mission progresses,
this will be converted to a value which is dependent on the initial flux
of the burst.]

The "phi" and "theta" are the BAT instrument coordinates of the trigger (burst)
as determined by the BAT flight software program.  Phi is measured phi=0
along the +Y-axis of the s/c and increasing towards the -Z axis,
and theta is measured from the boresight of the BAT instrument.

The "integ_time" is the duration of the sampling interval
of the successful trigger criteria.  The value is the number of 4msec ticks,
eg a value of 16 indicates a 64-msec trigger criterion.

The "lon_lat" is the Longitude and Lattitude of the s/c at the time of the BAT trigger.
This 4-byte quantity is composed of two 2-byte integers.  The high-order short
is the latitude multiplied by 100 and integerized; the low-order short
is the longitude multiplied by 100 and integerized; ie both values are in centi-degrees.
They will always be 0.00,0.00 for this Test notice.

The "trig_index" is the index value (row_number in the BAT on-board flight s/w table)
of the trigger criteria that was the highest-significance successful trigger.

The "soln_status" field contains flag bits on the type of solution that was
found in the FFT sky image and about whether it matches anything in the
on-board source catalog.
The bit packing of the "soln_status" entry in the packet is:
Bit_Number     Item_name      Description
----------     ---------      -----------
0              point_src      0=No or 1=Yes, a point source was found
1              grb            0=No or 1=Yes, it is a GRB
2              interesting    0=No or 1=Yes, it is an interesting src (else boring)
3              catalog_src    0=No or 1=Yes, it is in the catalog (else unknown src)
4              image_trig     0=No or 1=Yes, it is an image trigger (else rate trig)
13             near_brt_star  0=No or 1=Yes, there is a nearby bright star
For these Test Notices, the soln_status will always be 3 plus the bright_star flag.

The "misc" field contains miscellaneous flag bits.
The defined bits are:
2^0-10 Not assigned.
2^11  When this bit is a 1, it indicates that 1 (or more) of the RA,Dec,Roll,Theta,Phi values
      were out of valid range (eg RA was 361 deg).
2^12  Not assigned.
2^13  When this bit is a 1, it indicates that this position is near a bright star (mag less than 6.5).
2^14  When this bit is a 1, it indicates that the distance between the position and the center of the galaxy
      is less than the sum of the position error radius and the galaxy radius AND that the position error radius
      is smaller than the galaxy radius.  This position is inside the (circularized) NGC galaxy (or nearly in).
2^15  When this bit is a 1, it indicates that the distance between the position and the center of the galaxy
      is less than the sum of the position error radius and the galaxy radius AND that the galaxy radius
      is smaller than the position error radius.  This galaxy is inside the position error circle (or nearly in).
2^16-24  Not assigned.
2^25  This is an updated position.  This is a position notice that changes/improves
      the position specified in the original BAT_Position Notice.
2^26-28 Not assigned.
2^29  When this bit is a 1, then this Notice was reconstituted; else flight-generated.
      If a BAT_LC is received from the s/c before a BAT_Pos is received,
      then bits and pieces from other Notices will be used to generate a reconstituted
      BAT_Pos Notice and it will be issued (with this flag bit set to 1).
      Always 0 for Test Notices.
2^30  When this bit is a 1, then this Notice was ground-generated; else flight-generated.
      A Gnd-generated Notice is issued anytime humans during ground operations
      have determined if the Notice should be issued.  This will typically be used
      during the early phases of the mission before the automatic TDRSS capability is enabled.
      Once the burst position has been validated by humans, a BAT_Pos Notice will be issued
      (with this flag bit set to 1).
      Always 1 for Test Notices.
2^31  If there is a CRC error in one or more of the packets that went into
      the making of this Notice, then this bit is set to 1, else 0.

The "image_signif" is the significance level of the peak in the FFT image plane.
This is the signal/noise (ie number of sigma) of the grb peak detection.
It has NOT been pre-scaled -- the units are sigma.

The "rate_signif" is the significance level of the initial 'rate trigger'.
This is the signal/noise (ie number of sigma) of the rate-trigger detection
multiplied by 100 and then integerized (ie units are centi-sigma).

The "bkg_flue" is the total number of events in the time interval used
for the background subtraction of the detector_rate_map (used for the image calc).
The units are counts.

The "bkg_start" is the starting time of the background interval.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized).  Note that not all trigger criteria have a background interval
specified, so this "bkg_start" value may be zero for some bursts.

The "bkg_dur" is the duration of the background interval.
The units are in centi-seconds.

The "merit_0-3", "merit_4-7", and "merit_8-9" are the 10 merit parameters
for this burst.  Their values range from -127 to +127.  These 'merit params'
are different than the 'merit_value' that is in the SWIFT_FOM_OBS and SWIFT_SC_SLEW
messages.  There are 4, 4, & 2 values packed into these 3 longs.  The packing order
is "0" in the lowest_address byte in the long, "1" is in the next_lowest, etc.

There are a total of 12 "spare" items in the packet (usually filled
with zeros).  They are located at items 15 and 25-35.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.



PACKET ITEM DESCRIPTIONS for type=83:

The SWIFT_POINTDIR packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=83)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       trig_obs_num integers        Trigger num & Observation num
long         5       slew_tjd     [days]          Truncated Julian Day
long         6       slew_sod     [centi-sec]     (int)(sssss.sss *100)
long         7       point_ra     [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         8       point_dec    [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         9       point_roll   [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         10      spare        integer         4 bytes for the future
long         11      bat_mode     integer         BAT inst-mode for AT response
long         12      xrt_mode     integer         XRT inst-mode for AT response
long         13      uvot_mode    integer         UVOT inst-mode for AT response
long         14      obs_time     [centi-sec]     (int)(sssss.sss *100)
long         15      merit        [centi-merit]   Merit_value *100
long         16      lon_lat      2_shorts        (int)(Longitude,Lattitude *100)
long         17-18   spare[2]     integer         9 bytes for the future
long         19      misc         bits            Misc stuff packed in here
long         20-21   spare[2]     integer         8 bytes for the future
quad_char    22-38   tgtname[17]  chars           The name of the target
long         39      pkt_term     integer         Pkt Termination (always = \n)



PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 60, 61, 62, 63, ...84
that tells which type of packet this is and therefore what the contents are
(the contents change between the various types).  The SWIFT_POINTDIR type is 83.

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start from 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the burst_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "trig_obs_num" is a pair of uniquely identifying serial numbers (Trigger_num and Observation_num)
assigned by the on-board BAT flight software to each trigger.
The lower 24 bits contain the Trigger_number, and the upper 8 bits
contain the Observation_number. 

The "slew_tjd" is the TJD when the next s/c slew starts
eg. TJD=13370 is 01 Jan 2005.

The "slew_sod" is the seconds-of-day (SOD) when the next s/c slew starts.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized).

The "point_ra", "point_dec" and "point_roll" are the RA,Dec and Roll (J2000) coordinates
to which the spacecraft will point after this slew.  This is the pointing
direction of the s/c.  The target_destiation accuracy of the s/c is 3arcmin worst case.
So for any given slew, the s/c will be actually looking at an RA,Dec location
which is within 0-3 arcmin of the location given in this packet.
The units are 10^04 deg, ie the values where multiplied by 10000 and integerized.

The "bat_mode", "xrt_mode", and "uvot_mode" are the 3 sets of flagbits
the determine how each instrument responds for data-taking for each AT burst response.
Each mode is in the bottom 16 bits fo each 4-byte field.

The "obs_time" is the amount of time the sc will observe that new location.
Since the "slew_sod" is the start of the slew to the new location,
in practice the sc will actually be on target 50-100 sec less than the amount
of time specified in this field.  The units are in centi-seconds
(ie the fl.pt. seconds were multiplied by 100 and then integerized).

The "merit_value" is the total calculated value for the given burst.
It is a fl.pt. quantity which can theoretically range from -infinity to
+infinity, but given normal operations, we expect it to range from about
-100 to +300.  And for the first 6-12 months of flight-ops, it will be
forced to be a constant value of 100 for all bursts.

The "lon_lat" is the Longitude and Lattitude of the s/c at the time of the BAT trigger.
This 4-byte quantity is composed of two 2-byte integers.  The high-order short
is the latitude multiplied by 100 and integerized; the low-order short
is the longitude multiplied by 100 and integerized; ie both values are in centi-degrees.

The "misc" field contains miscellaneous flag bits.
The defined bits are:
2^0-12  Not assigned.
2^13    When this bit is 1, there is a nearby bright star (mag less than 6.5).
2^14-29 Not assigned.
2^30    When this bit is a 1, then this Notice was ground-generated; else flight-generated.
        This will always be 1 since the POINTDIR information is extracted from ground processing
        of ground files.
2^31    Not assigned.

The "tgtname" is the name of the object that the sc will be observing next.
It is a name chosen by the science planners at the MOC.  The name does not
always fit the IAU standard naming convention (due to limitations in the 
planning s/w).  This item in the packet spans 17 longwords (ie 68 bytes) and
contains the ASCII string of the target_name.  This string can be up to
67 characters long, and will always be null-terminated.

There are a total of 7 "spare" items in the packet (usually filled
with zeros).  They are located at items 10 and 16-21.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.





TYPE=84 PACKET CONTENTS:

The SWIFT_BAT_TRANS packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=84)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       trig_obs_num integers        Trigger num & Observation num
long         5       trans_tjd    [days]          Truncated Julian Day
long         6       trans_sod    [centi-sec]     (int)(sssss.sss *100)
long         7       trans_ra     [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         8       trans_dec    [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         9       trans_flue   [counts]        Num events during trig window, 0 to inf
long         10      trans_ipeak  [cnts*ff]       Counts in image-plane peak, 0 to infinity
long         11      trans_error  [0.0001-deg]    (int)(0.0 to 180.0 *10000)
long         12      phi          [centi-deg]     (int)(0.0 to 359.9999 *100)
long         13      theta        [centi-deg]     (int)(0.0 to +70.0 *100)
long         14      integ_time   [4mSec]         Duration of the trigger interval, 1 to inf
long         15      spare        integer         4 bytes for the future
long         16      lon_lat      2_shorts        (int)(Longitude,Lattitude *100)
long         17      trig_index   integer         Rate_Trigger index
long         18      soln_status  bits            Type of source found
long         19      misc         bits            Misc stuff packed in here
long         20      image_signif [centi-sigma]   (int)(sig2noise *100)
long         21      rate_signif  [centi-sigma]   (int)(sig2noise *100)
long         22      bkg_flue     [counts]        Num events during the bkg interval, 0 to inf
long         23      bkg_start    [centi-sec]     (int)(sssss.sss *100)
long         24      bkg_dur      [centi-sec]     (int)(0-80,000 *100)
long         25      cat_num      integer         On-board cat match ID number
long         26-35   spare[10]    integer         40 bytes for the future
long         36      merit_0-3    integers        Merit params 0,1,2,3 (-127 to +127)
long         37      merit_4-7    integers        Merit params 4,5,6,7 (-127 to +127)
long         38      merit_8-9    integers        Merit params 8,9     (-127 to +127)
long         39      pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 60, 61, 62, 63, ...84
that tells which type of packet this is and therefore what the contents are
(the contents change between the various types).  The SWIFT_BAT_TRANS type is 84. 

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start from 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the burst_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "trig_obs_num" is a pair of uniquely identifying serial numbers (Trigger_num and Observation_num)
assigned by the on-board BAT flight software to each trigger.
The lower 24 bits contain the Trigger_number, and the upper 8 bits
contain the Observation_number. 
The Observation_number is always 0 since these are always (by definition)
the first observation of this new trigger (this new burst).
The Trigger_number started with s/n=100,000 just after launch (Nov 04).

The "trans_tjd" is the Truncated Julian Day of the Swift-BAT transient trigger,
eg. TJD=13370 is 01 Jan 2005.

The "trans_sod" is the UT seconds-of-day (SOD) of the Swift-BAT transient trigger.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized).

The "trans_ra" and "trans_dec" are the RA,Dec coordinates of the transient (J2000 epoch)
as determined by the BAT flight software.  These intrinsically floating point quantities
have been multiplied by 10,000 and then integerized to make an integer quantity
with units of 0.0001-degrees.

The "trans_flue" is the number of events (trans+background events)
during the 'integ_time' time interval of the successful trigger criterion.

The "trans_ipeak" is the height of the peak in the sky-image plane (ie the result
of the FFT,MaskConvolution,InvFFT calculation).  The units are counts
after a multiplicative factor of about 0.5-0.7.

The "trans_error" is the radius of the circle that will contain on average
TBD% of the bursts.  The units are 0.0001-degrees (ie the fl.pt. error was multiplied
by 10000 and then integerized).  [Initially, this value will be hardwired 
at a constant value of 4arcmin (0.067deg) radius.  As the mission progresses,
this will be converted to a value which is dependent on the initial flux
of the burst.]

The "phi" and "theta" are the BAT instrument coordinates of the transient
as determined by the BAT flight software program.  Phi is measured phi=0
along the +Y-axis of the s/c and increasing towards the -Z axis,
and theta is measured from the boresight of the BAT instrument.

The "integ_time" is the duration of the sampling interval
of the successful trigger criteria.  The value is the number of 4msec ticks,
eg a value of 16 indicates a 64-msec trigger criterion.

The "lon_lat" is the Longitude and Lattitude of the s/c at the time of the BAT trigger.
This 4-byte quantity is composed of two 2-byte integers.  The high-order short
is the latitude multiplied by 100 and integerized; the low-order short
is the longitude multiplied by 100 and integerized; ie both values are in centi-degrees.

The "trig_index" is the index value (row_number in the BAT on-board flight s/w table)
of the trigger criteria that was the highest-significance successful trigger.

The "soln_status" (aka trigger_id) field contains (a) flag bits on the type of solution
that was found in the on-board image and about whether it matches anything in the
on-board source catalog, and (b) flag bits that are assigned in the GCN ground-processing.
The bit packing of the "soln_status" entry in the packet is:
BitNum  F-v-G  Item_name      Description
------  ----   ---------      -----------
0        F     point_src      0=No or 1=Yes, a point source was found
1        F     grb            0=No or 1=Yes, it is a GRB
2        F     interesting    0=No or 1=Yes, it is an interesting src (ie a flaring known src)
3        F     flt_cat_src    0=No or 1=Yes, it is in the flight catalog (else unknown src)
4        F     image_trig     0=No or 1=Yes, it is an image trigger (else rate trig)
5        G     def_not_grb    0=No or 1=Yes, it is definitely not a GRB (bad label; its really a Retraction)
6        G     uncert_grb     0=No or 1=Yes, it is probably not a GRBorTransient (hi bkg level, ie near SAA)
7        G     uncert_grb     0=No or 1=Yes, it is probably not a GRBorTransient (low image significance; less then 7.0 sigma)
8        G     gnd_cat_src    0=No or 1=Yes, it is in the ground catalog
9        G     uncert_grb     0=No or 1=Yes, it is probably not a GRBorTransient (negative bkg slope, exit SAA)
10       G     st_loss_lock   0=No or 1=Yes, StarTracker not locked so trigger probably bogus
11       G     uncert_grb     0=No or 1=Yes, it is really probably not a GRBorTransient (VERY low image significance; less than 6.5 sigma)
12       G     blk_cat_src    0=No or 1=Yes, it is in the catalog of sources to be blocked (internal use only)
13       G     near_brt_star  0=No or 1=Yes, there is a nearby bright star
14-27    -     spare          spare for future use
28       G     spatial_coinc  0=No or 1=Yes, there was a spatial coincidence with another event
29       G     temporal_coinc 0=No or 1=Yes, there was a temporal coincidence with another event
30       G     test_submit    0=No or 1=Yes, this is a test submission (internal use only)
31       -     spare          spare for future use
The columns are:  "BitNum" is the 2^N location within the 4-byte word; "F-v-G" is Flight-assigned vs
Ground-assigned;  "Item_name" is the short-description name of the quantity;  and
"Description" is the longer description of the quantity.

The "misc" field contains miscellaneous flag bits.
The defined bits are:
2^0-10 Not assigned.
2^11  When this bit is a 1, it indicates that 1 (or more) of the RA,Dec,Roll,Theta,Phi values
      were out of valid range (eg RA was 361 deg).
2^12  Not assigned.
2^13  When this bit is a 1, it indicates that this position is near a bright star (mag less than 6.5).
2^14  When this bit is a 1, it indicates that the distance between the position and the center of the galaxy
      is less than the sum of the position error radius and the galaxy radius AND that the position error radius
      is smaller than the galaxy radius.  This position is inside the (circularized) NGC galaxy (or nearly in).
2^15  When this bit is a 1, it indicates that the distance between the position and the center of the galaxy
      is less than the sum of the position error radius and the galaxy radius AND that the galaxy radius
      is smaller than the position error radius.  This galaxy is inside the position error circle (or nearly in).
2^16-20  Not assigned.
2^21  When this bit is a 1, it indicates that this is an Image-trigger AND that it occurred
      during a time when the StarTracker had a Loss-of-Lock.
2^22-24  Not assigned.
2^25  This is an updated position.  This is a position notice that changes/improves
      the position specified in the original BAT_TRANS_Position Notice.
2^26-27 Not assigned.
2^28  When this bit is a 1, then this Notice was derived from a SERS message; else flight-generated.
      If none of the regular position-containing messages are received via the real-time
      TDRSS comm channel (eg BAT_Pos_Ack, BAT_LC, etc), then the SERS messages
      (which come from pipe-line processing of the Malindi playback downlinked data)
      will be used to generate a reconstituted BAT_Pos_Ack Notice.  But since
      the SERS message only contains a subset of the information is a regular TDRSS
      messages, these SERS-reconstituted Notices will have (tbd) fields undefined.
      These undefined fields are: tbd, tbd, tbd....
2^29  When this bit is a 1, then this Notice was reconstituted; else flight-generated.
      If a BAT_LC is received from the s/c before a BAT_Pos is received,
      then bits and pieces from other Notices will be used to generate a reconstituted
      BAT_Pos Notice and it will be issued (with this flag bit set to 1).
2^30  When this bit is a 1, then this Notice was ground-generated; else flight-generated.
      A Gnd-generated Notice is issued anytime humans during ground operations
      have determined if the Notice should be issued.  This will typically be used
      during the early phases of the mission before the automatic TDRSS capability is enabled.
      Once the burst position has been validated by humans, a BAT_Trans Notice will be issued
      (with this flag bit set to 1).
2^31  If there is a CRC error in one or more of the packets that went into
      the making of this Notice, then this bit is set to 1, else 0.

The "image_signif" is the significance level of the peak in the FFT image plane.
This is the signal/noise (ie number of sigma) of the grb peak detection
multiplied by 100 and then integerized (ie units are centi-sigma).

The "rate_signif" is the significance level of the initial 'rate trigger'.
This is the signal/noise (ie number of sigma) of the rate-trigger detection
multiplied by 100 and then integerized (ie units are centi-sigma).

The "bkg_flue" is the total number of events in the time interval used
for the background subtraction of the detector_rate_map (used for the image calc).
The units are counts.

The "bkg_start" is the starting time of the background interval.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized).  Note that not all trigger criteria have a background interval
specified, so this "bkg_start" value may be zero for some bursts.

The "bkg_dur" is the duration of the background interval.
The units are in centi-seconds.

The "merit_0-3", "merit_4-7", and "merit_8-9" are the 10 merit parameters
for this burst.  Their values range from -127 to +127.  These 'merit params'
are different than the 'merit_value' that is in the SWIFT_FOM_OBS and SWIFT_SC_SLEW
messages.  There are 4, 4, & 2 values packed into these 3 longs.  The packing order
is "0" in the lowest_address byte in the long, "1" is in the next_lowest, etc.

The "cat_num" is the catalog number in the on-board catalog for sources
that match a known source in the on-board catalog.  (Typically it is a number
in the low 10,000s, 12,000s, or 15,000s.)

There are a total of 11 "spare" items in the packet (usually filled with zeros).

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.





TYPE=85 PACKET CONTENTS:   NOT AVAILABLE TO THE PUBLIC

The SWIFT_XRT_THRESHPIX packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=85)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       trig_obs_num integers        Trigger num & Observation num
long         5       burst_tjd    [days]          Truncated Julian Day
long         6       burst_sod    [centi-sec]     (int)(sssss.sss *100)
long         7       point_ra     [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         8       point_dec    [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         9       expo_time    [0.001-sec]     (int)(time*1000.0)
long         10      stop_date    [days]          Truncated Julian Day
long         11      stop_time    [centi-sec]     (int)(sssss.sss *100)
long         12      seq_num      integer         Serial number of message (1-3)
long         13-18   spare[6]     integer         24 bytes for the future
long         19      misc         bits            Misc stuff packed in here
long         20-21   spare[2]     integer         8 bytes for the future
quad_char    22-38   url[17]      chars           The URL for the image
long         39      pkt_term     integer         Pkt Termination (always = \n)



PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 60, 61, 62, 63, ...88
that tells which type of packet this is and therefore what the contents are
(the contents change between the various types).  The SWIFT_XRT_THRESHIX type is 85. 

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start from 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the burst_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "trig_obs_num" is a pair of uniquely identifying serial numbers (Trigger_num and Observation_num)
assigned by the on-board BAT flight software to each trigger.
The lower 24 bits contain the Trigger_number, and the upper 8 bits
contain the Observation_number. 
The Observation_number is always 0 since these are always (by definition)
the first observation of this new trigger (this new burst).
The Trigger_number started with s/n=100,000 just after launch (Nov 04).

The "burst_tjd" is the Truncated Julian Day of the Swift-BAT transient trigger,
eg. TJD=13370 is 01 Jan 2005.

The "burst_sod" is the UT seconds-of-day (SOD) of the Swift-BAT transient trigger.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized).

The "point_ra" and "point_dec" are the RA,Dec coordinates (J2000)
of the pointing direction of the s/c at the start?/stop???
of the integration interval for the image.  These intrinsically floating point
quantities have been multiplied by 10,000 and then integerized to make
an integer quantity with units of 0.0001-degrees.

The "misc" field contains miscellaneous flag bits.
The defined bits are:
2^0-10  Not assigned.
2^11  When this bit is a 1, it indicates that 1 (or more) of the RA,Dec,Roll,Theta,Phi values
      were out of valid range (eg RA was 361 deg).
2^12-19  Not assigned.
2^20  When this bit is a 1, it indicates that this was originally a SubThresh, but it is now converted to a real XRT_TP.
2^21  When this bit is a 1, it indicates the original BAT trigger was an ImageTrigger AND the Startracker was in Loss-of-Lock.
2^22  When this bit is a 1, it indicates that this Notice was generated as a result
      of an uploaded TOO sequence.  (Swift is commanded to look at this RA,Dec location
      and this Notice was generated as a direct result of that observation.)
2^23  Not assigned.
2^24  Not assigned.
2^25  Not assigned.
2^26  Not assigned.
2^27  Not assigned.
2^28  Not assigned.
2^29  This bit is set if this XRT_ThreshPix Noticed was forced-out via the watchdog timeout.
2^30  Not assigned.
2^31  If there is a CRC error in one or more of the packets that went into
      the making of this Notice, then this bit is set to 1, else 0.

The "url" is just the filename portion of the URL -- the leading path
for the full URL is "http://gcn.gsfc.nasa.gov/gcn/notices_s/".
This item in the packet spans 17 longwords (ie 68 bytes) and contains 
the ASCII string of the filename portion of the URL.  This string
can be up to 67 characters long, and will always be null-terminated.

There are a total of 29 "spare" items in the packet (usually filled
with zeros).  They are located at items 9-18 and 20-38.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.




TYPE=86 PACKET CONTENTS:   NOT AVAILABLE TO THE PUBLIC

The SWIFT_XRT_THRESHPIX_PROC packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=86)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       trig_obs_num integers        Trigger num & Observation num
long         5       burst_tjd    [days]          Truncated Julian Day
long         6       burst_sod    [centi-sec]     (int)(sssss.sss *100)
long         7       point_ra     [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         8       point_dec    [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         9       expo_time    [0.001-sec]     (int)(time*1000.0)
long         10      stop_date    [days]          Truncated Julian Day
long         11      stop_time    [centi-sec]     (int)(sssss.sss *100)
long         12      seq_num      integer         Serial number of message (1-3)
long         13-18   spare[6]     integer         24 bytes for the future
long         19      misc         bits            Misc stuff packed in here
long         20-21   spare[2]     integer         8 bytes for the future
quad_char    22-38   url[17]      chars           The URL for the image
long         39      pkt_term     integer         Pkt Termination (always = \n)



PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 60, 61, 62, 63, ...88
that tells which type of packet this is and therefore what the contents are
(the contents change between the various types).  The SWIFT_XRT_THRESHIX_PROC type is 86. 

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start from 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the burst_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "trig_obs_num" is a pair of uniquely identifying serial numbers (Trigger_num and Observation_num)
assigned by the on-board BAT flight software to each trigger.
The lower 24 bits contain the Trigger_number, and the upper 8 bits
contain the Observation_number. 
The Observation_number is always 0 since these are always (by definition)
the first observation of this new trigger (this new burst).
The Trigger_number started with s/n=100,000 just after launch (Nov 04).

The "burst_tjd" is the Truncated Julian Day of the Swift-BAT transient trigger,
eg. TJD=13370 is 01 Jan 2005.

The "burst_sod" is the UT seconds-of-day (SOD) of the Swift-BAT transient trigger.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized).

The "point_ra" and "point_dec" are the RA,Dec coordinates (J2000)
of the pointing direction of the s/c at the start?/stop???
of the integration interval for the image.  These intrinsically floating point
quantities have been multiplied by 10,000 and then integerized to make
an integer quantity with units of 0.0001-degrees.

The "misc" field contains miscellaneous flag bits.
The defined bits are:
2^0-10  Not assigned.
2^11  When this bit is a 1, it indicates that 1 (or more) of the RA,Dec,Roll,Theta,Phi values
      were out of valid range (eg RA was 361 deg).
2^12-19 Not assigned.
2^20  When this bit is a 1, it indicates that this was originally a SubThresh, but it is now converted to a real XRT_PROC_TP.
2^21  Not assigned.
2^22  When this bit is a 1, it indicates that this Notice was generated as a result
      of an uploaded TOO sequence.  (Swift is commanded to look at this RA,Dec location
      and this Notice was generated as a direct result of that observation.)
2^23  Not assigned.
2^24  Not assigned.
2^25  Not assigned.
2^26  Not assigned.
2^27  Not assigned.
2^28  Not assigned.
2^29  This bit is set if this original 'raw' ThreshPix Noticed was forced-out via the watchdog timeout.
2^30  Not assigned.
2^31  If there is a CRC error in one or more of the packets that went into
      the making of this Notice, then this bit is set to 1, else 0.

The "url" is just the filename portion of the URL -- the leading path
for the full URL is "http://gcn.gsfc.nasa.gov/gcn/notices_s/".
This item in the packet spans 17 longwords (ie 68 bytes) and contains 
the ASCII string of the filename portion of the URL.  This string
can be up to 67 characters long, and will always be null-terminated.

There are a total of 29 "spare" items in the packet (usually filled
with zeros).  They are located at items 9-18 and 20-38.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.





TYPE=87 PACKET CONTENTS:   NOT AVAILABLE TO THE PUBLIC

The SWIFT_XRT_SPER packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=87)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       trig_obs_num integers        Trigger num & Observation num
long         5       burst_tjd    [days]          Truncated Julian Day
long         6       burst_sod    [centi-sec]     (int)(sssss.sss *100)
long         7       point_ra     [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         8       point_dec    [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         9       expo_time    [0.001-sec]     (int)(time*1000.0)
long         10      stop_date    [days]          Truncated Julian Day
long         11      stop_time    [centi-sec]     (int)(sssss.sss *100)
long         12      num_pkt      integer         Number of pkts in integration
long         13-18   spare[6]     integer         24 bytes for the future
long         19      misc         bits            Misc stuff packed in here
long         20      num_evt      integer         Number of events in the integration
long         21      spare        integer         4 bytes for the future
quad_char    22-38   url[17]      chars           The URL for the image
long         39      pkt_term     integer         Pkt Termination (always = \n)



PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 60, 61, 62, 63, ...88
that tells which type of packet this is and therefore what the contents are
(the contents change between the various types).  The SWIFT_XRT_SPER type is 87. 

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start from 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the burst_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "trig_obs_num" is a pair of uniquely identifying serial numbers (Trigger_num and Observation_num)
assigned by the on-board BAT flight software to each trigger.
The lower 24 bits contain the Trigger_number, and the upper 8 bits
contain the Observation_number. 
The Observation_number is always 0 since these are always (by definition)
the first observation of this new trigger (this new burst).
The Trigger_number started with s/n=100,000 just after launch (Nov 04).

The "burst_tjd" is the Truncated Julian Day of the Swift-BAT transient trigger,
eg. TJD=13370 is 01 Jan 2005.

The "burst_sod" is the UT seconds-of-day (SOD) of the Swift-BAT transient trigger.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized).

The "point_ra" and "point_dec" are the RA,Dec coordinates (J2000)
of the pointing direction of the s/c at the start?/stop???
of the integration interval for the image.  These intrinsically floating point
quantities have been multiplied by 10,000 and then integerized to make
an integer quantity with units of 0.0001-degrees.

The "misc" field contains miscellaneous flag bits.
The defined bits are:
2^0-9 Not assigned.
2^10  When this bit is a 1, it indicates that StarTracker not locked (at the time the SPER is pushed out)
2^11  When this bit is a 1, it indicates that 1 (or more) of the RA,Dec,Roll,Theta,Phi values
      were out of valid range (eg RA was 361 deg).
2^12-20  Not assigned.
2^21  When this bit is a 1, it indicates the original BAT trigger was an ImageTrigger AND the Startracker was in Loss-of-Lock.
2^22  When this bit is a 1, it indicates that this Notice was generated as a result
      of an uploaded TOO sequence.  (Swift is commanded to look at this RA,Dec location
      and this Notice was generated as a direct result of that observation.)
2^23  Not assigned.
2^24  Not assigned.
2^25  Not assigned.
2^26  Not assigned.
2^27  Not assigned.
2^28  Not assigned.
2^29  This bit is set if this XRT_SPER Noticed was forced-out via the watchdog timeout.
2^30  Not assigned.
2^31  If there is a CRC error in one or more of the packets that went into
      the making of this Notice, then this bit is set to 1, else 0.

The "num_evt" is the number of photon events in the integration.

The "url" is just the filename portion of the URL -- the leading path
for the full URL is "http://gcn.gsfc.nasa.gov/gcn/notices_s/".
This item in the packet spans 17 longwords (ie 68 bytes) and contains 
the ASCII string of the filename portion of the URL.  This string
can be up to 67 characters long, and will always be null-terminated.

There are a total of 28 "spare" items in the packet (usually filled
with zeros).  They are located at items 9-18 and 20-38.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.






TYPE=88 PACKET CONTENTS:   NOT AVAILABLE TO THE PUBLIC

The SWIFT_XRT_SPER_PROC packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=88)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       trig_obs_num integers        Trigger num & Observation num
long         5       burst_tjd    [days]          Truncated Julian Day
long         6       burst_sod    [centi-sec]     (int)(sssss.sss *100)
long         7       point_ra     [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         8       point_dec    [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         9       expo_time    [0.001-sec]     (int)(time*1000.0)
long         10      stop_date    [days]          Truncated Julian Day
long         11      stop_time    [centi-sec]     (int)(sssss.sss *100)
long         12      seq_num      integer         Serial number of message (1-3)
long         13-18   spare[6]     integer         24 bytes for the future
long         19      misc         bits            Misc stuff packed in here
long         20      num_evt      integer         Number of events in the integration
long         21      spare        integer         4 bytes for the future
quad_char    22-38   url[17]      chars           The URL for the image
long         39      pkt_term     integer         Pkt Termination (always = \n)



PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 60, 61, 62, 63, ...88
that tells which type of packet this is and therefore what the contents are
(the contents change between the various types).  The SWIFT_XRT_SPER_PROC type is 88. 

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start from 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the burst_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "trig_obs_num" is a pair of uniquely identifying serial numbers (Trigger_num and Observation_num)
assigned by the on-board BAT flight software to each trigger.
The lower 24 bits contain the Trigger_number, and the upper 8 bits
contain the Observation_number. 
The Observation_number is always 0 since these are always (by definition)
the first observation of this new trigger (this new burst).
The Trigger_number started with s/n=100,000 just after launch (Nov 04).

The "burst_tjd" is the Truncated Julian Day of the Swift-BAT transient trigger,
eg. TJD=13370 is 01 Jan 2005.

The "burst_sod" is the UT seconds-of-day (SOD) of the Swift-BAT transient trigger.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized).

The "point_ra" and "point_dec" are the RA,Dec coordinates (J2000)
of the pointing direction of the s/c at the start?/stop???
of the integration interval for the image.  These intrinsically floating point
quantities have been multiplied by 10,000 and then integerized to make
an integer quantity with units of 0.0001-degrees.

The "misc" field contains miscellaneous flag bits.
The defined bits are:
2^0-21  Not assigned.
2^22  When this bit is a 1, it indicates that this Notice was generated as a result
      of an uploaded TOO sequence.  (Swift is commanded to look at this RA,Dec location
      and this Notice was generated as a direct result of that observation.)
2^23  Not assigned.
2^24  Not assigned.
2^25  Not assigned.
2^26  Not assigned.
2^27  Not assigned.
2^28  Not assigned.
2^29  This bit is set if this original 'raw' SPER Noticed was forced-out via the watchdog timeout.
2^30  Not assigned.
2^31  If there is a CRC error in one or more of the packets that went into
      the making of this Notice, then this bit is set to 1, else 0.

The "num_evt" is the number of photon events in the integration.

The "url" is just the filename portion of the URL -- the leading path
for the full URL is "http://gcn.gsfc.nasa.gov/gcn/notices_s/".
This item in the packet spans 17 longwords (ie 68 bytes) and contains 
the ASCII string of the filename portion of the URL.  This string
can be up to 67 characters long, and will always be null-terminated.

There are a total of 28 "spare" items in the packet (usually filled
with zeros).  They are located at items 9-18 and 20-38.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.




TYPE=89 PACKET CONTENTS:

The SWIFT_UVOT_POS_NACK packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=89)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       trig_obs_num integers        Trigger num & Observation num
long         5       trig_tjd     [days]          Truncated Julian Day
long         6       trig_sod     [centi-sec]     (int)(sssss.sss *100)
long         7       src_ra       [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         8       src_dec      [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         9       mag_lim      [arb]           (int)(uvot_mag_lim *100)
long         10      filter       integer         Filter_value: x-yy
long         11      src_error    [0.0001-deg]    (int)(0.0 to 180.0 *10000)
long         12      source       integer         1-3
long         13-17   spare[5]     integer         20 bytes for the future
long         18      soln_status  bits            Type of source found
long         19      misc         bits            Misc stuff packed in here
long         20      lim_sigma    [arb]           (int)(mag_lim_sigma *100)
long         21-38   spare[18]    integer         72 bytes for the future
long         39      pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 60, 61, 62, 63, ...84
that tells which type of packet this is and therefore what the contents are
(the contents change between the various types).

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start from 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the burst_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "trig_obs_num" is a pair of uniquely identifying serial numbers (Trigger_num and Observation_num)
assigned by the on-board BAT flight software to each trigger.
The lower 24 bits contain the Trigger_number, and the upper 8 bits
contain the Observation_number. 
The Observation_number is always 0 since these are always (by definition)
the first observation of this new trigger.
The Trigger_number started with s/n=100,000 just after launch (Nov 04).

The "trig_tjd" is the Truncated Julian Day of the Swift-UVOT data,
eg. TJD=13370 is 01 Jan 2005.

The "trig_sod" is the UT seconds-of-day (SOD) of the Swift-UVOT data.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized).

The "src_ra" and "src_dec" are the RA,Dec coordinates (J2000) of the center
of the error circle that was searched to determine this limiting magnitude from "source".
These intrinsically floating point quantities have been multiplied by 10,000 and
then integerized to make an integer quantity with units of 0.0001-degrees.

The "mag_lim" is the lower limit magnitude (in the specified filter)
of the burst observation (ie the afterglow has to be fainter than the 'mag_lim' value).
The units are in centi-mag (ie the fl.pt. magnitude is multiplied by 100
and then integerized).

The "filter" is the filter used for the image.  The values are:
0  Blocked
1  UV_Grism
2  UVW2
3  V
4  UVM2
5  Vis_Grism
6  UVW1
7  U
8  Magnifier
9  B
10 White
11 unknown

The "src_error" is the radius of the location error circle that was searched in UVOT
to yield the quoted limiting magnitude.

The "source" field indicates the source of the RA,Dec-location error circle
that was searched to yield the limiting magnitude for the UVOT non-detection.
The values are:
0  undefined
1  BAT
2  XRT
3  Other

The "soln_status" field contains flag bits on the type of solution that was found.
The bit packing of the "soln_status" entry in the packet is:
Bit_Number     Item_name      Description
----------     ---------      -----------
0              point_src      0=No or 1=Yes, a point source was found, 0=extended src
1              grb            0=No or 1=Yes, it is a GRB
2              undef          Not yet defined.
3              catalog_src    0=No or 1=Yes, it is in the catalog (else unknown src)
4              undef          Not yet defined.
5              def_not_grb    0=No or 1=Yes, it is definitely not a GRB (a Retraction)
6-29           undef          Not yet defined.
30             test_submit    0=No or 1=Yes, this is a test submission (internal use only)
31             undef          Not yet defined.

The "misc" field contains miscellaneous flag bits.
The defined bits are:
2^0-11  Not assigned.
2^12  When this bit is 1, it is in the catalog of sources to be blocked (internal use only).
2^13  When this bit is 1, there is a nearby bright star.
2^14-21  Not assigned.
2^22  When this bit is a 1, it indicates that this Notice was generated as a result
      of an uploaded TOO sequence.  (Swift is commanded to look at this RA,Dec location
      and this Notice was generated as a direct result of that observation.)
2^23  Not assigned.
2^24  Not assigned.
2^25  Not assigned.
2^26  Not assigned.
2^27  Not assigned.
2^28  Not assigned.
2^29  Not assigned.
2^30  When this bit is a 1, then this Notice was ground-generated; else flight-generated.
2^31  If there is a CRC error in one or more of the packets that went into
      the making of this Notice, then this bit is set to 1, else 0.

The "lim_sigma" is the number of simga that the mag_limit is quoted for.
The units are in centi-sigma (ie the fl.pt. mag_error is multiplied by 100
and then integerized).

There are a total of 28 "spare" items in the packet (usually filled
with zeros).  They are located at items 7-8, 11-18, and 21-38.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.



TYPE=97 PACKET CONTENTS:

The SWIFT_BAT_QUICKLOOK_POS packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=97)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       trig_obs_num integers        Trigger num & Observation num
long         5       burst_tjd    [days]          Truncated Julian Day
long         6       burst_sod    [centi-sec]     (int)(sssss.sss *100)
long         7       burst_ra     [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         8       burst_dec    [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         9       roll         [0.0001-deg]    (int)(0.0 to 360.0 *10000)
long         10      lon_lat      2_shorts        (int)(Longitude,Lattitude *100)
long         11      burst_error  [0.0001-deg]    (int)(0.0 to 180.0 *10000)
long         12-16   spare[5]     integer         20 bytes for the future
long         17      trig_index   integer         Rate_Trigger index
long         18      AT_slewFlags bits            Two flag bits packed here
long         19      misc         bits            Misc stuff packed in here
long         20      spare        integer         4 bytes for the future
long         21      rate_signif  [centi-sigma]   (int)(sig2noise *100)
long         22-37   spare[16]    integer         64 bytes for the future
long         38      merit        [centi-merit]   (int)(Merit_value *100)
long         39      pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 60, 61, 62, 63, ...84 ...97
that tells which type of packet this is and therefore what the contents are
(the contents change between the various types).

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start from 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the burst_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "trig_obs_num" is a pair of uniquely identifying serial numbers (Trigger_num and Observation_num)
assigned by the on-board BAT flight software to each trigger.
The lower 24 bits contain the Trigger_number, and the upper 8 bits
contain the Observation_number. 
The Observation_number is always 0 since these are always (by definition)
the first observation of this new trigger.
The Trigger_number started with s/n=100,000 just after launch (Nov 04).

The "burst_tjd" is the Truncated Julian Day of the Swift-BAT burst trigger,
eg. TJD=13370 is 01 Jan 2005.

The "burst_sod" is the UT seconds-of-day (SOD) of the Swift-BAT burst trigger.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized).  (For this packet type alone: Due to sequencing uncertainties
in the GCN ground s/w, this quantity does not contain the UT_Correction_Factor
about 50% of the instances.  It is a small correction; ~5 sec circa Nov 2009.)

The "burst_ra" and "burst_dec" are the RA,Dec coordinates of the burst (or transient)
as determined by the BAT flight software (J2000 epoch).  These intrinsically
floating point quantities have been multiplied by 10,000 and then integerized
to make an integer quantity with units of 0.0001-degrees.

The "roll" of the spacecraft (after the slew to this new burst).  This intrinsically
floating point quantities have been multiplied by 10,000 and then integerized
to make an integer quantity with units of 0.0001-degrees.

The "lon_lat" is the Longitude and Lattitude of the s/c at the time of the BAT trigger.
This 4-byte quantity is composed of two 2-byte integers.  The high-order short
is the latitude multiplied by 100 and integerized; the low-order short
is the longitude multiplied by 100 and integerized; ie both values are in centi-degrees.

The "burst_error" is the radius of the circle that will contain on average
TBD% of the bursts.  The units are 0.0001-degrees (ie the fl.pt. error was multiplied
by 10000 and then integerized).  [This value was hardwired at a constant value
of 3arcmin radius.]

The "trig_index" is the index value (row_number in the BAT on-board flight s/w table)
of the trigger criteria that was the highest-significance successful trigger.

The "at_slew_flags" field contains two flag bits related to the status
of the given burst wrt the Automated Target (AT) and Slewing.
The bits are:
2^0  Is this burst worthy of becoming the new Automated Target: 1=yes
2^1  Is this burst of sufficient merit to request a s/c slew: 1=yes
2^8  Is this location in the  Gnd_cat: 1=yes
The AT is a special target that the Swift Observatory will keep coming back to
orbit after orbit without any ground commanding intervention.  It is distinct
from the PrePlanned Targets (PPTs) that are in the uploaded stored observing table.
Whenever the AT is observable (ie no observing constraint is in effect),
then the AT will override any previous PPT.  The override lasts for up to
a commandable amount of time -- typically 60,000 observed-seconds.

The "misc" field contains miscellaneous flag bits.
The defined bits are:
2^0   When this bit is a 1, it indicates that the BAT_Pos TDRSS message was not received for this TrigNum.
2^1-10  Not assigned.
2^11  When this bit is a 1, it indicates that 1 (or more) of the RA,Dec,Roll,Theta,Phi values
      were out of valid range (eg RA was 361 deg).
2^12  Not assigned.
2^13  When this bit is a 1, it indicates that this position is near a bright star (mag less than 6.5).
2^14  When this bit is a 1, it indicates that the distance between the position and the center of the galaxy
      is less than the sum of the position error radius and the galaxy radius AND that the position error radius
      is smaller than the galaxy radius.  This position is inside the (circularized) NGC galaxy (or nearly in).
2^15  When this bit is a 1, it indicates that the distance between the position and the center of the galaxy
      is less than the sum of the position error radius and the galaxy radius AND that the galaxy radius
      is smaller than the position error radius.  This galaxy is inside the position error circle (or nearly in).
2^16-21  Not assigned.
2^22  When this bit is a 1, it indicates that this Notice was generated as a result
      of an uploaded TOO sequence.  (Swift is commanded to look at this RA,Dec location
      and this Notice was generated as a direct result of that observation.)
2^23-30  Not assigned.
2^31  If there is a CRC error in one or more of the packets that went into
      the making of this Notice, then this bit is set to 1, else 0.

The "rate_signif" is the significance level of the initial 'rate trigger'.
This is the signal/noise (ie number of sigma) of the rate-trigger detection
multiplied by 100 and then integerized (ie units are centi-sigma).

The "merit_value" is the total calculated value for the given burst.
It is a fl.pt. quantity which can theoretically range from -infinity to
+infinity, but given normal operations, we expect it to range from about
-100 to +300.  And for the first 6-12 months of flight-ops, it will be
forced to be a constant value of 100 for all bursts.

There are a total of 23 "spare" items in the packet (usually filled
with zeros).  They are located at items 9-14, 20, and 22-37.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.





TYPE=98 PACKET CONTENTS:

The SWIFT_BAT_SUBTHRESHOLD packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=98)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       trig_obs_num integers        Trigger num & Observation num
long         5       burst_tjd    [days]          Truncated Julian Day
long         6       burst_sod    [centi-sec]     (int)(sssss.sss *100)
long         7       burst_ra     [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         8       burst_dec    [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         9       burst_flue   [counts]        Num events during trig window, 0 to inf
long         10      burst_ipeak  [cnts*ff]       Counts in image-plane peak, 0 to infinity
long         11      burst_error  [0.0001-deg]    (int)(0.0 to 180.0 *10000)
long         12      phi          [centi-deg]     (int)(0.0 to 359.9999 *100)
long         13      theta        [centi-deg]     (int)(0.0 to +70.0 *100)
long         14      integ_time   [4mSec]         Duration of the trigger interval, 1 to inf
long         15      spare        integer         4 bytes for the future
long         16      lon_lat      2_shorts        (int)(Longitude,Lattitude *100)
long         17      trig_index   integer         Rate_Trigger index
long         18      soln_status  bits            Type of source/trigger found
long         19      misc         bits            Misc stuff packed in here
long         20      image_signif [centi-sigma]   (int)(sig2noise *100)
long         21      rate_signif  [centi-sigma]   (int)(sig2noise *100)
long         22      bkg_flue     [counts]        Num events during the bkg interval, 0 to inf
long         23      bkg_start    [centi-sec]     (int)(sssss.sss *100)
long         24      bkg_dur      [centi-sec]     (int)(0-80,000 *100)
long         25      cat_num      integer         On-board cat match ID number
long         26-35   spare[10]    integer         40 bytes for the future
long         36      merit_0-3    integers        Merit params 0,1,2,3 (-127 to +127)
long         37      merit_4-7    integers        Merit params 4,5,6,7 (-127 to +127)
long         38      merit_8-9    integers        Merit params 8,9     (-127 to +127)
long         39      pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 60, 61, 62, 63, ...98
that tells which type of packet this is and therefore what the contents are
(the contents change between the various types).  Please note that this type
is identical in format and content as type=61 (BAT_POS).  The only difference
is that these notices are due to the (so called) Sub-Threshold events.

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start from 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the burst_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "trig_obs_num" is a pair of uniquely identifying serial numbers (Trigger_num and Observation_num)
assigned by the on-board BAT flight software to each trigger.
The lower 24 bits contain the Trigger_number, and the upper 8 bits
contain the Observation_number. 
The Observation_number is always 0 since these are always (by definition)
the first observation of this new trigger (this new burst).
The Trigger_number started with s/n=100,000 just after launch (Nov 04).

The "burst_tjd" is the Truncated Julian Day of the Swift-BAT burst trigger,
eg. TJD=13370 is 01 Jan 2005.

The "burst_sod" is the UT seconds-of-day (SOD) of the Swift-BAT burst trigger.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized).  The trigger time can be off by +/- 320 msec
for events that have a trigger duration less than 64 msec.  This has to do with the way
that the flight software is optimized for execution speed for evaluating trigger criteria.

The "burst_ra" and "burst_dec" are the RA,Dec coordinates of the burst (or transient)
as determined by the BAT flight software (J2000 epoch).  These intrinsically
floating point quantities have been multiplied by 10,000 and then integerized
to make an integer quantity with units of 0.0001-degrees.

The "burst_flue" is the number of events (burst+background events)
during the 'integ_time' time interval of the successful trigger criterion.

The "burst_ipeak" is the height of the peak in the sky-image plane (ie the result
of the FFT,MaskConvolution,InvFFT calculation).  The units are counts
after a multiplicative factor of about 0.5-0.7.

The "burst_error" is the radius of the circle that will contain on average
TBD% of the bursts.  The units are 0.0001-degrees (ie the fl.pt. error was multiplied
by 10000 and then integerized).  [Initially, this value was hardwired 
at a constant value of 4arcmin (0.067deg) radius.  As the mission progressed,
this changed to 3 arcmin.  Eventually, it might become intensity-dependant.]

The "phi" and "theta" are the BAT instrument coordinates of the trigger (burst)
as determined by the BAT flight software program.  Phi is measured phi=0
along the +Y-axis of the s/c and increasing towards the -Z axis,
and theta is measured from the boresight of the BAT instrument.

The "integ_time" is the duration of the sampling interval
of the successful trigger criteria.  The value is the number of 4msec ticks,
eg a value of 16 indicates a 64-msec trigger criterion.

The "lon_lat" is the Longitude and Lattitude of the s/c at the time of the BAT trigger.
This 4-byte quantity is composed of two 2-byte integers.  The high-order short
is the latitude multiplied by 100 and integerized; the low-order short
is the longitude multiplied by 100 and integerized; ie both values are in centi-degrees.

The "trig_index" is the index value (row_number in the BAT on-board flight s/w table)
of the trigger criteria that was the highest-significance successful trigger.

The "soln_status" (aka trigger_id) field contains (a) flag bits on the type of solution
that was found in the on-board image and about whether it matches anything in the
on-board source catalog, and (b) flag bits that are assigned in the GCN ground-processing.
The bit packing of the "soln_status" entry in the packet is:
BitNum  F-v-G  Item_name      Description
------  ----   ---------      -----------
0        F     point_src      0=No or 1=Yes, a point source was found
1        F     grb            0=No or 1=Yes, it is a GRB
2        F     interesting    0=No or 1=Yes, it is an interesting src (ie a flaring known src)
3        F     flt_cat_src    0=No or 1=Yes, it is in the flight catalog (else unknown src)
4        F     image_trig     0=No or 1=Yes, it is an image trigger (else rate trig)
5        G     def_not_grb    0=No or 1=Yes, it is definitely not a GRB (bad label; its really a Retraction)
6        G     uncert_grb     0=No or 1=Yes, it is probably not a GRBorTransient (hi bkg level, ie near SAA)
7        G     uncert_grb     0=No or 1=Yes, it is probably not a GRBorTransient (low image significance; less then 7.0 sigma)
8        G     gnd_cat_src    0=No or 1=Yes, it is in the ground catalog
9        G     uncert_grb     0=No or 1=Yes, it is probably not a GRBorTransient (negative bkg slope, exit SAA)
10       G     st_loss_lock   0=No or 1=Yes, StarTracker not locked so trigger probably bogus
11       G     uncert_grb     0=No or 1=Yes, it is really probably not a GRBorTransient (VERY low image significance; less than 6.5 sigma)
12       G     blk_cat_src    0=No or 1=Yes, it is in the catalog of sources to be blocked (internal use only)
13       G     near_brt_star  0=No or 1=Yes, there is a nearby bright star (this is a duplicate of the "misc" 2^13).
14-27    -     spare          spare for future use
28       G     spatial_coinc  0=No or 1=Yes, there was a spatial coincidence with another event
29       G     temporal_coinc 0=No or 1=Yes, there was a temporal coincidence with another event
30       G     test_submit    0=No or 1=Yes, this is a test submission (internal use only)
31       -     spare          spare for future use
The columns are:  "BitNum" is the 2^N location within the 4-byte word; "F-v-G" is Flight-assigned vs
Ground-assigned;  "Item_name" is the short-description name of the quantity;  and
"Description" is the longer description of the quantity.

The "misc" field contains miscellaneous flag bits.
The defined bits are:
2^0-10  Not assigned.
2^11  When this bit is a 1, it indicates that 1 (or more) of the RA,Dec,Roll,Theta,Phi values
      were out of valid range (eg RA was 361 deg).
2^12  Not assigned.
2^13  When this bit is a 1, it indicates that this position is near a bright star (mag less than 6.5).
2^14  When this bit is a 1, it indicates that the distance between the position and the center of the galaxy
      is less than the sum of the position error radius and the galaxy radius AND that the position error radius
      is smaller than the galaxy radius.  This position is inside the (circularized) NGC galaxy (or nearly in).
2^15  When this bit is a 1, it indicates that the distance between the position and the center of the galaxy
      is less than the sum of the position error radius and the galaxy radius AND that the galaxy radius
      is smaller than the position error radius.  This galaxy is inside the position error circle (or nearly in).
2^16-20  Not assigned.
2^21  When this bit is a 1, it indicates that this is an Image-trigger AND that it occurred
      during a time when the StarTracker had a Loss-of-Lock.
2^22  When this bit is a 1, it indicates that this Notice was generated as a result
      of an uploaded TOO sequence.  (Swift is commanded to look at this RA,Dec location
      and this Notice was generated as a direct result of that observation.)
2^23  When this bit is a 1, it means the Trig_Time was zero and so the s/c system clock
      in the TDRSS Message SecondaryHeader was used to fill the Burst_TJD and Burst_SOD fields.
      (This 0-Trig_Time scenario occurs during a ground-commanded GRB_TOO.)
2^24  When this bit is a 1, it indicates that the TDRSS Message was received on the ground
      more than 60 sec after the BAT Trigger time.  The most likely scenario for this kind of transmission delay
      is that the BAT Trigger occured during a Malindi downlink pass (when TDRSS messages are blocked
      from immediate transmission; they are buffered onboard until the end of the Malindi Pass).
2^25  This is an updated position.  This is a position notice that changes/improves
      the position specified in the original BAT_Position Notice.
2^26  Not assigned.
2^27  Not assigned.
2^28  When this bit is a 1, then this Notice was derived from a SERS message; else flight-generated.
      If none of the regular position-containing messages are received via the real-time
      TDRSS comm channel (eg BAT_Pos_Ack, BAT_LC, etc), then the SERS messages
      (which come from pipe-line processing of the Malindi playback downlinked data)
      will be used to generate a reconstituted BAT_Pos_Ack Notice.  But since
      the SERS message only contains a subset of the information is a regular TDRSS
      messages, these SERS-reconstituted Notices will have (tbd) fields undefined.
      These undefined fields are: tbd, tbd, tbd....
2^29  When this bit is a 1, then this Notice was reconstituted; else flight-generated.
      If a BAT_LC is received from the s/c before a BAT_Pos is received,
      then bits and pieces from other Notices will be used to generate a reconstituted
      BAT_Pos Notice and it will be issued (with this flag bit set to 1).
      But since the LC message only contains a subset of the information in a regular
      TDRSS BAT_POS messages, these LC-reconstituted Notices will have (tbd) fields undefined.
      These undefined fields are: tbd, tbd, tbd....
2^30  When this bit is a 1, then this Notice was ground-generated; else flight-generated.
      A Gnd-generated Notice is issued anytime humans during ground operations
      have determined if this Notice should be issued.  This will typically be used
      during the early phases of the mission before the automatic real-time distribution
      capability is enabled.  Once the burst position has been validated by humans,
      a BAT_Pos Notice will be issued (with this flag bit set to 1).
      After the Verification phase, there may be rare occaisions when a BAT_POS notice
      is (re)created by human-processing and will be distributed (with this bit set to 1). 
2^31  If there is a CRC error in one or more of the packets that went into
      the making of this Notice, then this bit is set to 1, else 0.

The "image_signif" is the significance level of the peak in the FFT image plane.
This is the signal/noise (ie number of sigma) of the grb peak detection
multiplied by 100 and then integerized (ie units are centi-sigma).
By definition of this Notice type, the 'image_signif' value will never be 
greater than 6.50 sigma.

The "rate_signif" is the significance level of the initial 'rate trigger'.
This is the signal/noise (ie number of sigma) of the rate-trigger detection
multiplied by 100 and then integerized (ie units are centi-sigma).

The "bkg_flue" is the total number of events in the time interval used
for the background subtraction of the detector_rate_map (used for the image calc).
The units are counts.

The "bkg_start" is the starting time of the background interval.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized).  Note that not all trigger criteria have a background interval
specified, so this "bkg_start" value may be zero for some bursts.

The "bkg_dur" is the duration of the background interval.
The units are in centi-seconds.

The "cat_num" is the catalog number in the on-board catalog for sources
that match a known source in the on-board catalog.  (Typically it is a number
in the low 30,000's.)

The "merit_0-3", "merit_4-7", and "merit_8-9" are the 10 merit parameters
for this burst.  Their values range from -127 to +127.  These 'merit params'
are different than the 'merit_value' that is in the SWIFT_FOM_OBS and SWIFT_SC_SLEW
messages.  There are 4, 4, & 2 values packed into these 3 longs.  The packing order
is "0" in the lowest_address byte in the long, "1" is in the next_lowest, etc.
The definition of the 10 parameters is:
Param   Description
0       Flag bit indicating GRB or not (1 or 0, resp).
1       Flag bit indicating Transient or not (1 or 0, resp); unknown src with T_trig>64sec.
2       Flag bit indicating Known_src or not (1 or 0, resp); Known_src matches something in on-board catalog.
3       Trigger duration (log2(t_trig/1024msec)).
4       Trigger energy range (0=15-25,1=15-50,2=25-100,3=50-350).
5       Image significance
6       Is it observable (-1 if w/in 30deg of Moon; -5, 20deg Moon; -100, 45deg Sun).
7       Flag bit indicating in a confused or obscurred region (ie Gal Center or Plane).
8       Sun distance (-100*cos(sun_angular_dist)).
9       An offset param to bias for/against PPTs and TOOs wrt ATs (norm param).

There are a total of 13 "spare" items in the packet (usually filled
with zeros).  They are located at items 15-16 and 25-35.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.




TYPE=99 PACKET CONTENTS:

The SWIFT_BAT_SLEW_GRB_POSITION packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=99)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       id_num       integer         ID number
long         5       burst_tjd    [days]          Truncated Julian Day
long         6       burst_sod    [centi-sec]     (int)(sssss.sss *100)
long         7       burst_ra     [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         8       burst_dec    [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         9       burst_flue   [counts]        Num events during image window, 0 to inf
long         10      burst_ipeak  [cnts/sec]      Counts in image-plane peak, 0 to infinity
long         11      burst_error  [0.0001-deg]    (int)(0.0 to 180.0 *10000)
long         12      spare        integer         4 bytes for the future
long         13      spare        integer         4 bytes for the future
long         14      integ_time   [4mSec]         Duration of the trigger interval, 1 to inf
long         15      spare        integer         4 bytes for the future
long         16      spare        integer         4 bytes for the future
long         17      trig_index   integer         Trigger criterion index
long         18      soln_status  bits            Type of source/trigger found
long         19      misc         bits            Misc stuff packed in here
long         20      image_signif [centi-sigma]   (int)(sig2noise *100)
long         21-38   spare[18]    integer         72 bytes for the future
long         39      pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 60, 61, 62, 63, ...98, 99
that tells which type of packet this is and therefore what the contents are
(the contents change between the various types).  Please note that this type
is very similar but not identical in format and content as type=61 (BAT_POS).
These notices contain positions of bursts (and transients) detected by BAT
during intervals when Swift is slewing.  A separate notice type was created 
because there will be some delay (0.5 - 6 hrs: tlm downlink delays and 
ground-processing time) compared to the 10-20sec delay for the type=61 notices.
And there may well be a slightly larger position uncertainty compared
to type=61.

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start from 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the burst_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "id_num" is a uniquely identifying serial number assigned by Harvard.
It started at 1 with GRB 070326.

The "burst_tjd" is the Truncated Julian Day of the Swift-BAT burst trigger,
eg. TJD=13370 is 01 Jan 2005.

The "burst_sod" is the UT seconds-of-day (SOD) of the ground burst trigger.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized).

The "burst_ra" and "burst_dec" are the RA,Dec coordinates of the burst (or transient)
as determined by the BATSS ground-procesing software (J2000 epoch).  These intrinsically
floating point quantities have been multiplied by 10,000 and then integerized
to make an integer quantity with units of 0.0001-degrees.

The "burst_flue" is the number of events (burst+background events)
during the 'integ_time' time interval of the successful trigger criterion.
The units are counts.

The "burst_ipeak" is the height of the peak in the sky-image plane.
The units are counts/sec.
Always 0 for now.

The "burst_error" is the radius of the circle that will contain on average
TBD% of the bursts.  The units are 0.0001-degrees (ie the fl.pt. error was multiplied
by 10000 and then integerized).

The "integ_time" is the duration of the sampling interval
of the successful trigger criteria.  The value is the number of 4msec ticks,
eg a value of 16000 indicates a 64-sec trigger integration.

The "trig_index" is the value of the trigger criteria that had the highest evaluation.
The definitions of the various trigger criteria are listed in the table below:
INDEX  CRITERIA                                                              COMMENT
1-4    Undefined                                                             Undefined
5      S(15-50) > 6.5 sigma OR H(50-150) > 6.5 sigma                         High confidence of reality (silver-plated).
6      Non-simultaneous detections > 4.0 sigma over more than an orbit       Transient (ie too long for a GRB)
7      Non-simultaneous detections > 4.0 sigma within an orbit               Transient (ie too long for a GRB)
8      Simul (S(15-50) OR H(50-150)) > 6.0 sigma AND B(15-150) > 4.0 sigma   High confidence of reality (silver-plated).
9      Simul (S(15-50) OR H(50-150)) > 4.0 sigma AND B(15-150) > 6.0 sigma   High confidence of reality (silver-plated).
10     Simul S(15-50) > 4.0 sigma AND H(50-150) > 4.0 sigma                  Very high confidence of reality (gold-plated).

The "soln_status" (aka trigger_id) field contains (a) flag bits on the type of solution
that was found in the on-board image and about whether it matches anything in the
on-board source catalog, and (b) flag bits that are assigned in the GCN ground-processing.
The bit packing of the "soln_status" entry in the packet is:
BitNum  F-v-G  Item_name      Description
------  ----   ---------      -----------
0        G     point_src      0=No or 1=Yes, a point source was found
1        G     grb            0=No or 1=Yes, it is definitely a GRB (or a hard x-ray transient, ie astrophysical) Index=10)
2        -     spare          not defined for this notice type
3        -     spare          not defined for this notice type
4        G     image_trig     0=No or 1=Yes, it is an image trigger (else rate trig) Always I-trigger for BATSS.
5        G     def_not_grb    0=No or 1=Yes, it is definitely not a GRB (bad label; its really a Retraction)
6        -     spare          not defined for this notice type
7        -     spare          not defined for this notice type
8        G     gnd_cat_src    0=No or 1=Yes, it is in the ground catalog
9        -     spare          not defined for this notice type
10       -     spare          not defined for this notice type
11       -     spare          not defined for this notice type
12       G     blk_cat_src    0=No or 1=Yes, it is in the catalog of sources to be blocked (internal use only)
13       G     near_brt_star  0=No or 1=Yes, there is a nearby bright star (this is a duplicate of the "misc" 2^13).
14       G     prob_grb       0=No or 1=Yes, it is probably a GRB (or hard x-ray transient, ie astrophysical) (Index=8 or 9)
15-27    -     spare          spare for future use
28       G     spatial_coinc  0=No or 1=Yes, there was a spatial coincidence with another event
29       G     temporal_coinc 0=No or 1=Yes, there was a temporal coincidence with another event
30       G     test_submit    0=No or 1=Yes, this is a test submission (internal use only)
31       -     spare          spare for future use
The columns are:  "BitNum" is the 2^N location within the 4-byte word; "F-v-G" is Flight-assigned vs
Ground-assigned (they are ALL gnd-assigned for BAT_Slew_Pos);  "Item_name" is the short-description name of the quantity;  and
"Description" is the longer description of the quantity.

The "misc" field contains miscellaneous flag bits.
The defined bits are:
2^0   When this bit is a 1, it indicates that photons in the 15-50 keV (Soft) band where used in the detection.
2^1   When this bit is a 1, it indicates that photons in the 50-150 keV (Hard) band where used in the detection.
2^3   When this bit is a 1, it indicates that photons in the 15-150 keV (Broad) band where used in the detection.
2^3-10 Not assigned.
2^11  When this bit is a 1, it indicates that 1 (or more) of the RA,Dec,Roll,Theta,Phi values
      were out of valid range (eg RA was 361 deg).
2^12  Not assigned.
2^13  When this bit is a 1, it indicates that this position is near a bright star (mag less than 6.5).
2^14  When this bit is a 1, it indicates that the distance between the position and the center of the galaxy
      is less than the sum of the position error radius and the galaxy radius AND that the position error radius
      is smaller than the galaxy radius.  This position is inside the (circularized) NGC galaxy (or nearly in).
2^15  When this bit is a 1, it indicates that the distance between the position and the center of the galaxy
      is less than the sum of the position error radius and the galaxy radius AND that the galaxy radius
      is smaller than the position error radius.  This galaxy is inside the position error circle (or nearly in).
2^16-24  Not assigned.
2^25  This is an updated position.  This is a position notice that changes/improves
      the position specified in the original BAT_Slew_Position Notice.
2^26-29  Not assigned.
2^30  When this bit is a 1, then this Notice was ground-generated; else flight-generated.
      By definition, all BATSS-generated notices are ground-generated (via the pipeline
      processing at Harvard).
2^31  Not assigned.

The "image_signif" is the significance level of the peak in the image plane.
This is the signal/noise (ie number of sigma) of the grb peak detection
multiplied by 100 and then integerized (ie units are centi-sigma).

There are a total of 20 "spare" items in the packet (usually filled
with zeros).  They are located at items 15-16 and 21-38.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.




================================================================================

TYPE=100 PACKET CONTENTS:
TYPE=101 PACKET CONTENTS:
TYPE=102 PACKET CONTENTS:

The SuperAGILE_GRB_WAKEUP/_GROUND/_REFINED packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=100-102)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       trig_num     integer         Trigger number
long         5       burst_tjd    [days]          Truncated Julian Day
long         6       burst_sod    [centi-sec]     (int)(sssss.sss *100)
long         7       burst_ra     [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         8       burst_dec    [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         9       burst_intenX [0.001-cnts]    Num events in each X 1-D
long         10      burst_intenY [0.001-cnts]    Num events in each Y 1-D
long         11      burst_error  [0.0001-deg]    (int)(0.0 to 180.0 *10000)
long         12-17   spare[6]     integers        24 bytes for the future
long         18      trigger_id   bits            Type of source/trigger found
long         19      misc         bits            Misc stuff packed in here
long         20      signif       two_2byte_int   Misc stuff packed in here
long         21-38   spare[17]    integers        68 bytes for the future
long         39      pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 100, 101, 102, 107, 108, 109;
that tells which type of packet this is and therefore what the contents are
(the contents change between the various types).  The types are:
100 = SuperAGILE_GRB_WAKEUP, 101 = SuperAGILE_GRB_GROUND, 102 = SuperAGILE_REFINED,
107 = AGILE_POINTDIR, 108 = SuperAGILE_TRANS, and 109 = SuperAGILE_GRB_POS_TEST. 

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start at 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the burst_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "trig_num" is a uniquely identifying number for each trigger.
It is the number of 0.1-secs since 01 Jan 2001.

The "burst_tjd" is the Truncated Julian Day of the AGILE burst trigger,
eg. TJD=13370 is 01 Jan 2005.

The "burst_sod" is the UT seconds-of-day (SOD) of the AGILE burst trigger.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized).

The "burst_ra" and "burst_dec" are the RA,Dec coordinates of the burst (or transient)
as determined by the flight software (J2000 epoch).  These intrinsically
floating point quantities have been multiplied by 10,000 and then integerized
to make an integer quantity with units of 0.0001-degrees (over-precision,
but this is the new standard scaling for all missions/instruments).

The "burst_intenX" is the number of events in the 15-45 keV band
in the X 1-D localization.  This intrinsically floating point quantity
have been multiplied by 1000 and then integerized to make an integer quantity
with units of 0.001-counts.

The "burst_intenY" is the number of events in the 15-45 keV band
in the Y 1-D localization.  This intrinsically floating point quantity
have been multiplied by 1000 and then integerized to make an integer quantity
with units of 0.001-counts.

The "burst_error" is the radius of the circle that will contain on average
TBD% of the bursts.  It contains statical plus the systematic contribution.
The units are 0.0001-degrees (ie the fl.pt. error was multiplied by 10000 and
then integerized).

The "trigger_id" field contains (a) flag bits on the type of solution
that was found in the on-board (and SA-OPS) and (b) flag bits that are assigned
in the GCN ground-processing.
BitNum  F-v-G  Item_name      Description
------  ----   ---------      -----------
0        -     spare          spare for future use
1        F     grb            0=No or 1=Yes, it is a GRB
2-4      -     spare          spare for future use
5        G     def_not_grb    0=No or 1=Yes, it is definitely not a GRB (bad label; its really a Retraction)
6-12     -     spare          spare for future use
13       G     near_brt_star  0=No or 1=Yes, there is a nearby bright star (see MISC for correct location)
14-27    -     spare          spare for future use
28       G     spatial_coinc  0=No or 1=Yes, there was a spatial coincidence with another event
29       G     temporal_coinc 0=No or 1=Yes, there was a temporal coincidence with another event
30       G     test_submit    0=No or 1=Yes, this is a test submission (internal use only)
31       -     spare          spare for future use
The columns are:  "BitNum" is the 2^N location within the 4-byte word; "F-v-G" is Flight-assigned vs
Ground-assigned;  "Item_name" is the short-description name of the quantity;  and
"Description" is the longer description of the quantity.

The "misc" field contains miscellaneous flag bits.
2^0-12  Not assigned.
2^13    When this bit is a 1, it indicates that this position is near (lss than 0.3deg)  a bright star (mag less than 6.5).
2^14    When this bit is a 1, it indicates that the distance between the position and the center of the galaxy
        is less than the sum of the position error radius and the galaxy radius AND that the position error radius
        is smaller than the galaxy radius.  This position is inside the (circularized) NGC galaxy (or nearly in).
2^15    When this bit is a 1, it indicates that the distance between the position and the center of the galaxy
        is less than the sum of the position error radius and the galaxy radius AND that the galaxy radius
        is smaller than the position error radius.  This galaxy is inside the position error circle (or nearly in).
2^30    When this bit is a 1, then this Notice was ground-generated; else flight-generated.
        A Gnd-generated Notice is issued anytime humans during ground operations
        have determined if this Notice should be issued.  This will typically be used
        during the early phases of the mission before the automatic real-time distribution
        capability is enabled.  Once the burst position has been validated by humans,
        a GRB_Pos Notice will be issued (with this flag bit set to 1).
        After the Verification phase, there still may be occaisions when a GRB_POS notice
        is (re)created by human-processing and will be distributed (with this bit set to 1). 
2^31    Not assigned.

The "signif" is composed of two 16-bit quantitites packed into a single 32-bit field.
The two quantities are the X-axis and Y-axis 1-dimentional significance detections.
The X-direction significance is packed in the bottom 16 bits, and the Y-direction
is packed in the upper 16 bits.  Prior to packing, both significances were multiplied
by 100 and integerized (i.e centi-sigma)..

There are a total of 17(tbr) "spare" items in the packet (usually filled
with zeros).  They are located at items 15-16(tbd) and 21-35.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.




PACKET ITEM DESCRIPTIONS for type=103:

The SWIFT_ACTUAL_POINTDIR packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=103)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       trig_obs_num integers        Trigger num & Observation num
long         5       slew_tjd     [days]          Truncated Julian Day
long         6       slew_sod     [centi-sec]     (int)(sssss.sss *100)
long         7       point_ra     [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         8       point_dec    [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         9       point_roll   [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         10-15   spare[6]     integer         24 bytes for the future
long         16      lon_lat      2_shorts        (int)(Longitude,Lattitude *100)
long         17-18   spare[2]     integer         8 bytes for the future
long         19      misc         bits            Misc stuff packed in here
long         20-21   spare[2]     integer         8 bytes for the future
//quad_char    22-38   tgtname[17]  chars           The name of the target
long         39      pkt_term     integer         Pkt Termination (always = \n)



PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 60, 61, 62, 63, ...83... 103
that tells which type of packet this is and therefore what the contents are
(the contents change between the various types).  The SWIFT_ACTUAL_POINTDIR type is 103.

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start from 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the burst_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "trig_obs_num" is a pair of uniquely identifying serial numbers (Trigger_num and Observation_num)
assigned by the on-board BAT flight software to each trigger.
The lower 24 bits contain the Trigger_number, and the upper 8 bits
contain the Observation_number. 

The "slew_tjd" is the TJD when the next s/c slew ends
eg. TJD=13370 is 01 Jan 2005.

The "slew_sod" is the seconds-of-day (SOD) when the next s/c slew ends.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized).

The actual "point_ra", "point_dec" and "point_roll" are the RA,Dec and Roll (J2000) coordinates
of the spacecraft after the slew.  This is the pointing direction of the s/c.
The units are 10^04 deg, ie the values where multiplied by 10000 and integerized.

The "lon_lat" is the Longitude and Lattitude of the s/c at the time of the notice.
This 4-byte quantity is composed of two 2-byte integers.  The high-order short
is the latitude multiplied by 100 and integerized; the low-order short
is the longitude multiplied by 100 and integerized; ie both values are in centi-degrees.

The "misc" field contains miscellaneous flag bits.
The defined bits are:
2^0-12  Not assigned.
2^13    When this bit is 1, there is a nearby bright star (mag less than 6.5).
2^14-29 Not assigned.
2^30    When this bit is a 1, then this Notice was ground-generated; else flight-generated.
        This will always be 1 since the POINTDIR information is extracted from ground processing
        of ground files.
2^31    Not assigned.

//This might get added at a later date:
//The "tgtname" is the name of the object that the sc will be observing next.
//It is a name chosen by the science planners at the MOC.  The name does not
//always fit the IAU standard naming convention (due to limitations in the 
//planning s/w).  This item in the packet spans 17 longwords (ie 68 bytes) and
//contains the ASCII string of the target_name.  This string can be up to
//67 characters long, and will always be null-terminated.

There are a total of 10 "spare" items in the packet (usually filled
with zeros).  They are located at items 10-15, 17-18 and 20-21.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.







TYPE=107 PACKET CONTENTS:

The AGILE_POINTDIR packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=107)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       spare        integer         4 bytes for the future
long         5       start_tjd    [days]          Truncated Julian Day
long         6       start_sod    [centi-sec]     (int)(sssss.sss *100)
long         7       curr_ra      [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         8       curr_dec     [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         9-11    spare[3]     integer         12 bytes for the future
long         12      stop_tjd     [days]          Truncated Julian Day
long         13      stop_sod     [centi-sec]     (int)(sssss.sss *100)
long         14-18   spare[5]     integer         20 bytes for the future
long         19      misc         bits            Misc stuff packed in here
long         20-38   spare[19]    integer         48 bytes for the future
long         39      pkt_term     integer         Pkt Termination (always = \n)


The "pkt_type" is an integer with a value of 100, 101, 102, 107, 108, 109;
that tells which type of packet this is and therefore what the contents are
(the contents change between the various types).  The types are:
100 = SuperAGILE_GRB_WAKEUP, 101 = SuperAGILE_GRB_GROUND, 102 = SuperAGILE_REFINED,
107 = AGILE_POINTDIR, 108 = SuperAGILE_TRANS, and 109 = SuperAGILE_GRB_POS_TEST. 

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start at 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the burst_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "start_tjd" is the Truncated Julian Day of the start of the AGILE pointing session.
It is NOT the "current" TJD.  The distinction being that the AGILE spacecraft
does "pointing sessions" for 2-4 weeks (mostly), and during that session
the instantaneous RA,Dec direction slowly drifts about 1 deg per day.
The Start and Stop dates/times give the beginning and end of that pointing session,
and the "Curr_RA/_Dec" give the RA,Dec pointing direction of the spacecraft
at the time of the Notice.

The "start_sod" is the UT seconds-of-day (SOD) of the start of the AGILE pointing session.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized).

The "curr_ra" and "curr_dec" are the RA,Dec coordinates of the current 
pointing direction (bore sight) of AGILE (J2000 epoch).  These intrinsically
floating point quantities have been multiplied by 10,000 and then integerized
to make an integer quantity with units of 0.0001-degrees (over-precision,
but this is the new standard scaling for all missions/instruments).

The "stop_tjd" is the Truncated Julian Day of the end of the AGILE pointing session.

The "stop_sod" is the UT seconds-of-day (SOD) of the end of the AGILE pointing session.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized).

The "misc" field contains miscellaneous flag bits.
2^0-29 Not assigned.
2^30   When this bit is a 1, then this Notice was ground-generated; else flight-generated.
       This notice type will ALWAYS be ground-generated.
2^31   Not assigned.

There are a total of 28 "spare" items in the packet (usually filled
with zeros).  They are located at items 4, 9-11, 14-19, and 20-38.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.




TYPE=108 PACKET CONTENTS:

The SuperAGILE_TRANS packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=108)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       trig_num     integer         Trigger number
long         5       trans_tjd    [days]          Truncated Julian Day
long         6       trans_sod    [centi-sec]     (int)(sssss.sss *100)
long         7       trans_ra     [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         8       trans_dec    [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         9       trans_intenX [0.001-cnts]    Num events in each X 1-D
long         10      trans_intenY [0.001-cnts]    Num events in each Y 1-D
long         11      trans_error  [0.0001-deg]    (int)(0.0 to 180.0 *10000)
long         12-13   spare[2]     integers        8 bytes for the future
long         14      integ_time   [centi-sec]     Duration of the trigger interval, 1 to inf
long         15-17   spare[3]     integer         12 bytes for the future
long         18      trigger_id   bits            Type of source/trigger found
long         19      misc         bits            Misc stuff packed in here
long         20      signif       two_2byte_int   Misc stuff packed in here
long         21-38   spare[17]    integers        68 bytes for the future
long         39      pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 100, 101, 102, 107, 108, 109;
that tells which type of packet this is and therefore what the contents are
(the contents change between the various types).  The types are:
100 = SuperAGILE_GRB_WAKEUP, 101 = SuperAGILE_GRB_GROUND, 102 = SuperAGILE_REFINED,
107 = AGILE_POINTDIR, 108 = SuperAGILE_TRANS, and 109 = SuperAGILE_GRB_POS_TEST. 

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start at 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the burst_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "trig_num" is a uniquely identifying number for each trigger.
It is the number of 0.1-secs since 01 Jan 2001.

The "trans_tjd" is the Truncated Julian Day of the AGILE burst trigger,
eg. TJD=13370 is 01 Jan 2005.

The "trans_sod" is the UT seconds-of-day (SOD) of the AGILE burst trigger.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized).

The "trans_ra" and "trans_dec" are the RA,Dec coordinates of the burst (or transient)
as determined by the flight software (J2000 epoch).  These intrinsically
floating point quantities have been multiplied by 10,000 and then integerized
to make an integer quantity with units of 0.0001-degrees (over-precision,
but this is the new standard scaling for all missions/instruments).

The "trans_intenX" is the number of events in the 15-45 keV band
in the X 1-D localization.  This intrinsically floating point quantity
have been multiplied by 1000 and then integerized to make an integer quantity
with units of 0.001-counts.

The "trans_intenY" is the number of events in the 15-45 keV band
in the Y 1-D localization.  This intrinsically floating point quantity
have been multiplied by 1000 and then integerized to make an integer quantity
with units of 0.001-counts.

The "trans_error" is the radius of the circle that will contain on average
TBD% of the bursts.  The units are 0.0001-degrees (ie the fl.pt. error was multiplied
by 10000 and then integerized).

The "trigger_id" field contains (a) flag bits on the type of solution
that was found in the on-board (and SA-OPS) and (b) flag bits that are assigned
in the GCN ground-processing.
BitNum  F-v-G  Item_name      Description
------  ----   ---------      -----------
0-4      -     spare          spare for future use
5        G     def_not_grb    0=No or 1=Yes, it is definitely not a GRB (a Retraction)
6-7      -     spare          spare for future use
//8        G     gnd_cat_src    0=No or 1=Yes, it is in the ground catalog
9-12     -     spare          spare for future use
13       G     near_brt_star  0=No or 1=Yes, there is a nearby bright star
14-29    -     spare          spare for future use
30       G     test_submit    0=No or 1=Yes, this is a test submission (internal use only)
31       -     spare          spare for future use
The columns are:  "BitNum" is the 2^N location within the 4-byte word; "F-v-G" is Flight-assigned vs
Ground-assigned;  "Item_name" is the short-description name of the quantity;  and
"Description" is the longer description of the quantity.

The "misc" field contains miscellaneous flag bits.
2^0-12  Not assigned.
2^13    When this bit is a 1, it indicates that this position is near (less than0.3deg)  a bright star (mag less than 6.5).
2^14    When this bit is a 1, it indicates that the distance between the position and the center of the galaxy
        is less than the sum of the position error radius and the galaxy radius AND that the position error radius
        is smaller than the galaxy radius.  This position is inside the (circularized) NGC galaxy (or nearly in).
2^15    When this bit is a 1, it indicates that the distance between the position and the center of the galaxy
        is less than the sum of the position error radius and the galaxy radius AND that the galaxy radius
        is smaller than the position error radius.  This galaxy is inside the position error circle (or nearly in).
2^16-29 Not assigned.
2^30    When this bit is a 1, then this Notice was ground-generated; else flight-generated.
        A Gnd-generated Notice is issued anytime humans during ground operations
        have determined if this Notice should be issued.  This will typically be used
        during the early phases of the mission before the automatic real-time distribution
        capability is enabled.  Once the burst position has been validated by humans,
        a GRB_Pos Notice will be issued (with this flag bit set to 1).
        After the Verification phase, there still may be occaisions when a GRB_POS notice
        is (re)created by human-processing and will be distributed (with this bit set to 1). 
2^31    Not assigned.

The "signif" is composed of two 16-bit quantitites packed into a single 32-bit field.
The two quantities are the X-axis and Y-axis 1-dimentional significance detections.
The X-direction significance is packed in the bottom 16 bits, and the Y-direction
is packed in the upper 16 bits.  Prior to packing, both significances were multiplied
by 100 and integerized (i.e centi-sigma)..

There are a total of 17(tbr) "spare" items in the packet (usually filled
with zeros).  They are located at items 15-16(tbd) and 21-35.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.





TYPE=109 PACKET CONTENTS:

The SuperAGILE_GRB_POS_TEST packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=109)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       trig_num     integer         Trigger number
long         5       burst_tjd    [days]          Truncated Julian Day
long         6       burst_sod    [centi-sec]     (int)(sssss.sss *100)
long         7       burst_ra     [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         8       burst_dec    [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         9       burst_intenX [0.001-cnts]    Num events in each X 1-D
long         10      burst_intenY [0.001-cnts]    Num events in each Y 1-D
long         11      burst_error  [0.0001-deg]    (int)(0.0 to 180.0 *10000)
long         12-13   spare[2]     integers        8 bytes for the future
long         14      integ_time   [centi-sec]     Duration of the trigger interval, 1 to inf
long         15-17   spare[3]     integer         12 bytes for the future
long         18      trigger_id   bits            Type of source/trigger found
long         19      misc         bits            Misc stuff packed in here
long         20      signif       two_2byte_int   Misc stuff packed in here
long         21-38   spare[17]    integers        68 bytes for the future
long         39      pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 100, 101, 102, 107, 108, 109;
that tells which type of packet this is and therefore what the contents are
(the contents change between the various types).  The types are:
100 = SuperAGILE_GRB_WAKEUP, 101 = SuperAGILE_GRB_GROUND, 102 = SuperAGILE_REFINED,
107 = AGILE_POINTDIR, 108 = SuperAGILE_TRANS, and 109 = SuperAGILE_GRB_POS_TEST. 

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start at 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the burst_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "trig_num" is a uniquely identifying number for each trigger.
It is the number of 0.1-secs since 01 Jan 2001.

The "burst_tjd" is the Truncated Julian Day of the AGILE burst trigger,
eg. TJD=13370 is 01 Jan 2005.

The "burst_sod" is the UT seconds-of-day (SOD) of the AGILE burst trigger.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized).

The "burst_ra" and "burst_dec" are the RA,Dec coordinates of the burst (or transient)
as determined by the flight software (J2000 epoch).  These intrinsically
floating point quantities have been multiplied by 10,000 and then integerized
to make an integer quantity with units of 0.0001-degrees (over-precision,
but this is the new standard scaling for all missions/instruments).

The "burst_intenX" is the number of events in the 15-45 keV band
in the X 1-D localization.  This intrinsically floating point quantity
have been multiplied by 1000 and then integerized to make an integer quantity
with units of 0.001-counts.

The "burst_intenY" is the number of events in the 15-45 keV band
in the Y 1-D localization.  This intrinsically floating point quantity
have been multiplied by 1000 and then integerized to make an integer quantity
with units of 0.001-counts.

The "burst_error" is the radius of the circle that will contain on average
TBD% of the bursts.  The units are 0.0001-degrees (ie the fl.pt. error was multiplied
by 10000 and then integerized).

The "trigger_id" field contains (a) flag bits on the type of solution
that was found in the on-board (and SA-OPS) and (b) flag bits that are assigned
in the GCN ground-processing.
BitNum  F-v-G  Item_name      Description
------  ----   ---------      -----------
0-4      -     spare          spare for future use
5        G     def_not_grb    0=No or 1=Yes, it is definitely not a GRB (bad label; its really a Retraction)
6-7      -     spare          spare for future use
//8        G     gnd_cat_src    0=No or 1=Yes, it is in the ground catalog
9-12     -     spare          spare for future use
13       G     near_brt_star  0=No or 1=Yes, there is a nearby bright star
14-27    -     spare          spare for future use
28       G     spatial_coinc  0=No or 1=Yes, there was a spatial coincidence with another event (NOOP for TEST)
29       G     temporal_coinc 0=No or 1=Yes, there was a temporal coincidence with another event (NOOP for TEST)
30       G     test_submit    0=No or 1=Yes, this is a test submission (internal use only)
31       -     spare          spare for future use
The columns are:  "BitNum" is the 2^N location within the 4-byte word; "F-v-G" is Flight-assigned vs
Ground-assigned;  "Item_name" is the short-description name of the quantity;  and
"Description" is the longer description of the quantity.

The "misc" field contains miscellaneous flag bits.
2^0-12  Not assigned.
2^13    When this bit is a 1, it indicates that this position is near (less than 0.3deg)  a bright star (mag less than 6.5).
2^14    When this bit is a 1, it indicates that the distance between the position and the center of the galaxy
        is less than the sum of the position error radius and the galaxy radius AND that the position error radius
        is smaller than the galaxy radius.  This position is inside the (circularized) NGC galaxy (or nearly in).
2^15    When this bit is a 1, it indicates that the distance between the position and the center of the galaxy
        is less than the sum of the position error radius and the galaxy radius AND that the galaxy radius
        is smaller than the position error radius.  This galaxy is inside the position error circle (or nearly in).
2^16-29 Not assigned.
2^30    When this bit is a 1, then this Notice was ground-generated; else flight-generated.
        A Gnd-generated Notice is issued anytime humans during ground operations
        have determined if this Notice should be issued.  This will typically be used
        during the early phases of the mission before the automatic real-time distribution
        capability is enabled.  Once the burst position has been validated by humans,
        a GRB_Pos Notice will be issued (with this flag bit set to 1).
        After the Verification phase, there still may be occaisions when a GRB_POS notice
        is (re)created by human-processing and will be distributed (with this bit set to 1). 
2^31    Not assigned.

The "signif" is composed of two 16-bit quantitites packed into a single 32-bit field.
The two quantities are the X-axis and Y-axis 1-dimentional significance detections.
The X-direction significance is packed in the bottom 16 bits, and the Y-direction
is packed in the upper 16 bits.  Prior to packing, both significances were multiplied
by 100 and integerized (i.e centi-sigma)..

There are a total of 17(tbr) "spare" items in the packet (usually filled
with zeros).  They are located at items 15-16(tbd) and 21-35.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.







================================================================================

TYPE=110 (&116) PACKET CONTENTS:

The FERMI_GBM_ALERT (& FERMI_GBM_ALERT_INTERNAL) packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=110 or 116)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       trig_num     integer         Trigger number
long         5       trig_tjd     [days]          Truncated Julian Day
long         6       trig_sod     [centi-sec]     (int)(sssss.sss *100)
long         7-8     spare[2]     integer         8 bytes for the future
long         9       algorithm    integer         Which loc calc algorithm used
long         10-13   spare[4]     integer         20 bytes for the future
long         14      trig_dur     [milli-sec]     Trigger Integration time [msec]
long         15      lo_chan      integer         ???
long         16      hi_chan      integer         ???
long         17      adc_lochan   integer         ???
long         18      adc_hichan   integer         ???
long         19      misc         bits            Misc stuff packed in here
long         20      rec_seq_num  integer         SerNum of the messages (1-~20)
long         21      trig_signif  [deci-sigma]    (int)(Trigger significance *10)
long         22      longitude    [arcmin]        Geo (East) Longitude of s/c
long         23      latitude     [arcmin]        Geo Latitude of s/c
long         24-25   spare[2]     integer         8 bytes of the trigger criterion
long         26      dets         bits            The 2 dets that triggered
long         27-30   spare[4]     integer         16 bytes for the future
long         31      lc_yymmdd    integer         The YYMMDD for the lc_url
long         32      lc_fff       integer         The FFF fraction-of-day for the lc_url
long         33-38   spare[6]     integer         24 bytes for the future
long         39      pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 110, 111, 112, 117, 118, and 119;
that tells which type of packet this is and therefore what the contents are
(the contents change between the various types).

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start at 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the trig_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "trig_num" is a uniquely identifying number for each trigger.
It is the number of seconds since 01 Jan 2001.

The "trig_tjd" is the Truncated Julian Day of the FERMI-GBM trigger,
eg. TJD=13370 is 01 Jan 2005.  (Note that at the time when this Alert Notice
is issued, it is not yet known if this trigger is due to a real GRB,
or some non-GRB cause (eg. failed or false trigger).)

The "trig_sod" is the UT seconds-of-day (SOD) of the FERMI-GBM trigger.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized).  (Since this is the Alert Notice, it is not known yet
if this trigger is a real GRB or just a false trigger.)

The "algorithm" is the index value identifying which location_calculation
algorithm was used for this event.

The "trig_dur" is the duration of the trigger criteria that initiated this event.
The units are in milli-seconds (ie the fl.pt. seconds were multiplied by 1000
and then integerized).

The "lo_chan" and "hi_chan" are ....

The "adc_lochan" and "adc_hichan" are ...

The "misc" field contains miscellaneous flag bits.
2^0-27  Not assigned.
29       G     temporal_coinc 0=No or 1=Yes, there was a temporal coincidence with another event
               (note the non-standard location; these coinc bits are usually in the trigger_id slot)
2^30    When this bit is a 1, then this Notice was ground-generated; else flight-generated.
        A Gnd-generated Notice is issued anytime humans during ground operations
        have determined if this Notice should be issued.  This will typically be used
        during the early phases of the mission before the automatic real-time distribution
        capability is enabled.
2^31    Not assigned.

The "rec_seq_num" is the Record Sequence Number.  It is a packet serial number
spanning all the messages that come from the s/c during a given trigger.
It will start with a value of 1 for the FERMI_GBM_ALERT Notice and increment during 
the multiple GBM_POS Notices, etc.

The "trig_signif" is the significance of the trigger in units of deci-sigma.
It is the significance of the initial trigger (ie of the data in the trigger_time interval).
The value is multiplied by 10 and integerized (i.e. the units are deci-sigma).
(This should not be confused with the "Data Significance" in the FLT_POS and GND_POS notices,
which is the significance of the data in the ongoing integration time interval.)
(This is a deviation from the way 'sigma' is usually stored in GCN packets -- it is
in deci-sigma instead of the usual centi-sigma.)

The "longitude" is the East geographic longitude of the s/c at the time of the trigger.
The units are arcmin (ie divide the 4-byte integer by 60.0 to get fl.pt. degrees).
The "latitude" is the geographic latitude of the s/c at the time of the trigger.
The units are arcmin (ie divide the 4-byte integer by 60.0 to get fl.pt. degrees).

The "det" contains which two of the 14 detectors that resulted in this event.
There are 12 position-determining detectors and 2 spectrum-measuring detectors (BGO).
The packing sequence starts with Det0 in the 2^0 bit, Det1 in 2^1, ..., DetB in 2^11,
and the two BGO dets in 2^12 and 2^13.

The "lc_yymmdd" contains the YYMMDD information for building the lightcurve URL.
When forming the url string (to wget the lc file) this file should be converted
using the "%06d" (ie decimal form in the C-language) to for the "YYMMDD" portion of the string:
http://heasarc.gsfc.nasa.gov/FTP/fermi/data/gbm/triggers/20YY/bnYYMMDDFFF/quicklook/glg_lc_medres34_bnYYMMDDFFF.gif

The "lc_fff" contains the FFF fraction-of-day information for building the lightcurve URL.
When forming the url string (to wget the lc file) this file should be converted
using the "%03d" (ie decimal form in the C-language) to for the "fff" portion of the string:
http://heasarc.gsfc.nasa.gov/FTP/fermi/data/gbm/triggers/20YY/bnYYMMDDFFF/quicklook/glg_lc_medres34_bnYYMMDDFFF.gif

There are a total of 18 "spare" items in the packet (usually filled
with zeros).  They are located at items 7-8, 10-13, 24-25, 27-30, and 32-38.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.



TYPE=111 (&117) PACKET CONTENTS:

The FERMI_GBM_FLT_POS (& FERMI_GBM_FLT_INTERNAL) packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=111 or 117)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       trig_num     integer         Trigger number
long         5       burst_tjd    [days]          Truncated Julian Day
long         6       burst_sod    [centi-sec]     (int)(sssss.sss *100)
long         7       burst_ra     [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         8       burst_dec    [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         9       burst_flue   [counts]        Num events during trig window, 0 to inf
long         10      spare        integer         4 bytes for the future
long         11      burst_error  [0.0001-deg]    (int)(0.0 to 180.0 *10000)
long         12      phi          [centi-deg]     (int)(0.0 to 359.9999 *100)
long         13      theta        [centi-deg]     (int)(0.0 to +100.0 *100)
long         14      trigTmScale  [milli-sec]     Trigger Time Scale
long         15-16   spare[2]     integer         8 bytes for the future
long         17      dataTmScale  [milli-sec]     Data Time Scale
long         18      soln_status  bits            Type of source/trigger found
long         19      misc         bits            Misc stuff packed in here
long         20      rec_seq_num  integer         SerNum of the messages (1-~40)
long         21      data_signif  [centi-sigma]   (int)(sig2noise *100)
long         22      loc_algor    integer         Location algorithm (1-???)
long         23      most_likely  integer         Mostly class ID & probability
long         24      2most_likely integer         2nd Mostly class ID & probability
long         25      hard_ratio   [centi-dn]      (int)(hardness_ratio *100)
long         26      dets         bits            The 2 dets that triggered
long         26-30   spare[5]     integers        20 bytes for the future
long         31      lc_yymmdd    integer         The YYMMDD for the lc_url
long         32      lc_fff       integer         The FFF fraction-of-day for the lc_url
long         33-36   spare[4]     integer         16 bytes for the future
long         37      longitude    [arcmin]        Geo (East) Longitude of s/c
long         38      latitude     [arcmin]        Geo Latitude of s/c
long         39      pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 110, 111, 112, 117, 118, and 119;
that tells which type of packet this is and therefore what the contents are
(the contents change between the various types).

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start at 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the burst_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "trig_num" is a uniquely identifying number for each trigger.
It is the number of seconds since 01 Jan 2001.

The "burst_tjd" is the Truncated Julian Day of the FERMI-GBM burst trigger,
eg. TJD=13370 is 01 Jan 2005.

The "burst_sod" is the UT seconds-of-day (SOD) of the FERMI-GBM burst trigger.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized).

The "burst_ra" and "burst_dec" are the RA,Dec coordinates of the burst (or transient)
as determined by the GBM flight software (J2000 epoch).  These intrinsically
floating point quantities have been multiplied by 10,000 and then integerized
to make an integer quantity with units of 0.0001-degrees (over-precision,
but this is the new standard scaling for all missions/instruments).

The "burst_flue" is the number of events in the 50-300 keV band 
during the 'integ_time' time interval of the successful trigger criterion.
(is this bkg subtracted?????)

The "burst_error" is the radius of the circle that will contain on average
TBD% of the bursts.  The units are 0.0001-degrees (ie the fl.pt. error was multiplied
by 10000 and then integerized).

The "phi" and "theta" are the GBM instrument coordinates of the trigger (burst)
as determined by the GBM flight software program.  Phi is measured phi=0
along the +X-axis of the s/c and increasing towards the +Y axis,
and theta is measured from the boresight of the LAT instrument (the +Z axis).
The units are in centi-degees (ie the fl.pt. degrees were multiplied by 100
and then integerized).

The "trigTmScale" is the binning (ie timesampling) of the countrate lightcurve
used for the initial countrate lightcurve trigger detection.
The units are in milli-seconds (ie the fl.pt. seconds were multiplied by 1000
and then integerized).

The "dataTmScale" is the binning (ie timesampling) of the countrate lightcurve
used during the iterative location calculation phase of the processing.
The units are in milli-seconds (ie the fl.pt. seconds were multiplied by 1000
and then integerized).

The "soln_status" (aka trigger_id) field contains (a) flag bits on the type of solution
that was found in the on-board image and about whether it matches anything in the
on-board source catalog, and (b) flag bits that are assigned in the GCN ground-processing.
These bits will be determined largely (but not completely) by the ClassID in the Most_Likely fields.
The bit packing of the "soln_status" entry in the packet is:
BitNum  F-v-G  Item_name      Description
------  ----   ---------      -----------
0-4      -     spare          spares for future use
5        G     def_not_grb    0=No or 1=Yes, it is definitely not a GRB (bad label; its really a Retraction)
6-11     -     spare          spares for future use
12       G     blk_cat_src    0=No or 1=Yes, it is in the catalog of sources to be blocked (internal use only)
13-27    -     spare          spares for future use
28       G     spatial_coinc  0=No or 1=Yes, there was a spatial coincidence with another event
29       G     temporal_coinc 0=No or 1=Yes, there was a temporal coincidence with another event
30       G     test_submit    0=No or 1=Yes, this is a test submission (internal use only)
31       -     spare          spare for future use
The columns are:  "BitNum" is the 2^N location within the 4-byte word; "F-v-G" is Flight-assigned vs
Ground-assigned;  "Item_name" is the short-description name of the quantity;  and
"Description" is the longer description of the quantity.

The "misc" field contains miscellaneous flag bits.
2^0-10  Not asigned.
2^11    When this bit is a 1, it indicates that 1 (or more) of the RA,Dec,Lon,Lat values
        were out of valid range (eg RA was 361 deg).
2^12-23 Not assigned.
2^24    When this bit is a 1, then the Dets (slot 26) and Lon,Lat (slots 37,38) have been filled
        from the GBM_ALERT Notice.
2^25-29 Not assigned.
2^30    When this bit is a 1, then this Notice was ground-generated; else flight-generated.
        A Gnd-generated Notice is issued anytime humans during ground operations
        have determined if this Notice should be issued.  This will typically be used
        during the early phases of the mission before the automatic real-time distribution
        capability is enabled.  Once the burst position has been validated by humans,
        a GBM_Pos Notice will be issued (with this flag bit set to 1).
        After the Verification phase, there still may be occaisions when a GBM_POS notice
        is (re)created by human-processing and will be distributed (with this bit set to 1). 
2^31    Not assigned.

The "rec_seq_num" is the Record Sequence Number.  It is a serial number spanning
all the messages (of all types) that come from the s/c during a given trigger.
It will start with a value of 1 for the FERMI_GBM_ALERT Notice and increment during 
the multiple GBM_POS Notices, etc.

The "data_signif" is the significance level of the data in the on-going integrated data.
This is the signal/noise (ie number of sigma) of the rate-trigger detection
multiplied by 100 and then integerized (ie units are centi-sigma).
(This should not be confused with the "trigger Significance" in the ALERT Notice,
which is the significance of the data in a just the initial trigger interval.)

The "loc_algorithm" field contains the version_number of the Flt S/W Location Algorithm
that was used to derive this RA,Dec location for this event.  The values are:
1   (Only one is defined so far in the FSW)
2   (Future use as on-orbit experience dictates)
3   (Future use as on-orbit experience dictates)

The "most_likely" field contains two values (2 shorts into the 1 long).
The least significant two bytes contains an the index value identifying
the source_object class of this event (eg GRB, solar flare, electron precip, etc).
The most significant two bytes contain the probability (ie confidence level) that
this class assignment is correct.  The range of the probability is 0.0 to 1.0;
and the units are in centi-probability (ie prob*100).
CLASS TABLE:
    //  Classes 0 to 3 are "single classes": the probabilities of these
    //  will be either 0 or 1, never an in between value.
    0  ERROR                - ????????????precise wording needed!!!!!!
    1  UNRELIABLE_LOCATION  - ????????????precise wording needed!!!!!!
    2  LOCAL_PARTICLES      - Local particles, equal rates in opposite detectors
    3  BELOW_HORIZON        - Distant particles, assumed to come from the horizon
    //  The remaining classes have probabilities calculated
    //  from Bayes Theorem: any value from 0.0 to 1.0 is possible.
    4  GRB                  - For bursts with good localizations
    5  GENERIC_SGR          - This is any SGR (except 1806-20)
    6  GENERIC_TRANSIENT    - Any astrophysical transient not included elsewhere
    7  DISTANT_PARTICLES    - Particles at a distance
    8  SOLAR_FLARE          - This is a Solar Flare event
    9  CYG_X1               - This trigger was caused by a CygX1 fluctuation
    10 SGR_1806_20          - This trigger came from SGR 1806-20
    11 GROJ_0422_32         - This trigger came from GRO J0422-32
    12-18  undefined        - (spares for future use)
    19 TGF                  - Terrestial Gamma Flashes

The "2most_likely" field contains two values (2 shorts into the 1 long).
The least significant two bytes contains an the index value identifying
the second-most-likely source_object class of this event.
The most significant two bytes contain the probability (ie confidence level) that
this class assignment is correct.  The range of the probability is 0.0 to 1.0;
and the units are in centi-probability (ie prob*100).
The Class Table is the same as above.

The "hardness_ratio" is the hardness ratio (HR = Cnts(15-50keV) / Cnts(50-300keV)).
The units are in centi-dig_number (ie hard_ratio = (int)(HR*100)).

The "det" contains which two of the 14 detectors that resulted in this event.
There are 12 position-determining detectors and 2 spectrum-measuring detectors (BGO).
The packing sequence starts with Det0 in the 2^0 bit, Det1 in 2^1, ..., DetB in 2^11,
and the two BGO dets in 2^12 and 2^13.  This field is valid only if the 2^24 bit in MISC
is set.

The "lc_yymmdd" contains the YYMMDD information for building the lightcurve URL.
When forming the url string (to wget the lc file) this file should be converted
using the "%06d" (ie decimal form in the C-language) to for the "YYMMDD" portion of the string:
http://heasarc.gsfc.nasa.gov/FTP/fermi/data/gbm/triggers/20YY/bnYYMMDDFFF/quicklook/glg_lc_medres34_bnYYMMDDFFF.gif

The "lc_fff" contains the FFF fraction-of-day information for building the lightcurve URL.
When forming the url string (to wget the lc file) this file should be converted
using the "%03d" (ie decimal form in the C-language) to for the "fff" portion of the string:
http://heasarc.gsfc.nasa.gov/FTP/fermi/data/gbm/triggers/20YY/bnYYMMDDFFF/quicklook/glg_lc_medres34_bnYYMMDDFFF.gif

The "longitude" is the East geographic longitude of the s/c at the time of the trigger.
The units are arcmin (ie divide the 4-byte integer by 60.0 to get fl.pt. degrees).
The "latitude" is the geographic latitude of the s/c at the time of the trigger.
The units are arcmin (ie divide the 4-byte integer by 60.0 to get fl.pt. degrees).

There are a total of 12 "spare" items in the packet (usually filled
with zeros).  They are located at items 15-16, 25-30, and 33-36.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.



TYPE=112 PACKET CONTENTS:

The FERMI_GBM_GND_POS packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=112)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       trig_num     integer         Trigger number
long         5       burst_tjd    [days]          Truncated Julian Day
long         6       burst_sod    [centi-sec]     (int)(sssss.sss *100)
long         7       burst_ra     [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         8       burst_dec    [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         9-10    spare[2]     integers        8 bytes for the future
long         11      burst_error  [0.0001-deg]    (int)(0.0 to 180.0 *10000)
long         12      phi          [centi-deg]     (int)(0.0 to 359.9999 *100)
long         13      theta        [centi-deg]     (int)(0.0 to +100.0 *100)
long         14      data_interval[milli-sec]     Duration of the trigger interval, 1 to inf
long         15-17   spare[3]     integer         12 bytes for the future
long         18      soln_status  bits            Type of source/trigger found
long         19      misc         bits            Misc stuff packed in here
long         20      rec_seq_num  integer         SerNum of the messages (1-~40)
long         21      burst_signif [deci-sigma]    (int)(sig2noise *10)
long         22      loc_algor    integer         Location algorithm (1-???)
long         23-25   spare[3]     integers        12 bytes for the future
long         26      lo_E         integer         E_range_lo *1000 keV
long         27      hi_E         integer         E_range_hi *1000 keV
long         28      sc_x         integer         X-coordinate of s/c *4.0 km
long         29      sc_y         integer         Y-coordinate of s/c *4.0 km 
long         30      sc_z         integer         Z-coordinate of s/c *4.0 km 
long         31      lc_yymmdd    integer         The YYMMDD for the lc_url
long         32      lc_fff       integer         The FFF fraction-of-day for the lc_url
long         33-38   spare[6]     integer         24 bytes for the future
long         39      pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 110, 111, 112, 117, 118, and 119;
that tells which type of packet this is and therefore what the contents are
(the contents change between the various types).

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start at 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the burst_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "trig_num" is a uniquely identifying number for each trigger.
It is the number of seconds since 01 Jan 2001.

The "burst_tjd" is the Truncated Julian Day of the FERMI-GBM burst trigger,
eg. TJD=13370 is 01 Jan 2005.

The "burst_sod" is the UT seconds-of-day (SOD) of the FERMI-GBM burst trigger.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized).

The "burst_ra" and "burst_dec" are the RA,Dec coordinates of the burst (or transient)
as determined by the GBM flight software (J2000 epoch).  These intrinsically
floating point quantities have been multiplied by 10,000 and then integerized
to make an integer quantity with units of 0.0001-degrees (over-precision,
but this is the new standard scaling for all missions/instruments).

The "burst_error" is the radius of the circle that will contain on average
TBD% of the bursts.  The units are 0.0001-degrees (ie the fl.pt. error was multiplied
by 10000 and then integerized).

The "phi" and "theta" are the GBM instrument coordinates of the trigger (burst)
as determined by the GBM flight software program.  Phi is measured phi=0
along the +X-axis of the s/c and increasing towards the +Y axis,
and theta is measured from the boresight of the LAT instrument (the +Z axis).
The units are in centi-degees (ie the fl.pt. degrees were multiplied by 100
and then integerized).

The "data_interval" is the interval over which the data were accumulated in the MAXRATES packet,
also known as integration_interval.
The units are in milli-seconds (ie the fl.pt. seconds were multiplied by 1000
and then integerized).

The "soln_status" (aka trigger_id) field contains (a) flag bits on the type of solution
that was found in the on-board image and about whether it matches anything in the
on-board source catalog, and (b) flag bits that are assigned in the GCN ground-processing.
These bits will be determined largely (but not completely) by the ClassID in the Most_Likely fields.
The bit packing of the "soln_status" entry in the packet is:
BitNum  F-v-G  Item_name      Description
------  ----   ---------      -----------
0        G     LAT_FOV        In the LAT_FoV
1        G     Hi_E           GBM_hiE_inten_>_some_threshold
2-4      -     spare          spares for future use
5        G     def_not_grb    0=No or 1=Yes, it is definitely not a GRB (bad label; its really a Retraction)
6-11     -     spare          spares for future use
12       G     blk_cat_src    0=No or 1=Yes, it is in the catalog of sources to be blocked (internal use only)
13-25    -     spare          spares for future use
26-27    -     l-v-s          2 bits: 0=Undef,1=ShortGRB,2=LongGRB,3=not_used
28       G     spatial_coinc  0=No or 1=Yes, there was a spatial coincidence with another event
29       G     temporal_coinc 0=No or 1=Yes, there was a temporal coincidence with another event
30       G     test_submit    0=No or 1=Yes, this is a test submission (internal use only)
31       -     spare          spare for future use
The columns are:  "BitNum" is the 2^N location within the 4-byte word; "F-v-G" is Flight-assigned vs
Ground-assigned;  "Item_name" is the short-description name of the quantity;  and
"Description" is the longer description of the quantity.

The "misc" field contains miscellaneous flag bits.
2^0-10  spares.
2^11    When this bit is a 1, it indicates that 1 (or more) of the RA,Dec values
        were out of valid range (eg RA was 361 deg).
2^12-29 spares.
2^30    When this bit is a 1, then this Notice was ground-generated; else flight-generated.
        A Gnd-generated Notice is issued anytime humans during ground operations
        have determined if this Notice should be issued.  This will typically be used
        during the early phases of the mission before the automatic real-time distribution
        capability is enabled.  Once the burst position has been validated by humans,
        a GBM_Pos Notice will be issued (with this flag bit set to 1).
        After the Verification phase, there still may be occaisions when a GBM_POS notice
        is (re)created by human-processing and will be distributed (with this bit set to 1). 
2^31    spare.

The "rec_seq_num" is the Record Sequence Number.  It is a serial number spanning
all the messages (of all types) that come from the s/c during a given trigger.
It will start with a value of 1 for the FERMI_GBM_ALERT Notice and increment during 
the multiple GBM_POS Notices, etc.

The "burst_signif" is the significance level of the initial 'rate trigger'.
This is the signal/noise (ie number of sigma) of the rate-trigger detection
multiplied by 10 and then integerized (ie units are deci-sigma).  (Please note that
this is a deviation from the way 'sigma' is usually stored in GCN packets -- it is
in deci-sigma instead of the usual centi-sigma.)

The "loc_algorithm" field contains the version_number of the Location Algorithm Gnd s/w
that was used to derive this RA,Dec location for this event.  The values are:
1   (Only one is defined so far in the Gnd SW)
2   (Future use as on-orbit experience dictates)
3   (Future use as on-orbit experience dictates)

The 'lo_E' is the bottom of the energy range for ...
It is in eV -- keV/1000 and then intergerized.
The 'hi_E' is the bottom of the energy range for ...
It is in eV -- keV/1000 and then intergerized.

The "lc_yymmdd" contains the YYMMDD information for building the lightcurve URL.
When forming the url string (to wget the lc file) this file should be converted
using the "%06d" (ie decimal form in the C-language) to for the "YYMMDD" portion of the string:
http://heasarc.gsfc.nasa.gov/FTP/fermi/data/gbm/triggers/20YY/bnYYMMDDFFF/quicklook/glg_lc_medres34_bnYYMMDDFFF.gif

The "lc_fff" contains the FFF fraction-of-day information for building the lightcurve URL.
When forming the url string (to wget the lc file) this file should be converted
using the "%03d" (ie decimal form in the C-language) to for the "fff" portion of the string:
http://heasarc.gsfc.nasa.gov/FTP/fermi/data/gbm/triggers/20YY/bnYYMMDDFFF/quicklook/glg_lc_medres34_bnYYMMDDFFF.gif

There are a total of 18 "spare" items in the packet (usually filled
with zeros).  They are located at items 9-10, 15-17, and 32-38.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.




TYPE=113 PACKET CONTENTS:

The FERMI_GBM_LC packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=113)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       trig_num     integer         Trigger number
long         5       trig_tjd     [days]          Truncated Julian Day
long         6       trig_sod     [centi-sec]     (int)(sssss.sss *100)
long         7       burst_ra     [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         8       burst_dec    [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         9       burst_inten  [counts]        Num events during trig window, 0 to inf
long         10-15   spare[6]     integers        24 bytes for the future
//long       12      phi          [centi-deg]     (int)(0.0 to 359.9999 *100)
//long       13      theta        [centi-deg]     (int)(0.0 to +100.0 *100)
//long       16      lon_lat      2_shorts        (int)(Longitude,Lattitude *100)
long         17      spare        integer         4 bytes for the future
long         18      soln_status  bits            Type of source/trigger found
long         19      misc         bits            Misc stuff packed in here
long         20      rec_seq_num  integer         SerNum of the messages (1-~20)
long         21-38   spare[18]    integer         72 bytes for the future
long         39      pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 110, 111, 112, 117, 118, and 119;
that tells which type of packet this is and therefore what the contents are
(the contents change between the various types).

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start at 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the trig_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "trig_num" is a uniquely identifying number for each trigger.
It is the number of seconds since 01 Jan 2001.

The "burst_tjd" is the Truncated Julian Day of the FERMI-GBM trigger,
eg. TJD=13370 is 01 Jan 2005.  (Note that at the time when this Alert Notice
is issued, it is not yet known if this trigger is due to a real GRB,
or some non-GRB cause (eg. failed or false trigger).)

The "burst_sod" is the UT seconds-of-day (SOD) of the FERMI-GBM trigger.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized).  (Since this is the Alert Notice, it is not known yet
if this trigger is a real GRB or just a false trigger.)

The "burst_ra" and "burst_dec" are the RA,Dec coordinates of the burst (or transient)
as determined by the GBM flight software (J2000 epoch).  These intrinsically
floating point quantities have been multiplied by 10,000 and then integerized
to make an integer quantity with units of 0.0001-degrees.

The "burst_flue" is the number of events in the 50-300 keV band 
during the 'integ_time' time interval of the successful trigger criterion.
(is this bkg subtracted?????)

The "burst_inten" is the number of events in the 50-300 keV band
averaged over the two closed detectors and cosine-corrected.
(is this bkg subtracted?????)

The "burst_error" is the radius of the circle that will contain on average
TBD% of the bursts.  The units are 0.0001-degrees (ie the fl.pt. error was multiplied
by 10000 and then integerized).

//The "phi" and "theta" are the GBM instrument coordinates of the trigger (burst)
//as determined by the GBM flight software program.  Phi is measured phi=0
//along the (TBC)+Y-axis of the s/c and increasing towards the -Z axis,
//and theta is measured from the boresight of the GBM instrument.
//The units are in centi-degees (ie the fl.pt. degrees were multiplied by 100
//and then integerized).

??????????????stuff here???????????

The "soln_status" (aka trigger_id) field contains (a) flag bits on the type of solution
that was found in the on-board image and about whether it matches anything in the
on-board source catalog, and (b) flag bits that are assigned in the GCN ground-processing.
The bit packing of the "soln_status" entry in the packet is:
BitNum  F-v-G  Item_name      Description
------  ----   ---------      -----------
0-4      -     spare          spares for future use
5        G     def_not_grb    0=No or 1=Yes, it is definitely not a GRB (bad label; its really a Retraction)
6-11     -     spare          spares for future use
12       G     blk_cat_src    0=No or 1=Yes, it is in the catalog of sources to be blocked (internal use only)
13-29    -     spare          spares for future use
30       G     test_submit    0=No or 1=Yes, this is a test submission (internal use only)
31       -     spare          spare for future use
The columns are:  "BitNum" is the 2^N location within the 4-byte word; "F-v-G" is Flight-assigned vs
Ground-assigned;  "Item_name" is the short-description name of the quantity;  and
"Description" is the longer description of the quantity.

The "misc" field contains miscellaneous flag bits.
2^0-10  spares.
2^11    When this bit is a 1, it indicates that 1 (or more) of the RA,Dec values
        were out of valid range (eg RA was 361 deg).
2^12-29 spares.
2^30    When this bit is a 1, then this Notice was ground-generated; else flight-generated.
        A Gnd-generated Notice is issued anytime humans during ground operations
        have determined if this Notice should be issued.  This will typically be used
        during the early phases of the mission before the automatic real-time distribution
        capability is enabled.  Once the burst position has been validated by humans,
        a GBM_Pos Notice will be issued (with this flag bit set to 1).
        After the Verification phase, there may be occaisions when a GBM_LC notice
        is (re)created by human-processing and will be distributed (with this bit set to 1). 
2^31    spare.

The "rec_seq_num" is the Record Sequence Number.  It is a serial number spanning
all the messages that come from the s/c during a given trigger.  It will start
with a value of 1 for the FERMI_GBM_ALERT Notice and increment during 
the multiple GBM_POS Notices, etc.

There are a total of 13(tbr) "spare" items in the packet (usually filled
with zeros).  They are located at items 15-16 and 25-35.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.




TYPE=114 PACKET CONTENTS:

Not for public distribution -- Fermi-GBM team only.

The FERMI_GBM_GND_INTERNAL packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Same as 112 except slots 23 and 24 are filled.

long         23      most_likely  integer         Mostly class ID and probability
long         24      2most_likely integer         2nd Mostly class ID and probability





TYPE=115 (& 146) PACKET CONTENTS:

The FERMI_GBM_FINAL_POS (& FERMI_GBM_FINAL_POS_INTERNAL) packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=115)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       trig_num     integer         Trigger number
long         5       burst_tjd    [days]          Truncated Julian Day
long         6       burst_sod    [centi-sec]     (int)(sssss.sss *100)
long         7       burst_ra     [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         8       burst_dec    [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         9-10    spare[2]     integers        8 bytes for the future
long         11      burst_error  [0.0001-deg]    (int)(0.0 to 180.0 *10000)
long         12      phi          [centi-deg]     (int)(0.0 to 359.9999 *100)
long         13      theta        [centi-deg]     (int)(0.0 to +100.0 *100)
long         14-17   spare[4]     integer         16 bytes for the future
long         18      soln_status  bits            Type of source/trigger found
long         19      misc         bits            Misc stuff packed in here
long         20-21   spare[2]     integer         8 bytes for the future
long         22      loc_algor    integer         Location algorithm (1-???)
long         23-30   spare[8]     integers        32 bytes for the future
long         31      lc_yymmdd    integer         The YYMMDD for the lc_url
long         32      lc_fff       integer         The FFF fraction-of-day for the lc_url
long         33-38   spare[6]     integer         24 bytes for the future
long         39      pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 110, 111, 112, 117, 118, and 119;
that tells which type of packet this is and therefore what the contents are
(the contents change between the various types).

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start at 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the burst_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "trig_num" is a uniquely identifying number for each trigger.
It is the number of seconds since 01 Jan 2001.

The "burst_tjd" is the Truncated Julian Day of the FERMI-GBM burst trigger,
eg. TJD=13370 is 01 Jan 2005.

The "burst_sod" is the UT seconds-of-day (SOD) of the FERMI-GBM burst trigger.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized).

The "burst_ra" and "burst_dec" are the RA,Dec coordinates of the burst (or transient)
as determined by the GBM flight software (J2000 epoch).  These intrinsically
floating point quantities have been multiplied by 10,000 and then integerized
to make an integer quantity with units of 0.0001-degrees (over-precision,
but this is the new standard scaling for all missions/instruments).

The "burst_error" is the radius of the circle that will contain on average
TBD% of the bursts.  The units are 0.0001-degrees (ie the fl.pt. error was multiplied
by 10000 and then integerized).

The "phi" and "theta" are the GBM instrument coordinates of the trigger (burst)
as determined by the GBM flight software program.  Phi is measured phi=0
along the +X-axis of the s/c and increasing towards the +Y axis,
and theta is measured from the boresight of the LAT instrument (the +Z axis).
The units are in centi-degees (ie the fl.pt. degrees were multiplied by 100
and then integerized).

The "soln_status" (aka trigger_id) field contains (a) flag bits on the type of solution
that was found in the on-board image and about whether it matches anything in the
on-board source catalog, and (b) flag bits that are assigned in the GCN ground-processing.
These bits will be determined largely (but not completely) by the ClassID in the Most_Likely fields.
The bit packing of the "soln_status" entry in the packet is:
BitNum  F-v-G  Item_name      Description
------  ----   ---------      -----------
0        G     LAT_FOV        In the LAT_FoV
1        G     Hi_E           GBM_hiE_inten_>_some_threshold
2-4      -     spare          spares for future use
5        G     def_not_grb    0=No or 1=Yes, it is definitely not a GRB (bad label; its really a Retraction)
6-11     -     spare          spares for future use
12       G     blk_cat_src    0=No or 1=Yes, it is in the catalog of sources to be blocked (internal use only)
13-25    -     spare          spares for future use
26-27    -     l-v-s          2 bits: 0=Undef,1=ShortGRB,2=LongGRB,3=not_used
28       G     spatial_coinc  0=No or 1=Yes, there was a spatial coincidence with another event
29       G     temporal_coinc 0=No or 1=Yes, there was a temporal coincidence with another event
30       G     test_submit    0=No or 1=Yes, this is a test submission (internal use only)
31       -     spare          spare for future use
The columns are:  "BitNum" is the 2^N location within the 4-byte word; "F-v-G" is Flight-assigned vs
Ground-assigned;  "Item_name" is the short-description name of the quantity;  and
"Description" is the longer description of the quantity.

The "misc" field contains miscellaneous flag bits.
2^0     When this bit is a 1, then bits 2^0, 2^1, and 2^26-27 in slot 18 are assigned and valid;
        when not set, those 4 bits are undefined.
2^0-10  spares.
2^11    When this bit is a 1, it indicates that 1 (or more) of the RA,Dec values
        were out of valid range (eg RA was 361 deg).
2^12-29 spares.
2^30    When this bit is a 1, then this Notice was ground-generated; else flight-generated.
        A Gnd-generated Notice is issued anytime humans during ground operations
        have determined if this Notice should be issued.  This will typically be used
        during the early phases of the mission before the automatic real-time distribution
        capability is enabled.  Once the burst position has been validated by humans,
        a GBM_Pos Notice will be issued (with this flag bit set to 1).
        After the Verification phase, there still may be occaisions when a GBM_POS notice
        is (re)created by human-processing and will be distributed (with this bit set to 1). 
2^31    spare.

The "loc_algorithm" field contains the version_number of the Location Algorithm Gnd s/w
that was used to derive this RA,Dec location for this event.  The values are:
1   (Only one is defined so far in the Gnd SW)
2   (Future use as on-orbit experience dictates)
3   (Future use as on-orbit experience dictates)

The "lc_yymmdd" contains the YYMMDD information for building the lightcurve URL.
When forming the url string (to wget the lc file) this file should be converted
using the "%06d" (ie decimal form in the C-language) to for the "YYMMDD" portion of the string:
http://heasarc.gsfc.nasa.gov/FTP/fermi/data/gbm/triggers/20YY/bnYYMMDDFFF/quicklook/glg_lc_medres34_bnYYMMDDFFF.gif

The "lc_fff" contains the FFF fraction-of-day information for building the lightcurve URL.
When forming the url string (to wget the lc file) this file should be converted
using the "%03d" (ie decimal form in the C-language) to for the "fff" portion of the string:
http://heasarc.gsfc.nasa.gov/FTP/fermi/data/gbm/triggers/20YY/bnYYMMDDFFF/quicklook/glg_lc_medres34_bnYYMMDDFFF.gif

There are a total of 15(tbr) "spare" items in the packet (usually filled
with zeros).  They are located at items 15-16(tbd) and 25-35.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.





TYPE=118 PACKET CONTENTS:

The FERMI_GBM_TRANS packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=118)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       trig_num     integer         Trigger number
long         5       trig_tjd     [days]          Truncated Julian Day
long         6       trig_sod     [centi-sec]     (int)(sssss.sss *100)
long         7       trans_ra     [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         8       trans_dec    [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         9       trans_inten  [counts]        Num events during trig window, 0 to inf
long         10      spare        integer         4 bytes for the future
long         11      trans_error  [0.0001-deg]    (int)(0.0 to 180.0 *10000)
long         12      phi          [centi-deg]     (int)(0.0 to 359.9999 *100)
long         13      theta        [centi-deg]     (int)(0.0 to +100.0 *100)
long         14      integ_time   [centi-sec]     Duration of the trigger interval, 1 to inf
long         15      spare        integer         4 bytes for the future
long         16      lon_lat      2_shorts        (int)(Longitude,Lattitude *100)
long         17      spare        integer         4 bytes for the future
long         18      soln_status  bits            Type of source/trigger found
long         19      misc         bits            Misc stuff packed in here
long         20      rec_seq_num  integer         SerNum of the messages (1-~40)
long         21      burst_signif [deci-sigma]    (int)(sig2noise *10)
long         22      loc_algor    integer         Location algorithm (1-???)
long         23      most_likely  integer         Mostly class ID & probability
long         24      2most_likely integer         2nd Mostly class ID & probability
long         25-38   spare[14]    integers        56 bytes for the future
long         39      pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 110, 111, 112, 117, 118, and 119;
that tells which type of packet this is and therefore what the contents are
(the contents change between the various types).

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start at 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the trig_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "trig_num" is a uniquely identifying number for each trigger.
It is the number of seconds since 01 Jan 2001.

The "trig_tjd" is the Truncated Julian Day of the FERMI-GBM trigger,
eg. TJD=13370 is 01 Jan 2005.  (Note that at the time when this Alert Notice
is issued, it is not yet known if this trigger is due to a real GRB,
or some non-GRB cause (eg. failed or false trigger).)

The "trig_sod" is the UT seconds-of-day (SOD) of the FERMI-GBM trigger.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized).  (Since this is the Alert Notice, it is not known yet
if this trigger is a real GRB or just a false trigger.)

The "trans_ra" and "trans_dec" are the RA,Dec coordinates of the trans
as determined by the GBM flight software (J2000 epoch).  These intrinsically
floating point quantities have been multiplied by 10,000 and then integerized
to make an integer quantity with units of 0.0001-degrees.

The "trans_flue" is the number of events in the 50-300 keV band 
during the 'integ_time' time interval of the successful trigger criterion.
(is this bkg subtracted?????)

The "trans_inten" is the number of events in the 50-300 keV band
averaged over the two closed detectors and cosine-corrected.
(is this bkg subtracted?????)

The "trans_error" is the radius of the circle that will contain on average
TBD% of the transients.  The units are 0.0001-degrees (ie the fl.pt. error was multiplied
by 10000 and then integerized).

The "phi" and "theta" are the GBM instrument coordinates of the trigger (transient)
as determined by the GBM flight software program.  Phi is measured phi=0
along the (TBC)+Y-axis of the s/c and increasing towards the -Z axis,
and theta is measured from the boresight of the LAT instrument.
The units are in centi-degees (ie the fl.pt. degrees were multiplied by 100
and then integerized).

The "integ_time" is the duration of the sampling interval
of the successful trigger criteria.  The units are in centi-seconds
(ie the fl.pt. seconds were multiplied by 100 and then integerized).

The "loc_algorithm" field contains the index value of which Location Algorithm
was used to derive this RA,Dec location for this event.  The values are:
1   (Only one is defined so far in the FSW; it really does not have a name)
2   (Future use as on-orbit experience dictates)
3   (Future use as on-orbit experience dictates)

The "most_likely" field contains two values (2 shorts into the 1 long).
The least significant two bytes contains an the index value identifying
the source_object class of this event (eg GRB, solar flare, electron precip, etc).
The most significant two bytes contain the probability (ie confidence level) that
this class assignment is correct.  The range of the probability is 0.0 to 1.0;
and the units are in centi-probability (ie prob*100).
CLASS TABLE:
    //  Classes 0 to 3 are "single classes": the probabilities of these
    //  will be either 0 or 1, never an in between value.
    0  ERROR                - ????????????precice wording needed!!!!!!
    1  UNRELIABLE_LOCATION  - ????????????precice wording needed!!!!!!
    2  LOCAL_PARTICLES      - Local particles, equal rates in opposite detectors
    3  BELOW_HORIZON        - Distant particles, assumed to come from the horizon
    //  The remaining classes have probabilities calculated
    //  from Bayes Theorem: any value from 0.0 to 1.0 is possible.
    4  GRB                  - For bursts with good localizations
    5  GENERIC_SGR          - This is any SGR (except 1806-20)
    6  GENERIC_TRANSIENT    - Any astrophysical transient not included elsewhere
    7  DISTANT_PARTICLES    - ????????????precice wording needed!!!!!!
    8  SOLAR_FLARE          - This is a Solar Flare event
    9  CYG_X1               - This trigger was caused by a CygX1 fluctuation
    10 SGR_1806_20          - This trigger came from SGR 1806-20
    11 GROJ_0422_32         - This trigger came from GRO J0422-32
    12-18  undefined        - (spares for future use)
    19 TGF                  - Terrestial Gamma Flashes

The "2most_likely" field contains two values (2 shorts into the 1 long).
The least significant two bytes contains an the index value identifying
the second-most-likely source_object class of this event.
The most significant two bytes contain the probability (ie confidence level) that
this class assignment is correct.  The range of the probability is 0.0 to 1.0;
and the units are in centi-probability (ie prob*100).
The Class Table is the same as above.

The "soln_status" (aka trigger_id) field contains (a) flag bits on the type of solution
that was found in the on-board image and about whether it matches anything in the
on-board source catalog, and (b) flag bits that are assigned in the GCN ground-processing.
The bit packing of the "soln_status" entry in the packet is:
BitNum  F-v-G  Item_name      Description
------  ----   ---------      -----------
0-4      -     spare          spares for future use
5        G     def_not_grb    0=No or 1=Yes, it is definitely not a GRB (a Retraction)
6-11     -     spare          spares for future use
12       G     blk_cat_src    0=No or 1=Yes, it is in the catalog of sources to be blocked (internal use only)
13-27    -     spare          spares for future use
28       G     spatial_coinc  0=No or 1=Yes, there was a spatial coincidence with another event
29       G     temporal_coinc 0=No or 1=Yes, there was a temporal coincidence with another event
30       G     test_submit    0=No or 1=Yes, this is a test submission (internal use only)
31       -     spare          spare for future use
The columns are:  "BitNum" is the 2^N location within the 4-byte word; "F-v-G" is Flight-assigned vs
Ground-assigned;  "Item_name" is the short-description name of the quantity;  and
"Description" is the longer description of the quantity.

The "misc" field contains miscellaneous flag bits.
2^0-10  spares.
2^11    When this bit is a 1, it indicates that 1 (or more) of the RA,Dec values
        were out of valid range (eg RA was 361 deg).
2^12-29 spares.
2^30    When this bit is a 1, then this Notice was ground-generated; else flight-generated.
        A Gnd-generated Notice is issued anytime humans during ground operations
        have determined if this Notice should be issued.  This will typically be used
        during the early phases of the mission before the automatic real-time distribution
        capability is enabled.  Once the burst position has been validated by humans,
        a GBM_Pos Notice will be issued (with this flag bit set to 1).
        After the Verification phase, there may be rare occaisions when a GBM_POS notice
        is (re)created by human-processing and will be distributed (with this bit set to 1). 
2^31    spare.

The "rec_seq_num" is the Record Sequence Number.  It is a serial number spanning
all the messages that come from the s/c during a given trigger.  It will start
with a value of 1 for the FERMI_GBM_ALERT Notice and increment during 
the multiple GBM_POS Notices, etc.

The "burst_signif" is the significance level of the data in the on-going integrated data.
This is the signal/noise (ie number of sigma) of the rate-trigger detection
multiplied by 10 and then integerized (ie units are deci-sigma).
(This should not be confused with the "trigger Significance" in the ALERT Notice,
which is the significance of the data in a just the initial trigger interval.)
(This is a deviation from the way 'sigma' is usually stored in GCN packets -- it is
in deci-sigma instead of the usual centi-sigma.)

There are a total of 14(tbr) "spare" items in the packet (usually filled
with zeros).  They are located at items 15-16 and 25-35.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.




TYPE=119 PACKET CONTENTS:

The FERMI_GBM_POS_TEST packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=119)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       trig_num     integer         Trigger number
long         5       burst_tjd    [days]          Truncated Julian Day
long         6       burst_sod    [centi-sec]     (int)(sssss.sss *100)
long         7       burst_ra     [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         8       burst_dec    [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         9       burst_flue   [counts]        Num events during trig window, 0 to inf
long         10      spare        integer         4 bytes for the future
long         11      burst_error  [0.0001-deg]    (int)(0.0 to 180.0 *10000)
long         12      phi          [centi-deg]     (int)(0.0 to 359.9999 *100)
long         13      theta        [centi-deg]     (int)(0.0 to +100.0 *100)
long         14      trigTmScale  [milli-sec]     Trigger Time Scale
long         15-16   spare[2]     integer         8 bytes for the future
long         17      dataTmScale  [milli-sec]     Data Time Scale
long         18      soln_status  bits            Type of source/trigger found
long         19      misc         bits            Misc stuff packed in here
long         20      rec_seq_num  integer         SerNum of the messages (1-~40)
long         21      data_signif  [deci-sigma]    (int)(sig2noise *10)
long         22      loc_algor    integer         Location algorithm (1-???)
long         23      most_likely  integer         Mostly class ID & probability
long         24      2most_likely integer         2nd Mostly class ID & probability
long         25      hard_ratio   [centi-dn]      (int)(hardness_ratio *100)
long         26      dets         bits            The 2 dets that triggered
long         26-30   spare[8]     integers        32 bytes for the future
long         31      lc_yymmdd    integer         The YYMMDD for the lc_url
long         32      lc_fff       integer         The FFF fraction-of-day for the lc_url
long         33-36   spare[4]     integer         16 bytes for the future
long         37      longitude    [arcmin]        Geo (East) Longitude of s/c
long         38      latitude     [arcmin]        Geo Latitude of s/c
long         39      pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 110, 111, 112, 117, 118, and 119;
that tells which type of packet this is and therefore what the contents are
(the contents change between the various types).

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start at 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the burst_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "trig_num" is a uniquely identifying number for each trigger.
It is the number of seconds since 01 Jan 2001.

The "burst_tjd" is the Truncated Julian Day of the FERMI-GBM burst trigger,
eg. TJD=13370 is 01 Jan 2005.

The "burst_sod" is the UT seconds-of-day (SOD) of the FERMI-GBM burst trigger.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized).

The "burst_ra" and "burst_dec" are the RA,Dec coordinates of the burst (or transient)
as determined by the GBM flight software (J2000 epoch).  These intrinsically
floating point quantities have been multiplied by 10,000 and then integerized
to make an integer quantity with units of 0.0001-degrees (over-precision,
but this is the new standard scaling for all missions/instruments).

The "burst_flue" is the number of events in the 50-300 keV band 
during the 'integ_time' time interval of the successful trigger criterion.
(is this bkg subtracted?????)

The "burst_error" is the radius of the circle that will contain on average
TBD% of the bursts.  The units are 0.0001-degrees (ie the fl.pt. error was multiplied
by 10000 and then integerized).

The "phi" and "theta" are the GBM instrument coordinates of the trigger (burst)
as determined by the GBM flight software program.  Phi is measured phi=0
along the +X-axis of the s/c and increasing towards the +Y axis,
and theta is measured from the boresight of the LAT instrument (the +Z axis).
The units are in centi-degees (ie the fl.pt. degrees were multiplied by 100
and then integerized).

The "trigTmScale" is the binning (ie timesampling) of the countrate lightcurve
used for the initial countrate lightcurve trigger detection.
The units are in milli-seconds (ie the fl.pt. seconds were multiplied by 1000
and then integerized).

The "dataTmScale" is the binning (ie timesampling) of the countrate lightcurve
used during the iterative location calculation phase of the processing.
The units are in milli-seconds (ie the fl.pt. seconds were multiplied by 1000
and then integerized).

The "soln_status" (aka trigger_id) field contains (a) flag bits on the type of solution
that was found in the on-board image and about whether it matches anything in the
on-board source catalog, and (b) flag bits that are assigned in the GCN ground-processing.
These bits will be determined largely (but not completely) by the ClassID in the Most_Likely fields.
The bit packing of the "soln_status" entry in the packet is:
BitNum  F-v-G  Item_name      Description
------  ----   ---------      -----------
0-4      -     spare          spares for future use
5        G     def_not_grb    0=No or 1=Yes, it is definitely not a GRB (bad label; its really a Retraction)
6-11     -     spare          spares for future use
12       G     blk_cat_src    0=No or 1=Yes, it is in the catalog of sources to be blocked (internal use only)
13-27    -     spare          spares for future use
28       G     spatial_coinc  0=No or 1=Yes, there was a spatial coincidence with another event
29       G     temporal_coinc 0=No or 1=Yes, there was a temporal coincidence with another event
30       G     test_submit    0=No or 1=Yes, this is a test submission (internal use only)
31       -     spare          spare for future use
The columns are:  "BitNum" is the 2^N location within the 4-byte word; "F-v-G" is Flight-assigned vs
Ground-assigned;  "Item_name" is the short-description name of the quantity;  and
"Description" is the longer description of the quantity.

The "misc" field contains miscellaneous flag bits.
2^0-10  Not asigned.
2^11    When this bit is a 1, it indicates that 1 (or more) of the RA,Dec,Lon,Lat values
        were out of valid range (eg RA was 361 deg).
2^12-23 Not assigned.
2^24    When this bit is a 1, then the Dets (slot 26) and Lon,Lat (slots 37,38) have been filled
        from the GBM_ALERT Notice.
2^25-29 Not assigned.
2^30    When this bit is a 1, then this Notice was ground-generated; else flight-generated.
        A Gnd-generated Notice is issued anytime humans during ground operations
        have determined if this Notice should be issued.  This will typically be used
        during the early phases of the mission before the automatic real-time distribution
        capability is enabled.  Once the burst position has been validated by humans,
        a GBM_Pos Notice will be issued (with this flag bit set to 1).
        After the Verification phase, there still may be occaisions when a GBM_POS notice
        is (re)created by human-processing and will be distributed (with this bit set to 1). 
2^31    Not assigned.

The "rec_seq_num" is the Record Sequence Number.  It is a serial number spanning
all the messages (of all types) that come from the s/c during a given trigger.
It will start with a value of 1 for the FERMI_GBM_ALERT Notice and increment during 
the multiple GBM_POS Notices, etc.

The "data_signif" is the significance level of the data in the on-going integrated data.
This is the signal/noise (ie number of sigma) of the rate-trigger detection
multiplied by 10 and then integerized (ie units are deci-sigma).
(This should not be confused with the "trigger Significance" in the ALERT Notice,
which is the significance of the data in a just the initial trigger interval.)
(This is a deviation from the way 'sigma' is usually stored in GCN packets -- it is
in deci-sigma instead of the usual centi-sigma.)

The "loc_algorithm" field contains the version_number of the Flt S/W Location Algorithm
that was used to derive this RA,Dec location for this event.  The values are:
1   (Only one is defined so far in the FSW)
2   (Future use as on-orbit experience dictates)
3   (Future use as on-orbit experience dictates)

The "most_likely" field contains two values (2 shorts into the 1 long).
The least significant two bytes contains an the index value identifying
the source_object class of this event (eg GRB, solar flare, electron precip, etc).
The most significant two bytes contain the probability (ie confidence level) that
this class assignment is correct.  The range of the probability is 0.0 to 1.0;
and the units are in centi-probability (ie prob*100).
CLASS TABLE:
    //  Classes 0 to 3 are "single classes": the probabilities of these
    //  will be either 0 or 1, never an in between value.
    0  ERROR                - ????????????precise wording needed!!!!!!
    1  UNRELIABLE_LOCATION  - ????????????precise wording needed!!!!!!
    2  LOCAL_PARTICLES      - Local particles, equal rates in opposite detectors
    3  BELOW_HORIZON        - Distant particles, assumed to come from the horizon
    //  The remaining classes have probabilities calculated
    //  from Bayes Theorem: any value from 0.0 to 1.0 is possible.
    4  GRB                  - For bursts with good localizations
    5  GENERIC_SGR          - This is any SGR (except 1806-20)
    6  GENERIC_TRANSIENT    - Any astrophysical transient not included elsewhere
    7  DISTANT_PARTICLES    - Particles at a distance
    8  SOLAR_FLARE          - This is a Solar Flare event
    9  CYG_X1               - This trigger was caused by a CygX1 fluctuation
    10 SGR_1806_20          - This trigger came from SGR 1806-20
    11 GROJ_0422_32         - This trigger came from GRO J0422-32
    12-18  undefined        - (spares for future use)
    19 TGF                  - Terrestial Gamma Flashes

The "2most_likely" field contains two values (2 shorts into the 1 long).
The least significant two bytes contains an the index value identifying
the second-most-likely source_object class of this event.
The most significant two bytes contain the probability (ie confidence level) that
this class assignment is correct.  The range of the probability is 0.0 to 1.0;
and the units are in centi-probability (ie prob*100).
The Class Table is the same as above.

The "hardness_ratio" is the hardness ratio (HR = Cnts(15-50keV) / Cnts(50-300keV)).
The units are in centi-dig_number (ie hard_ratio = (int)(HR*100)).

The "det" contains which two of the 14 detectors that resulted in this event.
There are 12 position-determining detectors and 2 spectrum-measuring detectors (BGO).
The packing sequence starts with Det0 in the 2^0 bit, Det1 in 2^1, ..., DetB in 2^11,
and the two BGO dets in 2^12 and 2^13.  This field is valid only if the 2^24 bit in MISC
is set.

The "longitude" is the East geographic longitude of the s/c at the time of the trigger.
The units are arcmin (ie divide the 4-byte integer by 60.0 to get fl.pt. degrees).
The "latitude" is the geographic latitude of the s/c at the time of the trigger.
The units are arcmin (ie divide the 4-byte integer by 60.0 to get fl.pt. degrees).

The "lc_yymmdd" contains the YYMMDD information for building the lightcurve URL.
When forming the url string (to wget the lc file) this file should be converted
using the "%06d" (ie decimal form in the C-language) to for the "YYMMDD" portion of the string.

The "lc_fff" contains the FFF fraction-of-day information for building the lightcurve URL.
When forming the url string (to wget the lc file) this file should be converted
using the "%03d" (ie decimal form in the C-language) to for the "fff" portion of the string.

There are a total of 19(tbr) "spare" items in the packet (usually filled
with zeros).  They are located at items 15-16(tbd) and 25-36.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.




TYPE=120 PACKET CONTENTS:

The FERMI_LAT_GRB_POS_INI packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=120)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       trig_num     integer         Trigger number
long         5       trig_tjd     [days]          Truncated Julian Day
long         6       trig_sod     [centi-sec]     (int)(sssss.sss *100)
long         7       burst_ra     [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         8       burst_dec    [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         9       burst_inten  [counts]        Num events used in location calc, 0 to inf
long         10      inten_4      4_bytes         Evt_cnts in 4 energy bands
long         11      burst_error  [0.0001-deg]    (int)(0.0 to 180.0 *10000)
long         12      phi          [centi-deg]     (int)(0.0 to 359.9999 *100)
long         13      theta        [centi-deg]     (int)(0.0 to +100.0 *100)
long         14      int_time     [msec]          Integration time
long         15-16   spare[2]     integer         8 bytes for the future
long         17      trig_index   integer         Trigger index
long         18      Trig_ID      bits            LAT burst candidate, 1=yes,0=no
long         19      misc         bits            Misc stuff packed in here
long         20      rec_seq_num  integer         SerNum of the messages (1-~40)
long         21-24   spare[4]     integer         16 bytes for the future
long         25      temp_stat    integer         (int)(4*(-log10(probability)))
long         26      image_stat   integer         (int)(4*(-log10(probability)))
long         27-30   spare[4]     integer         16 bytes for the future
long         31      1st_ph_tjd   [days]          First photon used in loc calc
long         32      1st_ph_sod   [centi-sec]     (int)(sssss.sss *100)
long         33      last_ph_tjd  [days]          last photon used in loc calc
long         34      last_ph_sod  [centi-sec]     (int)(sssss.sss *100)
long         35-36   spare[2]     integer         8 bytes for the future
long         37      misc         bits            Misc stuff packed in here
long         38      loc_qual     [0.0001]        Location quality (0.0 - 1.0)
long         39      pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 120, 121, 122, and 123;
that tells which type of packet this is and therefore what the contents are
(the contents change between the various types).

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start at 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the trig_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "trig_num" is a uniquely identifying number for each trigger.
It is the number of seconds since 01 Jan 2001.

The "trig_tjd" is the Truncated Julian Day of the FERMI-LAT trigger,
eg. TJD=14466 is 01 Jan 2008.

The "trig_sod" is the UT seconds-of-day (SOD) of the FERMI-LAT trigger.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized). The trigger time can be off by TBD msec.  This uncertainty
is governed by the variable trigger window duration, which is defined in terms
of a fixed number of events rather than an interval of time.

The "burst_ra" and "burst_dec" are the RA,Dec coordinates of the burst (or transient)
as determined by the LAT flight software (J2000 epoch).  These intrinsically
floating point quantities have been multiplied by 10,000 and then integerized
to make an integer quantity with units of 0.0001-degrees.

The "burst_inten" is the subset of the total number of events in the standard 4 energy bands
that were used in the location calculation.

The "inten_4" contains 4 unsigned_char counts of the number of events
in the standard 4 energy bands during the 'int_time' interval.
The least significant byte (lsb) contains the number of events in the 0-100 MeV band,
the next lsb contains number of events in the 100MeV-1GeV band,
3rd has 1-10GeV, and the most significant byte has the number of events in the greater_than_10GeV band.

The "burst_error" is the radius of the circle that will contain on average
TBD% of the bursts.  The units are 0.0001-degrees (ie the fl.pt. error was multiplied
by 10000 and then integerized).

The "phi" and "theta" is the location of the source in LAT instrument coordinates.
Theta is the angle off the bore sight.  //And phi is the azimuthal angle
//with 0 starting at..... measured CW(tbd)....

The "int_time" is the time interval between the first and last photons
which went into the successful trigger criteria.  The units are milli-seconds.

The "trig_index" is the index value of the trigger criterion that provided
the successful trigger (precise definition is TBD).

The "trig_id" field contains (a) flag bits on the type of solution
that was found on-board and (b) flag bits that are assigned in the GCN ground-processing.
The bit packing of the "trig_id" entry in the packet is:
BitNum  F-v-G  Item_name      Description
------  ----   ---------      -----------
0        F     G-vs-L         The starting/seed location came from 0=GBM, 1=from_LAT
1        F     gam_used       0=all gammas used, or 1=only gammas above a cut used in loc method
2-4      -     spare          spare for future use
5        G     def_not_grb    0=No or 1=Yes, it is definitely not a GRB (bad label; its really a Retraction)
6-27     -     spare          spare for future use
28       G     spatial_coinc  0=No or 1=Yes, there was a spatial coincidence with another event
29       G     temporal_coinc 0=No or 1=Yes, there was a temporal coincidence with another event
30       G     test_submit    0=No or 1=Yes, this is a test submission (internal use only)
31       -     spare          spare for future use
The columns are:  "BitNum" is the 2^N location within the 4-byte word; "F-v-G" is Flight-assigned vs
Ground-assigned;  "Item_name" is the short-description name of the quantity;  and
"Description" is the longer description of the quantity.

The "misc" field contains miscellaneous flag bits.
2^0     A 1 means a Repoint_Request was made to the s/c. [Not defined for INI]
2^1-10  spare.
2^11    When this bit is a 1, it indicates that 1 (or more) of the RA,Dec values
        were out of valid range (eg RA was 361 deg).
2^12    spare.
2^13    When this bit is a 1, it indicates that this position is near (less then 0.3deg)  a bright star (mag less than 6.5).
2^14    When this bit is a 1, it indicates that the distance between the position and the center of the galaxy
        is less than the sum of the position error radius and the galaxy radius AND that the position error radius
        is smaller than the galaxy radius.  This position is inside the (circularized) NGC galaxy (or nearly in).
2^15    When this bit is a 1, it indicates that the distance between the position and the center of the galaxy
        is less than the sum of the position error radius and the galaxy radius AND that the galaxy radius
        is smaller than the position error radius.  This galaxy is inside the position error circle (or nearly in).
2^16-31 spare.

The "rec_seq_num" is the Record Sequence Number.  It is a packet serial number
spanning all the messages that come from the s/c during a given trigger.
It will start with a value of 1 for the FERMI_GBM_ALERT Notice and increment during 
the multiple GBM_POS Notices, etc.

The "temp_stat" is the measure of the Temporal Test Statistic.
The value is 4*(-log10(temporal_probability)).  The bigger the number,
the more likely the trigger is a real GRB.  The Fermi-LAT team
has determined that sum of the Temporal and Image Stats >120 is a good working threshold.

The "image_stat" is the measure of the Image Test Statistic.
The value is 4*(-log10(spatial_probability)).  The bigger the number,
the more likely the trigger is a real GRB.  The Fermi-LAT team
has determined that sum of the Temporal and Image Stats >120 is a good working threshold.

The "1st_ph_tjd" is the Truncated Julian Day of the first photon used
in the location calculation.

The "1st_ph_sod" is the UT seconds-of-day (SOD) of the first photon used
in the location calculation.  The units are in centi-seconds
(ie the fl.pt. seconds were multiplied by 100 and then integerized).

The "last_ph_tjd" is the Truncated Julian Day of the last photon used
in the location calculation.

The "last_ph_sod" is the UT seconds-of-day (SOD) of the last photon used
in the location calculation.  The units are in centi-seconds
(ie the fl.pt. seconds were multiplied by 100 and then integerized).

The "burst_id" is an index value specifying the source_type of the triggered event.

The "loc_qual" is the quality of the location determination.  It is a probability
that runs from 0.0 to 1.0 (0.0001 quantization), with a value of -1
if undefined. The bigger the number, the more likely that is a real event
(and converely, a small number indicates random photons in the LAT).

There are a total of 17(tbr) "spare" items in the packet (usually filled
with zeros).  They are located at items 12-13, 15, 20 and 22-37.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.




TYPE=121 PACKET CONTENTS:

The FERMI_LAT_GRB_POS_UPD packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=121)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       trig_num     integer         Trigger number
long         5       trig_tjd     [days]          Truncated Julian Day
long         6       trig_sod     [centi-sec]     (int)(sssss.sss *100)
long         7       burst_ra     [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         8       burst_dec    [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         9       burst_inten  [counts]        Num events used in location calc, 0 to inf
long         10      inten_4      4_bytes         Evt_cnts in 4 energy bands
long         11      burst_error  [0.0001-deg]    (int)(0.0 to 180.0 *10000)
long         12      phi          [centi-deg]     (int)(0.0 to 359.9999 *100)
long         13      theta        [centi-deg]     (int)(0.0 to +100.0 *100)
long         14      int_time     [msec]          Integration time
long         15-16   spare[2]     integer         8 bytes for the future
long         17      trig_index   integer         Trigger index
long         18      Trig_ID      bits            LAT burst candidate, 1=yes,0=no
long         19      misc         bits            Misc stuff packed in here
long         20      rec_seq_num  integer         SerNum of the messages (1-~40)
long         21-24   spare[4]     integer         16 bytes for the future
long         25      temp_stat    integer         (int)(4*(-log10(probability)))
long         26      image_stat   integer         (int)(4*(-log10(probability)))
long         27-30   spare[4]     integer         16 bytes for the future
long         31      1st_ph_tjd   [days]          First photon used in loc calc
long         32      1st_ph_sod   [centi-sec]     (int)(sssss.sss *100)
long         33      last_ph_tjd  [days]          last photon used in loc calc
long         34      last_ph_sod  [centi-sec]     (int)(sssss.sss *100)
long         35-36   spare[2]     integer         8 bytes for the future
long         37      misc         bits            Misc stuff packed in here
long         38      loc_qual     [0.0001]        Location quality (0.0 - 1.0)
long         39      pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 120, 121, 122, and 123;
that tells which type of packet this is and therefore what the contents are
(the contents change between the various types).

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start at 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the trig_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "trig_num" is a uniquely identifying number for each trigger.
It is the number of seconds since 01 Jan 2001.

The "trig_tjd" is the Truncated Julian Day of the FERMI-LAT trigger,
eg. TJD=14466 is 01 Jan 2008.

The "trig_sod" is the UT seconds-of-day (SOD) of the FERMI-LAT trigger.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized). The trigger time can be off by TBD msec.  This uncertainty
is governed by the variable trigger window duration, which is defined in terms
of a fixed number of events rather than an interval of time.

The "burst_ra" and "burst_dec" are the RA,Dec coordinates of the burst (or transient)
as determined by the LAT flight software (J2000 epoch).  These intrinsically
floating point quantities have been multiplied by 10,000 and then integerized
to make an integer quantity with units of 0.0001-degrees.

The "burst_inten" is the subset of the total number of events in the standard 4 energy bands
that were used in the location calculation.

The "inten_4" contains 4 unsigned_char counts of the number of events
in the standard 4 energy bands during the 'int_time' interval.
The least significant byte (lsb) contains the number of events in the 0-100 MeV band,
the next lsb contains number of events in the 100MeV-1GeV band,
3rd has 1-10GeV, and the most significant byte has the number of events in the greater_than_10GeV band.

The "burst_error" is the radius of the circle that will contain on average
TBD% of the bursts.  The units are 0.0001-degrees (ie the fl.pt. error was multiplied
by 10000 and then integerized).

The "phi" and "theta" is the location of the source in LAT instrument coordinates.
Theta is the angle off the bore sight.  //And phi is the azimuthal angle
//with 0 starting at..... measured CW(tbd)....

The "int_time" is the time interval between the first and last photons
which went into the successful trigger criteria.  The units are milli-seconds.

The "trig_index" is the index value of the trigger criterion that provided
the successful trigger (precise definition is TBD).

The "trig_id" field contains (a) flag bits on the type of solution
that was found on-board and (b) flag bits that are assigned in the GCN ground-processing.
The bit packing of the "trig_id" entry in the packet is:
BitNum  F-v-G  Item_name      Description
------  ----   ---------      -----------
0        F     G-vs-L         The starting/seed location came from 0=GBM, 1=from_LAT
1        F     gam_used       0=all gammas used, or 1=only gammas above a cut used in loc method
2-4      -     spare          spare for future use
5        G     def_not_grb    0=No or 1=Yes, it is definitely not a GRB (bad label; its really a Retraction)
6-27     -     spare          spare for future use
28       G     spatial_coinc  0=No or 1=Yes, there was a spatial coincidence with another event
29       G     temporal_coinc 0=No or 1=Yes, there was a temporal coincidence with another event
30       G     test_submit    0=No or 1=Yes, this is a test submission (internal use only)
31       -     spare          spare for future use
The columns are:  "BitNum" is the 2^N location within the 4-byte word; "F-v-G" is Flight-assigned vs
Ground-assigned;  "Item_name" is the short-description name of the quantity;  and
"Description" is the longer description of the quantity.

The "misc" field contains miscellaneous flag bits.
2^0     When this bit is a 1, then a Repoint_Request was made to the s/c.
2^1-10  spare.
2^11    When this bit is a 1, it indicates that 1 (or more) of the RA,Dec values
        were out of valid range (eg RA was 361 deg).
2^12    spare.
2^13    When this bit is a 1, it indicates that this position is near (less than 0.3deg)  a bright star (mag less than 6.5).
2^14    When this bit is a 1, it indicates that the distance between the position and the center of the galaxy
        is less than the sum of the position error radius and the galaxy radius AND that the position error radius
        is smaller than the galaxy radius.  This position is inside the (circularized) NGC galaxy (or nearly in).
2^15    When this bit is a 1, it indicates that the distance between the position and the center of the galaxy
        is less than the sum of the position error radius and the galaxy radius AND that the galaxy radius
        is smaller than the position error radius.  This galaxy is inside the position error circle (or nearly in).
2^16-31 spare.

The "rec_seq_num" is the Record Sequence Number.  It is a packet serial number
spanning all the messages that come from the s/c during a given trigger.
It will start with a value of 1 for the FERMI_GBM_ALERT Notice and increment during 
the multiple GBM_POS Notices, etc.

The "temp_stat" is the measure of the Temporal Test Statistic.
The value is 4*(-log10(temporal_probability)).  The bigger the number,
the more likely the trigger is a real GRB.  The Fermi-LAT team
has determined that sum of the Temporal and Image Stats >120 is a good working threshold.

The "image_stat" is the measure of the Image Test Statistic.
The value is 4*(-log10(spatial_probability)).  The bigger the number,
the more likely the trigger is a real GRB.  The Fermi-LAT team
has determined that sum of the Temporal and Image Stats >120 is a good working threshold.

The "1st_ph_tjd" is the Truncated Julian Day of the first photon used
in the location calculation.

The "1st_ph_sod" is the UT seconds-of-day (SOD) of the first photon used
in the location calculation.  The units are in centi-seconds
(ie the fl.pt. seconds were multiplied by 100 and then integerized).

The "last_ph_tjd" is the Truncated Julian Day of the last photon used
in the location calculation.

The "last_ph_sod" is the UT seconds-of-day (SOD) of the last photon used
in the location calculation.  The units are in centi-seconds
(ie the fl.pt. seconds were multiplied by 100 and then integerized).

The "burst_id" is an index value specifying the source_type of the triggered event.

The "loc_qual" is the quality of the location determination.  It is a probability
that runs from 0.0 to 1.0 (0.0001 quantization), with a value of -1
if undefined. The bigger the number, the more likely that is a real event
(and converely, a small number indicates random photons in the LAT).

There are a total of 17(tbr) "spare" items in the packet (usually filled
with zeros).  They are located at items 12-13, 15, 20 and 22-37.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.



TYPE=122 PACKET CONTENTS:

The FERMI_LAT_GRB_POS_DIAG packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=122)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       trig_num     integer         Trigger number
long         5       trig_tjd     [days]          Truncated Julian Day
long         6       trig_sod     [centi-sec]     (int)(sssss.sss *100)
long         7       burst_ra     [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         8       burst_dec    [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         9       burst_inten  [counts]        Num events used in location calc, 0 to inf
long         10      inten_4      4_bytes         Evt_cnts in 4 energy bands
long         11      burst_error  [0.0001-deg]    (int)(0.0 to 180.0 *10000)
long         12      phi          [centi-deg]     (int)(0.0 to 359.9999 *100)
long         13      theta        [centi-deg]     (int)(0.0 to +100.0 *100)
long         14      int_time     [msec]          Integration time
long         15-16   spare[2]     integer         8 bytes for the future
long         17      trig_index   integer         Trigger index
long         18      Trig_ID      bits            LAT burst candidate, 1=yes,0=no
long         19      misc         bits            Misc stuff packed in here
long         20      rec_seq_num  integer         SerNum of the messages (1-~40)
long         21-24   spare[4]     integer         16 bytes for the future
long         25      temp_stat    integer         (int)(4*(-log10(probability)))
long         26      image_stat   integer         (int)(4*(-log10(probability)))
long         27-30   spare[4]     integer         16 bytes for the future
long         31      1st_ph_tjd   [days]          First photon used in loc calc
long         32      1st_ph_sod   [centi-sec]     (int)(sssss.sss *100)
long         33      last_ph_tjd  [days]          last photon used in loc calc
long         34      last_ph_sod  [centi-sec]     (int)(sssss.sss *100)
long         35-36   spare[2]     integer         8 bytes for the future
long         37      misc         bits            Misc stuff packed in here
long         38      spare        integer         4 bytes for the future
long         39      pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 120, 121, 122, and 123;
that tells which type of packet this is and therefore what the contents are
(the contents change between the various types).

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start at 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the trig_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "trig_num" is a uniquely identifying number for each trigger.
It is the number of seconds since 01 Jan 2001.

The "trig_tjd" is the Truncated Julian Day of the FERMI-LAT trigger,
eg. TJD=14466 is 01 Jan 2008.

The "trig_sod" is the UT seconds-of-day (SOD) of the FERMI-LAT trigger.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized). The trigger time can be off by TBD msec.  This uncertainty
is governed by the variable trigger window duration, which is defined in terms
of a fixed number of events rather than an interval of time.

The "burst_ra" and "burst_dec" are the RA,Dec coordinates of the burst (or transient)
as determined by the LAT flight software (J2000 epoch).  These intrinsically
floating point quantities have been multiplied by 10,000 and then integerized
to make an integer quantity with units of 0.0001-degrees.

The "burst_inten" is the subset of the total number of events in the standard 4 energy bands
that were used in the location calculation.

The "inten_4" contains 4 unsigned_char counts of the number of events
in the standard 4 energy bands during the 'int_time' interval.
The least significant byte (lsb) contains the number of events in the 0-100 MeV band,
the next lsb contains number of events in the 100MeV-1GeV band,
3rd has 1-10GeV, and the most significant byte has the number of events in the greater_than_10GeV band.

The "burst_error" is the radius of the circle that will contain on average
TBD% of the bursts.  The units are 0.0001-degrees (ie the fl.pt. error was multiplied
by 10000 and then integerized).

The "phi" and "theta" is the location of the source in LAT instrument coordinates.
Theta is the angle off the bore sight.  //And phi is the azimuthal angle
//with 0 starting at..... measured CW(tbd)....

The "int_time" is the time interval between the first and last photons
which went into the successful trigger criteria.  The units are milli-seconds.

The "trig_index" is the index value of the trigger criterion that provided
the successful trigger (precise definition is TBD).

The "trig_id" field contains (a) flag bits on the type of solution
that was found on-board and (b) flag bits that are assigned in the GCN ground-processing.
The bit packing of the "trig_id" entry in the packet is:
BitNum  F-v-G  Item_name      Description
------  ----   ---------      -----------
0        F     G-vs-L         The starting/seed location came from 0=GBM, 1=from_LAT
1        F     gam_used       0=all gammas used, or 1=only gammas above a cut used in loc method
2-4      -     spare          spare for future use
5        G     def_not_grb    0=No or 1=Yes, it is definitely not a GRB (bad label; its really a Retraction)
6-27     -     spare          spare for future use
28       G     spatial_coinc  0=No or 1=Yes, there was a spatial coincidence with another event
29       G     temporal_coinc 0=No or 1=Yes, there was a temporal coincidence with another event
30       G     test_submit    0=No or 1=Yes, this is a test submission (internal use only)
31       -     spare          spare for future use
The columns are:  "BitNum" is the 2^N location within the 4-byte word; "F-v-G" is Flight-assigned vs
Ground-assigned;  "Item_name" is the short-description name of the quantity;  and
"Description" is the longer description of the quantity.

The "misc" field contains miscellaneous flag bits.
2^0     When this bit is a 1, then a Repoint_Request was made to the s/c.
2^1-10  spare.
2^11    When this bit is a 1, it indicates that 1 (or more) of the RA,Dec values
        were out of valid range (eg RA was 361 deg).
2^12    spare.
2^13    When this bit is a 1, it indicates that this position is near (less than 0.3deg)  a bright star (mag less than 6.5).
2^14    When this bit is a 1, it indicates that the distance between the position and the center of the galaxy
        is less than the sum of the position error radius and the galaxy radius AND that the position error radius
        is smaller than the galaxy radius.  This position is inside the (circularized) NGC galaxy (or nearly in).
2^15    When this bit is a 1, it indicates that the distance between the position and the center of the galaxy
        is less than the sum of the position error radius and the galaxy radius AND that the galaxy radius
        is smaller than the position error radius.  This galaxy is inside the position error circle (or nearly in).
2^16-31  spare.

The "rec_seq_num" is the Record Sequence Number.  It is a packet serial number
spanning all the messages that come from the s/c during a given trigger.
It will start with a value of 1 for the FERMI_GBM_ALERT Notice and increment during 
the multiple GBM_POS Notices, etc.

The "temp_stat" is the measure of the Temporal Test Statistic.
The value is 4*(-log10(temporal_probability)).  The bigger the number,
the more likely the trigger is a real GRB.  The Fermi-LAT team
has determined that sum of the Temporal and Image Stats >120 is a good working threshold.

The "image_stat" is the measure of the Image Test Statistic.
The value is 4*(-log10(spatial_probability)).  The bigger the number,
the more likely the trigger is a real GRB.  The Fermi-LAT team
has determined that sum of the Temporal and Image Stats >120 is a good working threshold.

The "1st_ph_tjd" is the Truncated Julian Day of the first photon used
in the location calculation.

The "1st_ph_sod" is the UT seconds-of-day (SOD) of the first photon used
in the location calculation.  The units are in centi-seconds
(ie the fl.pt. seconds were multiplied by 100 and then integerized).

The "last_ph_tjd" is the Truncated Julian Day of the last photon used
in the location calculation.

The "last_ph_sod" is the UT seconds-of-day (SOD) of the last photon used
in the location calculation.  The units are in centi-seconds
(ie the fl.pt. seconds were multiplied by 100 and then integerized).

The "burst_id" is an index value specifying the source_type of the triggered event.

There are a total of 17(tbr) "spare" items in the packet (usually filled
with zeros).  They are located at items 12-13, 15, 20 and 22-38.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.



TYPE=123 PACKET CONTENTS:

The FERMI_LAT_TRANS packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=123)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       ref_num      integer         Reference number
long         5       trig_tjd     [days]          Truncated Julian Day
long         6       trig_sod     [centi-sec]     (int)(sssss.sss *100)
long         7       trans_ra     [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         8       trans_dec    [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         9       trans_flux   [ph/cm*2]       (int)(flux*1e9 ph/cm2/sec)
long         10      base_flux    [ph/cm*2]       (int)(flux*1e9 ph/cm2/sec)
long         11      trans_error  [0.0001-deg]    (int)(0.0 to 180.0 *10000)
long         12      cur_flx_err  [ph/cm*2]       (int)(flux*1e9 ph/cm2/sec)
long         13      bas_flx_err  [ph/cm*2]       (int)(flux*1e9 ph/cm2/sec)
long         14      time_scale   [integer]       Index: 1=1day, 2=1week
long         15      e_band       [integer]       e1-e2 [GeV] energy_band
long         16-17   spare[2]     integer         8 bytes for the future
long         18      trig_id      bits            ????????
long         19      misc         bits            Misc stuff packed in here
long         20-24   spare[5]     integer         20 bytes for the future
long         25      signif       [centi-sigma]   (int)(100*sigma_signif)
long         26-31   spare[6]     integer         24 bytes for the future
quad_char    32-38   srcname[17]  chars           The SourceName
long         39      pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 120, 121, 122, and 123;
that tells which type of packet this is and therefore what the contents are
(the contents change between the various types). 

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start at 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "ref_num" is a uniquely identifying reference number for each trigger.
It is the number of seconds since 01 Jan 2001 (when the notice was created).

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the trig_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "trig_tjd" is the Truncated Julian Day of the transient trigger,
eg. TJD=14466 is 01 Jan 2008.

The "trig_sod" is the UT seconds-of-day (SOD) of the transient trigger.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized). 

The "trans_ra" and "trans_dec" are the RA,Dec coordinates of the transient
as determined by the ground processing software (J2000 epoch).  These intrinsically
floating point quantities have been multiplied by 10,000 and then integerized
to make an integer quantity with units of 0.0001-degrees.

The "trans_current_flux" is the flux of the transient in units of photon/cm2/sec
mulitplied by 1.0e9, and then integerized.  It is the current flux (ie at the
time of the notice).

//The "trans_base_flux" is the flux of the transient in units of photon/cm2/sec
//mulitplied by 1.0e9, and then integerized.  This is the baseline flux prior
//to the flare that triggered the ground-processing analysis that resulted
//in producing this notice.

The "trans_error" is the radius of the circle that will contain on average
TBD% of the bursts.  The units are 0.0001-degrees (ie the fl.pt. error was multiplied
by 10000 and then integerized).

The "current_flux_error" is the uncertainty on the "trans_current_flux" value.
Also in photons/cm2/sec multiplied by 1.0e9, and then integerized.

//The "base_flux_error" is the uncertainty on the "trans_current_flux" value.
//Also in photons/cm2/sec multiplied by 1.0e9, and then integerized.

The "time_scale" is the time interval of the trigger criteria that found the new source.
A value of 1 for 1-day timescale, and a value of 2 for a 1-week scale.

The "e_band" contains the E_lo - E_hi energy band used for this detection,
where E_lo is in the bottem 16 bits and E_hi is in the upper 16 bits.
Both have been multiplied by 10 and integerized.  The units are deci-GeV.

The "trig_id" field contains (a) flag bits on the type of solution
that was found on-board and (b) flag bits that are assigned in the GCN ground-processing.
The bit packing of the "trig_id" entry in the packet is:
BitNum  F-v-G  Item_name      Description
------  ----   ---------      -----------
0-4      -     spare          spare for future use
5        G     retract        This is a retraction (=1)
6-27     -     spare          spare for future use
28       G     spatial_coinc  0=No or 1=Yes, there was a spatial coincidence with another event
29       G     temporal_coinc 0=No or 1=Yes, there was a temporal coincidence with another event
30       G     test_submit    0=No or 1=Yes, this is a test submission (internal use only)
31       -     spare          spare for future use
The columns are:  "BitNum" is the 2^N location within the 4-byte word; "F-v-G" is Flight-assigned vs
Ground-assigned;  "Item_name" is the short-description name of the quantity;  and
"Description" is the longer description of the quantity.

The "misc" field contains miscellaneous flag bits.
2^0-10  spare.
2^11    When this bit is a 1, it indicates that 1 (or more) of the RA,Dec values
        were out of valid range (eg RA was 361 deg).
2^12    spare.
2^13    When this bit is a 1, it indicates that this position is near (less than 0.3deg)  a bright star (mag less than 6.5).
2^14    When this bit is a 1, it indicates that the distance between the position and the center of the galaxy
        is less than the sum of the position error radius and the galaxy radius AND that the position error radius
        is smaller than the galaxy radius.  This position is inside the (circularized) NGC galaxy (or nearly in).
2^15    When this bit is a 1, it indicates that the distance between the position and the center of the galaxy
        is less than the sum of the position error radius and the galaxy radius AND that the galaxy radius
        is smaller than the position error radius.  This galaxy is inside the position error circle (or nearly in).
2^1-31  spare.

The "signif" is the detection significance of the new unknown source.
The units are centi-sigma (divide by 100 to get S.SS value in units of sigma).

The "sourcename" is the name of the new source.
This item in the packet spans 7 longwords (ie 28 bytes) and contains 
the ASCII string of the name.  This string can be up to 27 characters long,
and will always be null-terminated.

There are a total of 21(tbr) "spare" items in the packet (usually filled
with zeros).  They are located at items 12-13, 15, 20 and 22-38.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.




TYPE=124 PACKET CONTENTS:

The FERMI_LAT_GRB_POS_TEST packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=124)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       trig_num     integer         Trigger number
long         5       trig_tjd     [days]          Truncated Julian Day
long         6       trig_sod     [centi-sec]     (int)(sssss.sss *100)
long         7       burst_ra     [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         8       burst_dec    [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         9       burst_inten  [counts]        Num events used in location calc, 0 to inf
long         10      inten_4      4_bytes         Evt_cnts in 4 energy bands
long         11      burst_error  [0.0001-deg]    (int)(0.0 to 180.0 *10000)
long         12      phi          [centi-deg]     (int)(0.0 to 359.9999 *100)
long         13      theta        [centi-deg]     (int)(0.0 to +100.0 *100)
long         14      int_time     [msec]          Integration time
long         15-16   spare[2]     integer         8 bytes for the future
//long       16      lon_lat      2_shorts        (int)(Longitude,Lattitude *100)
long         17      trig_index   integer         Trigger index
long         18      Trig_ID      bits            LAT burst candidate, 1=yes,0=no
long         19      misc         bits            Misc stuff packed in here
long         20      rec_seq_num  integer         SerNum of the messages (1-~40)
long         21-24   spare[4]     integer         16 bytes for the future
long         25      temp_signif  integer         (int)(4*(-log10(probability)))
long         26      image_signif integer         (int)(4*(-log10(probability)))
long         27-37   spare[12]    integer         48 bytes for the future
long         38      loc_qual     [0.0001]        Location quality (0.0 - 1.0)
long         39      pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 120, 121, 122, 123, and 124;
that tells which type of packet this is and therefore what the contents are
(the contents change between the various types).

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start at 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the trig_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "trig_num" is a uniquely identifying number for each trigger.
It is the number of seconds since 01 Jan 2001.
But for this test notice type, it is hardwired to be a fixed value of 99999.

The "trig_tjd" is the Truncated Julian Day of the FERMI-LAT trigger,
eg. TJD=14466 is 01 Jan 2008.

The "trig_sod" is the UT seconds-of-day (SOD) of the FERMI-LAT trigger.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized).  The trigger time can be off by TBD msec.  This uncertainty
is governed by the variable trigger window duration, which is defined in terms
of a fixed number of events rather than an interval of time.

The "burst_ra" and "burst_dec" are the RA,Dec coordinates of the burst (or transient)
as determined by the LAT flight software (J2000 epoch).  These intrinsically
floating point quantities have been multiplied by 10,000 and then integerized
to make an integer quantity with units of 0.0001-degrees.

The "burst_inten" is the subset of the total number of events in the standard 4 energy bands
that were used in the location calculation.

The "inten_4" contains 4 unsigned_char counts of the number of events
in the standard 4 energy bands during the 'int_time' interval.
The least significant byte (lsb) contains the number of events in the 0-100 MeV band,
the next lsb contains number of events in the 100MeV-1GeV band,
3rd has 1-10GeV, and the most significant byte has the number of events in the greater_than_10GeV band.

The "burst_error" is the radius of the circle that will contain on average
TBD% of the bursts.  The units are 0.0001-degrees (ie the fl.pt. error was multiplied
by 10000 and then integerized).

The "phi" and "theta" is the location of the source in LAT instrument coordinates.
Theta is the angle off the bore sight.  //And phi is the azimuthal angle
//with 0 starting at..... measured CW(tbd)....

The "int_time" is the time interval between the first and last photons
which went into the successful trigger criteria.  The units are milli-seconds.

The "trig_index" is the index value of the trigger criterion that provided
the successful trigger (precise definition is TBD).

The "trig_id" field contains (a) flag bits on the type of solution
that was found on-board and (b) flag bits that are assigned in the GCN ground-processing.
The bit packing of the "trig_id" entry in the packet is:
BitNum  F-v-G  Item_name      Description
------  ----   ---------      -----------
0        F     G-vs-L         The starting/seed location came from 0=GBM, 1=from_LAT
1        F     gam_used       0=all gammas used, or 1=only gammas above a cut used in loc method
2-4      -     spare          spare for future use
5        G     def_not_grb    0=No or 1=Yes, it is definitely not a GRB (bad label; its really a Retraction)
6-27     -     spare          spare for future use
28       G     spatial_coinc  0=No or 1=Yes, there was a spatial coincidence with another event (NOOP for TEST)
29       G     temporal_coinc 0=No or 1=Yes, there was a temporal coincidence with another event (NOOP for TEST)
30       G     test_submit    0=No or 1=Yes, this is a test submission (internal use only)
31       -     spare          spare for future use
The columns are:  "BitNum" is the 2^N location within the 4-byte word; "F-v-G" is Flight-assigned vs
Ground-assigned;  "Item_name" is the short-description name of the quantity;  and
"Description" is the longer description of the quantity.

The "misc" field contains miscellaneous flag bits.
2^0     When this bit is a 1, then a Repoint_Request was made to the s/c.
2^1-10  spare.
2^11    When this bit is a 1, it indicates that 1 (or more) of the RA,Dec values
        were out of valid range (eg RA was 361 deg).
2^12    spare.
2^13    When this bit is a 1, it indicates that this position is near (less than 0.3deg)  a bright star (mag less than 6.5).
2^14    When this bit is a 1, it indicates that the distance between the position and the center of the galaxy
        is less than the sum of the position error radius and the galaxy radius AND that the position error radius
        is smaller than the galaxy radius.  This position is inside the (circularized) NGC galaxy (or nearly in).
2^15    When this bit is a 1, it indicates that the distance between the position and the center of the galaxy
        is less than the sum of the position error radius and the galaxy radius AND that the galaxy radius
        is smaller than the position error radius.  This galaxy is inside the position error circle (or nearly in).
2^1-31  spare.

The "rec_seq_num" is the Record Sequence Number.  It is a packet serial number
spanning all the messages that come from the s/c during a given trigger.
It will start with a value of 1 for the FERMI_GBM_ALERT Notice and increment during 
the multiple GBM_POS Notices, etc.

The "temp_stat" is the measure of the Temporal Test Statistic.
The value is 4*(-log10(temporal_probability)).  The bigger the number,
the more likely the trigger is a real GRB.  The Fermi-LAT team
has determined that sum of the Temporal and Image Stats >120 is a good working threshold.

The "image_stat" is the measure of the Image Test Statistic.
The value is 4*(-log10(spatial_probability)).  The bigger the number,
the more likely the trigger is a real GRB.  The Fermi-LAT team
has determined that sum of the Temporal and Image Stats >120 is a good working threshold.

The "loc_qual" is the quality of the location determination.  It is a probability
that runs from 0.0 to 1.0 (0.0001 quantization), with a value of -1
if undefined. The bigger the number, the more likely that is a real event
(and converely, a small number indicates random photons in the LAT).

There are a total of 21(tbr) "spare" items in the packet (usually filled
with zeros).  They are located at items 12-13, 15, 20 and 22-37.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.



TYPE=125 PACKET CONTENTS:

The FERMI_LAT_MONITOR packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=125)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       ref_num      integer         Reference number
long         5       trig_tjd     [days]          Truncated Julian Day
long         6       trig_sod     [centi-sec]     (int)(sssss.sss *100)
long         7       trans_ra     [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         8       trans_dec    [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         9       trans_flux   [ph/cm*2]       (int)(flux*1e9 ph/cm2/sec)
long         10      base_flux    [ph/cm*2]       (int)(flux*1e9 ph/cm2/sec)
long         11      trans_error  [0.0001-deg]    (int)(0.0 to 180.0 *10000)
long         12      cur_flx_err  [ph/cm*2]       (int)(flux*1e9 ph/cm2/sec)
long         13      bas_flx_err  [ph/cm*2]       (int)(flux*1e9 ph/cm2/sec)
long         14      time_scale   [integer]       Index: 1=1day, 2=1week
long         15      e_band       [integer]       e1-e2 [GeV] energy_band
long         16-17   spare[2]     integer         8 bytes for the future
long         18      trig_id      bits            Trigger identification flags
long         19      misc         bits            Misc stuff packed in here
long         20-24   spare[5]     integer         20 bytes for the future
long         25      signif       [centi-sigma]   (int)(100*sigma_signif)
long         26-31   spare[6]     integer         24 bytes for the future
quad_char    32-38   src[17]      chars           The Source name
long         39      pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 125,
that tells which type of packet this is and therefore what the contents are
(the contents change between the various types).

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start at 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the point_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "ref_num" is a uniquely identifying reference number for each trigger.
It is the number of seconds since 01 Jan 2001 (when the notice was created).

The "trig_tjd" is the Truncated Julian Day of the FERMI-LAT_MONITOR flare trigger,
eg. TJD=14466 is 01 Jan 2008.

The "trig_sod" is the UT seconds-of-day (SOD) of the FERMI-LAT_MONITOR flare trigger.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized).

The "trans_ra" and "trans_dec" are the RA,Dec coordinates of the flare
as determined by the ground processing software (J2000 epoch).  These intrinsically
floating point quantities have been multiplied by 10,000 and then integerized
to make an integer quantity with units of 0.0001-degrees.

The "trans_current_flux" is the flux of the transient in units of photon/cm2/sec
mulitplied by 1.0e9, and then integerized.  It is the current flux (ie at the
time of the notice).

The "trans_base_flux" is the flux of the transient in units of photon/cm2/sec
mulitplied by 1.0e9, and then integerized.  This is the baseline flux prior
to the flare that triggered the ground-processing analysis that resulted
in producing this notice.

The "trans_error" is the radius of the circle that will contain on average
TBD% of the bursts.  The units are 0.0001-degrees (ie the fl.pt. error was multiplied
by 10000 and then integerized).

The "current_flux_error" is the uncertainty on the "trans_current_flux" value.
Also in photons/cm2/sec multiplied by 1.0e9, and then integerized.

The "base_flux_error" is the uncertainty on the "trans_current_flux" value.
Also in photons/cm2/sec multiplied by 1.0e9, and then integerized.

The "time_scale" is the time interval of the trigger criteria that detected te flare.
A value of 1 for 1-day timescale, and a value of 2 for a 1-week scale.

The "e_band" contains the E_lo - E_hi energy band used for this detection,
where E_lo is in the bottem 16 bits and E_hi is in the upper 16 bits.
Both have been multiplied by 10 and integerized.  The units are deci-GeV.

The "trig_id" field contains (a) flag bits on the type of solution
that was found in the Fermi ground processing and (b) flag bits that are assigned in
the GCN ground-processing.
The bit packing of the "trig_id" entry in the packet is:
BitNum  F-v-G  Item_name      Description
------  ----   ---------      -----------
0        G     not_mon        This is a known source, but is not Monitored (=1)
1-4      -     spare          spare for future use
5        G     retract        This is a retraction (=1)
6-27     -     spare          spare for future use
28       G     spatial_coinc  0=No or 1=Yes, there was a spatial coincidence with another event
29       G     temporal_coinc 0=No or 1=Yes, there was a temporal coincidence with another event
30-31    -     spare          spare for future use
The columns are:  "BitNum" is the 2^N location within the 4-byte word; "F-v-G" is Flight-assigned vs
Ground-assigned;  "Item_name" is the short-description name of the quantity;  and
"Description" is the longer description of the quantity.

The "misc" field contains miscellaneous flag bits.
2^0-10  spare.
2^11    When this bit is a 1, it indicates that 1 (or more) of the RA,Dec values
        were out of valid range (eg RA was 361 deg).
2^12    spare.
2^13    When this bit is a 1, it indicates that this position is near (less than 0.3deg)  a bright star (mag less than 6.5).
2^14    When this bit is a 1, it indicates that the distance between the position and the center of the galaxy
        is less than the sum of the position error radius and the galaxy radius AND that the position error radius
        is smaller than the galaxy radius.  This position is inside the (circularized) NGC galaxy (or nearly in).
2^15    When this bit is a 1, it indicates that the distance between the position and the center of the galaxy
        is less than the sum of the position error radius and the galaxy radius AND that the galaxy radius
        is smaller than the position error radius.  This galaxy is inside the position error circle (or nearly in).
2^1-31  spare.

The "signif" is the rate increase.
Both are in units of centi-sigma (divide by 100 to get S.SS value in units of sigma).

The "src" is just the SourceName of the object.  Most of the time it can be used
as the filename portion of the URL -- the leading path for the full URL is
"http://fermi.gsfc.nasa.gov/FTP/glast/data/lat/catalogs/asp/current/lightcurves/".
This item in the packet spans 7 longwords (ie 28 bytes) and contains 
the ASCII string of the filename portion of the lightcurve URL.  This string
can be up to 27 characters long, and will always be null-terminated.

There are a total of 21(tbr) "spare" items in the packet (usually filled
with zeros).  They are located at items 12-13, 15, 20 and 22-38.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.



TYPE=126 (&144) PACKET CONTENTS:

The FERMI_SC_SLEW (& FERMI_SC_SLEW_INTERNAL) packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=126 or 144)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       trig_num     integer         Trigger number
long         5       point_tjd    [days]          Truncated Julian Day
long         6       point_sod    [centi-sec]     (int)(sssss.sss *100)
long         7       point_ra     [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         8       point_dec    [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         9-13    spare[5]     integer         20 bytes for the future
long         14      dwell_time   [centi-sec]     (int)(sssss.sss *100)
long         15-17   spare[3]     integer         12 bytes for the future
long         18      trig_id      bits            Trigger identification flags
long         19      misc         bits            Misc stuff packed in here
long         20-38   spare[19]    integer         76 bytes for the future
long         39      pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 126
that tells which type of packet this is and therefore what the contents are
(the contents change between the various types).

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start at 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the point_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "trig_num" is a uniquely identifying number for each trigger.
It is the number of seconds since 01 Jan 2001.

The "point_tjd" is the Truncated Julian Day of the FERMI-LAT request to the s/c
to repoint to the newly detected target by the LAT (ie an autonomous report request (ARR),
eg. TJD=14466 is 01 Jan 2008.

The "point_sod" is the UT seconds-of-day (SOD) of the FERMI-LAT request to the s/c
to repoint to the newly detected target by the LAT (ie an autonomous report request (ARR),
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized).

The "point_ra" and "point_dec" are the RA,Dec coordinates of the request
as determined by the LAT flight software (J2000 epoch).  These intrinsically
floating point quantities have been multiplied by 10,000 and then integerized
to make an integer quantity with units of 0.0001-degrees.

The "dwell_time" is the observation duration of the request.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized).

The "trig_id" field contains (a) flag bits on the type of solution
that was found in the Fermi ground processing and (b) flag bits that are assigned in
the GCN ground-processing.  This "trig_id" field is a little out of place
for the SLEW notice, but it is maintained for continuity with the other notice types,
and there are two bitfields that are still appropriate for a slew notice.
The bit packing of the "trig_id" entry in the packet is:
BitNum  F-v-G  Item_name      Description
------  ----   ---------      -----------
0-4      -     spare          spare for future use
5        G     retract        This is a retraction (=1)
6-29     -     spare          spare for future use
30       G     test_submit    0=No or 1=Yes, this is a test submission (internal use only)
31       -     spare          spare for future use
The columns are:  "BitNum" is the 2^N location within the 4-byte word; "F-v-G" is Flight-assigned vs
Ground-assigned;  "Item_name" is the short-description name of the quantity;  and
"Description" is the longer description of the quantity.

The "misc" field contains miscellaneous flag bits.
2^0     When this bit is a 1, then the s/c decided to slew to the requested position.
2^1-10  spare.
2^11    When this bit is a 1, it indicates that 1 (or more) of the RA,Dec values
        were out of valid range (eg RA was 361 deg).
2^12-29 spare.
2^30    When this bit is a 1, then this Notice was ground-generated; else flight-generated.
        A Gnd-generated Notice is issued anytime humans during ground operations
        have determined if this Notice should be issued.
2^31    spare.

There are a total of 19(tbr) "spare" items in the packet (usually filled
with zeros).  They are located at items (tbr)10-12, 15-16, 20, and 25-37.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.



TYPE=127 PACKET CONTENTS:

The FERMI_LAT_GND packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=127)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       trig_num     integer         Trigger number
long         5       trig_tjd     [days]          Truncated Julian Day
long         6       trig_sod     [centi-sec]     (int)(sssss.sss *100)
long         7       burst_ra     [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         8       burst_dec    [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         9       tot_inten    integer         Total Intensity
long         10      spare        integer         4 bytes for the future
long         11      burst_error  [0.0001-deg]    (int)(0.0 to 180.0 *10000)
long         12      phi          [centi-deg]     (int)(0.0 to 359.9999 *100)
long         13      theta        [centi-deg]     (int)(0.0 to +100.0 *100)
long         //14    int_time     [msec]          Integration time
long         15-17   spare[3]     integer         12 bytes for the future
long         18      trig_id      bits            Trigger identification flags
long         19      misc         bits            Misc stuff packed in here
long         20-25   spare[6]     integer         24 bytes for the future
long         26      signif       integer         (int)(sqrt(TS)*100)
long         27      inten2       centi-cnts      Evt_cnts in the 0.1-1.0 GeV band
long         28      inten3       centi-cnts      Evt_cnts in the 1.0-10 GeV band
long         29      inten4       centi-cnts      Evt_cnts in the 10-inf GeV band
long         30-38   spare[9]     integer         36 bytes for the future
long         39      pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 127
that tells which type of packet this is and therefore what the contents are
(the contents change between the various types).

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start at 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the trig_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "trig_num" is a uniquely identifying number for each trigger.
It is the number of seconds since 01 Jan 2001.

The "trig_tjd" is the Truncated Julian Day of the FERMI-LAT trigger,
eg. TJD=14466 is 01 Jan 2008.

The "trig_sod" is the UT seconds-of-day (SOD) of the FERMI-LAT trigger.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized). The trigger time can be off by TBD msec.  This uncertainty
is governed by the variable trigger window duration, which is defined in terms
of a fixed number of events rather than an interval of time.

The "burst_ra" and "burst_dec" are the RA,Dec coordinates of the burst (or transient)
as determined by the LAT flight software (J2000 epoch).  These intrinsically
floating point quantities have been multiplied by 10,000 and then integerized
to make an integer quantity with units of 0.0001-degrees.

The "tot_inten" is the total number of events in the lowest 3 energy bands
during the 'int_time' time interval (ie the sum of the values in slots 27-29).

The "burst_error" is the radius of the circle that will contain on average
TBD% of the bursts.  The units are 0.0001-degrees (ie the fl.pt. error was multiplied
by 10000 and then integerized).

The "phi" and "theta" is the location of the source in LAT instrument coordinates.
Theta is the angle off the bore sight.  And phi is the azimuthal angle
with 0 deg starting at..... measured CW(tbd)....

//The "int_time" is the time interval between the first and last photons
//which went into the successful trigger criteria.  The units are milli-seconds.

The "trig_id" field contains (a) flag bits on the type of solution
that was found on-board and (b) flag bits that are assigned in the GCN ground-processing.
The bit packing of the "trig_id" entry in the packet is:
BitNum  F-v-G  Item_name      Description
------  ----   ---------      -----------
0-2      G     Seed_index     3 bits giving the index value of the starting/seed location:
                              1=GBM,2=LAT_blind,3=LAT_onboard,4=Swift,5=INTEGRAL,6=AGILE.
3-4      -     spare          spare for future use
5        G     def_not_grb    0=No or 1=Yes, it is definitely not a GRB (bad label; its really a Retraction)
6-27     -     spare          spare for future use
28       G     spatial_coinc  0=No or 1=Yes, there was a spatial coincidence with another event
29       G     temporal_coinc 0=No or 1=Yes, there was a temporal coincidence with another event
30-31    -     spare          spare for future use
The columns are:  "BitNum" is the 2^N location within the 4-byte word; "F-v-G" is Flight-assigned vs Ground-assigned
(they will all be ground for this notice type);
"Item_name" is the short-description name of the quantity;  and
"Description" is the longer description of the quantity.

The "misc" field contains miscellaneous flag bits.
2^0-12  spare.
2^13    When this bit is a 1, it indicates that this position is near (less than 0.3deg)  a bright star (mag less than 6.5).
2^14    When this bit is a 1, it indicates that the distance between the position and the center of the galaxy
        is less than the sum of the position error radius and the galaxy radius AND that the position error radius
        is smaller than the galaxy radius.  This position is inside the (circularized) NGC galaxy (or nearly in).
2^15    When this bit is a 1, it indicates that the distance between the position and the center of the galaxy
        is less than the sum of the position error radius and the galaxy radius AND that the galaxy radius
        is smaller than the position error radius.  This galaxy is inside the position error circle (or nearly in).
2^16-31 spare.


The "trig_signif" is the sqrt(TS) (with 2 dof) of the detection.
It has been multiplied by 100 and integerized.

The "inten2", "inten3", and "inten4" contain the number of counts
in the three specified energy bands.  The units are centi-counts,
ie the counts have been multiplied by 100 and interegerized.
(I am not sure how fractional counts come to be in the Fermi analysis.)
The analysis used for these ground-generated LAT notices does not allow
for the "inten1" to be calculated.

The "spare" items are usually filled with zeros.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.




TYPE=128 PACKET CONTENTS:

The FERMI_LAT_OFFLINE packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=128)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       trig_num     integer         Trigger number
long         5       trig_tjd     [days]          Truncated Julian Day
long         6       trig_sod     [centi-sec]     (int)(sssss.sss *100)
long         7       burst_ra     [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         8       burst_dec    [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         9-10    spare[2]     integer         8 bytes for the future
long         11      burst_error  [0.0001-deg]    (int)(0.0 to 180.0 *10000)
long         12-17   spare[6]     integer         24 bytes for the future
long         18      Trig_ID      bits            LAT burst candidate, 1=yes,0=no
long         19      misc         bits            Misc stuff packed in here
long         20-38   spare[19]    integer         76 bytes for the future
long         39      pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 120, 121, 122, and 123;
that tells which type of packet this is and therefore what the contents are
(the contents change between the various types).  The types are:
120 = FERMI_LAT_GRB_POS_INI, 121 = FERMI_LAT_POS_UPD. 
122 = FERMI_LAT_GRB_POS_DIAG, and 123 = FERMI_LAT_TRANS.

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start at 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the trig_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "trig_num" is a uniquely identifying number for each trigger.
It is the number of seconds since 01 Jan 2001.

The "trig_tjd" is the Truncated Julian Day of the FERMI-LAT trigger,
eg. TJD=14466 is 01 Jan 2008.

The "trig_sod" is the UT seconds-of-day (SOD) of the FERMI-LAT trigger.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized). The trigger time can be off by TBD msec.  This uncertainty
is governed by the variable trigger window duration, which is defined in terms
of a fixed number of events rather than an interval of time.

The "burst_ra" and "burst_dec" are the RA,Dec coordinates of the burst (or transient)
as determined by the LAT flight software (J2000 epoch).  These intrinsically
floating point quantities have been multiplied by 10,000 and then integerized
to make an integer quantity with units of 0.0001-degrees.

The "burst_error" is the radius of the circle that will contain on average
TBD% of the bursts.  The units are 0.0001-degrees (ie the fl.pt. error was multiplied
by 10000 and then integerized).

The "trig_id" field contains (a) flag bits on the type of solution
that was found on-board and (b) flag bits that are assigned in the GCN ground-processing.
The bit packing of the "trig_id" entry in the packet is:
BitNum  F-v-G  Item_name      Description
------  ----   ---------      -----------
0-2      G     Seed_index     3 bits giving the index value of the starting/seed location:
                              1=GBM,2=LAT_blind,3=LAT_onboard,4=Swift,5=INTEGRAL,6=AGILE.
3-4      -     spare          spare for future use
5        G     def_not_grb    0=No or 1=Yes, it is definitely not a GRB (bad label; its really a Retraction)
6-29     -     spare          spare for future use
30       G     test_submit    0=No or 1=Yes, this is a test submission (internal use only)
31       -     spare          spare for future use
The columns are:  "BitNum" is the 2^N location within the 4-byte word; "F-v-G" is Flight-assigned vs
Ground-assigned;  "Item_name" is the short-description name of the quantity;  and
"Description" is the longer description of the quantity.

The "misc" field contains miscellaneous flag bits.
2^0-12  spare.
2^13    When this bit is a 1, it indicates that this position is near (less than 0.3deg)  a bright star (mag less than 6.5).
2^14    When this bit is a 1, it indicates that the distance between the position and the center of the galaxy
        is less than the sum of the position error radius and the galaxy radius AND that the position error radius
        is smaller than the galaxy radius.  This position is inside the (circularized) NGC galaxy (or nearly in).
2^15    When this bit is a 1, it indicates that the distance between the position and the center of the galaxy
        is less than the sum of the position error radius and the galaxy radius AND that the galaxy radius
        is smaller than the position error radius.  This galaxy is inside the position error circle (or nearly in).
2^16-29 spare.
2^30    When this bit is a 1, then this Notice was ground-generated; else flight-generated.
        A Gnd-generated Notice is issued anytime humans during ground operations
        have determined if this Notice should be issued.
2^31    spare.

There are "spare" items in the packet (usually filled with zeros).
They are located at items 9-10, 12-17, and 20-38.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.





TYPE=129 PACKET CONTENTS:

The FERMI_POINTDIR packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=129)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       -
long         5       start_tjd    [days]          Truncated Julian Day
long         6       start_sod    [centi-sec]     (int)(sssss.sss *100)
long         7       start_ra     [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         8       start_dec    [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         9       delta_t      [sec]           Typically 120 sec
long         10      ra_dec_1dt   [two 0.1-deg]   (int)(RA*10) in Hi, (int)(Dec*10) in Lo
long         11      ra_dec_2dt   [two 0.1-deg]   (int)(RA*10) in Hi, (int)(Dec*10) in Lo
long         12      ra_dec_3dt   [two 0.1-deg]   (int)(RA*10) in Hi, (int)(Dec*10) in Lo
long         13      ra_dec_4dt   [two 0.1-deg]   (int)(RA*10) in Hi, (int)(Dec*10) in Lo
long         14      ra_dec_5dt   [two 0.1-deg]   (int)(RA*10) in Hi, (int)(Dec*10) in Lo
long         15      ra_dec_6dt   [two 0.1-deg]   (int)(RA*10) in Hi, (int)(Dec*10) in Lo
long         16      ra_dec_7dt   [two 0.1-deg]   (int)(RA*10) in Hi, (int)(Dec*10) in Lo
long         17      ra_dec_8dt   [two 0.1-deg]   (int)(RA*10) in Hi, (int)(Dec*10) in Lo
long         18      ra_dec_9dt   [two 0.1-deg]   (int)(RA*10) in Hi, (int)(Dec*10) in Lo
long         19      ra_dec_10dt  [two 0.1-deg]   (int)(RA*10) in Hi, (int)(Dec*10) in Lo
long         20      ra_dec_11dt  [two 0.1-deg]   (int)(RA*10) in Hi, (int)(Dec*10) in Lo
long         21      ra_dec_12dt  [two 0.1-deg]   (int)(RA*10) in Hi, (int)(Dec*10) in Lo
long         22      ra_dec_13dt  [two 0.1-deg]   (int)(RA*10) in Hi, (int)(Dec*10) in Lo
long         23      ra_dec_14dt  [two 0.1-deg]   (int)(RA*10) in Hi, (int)(Dec*10) in Lo
long         24      ra_dec_15dt  [two 0.1-deg]   (int)(RA*10) in Hi, (int)(Dec*10) in Lo
long         25      ra_dec_16dt  [two 0.1-deg]   (int)(RA*10) in Hi, (int)(Dec*10) in Lo
long         26      ra_dec_17dt  [two 0.1-deg]   (int)(RA*10) in Hi, (int)(Dec*10) in Lo
long         27      ra_dec_18dt  [two 0.1-deg]   (int)(RA*10) in Hi, (int)(Dec*10) in Lo
long         28      ra_dec_19dt  [two 0.1-deg]   (int)(RA*10) in Hi, (int)(Dec*10) in Lo
long         29      ra_dec_20dt  [two 0.1-deg]   (int)(RA*10) in Hi, (int)(Dec*10) in Lo
long         30      ra_dec_21dt  [two 0.1-deg]   (int)(RA*10) in Hi, (int)(Dec*10) in Lo
long         31      ra_dec_22dt  [two 0.1-deg]   (int)(RA*10) in Hi, (int)(Dec*10) in Lo
long         32      ra_dec_23dt  [two 0.1-deg]   (int)(RA*10) in Hi, (int)(Dec*10) in Lo
long         33      ra_dec_24dt  [two 0.1-deg]   (int)(RA*10) in Hi, (int)(Dec*10) in Lo
long         34      ra_dec_25dt  [two 0.1-deg]   (int)(RA*10) in Hi, (int)(Dec*10) in Lo
long         35      ra_dec_26dt  [two 0.1-deg]   (int)(RA*10) in Hi, (int)(Dec*10) in Lo
long         36      ra_dec_27dt  [two 0.1-deg]   (int)(RA*10) in Hi, (int)(Dec*10) in Lo
long         37      ra_dec_28dt  [two 0.1-deg]   (int)(RA*10) in Hi, (int)(Dec*10) in Lo
long         38      ra_dec_29dt  [two 0.1-deg]   (int)(RA*10) in Hi, (int)(Dec*10) in Lo
long         39      pkt_term     integer         Pkt Termination (always = \n)

PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 129
that tells which type of packet this is and therefore what the contents are
(the contents change between the various types).

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start at 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the "start_sod" (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "start_tjd" is the Truncated Julian Day of the of the first RA,Dec pointing sample
(ie "point_ra" and "point_dec").

The "start_sod" is the UT seconds-of-day (SOD) of the first RA,Dec pointing sample
(ie "point_ra" and "point_dec").  The units are in centi-seconds (ie the fl.pt. seconds
were multiplied by 100 and then integerized).

The "point_ra" and "point_dec" are the RA,Dec pointing coordinates of FERMI
at the time specified by "start_tjd" and "start_sod".  J2000 epoch.
These intrinsically floating point quantities have been multiplied by 10,000
and then integerized to make an integer quantity with units of 0.0001-degrees.

The "delta_time" is the amount of time between the ra,dec pointing samples.
The units are integer seconds.  (It is believed that this value will always
be 120 sec, but do not count on it, ie have your software check the delta_time value
for every packet received.)

The "saa_ra_dec_Ndt" are a series of 29 (N=1 to 29) SAA_flag,RA,Dec values
for the FERMI s/c pointing sampled every "delta_time" sec (SAA_flag indicates
the s/c is inside the SAA region, and the RA,Dec values (J2000) are the pointing
direction of the LAT instrument).  Example:
"saa_ra_dec_1dt" is the SAA_flag,RA,Dec pointing at "start_sod+1*delta_time";
"saa_ra_dec_2dt" would be the pointing direction at "start_sod+2*delta_time"; etc.
The SAA,RA,Dec values are packed into a single 4-byte slot.
The SAA_flag occurpies the 2^31 bit, RA occupies bits 30-16, and the Dec
is in the lower 2 bytes (bits 15-0).
The RA,Dec values are in units of deci-degrees (ie the fl.pt. degress
were multiplied by 10 and then integerized).

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.




================================================================================

TYPE=130 PACKET CONTENTS:

This packet is for GCN programs internal use only.  The packet is NEVER itself
distributed to socket sites.  This packet is used to signal to the GCN programs
to produce the ONLY full-format email Notice (to got to sites requesting that Notice
Type and the full-format email format).  [The variable number of "sources"
makes this notice type incompatible with the socket and voevent formats and
with the email formats other than the full-format.]

The SIMBAD-NED packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=130)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       trig_num     integer         Trigger number
long         5       trig_tjd     [days]          Truncated Julian Day
long         6       trig_sod     [centi-sec]     (int)(sssss.sss *100)
long         7       burst_ra     [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         8       burst_dec    [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         9-10    spare[2]     integer         8 bytes for the future
long         11      burst_error  [0.0001-deg]    (int)(0.0 to 180.0 *10000)
long         12      otype        integer         Original Type number
long         13-21   spare[9]     integer         36 bytes for the future
quad_char    22-38   fname[17]    chars           Filename (up to 68 chars long)
long         39      pkt_term     integer         Pkt Termination (always = \n)


The "pkt_type" is an integer with a value of 130.

The "pkt_sernum", "pkt_hop_cnt", and "pkt_sod" have their usual definitions.

The "burst_ra" and "burst_dec" are the RA,Dec coordinates of the object
that is to be searched in SIMBAD and NED.  These intrinsically
floating point quantities have been multiplied by 10,000 and then integerized
to make an integer quantity with units of 0.0001-degrees.

The "burst_error" is the radius of the circle to search.
The units are 0.0001-degrees (ie the fl.pt. error was multiplied
by 10000 and then integerized).

The "otype" is the "pkt_type" value of the notice that spawned this SIMBAD/NED search.

The "fname" is the full-path filename of the file containing the results
of the SIMBAD/NED search.

The "pkt_term" is the usual definition.



================================================================================

TYPE=131 PACKET CONTENTS:

The FERMI_GBM_SUBTHRESHOLD (there is no "internal" for of this type) packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=131)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       tran_num     integer         Transient number
long         5       event_tjd    [days]          Truncated Julian Day
long         6       event_sod    [centi-sec]     (int)(sssss.sss *100)
long         7       event_ra     [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         8       event_dec    [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         9-10    spare[2]     integer         8 bytes for the future
long         11      event_error  [0.0001-deg]    (int)(0.0 to 180.0 *10000)
long         12      evt_sod_subsec [micro-sec]   Event_SOD subsec portion
long         13      earth_angle  [centi-deg]     (int)(0.0 to +180.0 *100)
long         14      duration     [milli-sec]     Event duration (int)(sss.sss *1000)
long         15-17   spare[3]     integer         12 bytes for the future
long         18      soln_status  bits            Type of source/transient found
long         19      misc         bits            Misc stuff packed in here
long         20-22   spare[3]     integer         12 bytes for the future
long         23      reliability  integer         Mostly class ID & probability
long         24-26   spare[3]     integers        12 bytes for the future
quad_char    27-38   url[12]      chars           The filename portion of the URLs
long         39      pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 131;
that tells which type of packet this is and therefore what the contents are
(the contents change between the various types).

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start at 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the event_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "trans_num" is a uniquely identifying number for each transient.
It is the number of seconds since 01 Jan 2001.

The "event_tjd" is the Truncated Julian Day of the FERMI-GBM transient,
eg. TJD=13370 is 01 Jan 2005.

The "event_sod" is the UT seconds-of-day (SOD) of the FERMI-GBM transient.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized).

The "event_ra" and "event_dec" are the RA,Dec coordinates of the event (or grb, transient, or noise)
as determined by the ground-processing software (J2000 epoch).  These intrinsically
floating point quantities have been multiplied by 10,000 and then integerized
to make an integer quantity with units of 0.0001-degrees (over-precision,
but this is the new standard scaling for all missions/instruments).

The "event_error" is the radius of the circle that will contain on average
68% of the transients.  It includes both the statistical error plus the
systematic error.  The units are 0.0001-degrees (ie the fl.pt. error was multiplied
by 10000 and then integerized).

The "event_sod_subsec" is the UT micro-seconds portion of the GBM transient time (event_sod).
The units are in micro-seconds (ie the fl.pt. subseconds portior were multiplied by 1e6
and then integerized).

The "earth_angle" is the angle between the event direction and the center of the Earth.
The units are in centi-degees (ie the fl.pt. degrees were multiplied by 100
and then integerized).

The "duration" is the length of the event.
The units are in milli-seconds (ie the fl.pt. seconds were multiplied by 1000
and then integerized).

The "soln_status" (aka trans_id) field contains (a) flag bits on the type of solution
that was found in the on-board image and about whether it matches anything in the
on-board source catalog, and (b) flag bits that are assigned in the GCN ground-processing.
These bits will be determined largely (but not completely) by the ClassID in the Most_Likely fields.
The bit packing of the "soln_status" entry in the packet is:
BitNum  F-v-G  Item_name      Description
------  ----   ---------      -----------
0-4      -     spare          spares for future use
5        G     def_not_grb    0=No or 1=Yes, it is definitely not a GRB (bad label; its really a Retraction)
6-11     -     spare          spares for future use
12       G     blk_cat_src    0=No or 1=Yes, it is in the catalog of sources to be blocked (internal use only)
13-27    -     spare          spares for future use
28       G     spatial_coinc  0=No or 1=Yes, there was a spatial coincidence with another event
29       G     temporal_coinc 0=No or 1=Yes, there was a temporal coincidence with another event
30       G     test_submit    0=No or 1=Yes, this is a test submission (internal use only)
31       -     spare          spare for future use
The columns are:  "BitNum" is the 2^N location within the 4-byte word; "F-v-G" is Flight-assigned vs
Ground-assigned;  "Item_name" is the short-description name of the quantity;  and
"Description" is the longer description of the quantity.

The "misc" field contains miscellaneous flag bits.
0       Contains spectral classification: 0=soft,  1=hard
1       Not asigned.
2       Contains type classification: 0=short,  1=long
2^3-10  Not asigned.
2^11    When this bit is a 1, it indicates that 1 (or more) of the RA,Dec,Lon,Lat values
        were out of valid range (eg RA was 361 deg).
2^12-29 Not assigned.
2^30    When this bit is a 1, then this Notice was ground-generated; else flight-generated.
        A Gnd-generated Notice is issued anytime humans during ground operations
        have determined if this Notice should be issued.  This will typically be used
        during the early phases of the mission before the automatic real-time distribution
        capability is enabled.  Once the burst position has been validated by humans,
        a GBM_Pos Notice will be issued (with this flag bit set to 1).
        After the Verification phase, there still may be occaisions when a GBM_POS notice
        is (re)created by human-processing and will be distributed (with this bit set to 1). 
2^31    Not assigned.

The "reliability" is a measure of confidence on the detection.
It is a number from 1 to 10, where 2 is "low", 5 is "medium", and 8 is "high.

The "url" is just the filename portion of the URL, where the leading path portion
for the URL is "https://gcn.gsfc.nasa.gov/notices_gcb_sub/".
So the full URL would look like:
"https://gcn.gsfc.nasa.gov/notices_gcb_sub/gbm_subthresh_NNNNNNNNN.nnnnnn_healpix.fits".
The other two URLs can be made by appropriate modification:
"https://gcn.gsfc.nasa.gov/notices_gcb_sub/gbm_subthresh_NNNNNNNNN.nnnnnn_map.png".
"https://gcn.gsfc.nasa.gov/notices_gcb_sub/gbm_subthresh_NNNNNNNNN.nnnnnn_lc.pdf".
This "url" item in the packet spans 12 longwords (ie 48 bytes) and contains 
the ASCII string of the filename portion of the URL.  This string
can be up to 47 characters long, and will always be null-terminated.

There are a total of 21 "spare" items in the packet (usually filled
with zeros).  They are located at items 9-10, 15-17, 20-22, and 24-26.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.



================================================================================

TYPE=133 PACKET CONTENTS:

The SWIFT_BAT_MONITOR packet contains detections of increases in known sources
periodically monitored by Swift-BAT.

The SWIFT_BAT_MONITOR packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=133)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       ref_num      integer         Some reference number
long         5       tjd          [days]          Truncated Julian Day
long         6       sod          [centi-sec]     (int)(sssss.sss *100)
long         7       ra           [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long         8       dec          [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long         9       crab_flux    [centi-mCrab]   (int)(mm.mm *100)
long        10       curr_flux    [10^-4-cnts/cm2/sec] (int)(c.cccc *10000)
long        11       base_flux    [10^-4-cnts/cm2/sec] (int)(c.cccc *10000)
long        12       signif       [centi-sigma]   (int)(ss.ss *100)
quad_char   13-17    srcname[5]   chars           The name of the source
long        18       flags        bits            Flag bits on source-type
long        19       misc         bits            Misc flag bits and fields
long        20-21    spare[2]     integer         8 bytes for future use
quad_char   22-38    lc_url[17]   chars           The URL for the lightcurve
long        39       pkt_term     integer         Pkt Termination (always = \n)

PACKET ITEM DESCRIPTIONS (type=133):


The "pkt_type" is an integer with a value of 133, which tells that this packet
is of type SWIFT_BAT_MONITOR and therefore indicates the contents and format
for the other 39 fields.

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start from 1 when
the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 1.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the burst_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "ref_num" is a reference number.  This is a sequential number that is assigned;
it is not guarenteed to be unique.

The "tjd" is the Truncated Julian Day of the strt of the outburst,
eg. TJD=12275 is 01 Jan 2002.  Please note that "start" is a somewhat
poorly determined quantity given the sampling of BAT Monitor program (orbital or daily)
and that the algorithm used to detect these outbursts is simplistic and biased
towards getting earlier detection (rather than more accurate but later detections).

The "sod" is the UT seconds-of-day (SOD) of the outburst.  The SOD
is multiplied by 100, and converted to a 32-bit integer for the packet
(i.e the units are centi-sec; 10 msec),  eg. 4689568 = 46895.68 sec-of-day =
13:01:35.68 UT.

The "ra" and "dec" fields specify the RA,Dec location of the source object
undergoing the transient outburst (in J2000 epoch).  These intrinsically floating point
quantities have been multiplied by 10,000 and then integerized to make
an integer quantity with units of 10^-4-degrees.

The "crab_flux" is the current intensity of the source object in mCrab units
(where the cnt/cm2/sec units have been converted to mCrab units with the usual
qualifier that spectral index differences between the source and the Crab make
this measure of intensity a semi-quantitative measure at best).  The units
are mCrab multiplied by 100 and then integerized.

The "curr_flux" is the current intensity of the source.  The units are cnts/cm2/sec
multiplied by 10000 and then integerized.

The "base_flux" is the baseline intensity of the source before the transient outburst occurred.
The units are cnts/cm2/sec multiplied by 10000 and then integerized.

The "signif" is the significance of the increase of the source outburst.  It is the
difference between "curr_flux" and "base_flux" in units of sigma (multiplied by 100
and then integerized, ie centi-sigma).

The "srcname" is the name of the object that is undergoing the transient outburst.
This item in the packet spans 5 longwords (ie 20 bytes) and contains the ASCII string
of the source_name.  This string can be up to 19 characters long, and will always be null-terminated.

The "flags" field contains flag bit bits relating to the source-type identity.
Bit_Number     Item_name      Description
----------     ---------      -----------
0              test_blank_sky This is a test "blank" skypatch "source"
1              unknown_src    This is a new, previously unknown source
2              known_src      This is an outburst/flare from a previously known source
3              prov_src       This is a marginal detection, aka a provisional source
2-31           x_x_x_x        spare

The "misc" field contains various miscelaneous flags and fields.
The bit packing of the "misc" entry in the packet is:
Bit_Number     Item_name      Description
----------     ---------      -----------
0-3            n/a            Sampling index (0=Orbital, 1=Daily)
4-12  spare.
2^13  When this bit is a 1, it indicates that this position is near (less than 0.3deg)  a bright star (mag less than 6.5).
2^14  When this bit is a 1, it indicates that the distance between the position and the center of the galaxy
      is less than the sum of the position error radius and the galaxy radius AND that the position error radius
      is smaller than the galaxy radius.  This position is inside the (circularized) NGC galaxy (or nearly in).
2^15  When this bit is a 1, it indicates that the distance between the position and the center of the galaxy
      is less than the sum of the position error radius and the galaxy radius AND that the galaxy radius
      is smaller than the position error radius.  This galaxy is inside the position error circle (or nearly in).
16-29 spare.
30    When this bit is a 1, then this Notice was ground-generated; else flight-generated.
      For this Notice Type, they are always ground-generated.
31    spare.

The "lc_url" is just the filename portion of the URL -- the leading path
for the full URL is "http://swift.gsfc.nasa.gov/docs/swift/results/transients/".
This item in the packet spans 17 longwords (ie 68 bytes) and contains 
the ASCII string of the filename portion of the URL.  This string
can be up to 67 characters long, and will always be null-terminated.


There are a total of 2 "spare" items in the packet (usually filled
with zeros).  They are located at items 20 and 21.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long in
the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of GCN operation
(called BACODINE in those days).  It has since become a permanent fixture
as multiple sites are now operating, and it is not wise to play
with format/content changes.  And it still does function as a fixed-content
termination value for processing/transmission checking.




================================================================================

TYPE=134 PACKET CONTENTS:

The MAXI_UNKNOWN_SOURCE packet contains detections of increases of previously uncataloged sources
on the sky periodically monitored by MAXI instrument on the ISS.

The MAXI_UNKNOWN_SOURCE packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------  
long         0       pkt_type     integer         Packet type number (=134)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       trig_num     integer         A unique ID number
long         5       event_tjd    [days]          Truncated Julian Day
long         6       event_sod    [centi-sec]     (int)(sssss.sss *100)
long         7       event_ra     [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         8       event_dec    [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         9       event_flux   [deci-mCrab]    Flux in mCrabs during trig window, 0 to inf
long         10      flux_err     [deci-mCrab]    Flux_Error in mCrabs during trig window
long         11      event_error  [0.0001-deg]    (int)(0.0 to 180.0 *10000)
//long         12      src_bkg_lo   [deci-mCrab]    Source and Bkg flux per scan in Lo band *10
//long         13      src_bkg_med  [deci-mCrab]    Source and Bkg flux per scan in Med band *10
//long         14      src_bkg_hi   [deci-mCrab]    Source and Bkg flux per scan in Hi band *10
long         15      t_and_e      integer         timescale and e_band of the detection trigger criteria
//long         16      lon_lat      two_shorts      (int)(Latitude *100; ditto Lon
long         17      spare        integer         4 bytes for future use
long         18      event_flags  bits            Type of source/trigger found
long         19      misc         bits            Misc flag bits and fields
long         20-38   spare[19]    integer         76 bytes for future use
long         39      pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS (type=134):


The "pkt_type" is an integer with a value of 134, which tells that this packet
is of type MAXI_UNKNOWN_SOURCE and therefore indicates the contents and format
for the other 39 fields.

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start from 1 when
the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 1.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the burst_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

As of 31Aug2011, the MAXI Team changed the way these Unknown transients are identified.
//The "trig_num" is a number which uniquely identifies this trigger.
//If is composed of 4 parts DDDdOPPPPP, where "DDDd" is the day_number of the event,
//and where the "d" is the units portion of the total 4-digit day_number.  For space limitations
//the full 10-digit ID number can not be stored in a 4-byte packet slot.  The 'd' is stored
//in slot 15 (see "misc" below).  The "O" is the orbit number for that day.
//And "PPPPP" is the position.  This 10-digit umber provides a unique ID number for each event.
The "trig_num" is a number which uniquely identifies this trigger.
It is a sequential serial number assigned by the analysis processing.
As of 24Sep2011, it is 30.

The "event_tjd" is the Truncated Julian Day of the MAXI trigger,
eg. TJD=15197 is 01 Jan 2010.

The "event_sod" is the UT seconds-of-day (SOD) of the outburst.  The SOD
is multiplied by 100, and converted to a 32-bit integer for the packet
(i.e the units are centi-sec; 10 msec),  eg. 4689500 = 46895.00 sec-of-day =
13:01:35.00 UT.

The "event_ra" and "dec" fields specify the RA,Dec location of the source object
undergoing the transient outburst (in J2000 epoch).  These intrinsically floating point
quantities have been multiplied by 10,000 and then integerized to make
an 32-bit integer quantity with units of 10^-4-degrees.

The "event_flux" is the flux (in mCrab units) of the event
during the time duration of the successful trigger criterion.
The units are deci-counts (the counts value has been multiplied by 10
and then integerized).
As of 31Aug2011, the MAXI Team added this flux_error field.
The "event_flux_error" is the uncertainty on the flux (in mCrab units).
The units are deci-counts (the counts value has been multiplied by 10
and then integerized).

The "event_error" is the radius of the circle that will contain on average
TBD% of the bursts.  The units are 0.0001-degrees (ie the fl.pt. error was multiplied
by 10000 and then integerized).  The MAXI team has hardwire fixed this value
to be 0.5 deg.  It contains the statistical error and the systematic error.

As of 31Aug2011, the MAXI Team stopped giving these value for the Unknown Transients.
//The "src_bkg_lo", "src_bkg_med", and "src_bkg_hi" each contain the 'source' and
//'background' mCrabs in the lo/medium/high energy_bands.  The 'source'
//has been multiplied by 10 and up-shifted into the most significant two bytes
//of the 4-byte slot.  The 'background' counts has been multiplied by 10
//and included in the least significant two bytes of the 4-byte slot.
//(ie the 'source' (total-bkg) for each band is in units of deci-nCrabs in the hi-word,
//and the 'background' also in deci-nCrabs is in the lo-word for each of the
//3 energy_bands in each of the 3 4-byte slots.)

The 't_and_e' contains the index values indicating which timescale and energy_band
for the sucessful trigger criteria.  The timescale index value occupies the first 4 bits (2^0 - 2^3),
and has values defined for 0 to 8 representing 'undefined/unavailable', 1 sec, 3 sec, 10 sec,
30 sec, 1 scan, 1 orbit, 4 orbits, 1 day; respectively.
The energy_band occupies bits 2^8 through 2^11, and has values defined for 0 to 4
representing "undefined/unavailable", "Low, 2-4 keV","Medium, 4-10 keV","High, 10-20 keV", "L+M, 2-10 keV".

As of 31Aug2011, the MAXI Team stopped giving these Lon,Lat values for the Unknown Transients.
//The "lon_lat" is the Longitude and Lattitude of the s/c at the time of the MAXI trigger.
//This 4-byte quantity is composed of two 2-byte integers.  The high-order short
//is the latitude multiplied by 100 and integerized; the low-order short
//is the longitude multiplied by 100 and integerized; ie both values are in centi-degrees.

The "event_flags" field contains flag bit bits relating to the trigger solution and the source-type identity.
Bit_Number     Item_name      Description
----------     ---------      -----------
0-3            x_x_x_x        spare
4              neg_flux       Indicates message received from MAXI team contained a negative flux value(s)
5-31           x_x_x_x        spare

The "misc" field contains various miscelaneous flags and fields.
The bit packing of the "misc" entry in the packet is:
Bit_Number     Item_name      Description
----------     ---------      -----------
0-3            day_units      The most-signif digit of the DDDDD portion of the Trigger_ID_Number
4-29  spare.
30    When this bit is a 1, then this Notice was ground-generated; else flight-generated.
      For this Notice Type, they are always ground-generated.
31    spare.

There are a total of 24 "spare" items in the packet (usually filled with zeros).
They are located at items 12-14, 16-17 and 20-38.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long in
the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of GCN operation
(called BACODINE in those days).  It has since become a permanent fixture
as multiple sites are now operating, and it is not wise to play
with format/content changes.  And it still does function as a fixed-content
termination value for processing/transmission checking.



 




================================================================================

TYPE=135 PACKET CONTENTS:

The MAXI_KNOWN_SOURCE packet contains detections of flux increases in known sources
periodically monitored by MAXI instrument on the ISS.

The MAXI_KNOWN_SOURCE packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------  
long         0       pkt_type     integer         Packet type number (=135)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       trig_num     integer         A unique ID number
long         5       event_tjd    [days]          Truncated Julian Day
long         6       event_sod    [centi-sec]     (int)(sssss.sss *100)
long         7       event_ra     [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         8       event_dec    [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         9       event_flux   [deci-mCrab]    Flux in mCrabs during trig window, 0 to inf
long         10      spare        integer         4 bytes for future use
long         11      event_error  [0.0001-deg]    (int)(0.0 to 180.0 *10000)
long         12      src_bkg_lo   [deci-mCrab]    Source and Bkg flux per scan in Lo band *10
long         13      src_bkg_med  [deci-mCrab]    Source and Bkg flux per scan in Med band *10
long         14      src_bkg_hi   [deci-mCrab]    Source and Bkg flux per scan in Hi band *10
long         15      t_and_e      integer         timescale and e_band of the detection trigger criteria
long         16      lon_lat      two_shorts      (int)(Latitude *100; ditto Lon
long         17      spare        integer         4 bytes for future use
long         18      event_flags  bits            Type of source/trigger found
long         19      misc         bits            Misc flag bits and fields
long         20-21   spare[2]     integer         8 bytes for future use
quad_char    22-29   srcname[8]   chars           Filename (up to 32 chars long)
long         30-38   spare[9]     integer         36 bytes for future use
long         39      pkt_term     integer         Pkt Termination (always = \n)

PACKET ITEM DESCRIPTIONS (type=135):


The "pkt_type" is an integer with a value of 135, which tells that this packet
is of type maXI_KNOWN_SOURCE and therefore indicates the contents and format
for the other 39 fields.

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start from 1 when
the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 1.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the burst_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "trig_num" is a number which uniquely identifies this trigger.
If is composed of 4 parts DDDdOPPPPP, where "DDDd" is the day_number of the event,
and where the "d" is the units portion of the total 4-digit day_number.  For space limitations
the full 10-digit ID number can not be stored in a 4-byte packet slot.  The 'd' is stored
in slot 15 (see "misc" below).  The "O" is the orbit number for that day.
And "PPPPP" is the position.  This 10-digit umber provides a unique ID number for each event.

The "event_tjd" is the Truncated Julian Day of the MAXI trigger,
eg. TJD=15197 is 01 Jan 2010.

The "event_sod" is the UT seconds-of-day (SOD) of the outburst.  The SOD
is multiplied by 100, and converted to a 32-bit integer for the packet
(i.e the units are centi-sec; 10 msec),  eg. 4689500 = 46895.00 sec-of-day =
13:01:35.00 UT.

The "event_ra" and "dec" fields specify the RA,Dec location of the source object
undergoing the transient outburst (in J2000 epoch).  These intrinsically floating point
quantities have been multiplied by 10,000 and then integerized to make
an 32-bit integer quantity with units of 10^-4-degrees.

The "event_flux" is the flux of the source (in mCrab units)
during the time duration of the successful trigger criterion.
The units are deci-counts (the counts value has been multiplied by 10
and then integerized).

The "event_error" is the radius of the circle that will contain on average
TBD% of the bursts.  The units are 0.0001-degrees (ie the fl.pt. error was multiplied
by 10000 and then integerized).  The MAXI team has hardwire fixed this value
to be 0.5 deg.  It contains the statistical error and the systematic error.

The "src_bkg_lo", "src_bkg_med", and "src_bkg_hi" each contain the 'source' and
'background' mCrabs in the lo/medium/high energy_bands.  The 'source'
has been multiplied by 10 and up-shifted into the most significant two bytes
of the 4-byte slot.  The 'background' counts has been multiplied by 10
and included in the least significant two bytes of the 4-byte slot.
(ie the 'source' (total-bkg) for each band is in units of deci-nCrabs in the hi-word,
and the 'background' also in deci-nCrabs is in the lo-word for each of the
3 energy_bands in each of the 3 4-byte slots.)

The 't_and_e' contains the index values indicating which timescale and energy_band
for the sucessful trigger criteria.  The timescale index value occupies the first 4 bits (2^0 - 2^3),
and has values defined for 0 to 8 representing 'undefined/unavailable', 1 sec, 3 sec, 10 sec,
30 sec, 1 scan, 1 orbit, 4 orbits, 1 day; respectively.
The energy_band occupies bits 2^8 through 2^11, and has values defined for 0 to 4
representing "undefined/unavailable", "Low, 2-4 keV","Medium, 4-10 keV","High, 10-20 keV", "L+M, 2-10 keV".

The "lon_lat" is the Longitude and Lattitude of the s/c at the time of the MAXI trigger.
This 4-byte quantity is composed of two 2-byte integers.  The high-order short
is the latitude multiplied by 100 and integerized; the low-order short
is the longitude multiplied by 100 and integerized; ie both values are in centi-degrees.

The "event_flags" field contains flag bit bits relating to the trigger solution and the source-type identity.
Bit_Number     Item_name      Description
----------     ---------      -----------
0-3            class          Undef(=0), GRB(=1), nova-cv(=2), SN(=3), x-ray-star(=4), AGN(=5)
4              neg_flux       Indicates messages received from MAXI team containing negative flux value(s)
5-31           x_x_x_x        spare

The "misc" field contains various miscelaneous flags and fields.
The bit packing of the "misc" entry in the packet is:
Bit_Number     Item_name      Description
----------     ---------      -----------
0-3            day_units      The most-signif digit of the DDDDD portion of the Trigger_ID_Number
4-29  spare.
30    When this bit is a 1, then this Notice was ground-generated; else flight-generated.
      For this Notice Type, they are always ground-generated.
31    spare.

The "srcname" is the name of the object that is undergoing the transient outburst.
This item in the packet spans 8 longwords (ie 32 bytes) and contains the ASCII string
of the source_name.  This string can be up to 31 characters long, and will always be null-terminated.

There are a total of 13 "spare" items in the packet (usually filled with zeros).
They are located at items 10, 17, 20-21, and 30-38.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long in
the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of GCN operation
(called BACODINE in those days).  It has since become a permanent fixture
as multiple sites are now operating, and it is not wise to play
with format/content changes.  And it still does function as a fixed-content
termination value for processing/transmission checking.





================================================================================

TYPE=136 PACKET CONTENTS:

The MAXI_TEST packet duplicates/mimicks the MAXI_UNKNOWN_SOURCE packet.

The MAXI_TEST packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------  
long         0       pkt_type     integer         Packet type number (=136)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       trig_num     integer         A unique ID number
long         5       event_tjd    [days]          Truncated Julian Day
long         6       event_sod    [centi-sec]     (int)(sssss.sss *100)
long         7       event_ra     [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         8       event_dec    [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         9       event_flux   [deci-mCrab]    Flux mCrab during trig window, 0 to inf
long         10      spare        integer         4 bytes for future use
long         11      event_error  [0.0001-deg]    (int)(0.0 to 180.0 *10000)
long         12      src_bkg_lo   [deci-mCrab]    Source and Bkg flux per scan in Lo band *10
long         13      src_bkg_med  [deci-mCrab]    Source and Bkg flux per scan in Med band *10
long         14      src_bkg_hi   [deci-mCrab]    Source and Bkg flux per scan in Hi band *10
long         15      t_and_e      integer         timescale and e_band of the detection trigger criteria
long         16      lon_lat      two_shorts      (int)(Latitude *100; ditto Lon
long         17      spare        integer         4 bytes for future use
long         18      event_flags  bits            Type of source/trigger found
long         19      misc         bits            Misc flag bits and fields
long         20-38   spare[19]    integer         76 bytes for future use
long         39      pkt_term     integer         Pkt Termination (always = \n)

PACKET ITEM DESCRIPTIONS (type=136):


The "pkt_type" is an integer with a value of 136, which tells that this packet
is of type SWIFT_BAT_MONITOR and therefore indicates the contents and format
for the other 39 fields.

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start from 1 when
the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 1.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the burst_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "trig_num" is a number which uniquely identifies this trigger.
If is composed of 4 parts DDDdOPPPPP, where "DDDd" is the day_number of the event,
and where the "d" is the units portion of the total 4-digit day_number.  For space limitations
the full 10-digit ID number can not be stored in a 4-byte packet slot.  The 'd' is stored
in slot 15 (see "misc" below).  The "O" is the orbit number for that day.
And "PPPPP" is the position.  This 10-digit umber provides a unique ID number for each event.

The "event_tjd" is the Truncated Julian Day of the MAXI trigger,
eg. TJD=15197 is 01 Jan 2010.

The "event_sod" is the UT seconds-of-day (SOD) of the outburst.  The SOD
is multiplied by 100, and converted to a 32-bit integer for the packet
(i.e the units are centi-sec; 10 msec),  eg. 4689500 = 46895.00 sec-of-day =
13:01:35.00 UT.

The "event_ra" and "dec" fields specify the RA,Dec location of the source object
undergoing the transient outburst (in J2000 epoch).  These intrinsically floating point
quantities have been multiplied by 10,000 and then integerized to make
an 32-bit integer quantity with units of 10^-4-degrees.

The "event_flux" is the source flux (in mCrab units)
during the time duration of the successful trigger criterion.
The units are deci-mCrabs (the value has been multiplied by 10
and then integerized).

The "event_error" is the radius of the circle that will contain on average
TBD% of the bursts.  The units are 0.0001-degrees (ie the fl.pt. error was multiplied
by 10000 and then integerized).  The MAXI team has hardwire fixed this value
to be 0.5 deg.  It contains the statistical error and the systematic error.

The "src_bkg_lo", "src_bkg_med", and "src_bkg_hi" each contain the 'source' and
'background' mCrabs in the lo/medium/high energy_bands.  The 'source'
has been multiplied by 10 and up-shifted into the most significant two bytes
of the 4-byte slot.  The 'background' counts has been multiplied by 10
and included in the least significant two bytes of the 4-byte slot.
(ie the 'source' (total-bkg) for each band is in units of deci-mCrabs in the hi-word,
and the 'background' also in deci-mCrabs is in the lo-word for each of the
3 energy_bands in each of the 3 4-byte slots.)

The 't_and_e' contains the index values indicating which timescale and energy_band
for the sucessful trigger criteria.  The timescale index value occupies the first 4 bits (2^0 - 2^3),
and has values defined for 0 to 8 representing 'undefined/unavailable', 1 sec, 3 sec, 10 sec,
30 sec, 1 scan, 1 orbit, 4 orbits, 1 day; respectively.
The energy_band occupies bits 2^8 through 2^11, and has values defined for 0 to 4
representing "undefined/unavailable", "Low, 2-4 keV","Medium, 4-10 keV","High, 10-20 keV", "L+M, 2-10 keV".

The "lon_lat" is the Longitude and Lattitude of the s/c at the time of the MAXI trigger.
This 4-byte quantity is composed of two 2-byte integers.  The high-order short
is the latitude multiplied by 100 and integerized; the low-order short
is the longitude multiplied by 100 and integerized; ie both values are in centi-degrees.

The "event_flags" field contains flag bit bits relating to the trigger solution and the source-type identity.
Bit_Number     Item_name      Description
----------     ---------      -----------
0-3            x_x_x_x        spare
4              neg_flux       Indicates messages received from MAXI team containing negative flux value(s)
5-31           x_x_x_x        spare

The "misc" field contains various miscelaneous flags and fields.
The bit packing of the "misc" entry in the packet is:
Bit_Number     Item_name      Description
----------     ---------      -----------
0-3            day_units      The most-signif digit of the DDDDD portion of the Trigger_ID_Number
4-29  spare.
30    When this bit is a 1, then this Notice was ground-generated; else flight-generated.
      For this Notice Type, they are always ground-generated.
31    spare.

There are a total of 21 "spare" items in the packet (usually filled with zeros).
They are located at items 10, 17 and 20-38.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long in
the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of GCN operation
(called BACODINE in those days).  It has since become a permanent fixture
as multiple sites are now operating, and it is not wise to play
with format/content changes.  And it still does function as a fixed-content
termination value for processing/transmission checking.




================================================================================

TYPE=137 PACKET CONTENTS:

UNDER CONSTRUCTION!!!!!  UNDER CONSTRUCTION!!!!!  UNDER CONSTRUCTION!!!!!

The OGLE packet contains detections of gravitational lensing events.

The OGLE packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=137)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       id_num       integer         ID number
long         5       det_tjd      [days]          Truncated Julian Day
long         6       det_sod      [centi-sec]     (int)(sssss.sss *100)
long         7       src_ra       [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long         8       src_dec      [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long         9       max_mag      [centi-mag]     (int)(mm.mm *100)
long        10       spare        integer         4 bytes for future use
long        11       base_mag     [centi-mag]     (int)(0.0 to 180.0 *10000)
long        12       b_mag_err    [centi-mag]     (int)(mm.mm *100)
long        13       max_tjd      [days]          Truncated Julian Day
long        14       max_sod      [centi-sec]     (int)(ss.ss *100)
long        15       max_t_err    [centi-days]    (int)(dd.dd *100)
long        16       umin         [d/n]           Impact Param [Einstein ring units]
long        17       umin_err     [d/n]           Impact Param error
long        18       flags        bits            Flag bits on source-type
long        19       misc         bits            Misc flag bits and fields
long        20       ampl-n-err   [centi]         Amplification factor and error
long        21       tau_width    [centi-days]    Width of the cusp
long        22       tau_err      [centi-days]    Error in the wifth
long        23-24    spare[2]     integer         8 bytes for the future
long        25       LeadTime     [centi-days]    Time until the maximum
long        26-29    spare[4]     integer         16 bytes for the future
quad_char   30-38    lc_url[9]    chars           The URL to the OGLE webpage on this event
long        39       pkt_term     integer         Pkt Termination (always = \n)



PACKET ITEM DESCRIPTIONS (type=137):


The "pkt_type" is an integer with a value of 137, which tells that this packet
is of type OGLE and therefore indicates the contents and format
for the other 39 fields.

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start from 1 when
the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 1.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the burst_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "id_num" is a reference number for the event.  It is the year-field*100000 plus
the serial_number-field from their "unique_ID" (eg "2011-GLG-231").

The "det_tjd" is the Truncated Julian Day of the detection/declaration of the event.
eg. TJD=15726 is 14 Uun 2011.

The "det_sod" is the UT seconds-of-day (SOD) of the detection/declaration of the event.
The SOD is multiplied by 100, and converted to a 32-bit integer for the packet
(i.e the units are centi-sec; 10 msec),  eg. 4689568 = 46895.68 sec-of-day =
13:01:35.68 UT.

The "src_ra" and "src_dec" fields specify the RA,Dec location of the detected
optical transient (in J2000 epoch).  These intrinsically floating point
quantities have been multiplied by 10,000 and then integerized to make
an integer quantity with units of 10^-4-degrees.

The "max_mag" field contains the projected maximum magnitude of the lensed star.
The units are centi-magnitudes (ie the magnitude was multiplied by 100 and
then integerized).

The "base_mag" field contains the baseline/quiescent magnitude of the lensed star.
The units are centi-magnitudes (ie the magnitude was multiplied by 100 and
then integerized).

The "base_mag_err" field contains the error in the baseline/quiescent magnitude value.
The units are centi-magnitudes (ie the magnitude was multiplied by 100 and
then integerized).

The "max_tjd" is the Truncated Julian Day of the maximum of the lightcurve (the cusp).

The "max_sod" is the UT seconds-of-day (SOD) of the maximum of the lightcurve (the cusp).
The SOD is multiplied by 100, and converted to a 32-bit integer for the packet
(i.e the units are centi-sec; 10 msec).

The "umin" (aka "u0") is the impact parameter [in Einstein Ring units] for the lensing compact object.
The value has been multiplied by 100 and integerized.

The "umin_err" is the uncertainty on "umin". 
The value has been multiplied by 100 and integerized.

The "flags" field contains flag bit bits relating to the source-type identity.
Bit_Number     Item_name      Description
----------     ---------      -----------
0-2            type_id        3 bits to indicate the source type: 0=grav_lens,1=CV,2=???
3-4            x_x_x_x        spare
//5              def_not_evt    0=No or 1=Yes, it is definitely not a Lensing Event (bad label; its really a Retraction).
//6              not_astro      When set, this trigger has been determined to not be anything astrophysical.
7-31           x_x_x_x        spare

The "misc" field contains various miscelaneous flags and fields.
The bit packing of the "misc" entry in the packet is:
Bit_Number     Item_name      Description
----------     ---------      -----------
//0            Amax_0         Flagbit is set when Amax was equal to 0.
//1            Amax_1         Flagbit is set when Amax is equal to unity.
//2            U0_0           Flagbit is set when u0 was equal to 0.
//3            spare          spare bit, not yet assigned.
//4            tE             Flagbit is set when tE or tEErr was 'nan'.
//5            Ibase          Flagbit is set when Ibase was 'nan'.
6              Amax_err_error Flagbit is set when Amax_err was 'nan'.
7-24           spares         spare bits, not yet assigned.
//25             Update         Flagbit is set when this is an updated/improved notice of a previous notice.
//26             Ampl_inc       Flagbit is set when the Amplitude value has increased from the original notice.
//27             Ampl_dec       Flagbit is set when the Amplitude value has decreased from the original notice.
//28             Hi-Ampl        Flagbit is set when the has identified as a high-amplitude event.
//29             Planet         Flagbit is set when the lightcurve fit solution now includes a planet.
29-31          spares         spare bits, not yet assigned.

The "ampl-n-err" has two values combined into a single 4-byte slot in the packet.
The "amplification" is the projected (based on a fit to the lightcurve)
of how bright the lensed star will get (ie the ratio of the maximum flux over
the baseline flux).  It is multiplied by 100, integerized, and packed into the
two most-significant bytes in the 4-byte slot.  The lower two bytes
contain the uncertainty in the applification factor.  It is also multiplied
by 100, integerized, and packed into the 2 least-significant bytes.

The "tau" (aka "tE_width") is the "Einstein crossing time"
The value is multiplied by 100, and converted to a 32-bit integer for the packet.

The "tau_err" (aka "tE_err") is the uncertainty in the "tau".
The value is multiplied by 100, and converted to a 32-bit integer for the packet.

The "LeadTime" is the difference between the "max_tjd+sod" and the "det_tjd+sod".
It is provided here as a convenience so that the calculation with the above
four quantities need not be repeated by the recipient ('max' minus 'det').
Positive values indicate that the maximum has not occurred yet; negative values
indicate the maximum has already passed (according to the fit of the lightcurve
at the time of issuance of the OGLE Event Message).
The value is multiplied by 100, and converted to a 32-bit integer for the packet.
As such, the quantization is 0.01 days (ie 14.4 minutes).

The "lc_url" is just the filename portion of the URL -- the leading path
for the full URL is "http://www.astrouw.edu.pl/~ogle/ogle3/ews/"
The full URL points to a OGLE webpage containing the initial and on-going information
about this lensing event (lightcurves, finding chart, reference stars, and data files).
This item in the packet spans 9 longwords (ie 36 bytes) and contains 
the ASCII string of the filename portion of the URL.  This string
can be up to 35 characters long, and will always be null-terminated.

There are a total of 21(tbr) "spare" items in the packet (usually filled
with zeros).  They are located at items (tbr)17 and 21-38.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long in
the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of GCN operation
(called BACODINE in those days).  It has since become a permanent fixture
as many sites are now operating, and it is not wise to play
with format/content changes.  And it still does function as a fixed-content
termination value for processing/transmission checking.



 
================================================================================

TYPE=139 PACKET CONTENTS:

The MOA packet contains detections of gravitational lensing events.

The MOA packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=139)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       id_num       integer         ID number
long         5       det_tjd      [days]          Truncated Julian Day
long         6       det_sod      [centi-sec]     (int)(sssss.sss *100)
long         7       src_ra       [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long         8       src_dec      [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long         9       max_mag      [centi-mag]     (int)(mm.mm *100)
long        10       spare        integer         4 bytes for future use
long        11       base_mag     [centi-mag]     (int)(0.0 to 180.0 *10000)
long        12       b_mag_err    [centi-mag]     (int)(mm.mm *100)
long        13       max_tjd      [days]          Truncated Julian Day
long        14       max_sod      [centi-sec]     (int)(ss.ss *100)
long        15       max_t_err    [centi-days]    (int)(dd.dd *100)
long        16       u0           [d/n]           Impact Param [Einstein ring units]
long        17       u0_err       [d/n]           Impact Param error
long        18       flags        bits            Flag bits on source-type
long        19       misc         bits            Misc flag bits and fields
long        20       ampl-n-err   [centi]         Amplification factor and error
long        21       tE_width     [centi-days]    Width of the cusp
long        22       tE_err       [centi-days]    Error in the width
long        23-24    spare[2]     integer         8 bytes for the future
long        25       LeadTime     [centi-days]    Time until the maximum
long        26-29    spare[4]     integer         16 bytes for the future
quad_char   30-38    lc_url[9]    chars           The URL to the MOA webpage on this event
long        39       pkt_term     integer         Pkt Termination (always = \n)



PACKET ITEM DESCRIPTIONS (type=139):


The "pkt_type" is an integer with a value of 139, which tells that this packet
is of type MOA and therefore indicates the contents and format
for the other 39 fields.

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start from 1 when
the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 1.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the burst_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "id_num" is a reference number for the event.  It is the year-field*100000 plus
the serial_number-field from their "unique_ID" (eg "2011-GLG-231").

The "det_tjd" is the Truncated Julian Day of the detection/declaration of the event.
eg. TJD=15726 is 14 Uun 2011.

The "det_sod" is the UT seconds-of-day (SOD) of the detection/declaration of the event.
The SOD is multiplied by 100, and converted to a 32-bit integer for the packet
(i.e the units are centi-sec; 10 msec),  eg. 4689568 = 46895.68 sec-of-day =
13:01:35.68 UT.

The "src_ra" and "src_dec" fields specify the RA,Dec location of the detected
optical transient (in J2000 epoch).  These intrinsically floating point
quantities have been multiplied by 10,000 and then integerized to make
an integer quantity with units of 10^-4-degrees.

The "max_mag" field contains the projected maximum magnitude of the lensed star.
The units are centi-magnitudes (ie the magnitude was multiplied by 100 and
then integerized).

The "base_mag" field contains the baseline/quiescent magnitude of the lensed star.
The units are centi-magnitudes (ie the magnitude was multiplied by 100 and
then integerized).

The "base_mag_err" field contains the error in the baseline/quiescent magnitude value.
The units are centi-magnitudes (ie the magnitude was multiplied by 100 and
then integerized).

The "max_tjd" is the Truncated Julian Day of the maximum of the lightcurve (the cusp).

The "max_sod" is the UT seconds-of-day (SOD) of the maximum of the lightcurve (the cusp).
The SOD is multiplied by 100, and converted to a 32-bit integer for the packet
(i.e the units are centi-sec; 10 msec).

The "u0" is the impact parameter [in Einstein Ring units] for the lensing compact object.
The value has been multiplied by 100 and integerized.

The "u0_err" is the uncertainty on "u0". 
The value has been multiplied by 100 and integerized.

The "flags" field contains flag bit bits relating to the source-type identity.
Bit_Number     Item_name      Description
----------     ---------      -----------
0-2            type_id        3 bits to indicate the source type: 0=grav_lens,1=CV,2=???
3-4            x_x_x_x        spare
//5              def_not_evt    0=No or 1=Yes, it is definitely not a Lensing Event (bad label; its really a Retraction).
//6              not_astro      When set, this trigger has been determined to not be anything astrophysical.
7-31           x_x_x_x        spare

The "misc" field contains various miscelaneous flags and fields.
The bit packing of the "misc" entry in the packet is:
Bit_Number     Item_name      Description
----------     ---------      -----------
0              Amax_0         Flagbit is set when Amax was equal to 0.
1              Amax_1         Flagbit is set when Amax is equal to unity.
2              U0_0           Flagbit is set when u0 was equal to 0.
3              spare          spare bit, not yet assigned.
4              tE             Flagbit is set when tE or tEErr was 'nan'.
5              Ibase          Flagbit is set when Ibase was 'nan'.
6-24           spares         spare bits, not yet assigned.
//25             Update         Flagbit is set when this is an updated/improved notice of a previous notice.
//26             Ampl_inc       Flagbit is set when the Amplitude value has increased from the original notice.
//27             Ampl_dec       Flagbit is set when the Amplitude value has decreased from the original notice.
//28             Hi-Ampl        Flagbit is set when the has identified as a high-amplitude event.
//29             Planet         Flagbit is set when the lightcurve fit solution now includes a planet.
29-31          spares         spare bits, not yet assigned.

The "ampl-n-err" has two values combined into a single 4-byte slot in the packet.
The "amplification" is the projected (based on a fit to the lightcurve)
of how bright the lensed star will get (ie the ratio of the maximum flux over
the baseline flux).  It is multiplied by 100, integerized, and packed into the
two most-significant bytes in the 4-byte slot.  The lower two bytes
contain the uncertainty in the applification factor.  It is also multiplied
by 100, integerized, and packed into the 2 least-significant bytes.

The "tE_width" is the "Einstein crossing time"
The value is multiplied by 100, and converted to a 32-bit integer for the packet.

The "tE_err" is the uncertainty in the "tE_width".
The value is multiplied by 100, and converted to a 32-bit integer for the packet.

The "LeadTime" is the difference between the "max_tjd+sod" and the "det_tjd+sod".
It is provided here as a convenience so that the calculation with the above
four quantities need not be repeated by the recipient ('max' minus 'det').
Positive values indicate that the maximum has not occurred yet; negative values
indicate the maximum has already passed (according to the fit of the lightcurve
at the time of issuance of the MOA Event Message).
The value is multiplied by 100, and converted to a 32-bit integer for the packet.
As such, the quantization is 0.01 days (ie 14.4 minutes).

The "lc_url" is just the filename portion of the URL -- the leading path
for the full URL is "https://it019909.massey.ac.nz/moa/alert/display.php?id="
The full URL points to a MOA webpage containing the initial and on-going information
about this lensing event (lightcurves, finding chart, reference stars, and data files).
This item in the packet spans 9 longwords (ie 36 bytes) and contains 
the ASCII string of the filename portion of the URL.  This string
can be up to 35 characters long, and will always be null-terminated.

There are a total of 21(tbr) "spare" items in the packet (usually filled
with zeros).  They are located at items (tbr)17 and 21-38.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long in
the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of GCN operation
(called BACODINE in those days).  It has since become a permanent fixture
as many sites are now operating, and it is not wise to play
with format/content changes.  And it still does function as a fixed-content
termination value for processing/transmission checking.

 

================================================================================

TYPE=140 and 141 PACKET CONTENTS:

The SWIFT_BAT_SUB_SUB_THRESH_POS & SWIFT_BAT_KNOWN_SRC_POS packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=140/141)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       ref_num      integer         Some reference number
long         5       evt_tjd      [days]          Truncated Julian Day
long         6       evt_sod      [centi-sec]     (int)(sssss.sss *100)
long         7       evt_ra       [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         8       evt_dec      [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         9       evt_inten    [counts]        Num events during trig window, 0 to inf
long         10      spare        integer         4 bytes for the future
long         11      evt_error    [0.0001-deg]    (int)(0.0 to 180.0 *10000)
long         12      phi          [centi-deg]     (int)(0.0 to 359.9999 *100)
long         13      theta        [centi-deg]     (int)(0.0 to +70.0 *100)
long         14      int_time     [milli-sec]     (int)(integration_time *100)
long         15-17   spare[4]     integers        16 bytes for the future
long         18      soln_status  bits            Type of source/trigger found
long         19      misc         bits            Misc stuff packed in here
long         20      image_signif [centi-sigma]   (int)(sig2noise *100)
long         21-24   spare[4]     integer         16 bytes for the future
long         25      cat_num      integer         On-board cat match ID number (only 141)
long         26      apid         integer         The APID this came from
long         27-38   spare[12]    integer         48 bytes for the future
long         39      pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 140 or 141, which tells that this packet
is of type SWIFT_BAT_SUB_SUB_THRESH_POS or SWIFT_BAT_KNOWN_SRC_POS and
therefore indicates the contents and format for the other 39 fields.

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start from 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the evt_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "ref_num" is a uniquely identifying reference number for each trigger.
It is the number of seconds since 01 Jan 2001 (when the notice was created).

The "evt_tjd" is the Truncated Julian Day of the Swift-BAT burst trigger,
eg. TJD=13370 is 01 Jan 2005.

The "evt_sod" is the UT seconds-of-day (SOD) of the Swift-BAT burst trigger.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized).  The trigger time can be off by +/- 320 msec
for events that have a trigger integration duration less than 64 msec.
This has to do with the way that the flight software is optimized
for execution speed for evaluating trigger criteria.
There is an additional +/- 20 msec of uncertainty in applying the UT Correction Factor.
Note that both of these uncertainties can be reduced by off-line/ground processing,
but for these automated notices, these large uncertainties exist.
This is important if you are using these SubSub triggers for temporal
correlations with other data sets.

The "evt_ra" and "evt_dec" are the RA,Dec coordinates of the burst (or transient)
as determined by the BAT flight software (J2000 epoch).  These intrinsically
floating point quantities have been multiplied by 10,000 and then integerized
to make an integer quantity with units of 0.0001-degrees.

The "evt_inten" is the height of the peak in the sky-image plane (ie the result
of the FFT,MaskConvolution,InvFFT calculation).  The units are counts
after a multiplicative factor of about 0.5-0.7.

The "evt_error" is the radius of the circle that will contain on average
TBD% of the bursts.  The units are 0.0001-degrees (ie the fl.pt. error was multiplied
by 10000 and then integerized).  [Initially, this value was hardwired 
at a constant value of 4arcmin (0.067deg) radius.  As the mission progressed,
this changed to 3 arcmin.  Eventually, it might become intensity-dependant.]

The "phi" and "theta" are the BAT instrument coordinates of the trigger (event)
as determined by the BAT flight software program.  Phi is measured phi=0
along the +Y-axis of the s/c and increasing towards the -Z axis,
and theta is measured from the boresight of the BAT instrument (it would have
been better to call this the 'zenith' angle).

The "int_time" is the intgration time of the image from which these "peaks"
were extracted.  The units are seconds.  This intrinsically
floating point quantity has been multiplied by 1000 and then integerized
to make an integer quantity with units of 0.001-seconds.

The "soln_status" (aka trigger_id) field contains (a) flag bits on the type of solution
that was found in the on-board image and about whether it matches anything in the
on-board source catalog, and (b) flag bits that are assigned in the GCN ground-processing.
The bit packing of the "soln_status" entry in the packet is:
BitNum  F-v-G  Item_name      Description
------  ----   ---------      -----------
0        F     point_src      0=No or 1=Yes, a point source was found
1-7      -     spare          spare for future use
8        G     gnd_cat_src    0=No or 1=Yes, it is in the ground catalog
9-10     -     spare          spare for future use
11       G     uncert_grb     0=No or 1=Yes, it is really probably not a GRBorTransient (VERY low image significance; less than 6.5 sigma)
12       -     spare          spare for future use
13       G     near_brt_star  0=No or 1=Yes, there is a nearby bright star (this is a duplicate of the "misc" 2^13).
14-27    -     spare          spare for future use
28       G     spatial_coinc  0=No or 1=Yes, there was a spatial coincidence with another event
29       G     temporal_coinc 0=No or 1=Yes, there was a temporal coincidence with another event
30       G     test_submit    0=No or 1=Yes, this is a test submission (internal use only)
31       -     spare          spare for future use
The columns are:  "BitNum" is the 2^N location within the 4-byte word; "F-v-G" is Flight-assigned vs
Ground-assigned;  "Item_name" is the short-description name of the quantity;  and
"Description" is the longer description of the quantity.

The "misc" field contains miscellaneous flag bits.
The defined bits are:
2^0-11  Not assigned.
2^11  When this bit is a 1, it indicates that 1 (or more) of the RA,Dec,Theta,Phi values
      were out of valid range (eg RA was 361 deg).
2^12  Not assigned.
2^13  When this bit is a 1, it indicates that this position is near (less than 0.3deg)  a bright star (mag less than 6.5).
2^14  When this bit is a 1, it indicates that the distance between the position and the center of the galaxy
      is less than the sum of the position error radius and the galaxy radius AND that the position error radius
      is smaller than the galaxy radius.  This position is inside the (circularized) NGC galaxy (or nearly in).
2^15  When this bit is a 1, it indicates that the distance between the position and the center of the galaxy
      is less than the sum of the position error radius and the galaxy radius AND that the galaxy radius
      is smaller than the position error radius.  This galaxy is inside the position error circle (or nearly in).
2^16-31  Not assigned.

The "image_signif" is the significance level of the peak in the FFT image plane.
This is the signal/noise (ie number of sigma) of the grb peak detection
multiplied by 100 and then integerized (ie units are centi-sigma).

The "cat_num" is the catalog number in the on-board catalog for sources
that match a known source in the on-board catalog.  (Typically it is a number
in the low 10,000s, 12,000s, or 15,000s.)  This is true only for Type=141;
for Type=140, this field will always be 0.

The "apid" is the APID that this particular sub_sub event came from;
either 0x23b or 0x129.

There are a total of 21 "spare" items in the packet (usually filled with zeros).

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.

 

================================================================================

TYPE=145 PACKET CONTENTS:

The COINCIDENCE packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=145)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       c1_fields    bits            Flags, mission, type
long         5       c1_trignum   [dn]            Trigger Number ID
long         6       c1_tjd       [integer]       Truncated Julian Day
long         7       c1_sod       [0.001-sec]     (int)(0.0 to 86400 *1000)
long         8       c1_ra        [0.0001-deg]    (int)(0.0 to 360.0 *10000)
long         9       c1_dec       [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         10      c2_fields    bits            Flags, mission, type
long         11      c2_trignum   [dn]            Trigger Number ID
long         12      c2_tjd       [integer]       Truncated Julian Day
long         13      c2_sod       [0.001-sec]     (int)(0.0 to 86400 *1000)
long         14      c2_ra        [0.0001-deg]    (int)(0.0 to 360.0 *10000)
long         15      c2_dec       [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         16      c3_fields    bits            Flags, mission, type
long         17      c3_trignum   [dn]            Trigger Number ID
long         18      c3_tjd       [integer]       Truncated Julian Day
long         19      c3_sod       [0.001-sec]     (int)(0.0 to 86400 *1000)
long         20      c3_ra        [0.0001-deg]    (int)(0.0 to 360.0 *10000)
long         21      c3_dec       [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         22      c4_fields    bits            Flags, mission, type
long         23      c4_trignum   [dn]            Trigger Number ID
long         24      c4_tjd       [integer]       Truncated Julian Day
long         25      c4_sod       [0.001-sec]     (int)(0.0 to 86400 *1000)
long         26      c4_ra        [0.0001-deg]    (int)(0.0 to 360.0 *10000)
long         27      c4_dec       [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         28      c5_fields    bits            Flags, mission, type
long         29      c5_trignum   [dn]            Trigger Number ID
long         30      c5_tjd       [integer]       Truncated Julian Day
long         31      c5_sod       [0.001-sec]     (int)(0.0 to 86400 *1000)
long         32      c5_ra        [0.0001-deg]    (int)(0.0 to 360.0 *10000)
long         33      c5_dec       [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         34      c6_fields    bits            Flags, mission, type
long         35      c7_fields    bits            Flags, mission, type
long         36      total_cnt    integer         Count of the total number of participants
long         37      trig_id      bits            Flags about the triggers
long         38      misc         bits            Flags from ground analysis
long         39      pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 145 , which tells that this packet
is of type TYPE_COINCIDENCE and
therefore indicates the contents and format for the other 39 fields.

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start from 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the evt_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "c1_fields" entry 3 quantities: type, mission_id, and spatial/temporal coincidence flags.
The "type" field gives the Noice Type number (1-170...) of the notice that contributed
to this multi-mission/instrument coincidence.
The "mission_id" field specifics which mission (and instrument) that contributed;
the values (1-~50) are defined in the table below.
And the "Spatial" and "Temporal" flag bits indicate which coincidence type occurred
(one or both flags can be set to 1 to indicate the respective coincidence).
BitNum  Item_name      Description
------  ---------      -----------
0-15    type           The Notice type number of this event that contributed
16-23   mission_id     The Mission_ID number of this event that contributed
24-29   spare          spare bits, unused
30      Spatial        Indicates that this contribution was a spatial match
31      Temporal       Indicates that this contribution was a temporal match

The "c1_trignum" is a uniquely identifying reference number for each trigger.
It is assigned by each contributing mission/instrument.
It is the number of seconds since 01 Jan 2001 (when the notice was created).

The "c1_tjd" is the Truncated Julian Day of the contributing trigger,
eg. TJD=13370 is 01 Jan 2005.

The "c1_sod" is the UT seconds-of-day (SOD) of the trigger time of the contributor.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized).

The "c1_ra" and "c1_dec" fields of the event.  These are only valid for
notice types that have position information.  These fields are set to 0.0 and 0.0
for those notice types that do not contain event position information.
These intrinsically floating point quantities have been multiplied by 10,000
and then integerized to make an integer quantity with units of 0.0001-degrees.

The "c2_fields", "c2_trignum", "c2_tjd", "c2_sod", "c2_ra", and "c2_dec" fields are identical
to the definitions listed ifor "c1_*" above, but for the 2nd contributing mission/instrument.
Obviously, since this is a coincidence notice type, there will always have to be
at least two contributors.

The "c3_fields", "c3_trignum", "c3_tjd", "c3_sod", "c3_ra", and "c3_dec" fields are identical
to the defintions listed ifor "c1_*" above, but for the 3rd contributing mission/instrument.
These fields will be 0 if there was no 3rd contributor.

The "c4_fields", "c4_trignum", "c4_tjd", "c4_sod", "c4_ra", and "c4_dec" fields are identical
to the definitions listed ifor "c1_*" above, but for the 4th contributing mission/instrument.
These fields will be 0 if there was no 4th contributor.

The "c5_fields", "c5_trignum", "c5_tjd", "c5_sod", "c5_ra", and "c5_dec" fields are identical
to the definitions listed ifor "c1_*" above, but for the 5th contributing mission/instrument.
These fields will be 0 if there was no 5th contributor.

The "c6_fields" slot is filled if there is a 6th participant in the coincidence (else 0).
It has the same content/definition as the "c1_fields" (but for the 6th participant).
These fields will be 0 if there was no 6th contributor.  For the 6th and 7th contributors 
there are no "cN_trignum", "cN_tjd", "cN_sod", "cN_ra", and "cN_dec" fields
because there was not suficient space in the 60-byte packet to holds these fields.
This is acceptable because it is extremely rare that there are more than 5 participants.

The "c7_fields" slot is filled if there is a 7th participant in the coincidence (else 0).
It has the same content/definition as the "c1_fields" (but for the 7th participant).
These fields will be 0 if there was no 7th contributor.

The "total_cnt" is the total number of mission/instruments that participated
in the coincidence.

The "trigger_id"  field contains (a) flag bits on the type of solution(s)
that was found in the on-board image and about whether it matches anything in the
on-board source catalog, and (b) flag bits that are assigned in the GCN ground-processing.
The bit packing of the "soln_status" entry in the packet is:
BitNum  F-v-G  Item_name      Description
------  ----   ---------      -----------
0        -     spare          spare for future use
1        G     def_not_evt    0=No or 1=Yes, it is definitely not a real Event (it is a Test event).
2-7      -     spare          spare for future use
8        G     gnd_cat_src    0=No or 1=Yes, it is in the ground catalog
9-29     -     spare          spare for future use
30       G     test_submit    0=No or 1=Yes, this is a test submission (internal use only)
31       -     spare          spare for future use
The columns are:  "BitNum" is the 2^N location within the 4-byte word; "F-v-G" is Flight-assigned vs
Ground-assigned;  "Item_name" is the short-description name of the quantity;  and
"Description" is the longer description of the quantity.

The "misc" field contains miscellaneous flag bits.
The defined bits are:
2^0-11  Not assigned.
2^11    When this bit is a 1, it indicates that 1 (or more) of the RA,Dec,Theta,Phi values
        were out of valid range (eg RA was 361 deg).
2^12-29  Not assigned.
30    When this bit is a 1, then this Notice was ground-generated; else flight-generated.
      For this Notice Type, they are always ground-generated.
31    spare.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.


MISSION_ID DEFINTION TABLE
The Mission-Instrument IDs are:
#define MI_ID_UNDEF                0   // This Mission-Inst type is undefined
#define MI_ID_GRO_BATSE            1   // GRO-BATSE
#define MI_ID_GRO_COMPTEL          2   // GRO-COMPTEL
#define MI_ID_GRO_OSSET            3   // GRO-OSSE
#define MI_ID_GRO_EGRET            4   // GRO-EGRET
#define MI_ID_IPN                  5   // IPN (for all 3 variants)
#define MI_ID_BRADFORD             6   // BRADFORD (the test pkt only)
#define MI_ID_ALEXIS               7   // ALEXIS
#define MI_ID_XTE_PCA              8   // XTE-PCA
#define MI_ID_XTE_ASM              9   // XTE-ASM
#define MI_ID_SAX_WFC              10  // SAX-WFC
#define MI_ID_SAX_NFI              11  // SAX-NFI
#define MI_ID_HETE                 12  // HETE (both WXM and SXC)
#define MI_ID_CNTRPART             13  // Counterpart (can be any gnd-gen)
#define MI_ID_INTEGRAL             14  // INTEGRAL (99% IBIS, 1% SPI)
#define MI_ID_INTEGRAL_SPIACS      15  // INTEGRAL_SPIACS
#define MI_ID_AAVSO                16  // AAVSO
#define MI_ID_MILAGRO              17  // MILAGRO
#define MI_ID_KONUS                18  // KONUS
#define MI_ID_SWIFT_BAT            19  // SWIFT-BAT (includes FOM and SC_SLEW)
#define MI_ID_SWIFT_XRT            20  // SWIFT-XRT
#define MI_ID_SWIFT_UVOT           21  // SWIFT-UVOT
#define MI_ID_AGILE                22  // AGILE
#define MI_ID_GLAST_GBM            23  // Fermi-GBM
#define MI_ID_GLAST_LAT            24  // Fermi-LAT
#define MI_ID_OGLE                 25  // OGLE
#define MI_ID_PIOTS                26  // PIOTS
#define MI_ID_KAIT                 27  // KAIT
#define MI_ID_MAXI                 28  // MAXI
#define MI_ID_NEAR                 29  // NEAR
#define MI_ID_CBAT                 30  // CBAT
#define MI_ID_MOA                  31  // MOA
#define MI_ID_LVC                  32  // LVC = LIGO+Virgo Consortium
#define MI_ID_CRTS                 33  // CRTS
#define MI_ID_SUZAKU               34  // Suzaku
#define MI_ID_SNEWS                35  // SNEWS
#define MI_ID_AMON_ICECUBE_HESE    36  // AMON IceCube HESE
#define MI_ID_CALET                37  // CALET_GBM
#define MI_ID_AMON_ICECUBE_EHE     38  // AMON IceCube EHE
#define MI_ID_GWHEN                39  // GW_HEN Coincidence
#define MI_ID_AMON_ICECUBE_COINC   40  // AMON IceCube COINC
#define MI_ID_POLAR                41  // POLAR
#define MI_ID_AMON_ICECUBE_CLUSTER 42  // AMON IceCube COLUSTER



================================================================================

TYPE=148 PACKET CONTENTS:

The SUZAKU_LIGHTCURVE packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=148)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       spare        integer         4 bytes for the future
long         5       trigger_tjd  [days]          Truncated Julian Day
long         6       trigger_sod  [centi-sec]     (int)(sssss.sss *100)
long         7-17    spare[11]    integer         44 bytes for the future
long         18      trigger_id   bits            Trigger identification flags
long         19-21   spare[3]     integer         12 bytes for the future
quad_char    22-38   url[17]      chars           The URL for the lightcurve
long         39      pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 148, which tells that this packet
is of type SUZAKU_LC and therefore indicates the contents and format
for the other 39 fields.

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start from 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the burst_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

[There is no "trig_num" for SUZAKU_LC at the time the notice is sent to GCN for distribution.]

The "trigger_tjd" is the Truncated Julian Day of the trigger,
eg. TJD=12640 is 01 Jan 2003.

The "trigger_sod" is the UT seconds-of-day (SOD) of the trigger.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized).

The "trigger_id" field contains bit_flags which attempt to identify
the type of event that triggered the KONUS instrument.
The bit packing of the "trigger_id" entry in the packet is:
Bit_Number     Item_name      Description
----------     ---------      -----------
0              times_only     0=No or 1=Yes, this packet contains only the times (trig,window,duration; no ra,dec,err) (from IPN_RAW copy)
1-8            x_x_x_x        spare
9              trig_time_only 0=No or 1=Yes, this packet contains only the trigger_time (no t_window, t_duration)) (from IPN_RAW copy)
10-14          x_x_x_x        spare
//15           def_not_grb    0=No or 1=Yes, it is definitely not a GRB
16-19          x_x_x_x        spare
20             det_BST01      0=No or 1=Yes, this was triggered from the BST01 detector
21             det_BST02      0=No or 1=Yes, this was triggered from the BST02 detector
22             transient      0=No or 1=Yes, this was triggered from the (ground) processing transient detection s/w
//29           temporal_coinc 0=No or 1=Yes, there was a temporal coincidence with another event
30-31          x_x_x_x        spare

The "url" is just the filename portion of the URL -- the leading path
for the full URL is "http://gcn.gsfc.nasa.gov/gcn/notices_suz".
This item in the packet spans 17 longwords (ie 68 bytes) and contains 
the ASCII string of the filename portion of the URL.  This string
can be up to 67 characters long, and will always be null-terminated.

There are a total of 15 "spare" items in the packet (usually filled
with zeros).  They are located at items 7-17 and 19-21.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.


================================================================================

TYPE=149 PACKET CONTENTS:

The SNEWS packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=149)   
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100)
long         4       trig_num     integer         Trigger num
long         5       event_tjd    [days]          Truncated Julian Day
long         6       event_sod    [centi-sec]     (int)(sssss.sss *100)   
long         7       event_ra     [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         8       event_dec    [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         9       event_flue   [counts]        Number of neutrinos (0 to infinity)
long         10      spare        integer         4 bytes for the future
long         11      event_error  [0.0001-deg]    (int)(0.0 to 180.0 *10000)
long         12      containment  [centi-%]       (int)(0.00 to 100.00 *100)
long         13      duration     [centi-sec]     (int)(0.0 to inf *100)
long         14-17   spare[4]     integer         16 bytes for the future
long         18      trig_id      bits            Trigger identification flags
long         19      misc         bits            Misc stuff packed in here
long         20-38   spare[19]    integer         76 bytes for the future
long         39      pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 149, which tells that this packet
is of type SNEWS and therefore indicates the contents and format
for the other 39 fields.

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start from 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the event_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "trig_num" is the uniquely identifying serial number
assigned by the SNEWS software to each trigger.

The "event_tjd" is the Truncated Julian Day of the trigger,
eg. TJD=12640 is 01 Jan 2003.

The "event_sod" is the UT seconds-of-day (SOD) of the trigger.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized).  This is the beginning of of the 'duration' search window
of the successful trigger.

The "event_ra" and "event_dec" are the RA,Dec coordinates (J2000 epoch) of the trigger/event
as determined by the detector in the coincidence with the best directional information.
These intrinsically floating point quantities have been multiplied by 10,000 and then integerized
to make an integer quantity with units of 0.0001-degrees.
These two fields are undefined for any trigger that dow not have Super-K as a contributor.

The "event_fluence" is actually the integrated fluence of the event
over the duration of the trigger.  The units are counts (ie number of neutrinos).
If this is an 'individual' notice sub-type, then it is the number of neutrinos
detected in that detector.  If it is a 'coincidence' sub-type, then it is
the number neutrinos detected by the first detector contributing to the coincidence.
More detailed information will be provided at snews.bnl.gov.

The "event_error" is the radius of the circle that will contain on average
containment_percentage of the event locations.  The units are 0.0001-deg (ie the error was multiplied
by 10000 and then integerized).

The "event_cont" is the containment percentage for the quoted event_error value.
It is stored in the packet in centi-percentage (e.g. 98.5% would be an integer value of 9850).

The "duration" is the time interval of the search window of the event_times stream
looking for an increase.  The units are seconds, quantized at the 0.1 sec level.
For the "coincidence" sub-type, this will always be 10.0 sec;<
and for the "individual" sub-type, it will be variable and will typically be the delta_time
between the first neutrino detected in that detector and the last detected;
each detector may have a different definition.

The "trigger_id" field contains bit_flags which attempt to identify
the type of transient event that triggered the SNEWS system.
The bit packing of the "trigger_id" entry in the packet is:
Bit_Number     Item_name      Description
----------     ---------      -----------
0              Subtype        Notice sub-type: 0=Individual, 1=Coincidence   
1              Test           Notice: 0=Real, 1=Test   
2              Radec_undef    RA,Dec knowledge: 0=defined, 1=Undefined (no position known)
3-4            spare          spare for future use
5              Retract        0=No or 1=Yes, it is definitely not a core-collapse supernova (a Retraction)
6-29           spare          spare for future use
30             Test_submit    0=No or 1=Yes, this is a test submission (internal use only)
31             spare          spare for future use

The "misc" field contains various bit_flags (ie which detectors contributed to the coincidence and.
if it was a 'possible' or 'good' contribution).
The bit packing is:
Bit_Number     Item_name      Description
----------     ---------      -----------
0              Super-K_part   Super-K contribution flag (aka 'participated')
1              Super-K_good   Quality_factor is 'possible'
2              Super-K_poss   Quality_factor is 'good'
3              Super-K_oride  Quality_factor is 'override'
4              LVD_part       LVD contribution flag
5              LVD_good       Quality_factor is 'possible'
6              LVD_poss       Quality_factor is 'good'
7              LVD_oride      Quality_factor is 'override'
8              IceCube_part   IceCube contribution flag
9              IceCube_good   Quality_factor is 'possible'
10             IceCube_poss   Quality_factor is 'good'
11             IceCube_oride  Quality_factor is 'override'
12             KamLAND_part   KamLAND contribution flag
13             KamLAND_good   Quality_factor is 'possible'
14             KamLAND_poss   Quality_factor is 'good'
15             KamLAND_oride  Quality_factor is 'override'
16             Borexino_part  Borexino instrument contribution flag
17             Borexino_good  Quality_factor is 'possible'
18             Borexino_poss  Quality_factor is 'good'
19             Borexino_oride Quality_factor is 'override'
20             Daya_Bay_part  Daya Bay instrument contribution flag
21             Daya_Bay_good  Quality_factor is 'possible'
22             Daya_Bay_poss  Quality_factor is 'good'
23             Daya_Bay_oride Quality_factor is 'override'
24             HALO_part      HALO instrument contribution flag
25             HALO_good      Quality_factor is 'possible'
26             HALO_poss      Quality_factor is 'good'
27             HALO_oride     Quality_factor is 'override'
28             TBD_part       TBD instrument contribution flag
29             TBD_good       Quality_factor is 'possible'
30             TBD_poss       Quality_factor is 'good'
31             TBD_oride      Quality_factor is 'override'

The 4 bits:
a) The 'contribution' flag means that that detector contributed to the notice.
b) The 'possible' flag means the trigger observed by that detector was possibly real.
c) The 'good' flag means the trigger within that detector was very likely real.
d) The 'override' flag means the trigger observed by that detector human-checked and is real with extremely high confiden.


There are a total of 24 "spare" items in the packet (usually filled
with zeros).  They are located at items 10, 14-17, and 20-38.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.




================================================================================

TYPE=150-153 PACKET CONTENTS:

The LVC Preliminary(150), Initial(151), Update(152), and Test(153)
packets consist of 40 four-byte quantities.
The order and contents are listed in the table below:

For Type = 150 (Preliminary):
Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=150)   
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100)
long         4       trig_num     integer         Trigger num
long         5       event_tjd    [days]          Truncated Julian Day
long         6       event_sod    [centi-sec]     (int)(sssss.sss *100)   
long         7-8     spare[2]     integer         8 bytes for the future
long         9       fluence      [evts]          (int)(log10(fluence)*10000)
long         10      peak_freq    [centi-Hz]      (int)(Hz*100.0)
long         11      FAR          [Hz]            (int)(log10(far)*10000)
long         12      event_type   bit_fields      Search/Pipeline/Group index values
long         13      trig_subsec  [micro-sec]     (int)(0.0 to 0.999999)
long         14      duration     [centi-sec]     (int)(0 to NNN.NN *100)
//long       15      Eta          [milli-dn]      (int)(0.00 to 0.NN *1000)
//long       16      ChipMass     [0.0001-Mo]     (int)(0.00 to 1.50 *10000)
//long       17      MaxDistance  [centi-Mpc]     (int)(0.00 to 600.00 *100)
long         18      trig_id      bits            Trigger identification flags
long         19      misc         bits            Misc stuff packed in here
long         20-28   spare[9]     integer         36 bytes for the future
[quad_char   29-38   url[10]      chars           The URL for the skymap]
long         39      pkt_term     integer         Pkt Termination (always = \n)


For Type = 151-153 (Initial,Update,Test):
(The only difference from 150/Preliminary is that now slots 29-38 are filled with the URL.)
Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=151-153)   
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100)
long         4       trig_num     integer         Trigger num
long         5       event_tjd    [days]          Truncated Julian Day
long         6       event_sod    [centi-sec]     (int)(sssss.sss *100)   
long         7-8     spare[2]     integer         8 bytes for the future
long         9       fluence      [J/m2]]         (int)(log10(fluence)*10000)
long         10      peak_freq    [centi-Hz]      (int)(Hz*100.0)
long         11      FAR          [Hz]            (int)(log10(far)*10000)
long         12      event_type   bit_fields      Search/Pipeline/Group index values
long         13      trig_subsec  [micro-sec]     (int)(0 to 999999)
long         14      duration     [centi-sec]     (int)(0 to NNN.NN *100)
//long       15      Eta          [milli-dn]      (int)(0.00 to 0.NN *1000)
//long       16      ChipMass     [0.0001-Mo]     (int)(0.00 to 1.50 *10000)
//long       17      MaxDistance  [centi-Mpc]     (int)(0.00 to 600.00 *100)
long         18      trig_id      bits            Trigger identification flags
long         19      misc         bits            Misc stuff packed in here
long         20      ProbNS,Rem   bits            (ProbNS*2^16)+ProbRemnant)*100
long         21-28   spare[8]     integer         32 bytes for the future
quad_char    29-38   url[10]      chars           The URL for the skymap
long         39      pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS (for 150-153):

The "pkt_type" is an integer with a value of 150-153 for the Preliminary,
Initial, Update, and Test LVC types.  (The Counterpart type will be described later.)
It tells which type of packet this is and therefore what the contents are
(the contents change somewhat between the various types).

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start at 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the trig_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "trig_num" is a uniquely identifying number for each trigger.

The "trig_tjd" is the Truncated Julian Day of the LVC event,
eg. TJD=14466 is 01 Jan 2008.

The "trig_sod" is the UT seconds-of-day (SOD) of the LVC event.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized). The trigger time can be off by TBD msec.
LVC has higher precission for the actual trigger SOD, but this field is kept
for LVC to be uniform with all the other GCN Notice types.  See slot 13
for the full 6 digits of precision (ie micro-sec) for the LVC Trigger SOD.

The normal slots 7 and 8 for the RA,Dec will contain nothing for 150-153.

The "fluence" is the estimated fluence of the GW burst signal.  The units are [J/m2].
It is defined (ie filled) only for the 'burst' class of trigger events.
The quantity stored in slot 9 is the log10() of the fleunce value from LVC
multiplied by 10,000 and then intergerized.  To retrieve the fluence value at the
receiving end, call fluence=pow(10.0,slot[9]/10000.0).

The "peak_freq" is peak frequency of the GW burst signal.  The units are [Hz].
It is defined (ie filled) only for the 'burst' class of trigger events.

The "FAR" is the false alarm rate.  The units are [Hz].
The quantity stored in slot 11 is the log10() of the FAR value from LVC
multiplied by 10,000 and then intergerized.  To retrieve the FAR value at the
receiving end, call FAR=pow(10.0,slot[11]/10000.0).

The "event_type" is a combination of the group-type, search-method and pipeline used.
The pipeline is packed in the bottom 8 bits (2^0 - 2^7), and the search-method is
packed into the next higher 8 bits (2^8 - 2^15), and group-type is packed
in the next higher 8 bits (2^16 - 2^23).
Pipeline: 
  0     undefined/illegal
  1     MTBAOnline
  2     CWB
  3     CWB2G
  4     GSTLAL
  5     GSTLAL_Spiir
  6     Hardware Injection
  7     X
  8     Q
  9     Omega
  10    Ringdown
  11    LIB
  12    Fermi
  13    Swift
  14    SNEWS
  15    pycbc
Search:
  0     undefined/illegal
  1     AllSky
  2     LowMass
  3     HighMass
  4     GRB
  5     Supernova
  6     MockDataChallenge
  7     AllSkyLong
Group:
  0     undefined/illegal
  1     CBC
  2     Burst

The "trig_sod_usec" is the sub-seconds portion of the Trigger_SOD.
The units are [microsec].

The "duration" is duration of the GW burst signal.  The units are [sec].
It is defined (ie filled) only for the 'burst' class of trigger events.

//These 3 Params are no longer reported for CBC-type events:
//The "Eta" is the ratio of the reduced mass to the total mass.  The units are [dn].
//It is defined (ie filled) only for the 'LowMass/CBC/inspiral' class of trigger events.
//The quantity stored in slot 15 is the eta value multiplied by 1000 and then intergerized.

//The "ChirpMass" is ((m1*m2)^3/5)/(m1+m2)^1/5.  The units are [M_solar].
//The chirp mass determines the leading-order amplitude and frequency evolution
//of the gravitational-wave signal emitted by the binary during its inspiral.
//It is defined (ie filled) only for the 'LowMass/CBC/inspiral' class of trigger events.
//The units are 0.0001-solar_mass (ie the value was multiplied by 10,000 and then integerized).

//The "MaxDistance" is the maximum distance this could be detected.  The units are [Mpc].
//It is defined (ie filled) only for the 'LowMass/CBC/inspiral' class of trigger events.

The "trig_id" field contains (a) flag bits on the type of solution
that was found by the LVC pipeline processing and
(b) flag bits that are assigned in the GCN ground-processing.
The bit packing of the "trig_id" entry in the packet is:
BitNum  F-v-G  Item_name      Description
------  ----   ---------      -----------
0        G     internal_flag  LCV_Team_internal distribution(=1), Full (Priv Phase) LV-EM-Forum_sign-up distribution(=0)
1        G     test_flag      Test-type: 0=Real, 1=Test
2        G     hardware_inj   Is this a hardware injection event: 0=no, 1=yes
3        G     vetted         Has this event been vetted by a human: 0=no, 1=yes
4        G     open_alert     Is this event an open alert: 0=no, 1=yes
5        G     retraction     0=No or 1=Yes, it is definitely not real, a Retraction
6-28     -     spare          spare for future use
29       G     temporal_coinc 0=No or 1=Yes, there was a temporal coincidence with another event
30             test_submit    0=No or 1=Yes, this is a test submission (internal use only)
31             spare          spare for furture use
The columns are:  "BitNum" is the 2^N location within the 4-byte word;
"Item_name" is the short-description name of the quantity;  and
"Description" is the longer description/definition of the quantity.

The "misc" field contains various bit_flags.
The bit packing is:
Bit_Number     Item_name      Description
----------     ---------      -----------
0              LHO            LIGO-Handford Obs contribution flag (aka 'participated')
1              LLO            LIGO-Livingston Obs contribution flag (aka 'participated')
2              Virgo          Virgo Obs contribution flag (aka 'participated')
3              GEO600         GEO600 Obs contribution flag (aka 'participated')
4              KAGRA          KAGRA Obs contribution flag (aka 'participated')
5              LIO            LIGO-India Obs contribution flag (aka 'participated')
6              spare_inst1    TBD1 Obs contribution flag (aka 'participated')
7              spare_inst2    TBD2 Obs contribution flag (aka 'participated')
8-9            cite_type      Citation type (0=undef,1=followup,2=supersedes,3=retraction)
10-18          spare          Spare bits within 'misc'.
19             Gnd_gen        When this bit is a 1, then this Notice was ground-generated; else flight-generated.
                              Always 1 for Test Notices.
20-23          Letter         The Letter prefix on the ID_number: 1='G', 2='T', 3='M', 4='Y', 5='H', 6='E', 7='K'.
24-31          serial         LVC-generated serial number for message sequence in this trig_num.

The "ProbNS" field contains the probability that at least one object in the event
has a mass less than 3 solar masses.  It is located in the upper 16 bits of the 32-bit quantity.
The "ProbRemnant" is in the lower 16 bits has the probability that there is matter in the surrounding central object.
Both quantities are in units of centi-percent (ie they have been multiplied by 100 and intergerized).
It is present on only Initial and Update notice types (but not the Preliminary notice type).

The "url_c0-3" is the start of slot 29 for Types I/U/T, and someties for P).
It is present in second issuance of the Preliminary notice (but not the 1st issuance of Preliminary).
It is present in both the Initial and Update notices.
It is not defined for the Counterpart Notice type.
It is a string that looks like: M131119/files/lalinference_nest.fits.gz
It needs to be combined with:
https://gracedb.ligo.org/apibasic/events
[Note that prior to 24jul15, it was recommended that  https://gracedb.ligo.org/events/
be used as the prefix string.  But this is the x509-based url string, and x509
requires that the recipient have an x509 certificate, and be able to launch a java script,
and this was just too complicated.  So LVC created a simplier web-based interface.]
to make the full URL:
https://gracedb.ligo.org/apibasic/events/M131119/files/lalinference_nest.fits.gz

For Preliminary, there are a total of 21 "spare" items in the packet (usually filled
with zeros).  They are located at items 7-8 and 20-38.
For Initial and Update, there are a total of 11 "spare" items in the packet (usually filled
with zeros).  They are located at items 7-8 and 30-38.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.

 

================================================================================

TYPE=154 PACKET CONTENTS:

The LVC_COUNTERPART packet contains detections of the burst counterpart
in both the optical, radio, and other (x-ray, TeV, neutrinos, etc ...) bandpasses.
However, to make the field_names shorter, they were given names
with an optical bias.  Only the name has this bias, the contents
cover the optical, the radio, the x-ray, particles, etc.

The LVC_COUNTERPART packet consists of 40 four-byte quantities.
The order and contents are listed in the table below:
This is intensionally identical to the Type=45 GRB_COUNTERPART packet.

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=154)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100) 
long         4       ref_num      integer         Some reference number (eg trig_num)
long         5       burst_tjd    [days]          Truncated Julian Day
long         6       burst_sod    [centi-sec]     (int)(sssss.sss *100)
long         7       cp_ra        [10^-4-deg]     (int)(0.0 to 359.9999 *10000)
long         8       cp_dec       [10^-4-deg]     (int)(-90.0 to +90.0 *10000)
long         9       cp_inten     [centi-mag]     (int)(mm.mm *100)
long        10       inten_err    [centi-mag]     (int)(mm.mm *100)
long        11       cp_error     [10^-4-deg]     (int)(0.0 to 180.0 *10000)
long        12       filter       integer         Opt_fltr, Radio_freq, Oth_??
long        13       seeing       [centi-arcsec]  (int)(ss.ss *100)
long        14       obs_tjd      [days]          Truncated Julian Day
long        15       obs_time     [centi-sec]     (int)(sssss.sss *100)
long        16       obs_dur      [centi-sec]     (int)(sssss.sss *100)
long        17       id_conf      [centi-%]       (int)(0.0 to 100.00 *100)
long        18       flags        bits            Flag bits on source-type
long        19       misc         bits            Misc flag bits and fields
quad_char   20-23    telescope[4] chars           The name of the telescope
quad_char   24-38    name[15]     chars           The name of the submittor
long        39       pkt_term     integer         Pkt Termination (always = \n)

PACKET ITEM DESCRIPTIONS (type=154):

The "pkt_type" is an integer with a value of 154, which tells that this packet
is of type LVC_COUNTERPART and therefore indicates the contents and format
for the other 39 fields.

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start from 1 when
the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 1.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the burst_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "ref_num" is a reference number for that burst.  Typically it is a trigger_number
of the burst for that mission that made the original detection of the burst.

The "burst_tjd" is the Truncated Julian Day of the burst,
eg. TJD=12275 is 01 Jan 2002.

The "burst_sod" is the UT seconds-of-day (SOD) of the burst.  The SOD
is multiplied by 100, and converted to a 32-bit integer for the packet
(i.e the units are centi-sec; 10 msec),  eg. 4689568 = 46895.68 sec-of-day =
13:01:35.68 UT.

The "cp_ra" and "cp_dec" fields specify the RA,Dec location of the detected
GRB counterpart (in J2000 epoch).  These intrinsically floating point
quantities have been multiplied by 10,000 and then integerized to make
an integer quantity with units of 10^-4-degrees.

The "cp_inten" field contains the magnitude of the detected optical transient.
The units are centi-magnitudes (ie the magnitude was multiplied by 100 and
then integerized).  If the packet is for a counterpart detection
in the radio band (see the misc flag bits field), then the "cp_mag" field
contains the flux in centi-uJy (ie 10^-8 Janskies).  If the counterpart
detection in the "other" (x-/g-ray) category, then the units type are specified
by the 'expo' subfield in the 'filter' field and then multiplied by 100 and integerized.

The "inten_error" field contains the uncertainty (statistical plus systematic)
of the cp_inten value.  The units are also centi-magnitudes or centi-uJy or
centi-units for 'other' (eg centi-kHz, centi-MHz, centi-GHz, centi-keV,...,centi-TeV).
(See the 'expo' subfield in the 'filter' value.)

The "cp_error" is the radius of the circle that will contain on average
68% of the counterpart locations.  The units are 10^-4-degrees.

The "filter" field specifies the filter that was used for the observation.
The filter_index comes from a predefined list [see below].
  0 n/a
  1 U
  2 B
  3 V
  4 R
  5 I
  6 J
  7 K
  8 L
  9 M
 10 r
The upper 8 bits (of this 32-bit field) contain the 'exponent' of the 'band'
if this is a 'radio' or 'other' counterpart observation:
RADIO:
  3 kHz
  6 MHz
  9 GHz
OTHER:
  3  keV
  6  MeV
  9  GeV
  12 TeV
  15 PeV

The "seeing" field contains the seeing during the observation.  The units
are centi-arcsec.

The "obs_tjd" is the Truncated Julian Day of the start of the observation.
The "obs_sod" is the Seconds-of-Day of the start of the observation.
The "obs_dur" is the duration in centi-seconds of the observation.  The seconds
is multiplied by 100 and then integerized (i.e. the units are centi-sec; 10 msec).

The "id_conf" is the confidence level of the identification of the counterpart to the GRB.
The percentage_value is multiplied by 100 and then integerized (i.e. the units are centi-%).

The "trig_id" field contains flag bit bits relating to the source-type identity.
Bit_Number     Item_name      Description
----------     ---------      -----------
0              n/a            n/a
1              GRB_DEF        This is definitely related to a LVC_Event/GRB
2-19           n/a            spare
20-23          let_code       The Letter Code (eg 1=GW_event(G-series), 2=Test, 3=Mock_data_challenge(M-series), 4, HW_injection, ...
24-27          n/a            spare
28             spatial_coinc  0=No or 1=Yes, there was a spatial coincidence with another event
29             temporal_coinc 0=No or 1=Yes, there was a temporal coincidence with another event
30-31          n/a            spare

The "misc" field contains various miscellaneous flags and fields.
The bit packing of the "misc" entry in the packet is:
Bit_Number     Item_name      Description
----------     ---------      -----------
0-2            xrt_rank       Detection ranking value {1-4} (only for XRT submissions)
3              xrt_warn       warning flag about the given source detection
4              n/a            spare
5-6            type           optical=1/radio=2/other=3
7-23           src_sn         source serial number
24-31          iexpo          intensity exponent

The "telescope" is the name of the telescope that made the detection.
This item in the packet spans 4 longwords (ie 16 bytes) and contains 
the ASCII string of the name of the telescope.  This string can be
up to 15 characters long, and will always be null-terminated.

The "name" is the name of the person submitting this counterpart detection notice.
This item in the packet spans 15 longwords (ie 60 bytes) and contains 
the ASCII string of the name of the submitter.  This string can be
up to 59 characters long, and will always be null-terminated.

There are no "spare" items in this packet.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long in
the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of GCN operation
(called BACODINE in those days).  It has since become a permanent fixture
as multiple sites are now operating, and it is not wise to play
with format/content changes.  And it still does function as a fixed-content
termination value for processing/transmission checking.



================================================================================

TYPE=157 PACKET CONTENTS:

UNDER CONSTRUCTION!!!!!  UNDER CONSTRUCTION!!!!!  UNDER CONSTRUCTION!!!!!

The AMON_ICECUBE_COINC Event packets consist of 40 four-byte quantities.
The order and contents are listed in the table below:

Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=157)   
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100)
long         4       id_num       integer         Trigger num
long         5       event_tjd    [days]          Truncated Julian Day
long         6       event_sod    [centi-sec]     (int)(sssss.sss *100)   
long         7       event_ra     [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         8       event_dec    [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         9       n_events     [evts]          Events [counts] (0 to infinity)
long         10      false_pos    [s^-1 sr^-1]    (int)(log10(far)*10000)
long         11      evt_error90  [0.0001-deg]    (int)(0.0 to 180.0 *10000) 90% Conf
//long         12      event_type   bit_fields      index value
long         12      spare        integer         4 bytes for the future
long         13      deltaT       [centi-sec]     (int)(0.0 to 0.999999 *100)
long         14      sigmaT       [centi-sec]     (int)(0 to NNN.NN *100)
long         15      observing    integer         bitmask of detectors
long         16      stream       integer         Which instrument/data source
long         17      rev          [dn]            0-N (revision/update serial number)
long         18      trig_id      bits            Trigger identification flags
long         19      misc         bits            Misc stuff packed in here
long         20      pvalue       [dn]            (int)(log10(pvalue)*10000)
long         21      substream    integer         Which instrument gets it
long         22-23   spare[2]     integer         8 bytes for the future
long         24      evt_error50  [0.0001-deg]    (int)(0.0 to 180.0 *10000) 50% Conf
long         25      sub_sec      [micro-sec]     (int)(0.sssssss *1e6)
long         26-38   spare[12]    integer         52 bytes for the future
long         39      pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 157, which tells that this packet
is of type AMON_ICECUBE_COINC and therefore indicates the contents and format
for the other 39 fields.

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start at 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the trig_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "trig_num" is a uniquely identifying number for each trigger
within each given 'stream'.

The "trig_tjd" is the Truncated Julian Day of the AMON event,
eg. TJD=14466 is 01 Jan 2008.

The "trig_sod" is the UT seconds-of-day (SOD) of the AMON event.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized). The trigger time can be off by TBD msec.

The "ra" and "dec" fields specify the RA,Dec location of the multiply detected
event (in J2000 epoch).  These intrinsically floating point
quantities have been multiplied by 10,000 and then integerized to make
an integer quantity with units of 10^-4-degrees.

The "n_events" field contains the number of events (ie neutrinos)
that occurred during the deltaT window.

The "false_pos" (flase positive rate) field contains the expected number of this type
of alerts per steradian per second in the vicinity of a given alert.
The quantity stored in slot 10 is the log10() of the False_pos value from AMON
multiplied by 10,000 and then intergerized.  To retrieve the False_pos value at the
receiving end, call false_pos=pow(10.0,slot[10]/10000.0).

The "event_error90" is the radius of the circle that will contain on average
90% of the sources.  The units are 0.0001-degrees (ie the fl.pt. error was multiplied
by 10000 and then integerized).
It contains the statistical error and the systematic error(TBD).

//The "event_type" field contains the  ?????????????.
//The ??????

The "deltaT" field contains the Time window of the burst.
The units are centi-sec (ie the deltaT was multiplied by 100 and
then integerized).

The "sigmaT" field contains the Uncertainty of the time window.
The units are centisec (ie the sigmaT was multiplied by 100 and
then integerized).

The "observing" is a bitmask field containing which detectors/instrument/observatories
which had within their field of view a part of the sky where a given alert originated.
This number is computed as a Sum(2^k), where k is observatory stream number.
For example, IceCube is stream 0 for AMON, ANTARES is stream 1, so if both
of these observatories are observing, this number would be 2^0+2^1=3.)
Bit_Number     Description
----------     -----------
0              IceCube
1              ANTARIES
2              ????
3              ????
4              ????

The "stream" field contains the index number identifying
which analysis stream produced this trigger.  It ranges from 1 - NN.
"Stream" (aka search_criteria_index / trigger_criteria_index)
for the AMON alert means a specific set of search/analysis criteria
used to define set of events as a special/temporal coincidence
(maybe something as "trigger").  For example, the default stream looks for
100 sec time window and 2 sigma angular separation in order to accept
events from different experiments as being in coincidence. There could be
other analysis streams looking for coincidences using different search criteria
(e.g. smaller or larger time window, requiring specific events to occur:
e.g. requiring to have the IceCube HESE neutrinos in an alert).
The list of streams are: 0=undefined, 1=ICECUBE_HESE, 2=abc, 3=def, ...

The "rev" field contains the number of times this notice was revised (ie updated).
rev=0 means the first notice, 1 is the 1st revision, 2 is 2nd revision, etc.
Revisions can happen when (a) one of the original detectors updates their
information (eg change in the number of neutrinos in their event), or
(b) a 3rd detector now has a coincidence with the original two detectors
(ie some detectors have longer data analysis times).

The "trig_id" field contains (a) flag bits on the type of solution
that was found by the AMON processing and
(b) flag bits that are assigned in the GCN ground-processing.
The bit packing of the "trig_id" entry in the packet is:
BitNum  F-v-G  Item_name      Description
------  ----   ---------      -----------
0-4      -     spare          spare for future use
5        G     retraction     0=No or 1=Yes, it is definitely not real, a Retraction
6-27     -     spare          spare for future use
28       G     spatial_coinc  0=No or 1=Yes, there was a spatial coincidence with another event
29       G     temporal_coinc 0=No or 1=Yes, there was a temporal coincidence with another event
30-31    -     spare          spare for future use
The columns are:  "BitNum" is the 2^N location within the 4-byte word;
"Item_name" is the short-description name of the quantity;  and
"Description" is the longer description of the quantity.

The "misc" field contains various bit_flags.
The bit packing is:
Bit_Number     Item_name      Description
----------     ---------      -----------
0              IceCube        They constributed to the AMON coincidence event
1              ANTARES        They constributed to the AMON coincidence event
2-7            spare          Spare bits within 'misc'.
8-9            cite_type      Citation type (0=undef,1=followup,2=supersedes,3=retraction)
10-31          spare          Spare bits within 'misc'.

The "pvalue" field contains the probability value that the trigger is real
(ie astrophysical, and not a chance coincidence).
This alert p-value is calculated by combining p-values of each individual neutrino
contributing to a given alert.  Observatories should send AMON the p-values for their events.
These event p-values measure some event quality from each observatory
(e.g., for IceCube subthrehsold stream event p-value is a probability of an event
being atmospheric neutrino of energy larger or equal than energy observed P(E>=E_observed)).
In case that some observatory does not send AMON their event p-value,
we can only report false alarm rate, but not valid combined p-value
associated with the corresponding alert.
The range of valuesi 0.0 to 1.0.
The quantity stored in slot 20 is the log10() of the Prob_value value from AMON
multiplied by 10,000 and then intergerized.  To retrieve the Prob_value value at the
receiving end, call pvalue=pow(10.0,slot[20]/10000.0).

The "substream" field contains the index number identifying to which instrument
should receive it during the proprietary phase.  It ranges from 1 - NN.

The "event_error50" is the radius of the circle that will contain on average
50% of the sources.  The units are 0.0001-degrees (ie the fl.pt. error was multiplied
by 10000 and then integerized).
It contains the statistical error and the systematic error(TBD).

The "sub_sec" field contains the subseconds portion of the trigger_sod.
It contains the fractional-seconds portion to 6 significant digits;
0.123456*1e6 and then integerized.

There are a total of 17 "spare" items in the packet (usually filled
with zeros).  They are located at items 21-88.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.



================================================================================

TYPE=158 PACKET CONTENTS:

The AMON_ICECUBE_HESE Event packets consist of 40 four-byte quantities.
The order and contents are listed in the table below:

For Type = 158
Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=158)   
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100)
long         4       event_num    integer         Event Trigger/ID number
long         5       event_tjd    [days]          Truncated Julian Day
long         6       event_sod    [centi-sec]     (int)(sssss.sss *100)   
long         7       event_ra     [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         8       event_dec    [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         9       n_events     [evts]          Events [counts] (always 1)
long         10      false_pos    [s^-1 sr^-1]    (int)(log10(far)*10000)
long         11      evt_error90  [0.0001-deg]    (int)(0.0 to 180.0 *10000) 90% Conf
long         12      spare        integer         4 bytes for the future
long         13      deltaT       [centi-sec]     (int)(0.0 to 0.999999 *100)
long         14      sigmaT       [centi?-sec]    (int)(0 to NNN.NN *100)
long         15      spare        integer         4 bytes for the future
long         16      stream       integer         Which instrument/data source
long         17      rev          [dn]            0-N (revision/update serial number)
long         18      trig_id      bits            Trigger identification flags
long         19      misc         bits            Misc stuff packed in here
long         20      pvalue       [dn]            (int)(log10(pvalue)*10000)
long         21      charge       [centi-pe]      (int)(0.0 to NNNN.N *100
long         22      signal_track [0.01-1.0]      (int)(0.0 to NNNN.N *100 ???
long         23      run_num      [dn]            Run number in which the Event was found
long         24      evt_error50  [0.0001-deg]    (int)(0.0 to 180.0 *10000) 50% Conf
long         25      sub_sec      [micro-sec]     (int)(0.sssssss *1e6)
long         26-38   spare[12]    integer         52 bytes for the future
long         39      pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 158, which tells that this packet
is of type AMON_ICECUBE_HESE and therefore indicates the contents and format
for the other 39 fields.

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start at 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the trig_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "event_num" is a uniquely identifying number for each event/trigger.
(See also 'run_num' below.)

The "trig_tjd" is the Truncated Julian Day of the AMON event,
eg. TJD=14466 is 01 Jan 2008.

The "trig_sod" is the UT seconds-of-day (SOD) of the AMON event.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized). The trigger time can be off by TBD msec.

The "ra" and "dec" fields specify the RA,Dec location of the multiply detected
event (in J2000 epoch).  These intrinsically floating point quantities
have been multiplied by 10,000 and then integerized to make
an integer quantity with units of 10^-4-degrees.

The "n_events" field contains the number of events (ie neutrinos)
that occurred during the deltaT window.  For ICECUBE_HESE type,
this number will always be 1 (ie one PeV neutrino).   (Alway one, but this field
is still included in the HESE type to be symetric with the EHE and COINC types.)

The "false_pos" (flase positive rate) field specifies how often a trigger of this intensity
would happen from just noise fluctuations (an identically 0 value  means it was not calculated).
The units are number per steradian per second.
The range of valuesi 0.0 to 1.0.
The quantity stored in slot 10 is the log10() of the False_pos value from AMON
multiplied by 10,000 and then intergerized.  To retrieve the False_pos value at the
receiving end, call false_pos=pow(10.0,slot[10]/10000.0).

The "event_error90" is the radius of the circle that will contain on average
90% of the sources.  The units are 0.0001-degrees (ie the fl.pt. error was multiplied
by 10000 and then integerized).
It contains the statistical error and the systematic error(TBD).

The "deltaT" field contains the Time window of the burst.
The units are centi-sec (ie the deltaT was multiplied by 100 and
then integerized).
This will always be identically 0 for HESE notice type.

The "sigmaT" field contains the Uncertainty of the time window.
The units are centi-sec (ie the sigmaT was multiplied by 100 and
then integerized).
This will always be identically 0 for HESE notice type.

The "stream" field contains the index number identifying
which analysis stream produced this trigger.  It ranges from 1 - NN.
The list of streams are: 0=undefined, 1=ICECUBE_HESE, 2=abc, 3=def, ...

The "rev" field contains the number of times this notice was revised (ie updated).
rev=0 means the first notice, 1 is the 1st revision, 2 is 2nd revision, etc.
For AMON_ICECUBE_HESE type notices, the changes are most likely due to the RA,Dec,Error
values changing.

The "trig_id" field contains (a) flag bits on the type of solution
that was found by the AMON processing and
(b) flag bits that are assigned in the GCN ground-processing.
The bit packing of the "trig_id" entry in the packet is:
BitNum  F-v-G  Item_name      Description
------  ----   ---------      -----------
0        -     spare          spare for future use
1        G     role=test      Received with role="test"
2-4      -     spare          spare for future use
5        G     retraction     0=No or 1=Yes, it is definitely not real, a Retraction
6-27     -     spare          spare for future use
28       G     spatial_coinc  0=No or 1=Yes, there was a spatial coincidence with another event
29       G     temporal_coinc 0=No or 1=Yes, there was a temporal coincidence with another event
30-31    -     spare          spare for future use
The columns are:  "BitNum" is the 2^N location within the 4-byte word;
"Item_name" is the short-description name of the quantity;  and
"Description" is the longer description of the quantity.

The "misc" field contains various bit_flags.
The bit packing is:
Bit_Number     Item_name      Description
----------     ---------      -----------
0-7            spare          Spare bits within 'misc'.
8-9            cite_type      Citation type (0=undef,1=followup,2=supersedes,3=retraction)
10-23          spare          Spare bits within 'misc'.
24-31          sernum         8 bits containing the type of the HESE serial number(???)

P-value is a chance probability of alert being an atmospheric neutrino.
Low p-value (p less than 0.0027 (3 sigma) indicates a given alert is less likely to be background.
The range is from 0 to 1 (an identically 0 value means it was not calculated).
The quantity stored in slot 20 is the log10() of the Prob_value value from AMON
multiplied by 10,000 and then intergerized.  To retrieve the Prob_value value at the
receiving end, call pvalue=pow(10.0,slot[20]/10000.0).

The "charge" field contains the total neutrino event charge (in units of photo-electrons).
The units are centi-pe (ie the pe-count was multiplied by 100.0 and
then integerized).
[Note that this is not the same packing method as 'charge' in the AMON_ICECUBE_EHE packet.]

The "signal_trackness" field contains the probability that the neutrino event was signal and track-like.
The range is 0.0 to 1.0.
The units are centi-probability (ie the magnitude was multiplied by 100.0 and
then integerized).

The 'run_num' is the 2nd of two parts ('event_num' is first part, see above)
that uniquely identifies this event.

The "event_error50" is the radius of the circle that will contain on average
50% of the sources.  The units are 0.0001-degrees (ie the fl.pt. error was multiplied
by 10000 and then integerized).
It contains the statistical error and the systematic error(TBD).

The "sub_sec" field contains the subseconds portion of the trigger_sod.
It contains the fractional-seconds portion to 6 significant digits;
0.123456*1e6 and then integerized.

There are a total of TBD "spare" items in the packet (usually filled
with zeros).  They are located at items TBD-TBD.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.



================================================================================

TYPE=159 PACKET CONTENTS:

The AMON_ICECUBE_TEST Event packets consist of 40 four-byte quantities.
The exact same contents as type 158, except that slot[0] is filled with a value of 159.


================================================================================

TYPE=160 PACKET CONTENTS:

The CALET_GBM_FLT_LC Event packets consist of 40 four-byte quantities.
The order and contents are listed in the table below:

For Type = 160
Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=160)   
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100)
long         4       id_num       integer         Trigger num
long         5       trig_tjd     [days]          Truncated Julian Day
long         6       trig_sod     [centi-sec]     (int)(sssss.sss *100)   
long         7       point_ra     [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         8       point_dec    [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         9       lc_signif    [centi-sigma]   (int)(0.0 to inf *100)
long         10      fore_dur     [centi-sec]     (int)(log10(far)*100)
long         11      bkg_dur1     [centi-sec]     (int)(0.0 to 180.0 *100)
long         12      bkg_dur2     [centi-sec]     Always 0 for FLT_LC
long         13-15   spare[3]     integer         12 bytes for the future
long         16      lon_lat      two_shorts      (int)(Latitude *100); ditto Lon
long         17      lo_e_hi_e    two_shorts      (int)(Lo_energy *100); ditto Hi_energy
long         18      trig_id      bits            Trigger identification flags
long         19      misc         bits            Misc stuff packed in here
long         20-28   spare[9]     integer         36 bytes for the future
quad_char    29-38   url[10]      chars           The URL for the lightcurve
long         39      pkt_term     integer         Pkt Termination (always = \n)



PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 160, which tells that this packet
is of type CALET_GBM_FLT_LC and therefore indicates the contents and format
for the other 39 fields.

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start at 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the trig_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "trig_num" is a uniquely identifying number for each trigger.
It is the on-board MET clock value at the time of the trigger.
The URL for retrieving the lightcurve can be generated by inserting trig_num into:
http://cgbm.calet.jp/cgbm_trigger/flight/%d/index.html

The "trig_tjd" is the Truncated Julian Day of the CALET-GBM trigger,
eg. TJD=17023 is 01 Jan 2015.

The "trig_sod" is the UT seconds-of-day (SOD) of the CALET-GBM trigger.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized). The trigger time can be off by TBD msec.
(CALET has higher precission for the actual trigger SOD(tbd), but this field is kept
at centi-sec for CALET to be uniform with all the other GCN Notice types.)

The "point_ra" and "point_dec" fields specify the RA,Dec pointing direction
of the boresight of the spacecraft (in J2000 epoch).  These intrinsically
floating point quantities have been multiplied by 10,000 and then integerized
to make an integer quantity with units of 10^-4-degrees.

The "lc_signif" signal-to-noise ratio (in units of centi-sigma) during the foreground interval
over the background interval(s).  The intrinsically floating point quantity
has been multiplied by 100 and then integerized.

The "fore_dur" the amount of time (in units of centi-sec) the defines the triggering interval.
There are only 4 foreground/triggering intervals 0.25s, 0.5s, 1s and 4s defined
in the on-board triggering logic.  This intrinsically floating point quantity
has been multiplied by 100 and then integerized.

The "bkg_dur1" is the amount of time (units centi-sec) used to determine the background countrate
when used for the lc_significance.  For the (on-board) Flight triggers, there is only
one background interval used, and it occurs before the foreground interval
(with a separation of X.X sec between the end of the background interval and
the beginning of the foreground interval).  This intrinsically floating point quantity
has been multiplied by 100 and then integerized.

The "bkg_dur2" is the amount of time (units sec) used to determine the background countrate
when used for the lc_significance.  This is NOT defined (empty) for the Flight Lightcurve notice type.

The "lon_lat" is the Longitude and Latitude of the CALET s/c at the time of the CALET-GBM trigger.
This 4-byte quantity is composed of two 2-byte integers.  The high-order 2-byte word
is the latitude multiplied by 100 and integerized; the low-order 2-byte word
is the longitude multiplied by 100 and integerized; ie both values are in centi-degrees.

The "lo_e_hi_e" is the Lo_energy band limit and Hi_energy band limit of the trigger criteria.
This 4-byte quantity is composed of two 2-byte integers.  The high-order 2-byte word
is the hi_energy multiplied by 100 and integerized; the low-order 2-byte word
is the lo_energy multiplied by 100 and integerized; ie both values are in centi-keV.

The "trig_id" field contains (a) flag bits on the type of solution
that was found by the CALET pipeline processing and
(b) flag bits that are assigned in the GCN ground-processing.
The bit packing of the "trig_id" entry in the packet is:
BitNum  F-v-G  Item_name      Description
------  ----   ---------      -----------
0-4      -     spare          spare for future use
5        G     retraction     0=No or 1=Yes, it is definitely not real, a Retraction
6-28     -     spare          spare for future use
29       G     temporal_coinc 0=No or 1=Yes, there was a temporal coincidence with another event
30             test_submit    0=No or 1=Yes, this is a test submission (internal use only)
31       -     spare          spare for future use
The columns are:  "BitNum" is the 2^N location within the 4-byte word;
"Item_name" is the short-description name of the quantity;  and
"Description" is the longer description of the quantity.

The "misc" field contains various bit_flags.
The bit packing is:
Bit_Number     Item_name      Description
----------     ---------      -----------
0              HMX1_det       HMX1 detector participated in the trigger
1              HMX2_det       HMX2 detector participated in the trigger
2              SGM_det        SGM detector participated in the trigger
3-31           spare          spare

The URL for retrieving the lightcurve can be generated by appending 
the string stored at starting location url_c0_3 onto:
http://cgbm.calet.jp/cgbm_trigger/flight/
You can then use wget or curl (or similar) to retrieve
the lightcurve file from the CALET science data center.

There are a total of 13 "spare" items in the packet (usually filled
with zeros).  They are located at items 13-15, and 20-28.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.




================================================================================

TYPE=161 PACKET CONTENTS:

The CALET_GBM_GND_LC Event packets consist of 40 four-byte quantities.
The order and contents are listed in the table below:

For Type = 161
Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=161)   
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100)
long         4       id_num       integer         Trigger num
long         5       trig_tjd     [days]          Truncated Julian Day
long         6       trig_sod     [centi-sec]     (int)(sssss.sss *100)   
long         7       point_ra     [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         8       point_dec    [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         9       lc_signif    [centi-sigma]   (int)(0.0 to inf *100)
long         10      fore_dur     [centi-sec]     (int)(log10(far)*100)
long         11      bkg_dur1     [centi-sec]     (int)(0.0 to 180.0 *100)
long         12      bkg_dur2     [centi-sec]     (int)(0.0 to 180.0 *100)
long         13-15   spare[3]     integer         12 bytes for the future
long         16      lon_lat      two_shorts      (int)(Latitude *100); ditto Lon
long         17      lo_e_hi_e    two_shorts      (int)(Lo_energy *100); ditto Hi_energy
long         18      trig_id      bits            Trigger identification flags
long         19      misc         bits            Misc stuff packed in here
long         20-28   spare[9]     integer         36 bytes for the future
quad_char    29-38   url[10]      chars           The URL for the lightcurve
long         39      pkt_term     integer         Pkt Termination (always = \n)



PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 160, which tells that this packet
is of type CALET_GBM_GND_LC and therefore indicates the contents and format
for the other 39 fields.

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start at 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the trig_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "trig_num" is a uniquely identifying number for each trigger.
It is the on-board MET clock value at the time of the trigger.
The URL for retrieving the lightcurve can be generated by inserting trig_num into:
http://cgbm.calet.jp/cgbm_trigger/ground/%d/index.html

The "trig_tjd" is the Truncated Julian Day of the CALET-GBM trigger,
eg. TJD=17023 is 01 Jan 2015.

The "trig_sod" is the UT seconds-of-day (SOD) of the CALET-GBM trigger.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized). The trigger time can be off by TBD msec.
(CALET has higher precission for the actual trigger SOD(tbd), but this field is kept
at centi-sec for CALET to be uniform with all the other GCN Notice types.)

The "point_ra" and "point_dec" fields specify the RA,Dec pointing direction
of the boresight of the spacecraft (in J2000 epoch).  These intrinsically
floating point quantities have been multiplied by 10,000 and then integerized
to make an integer quantity with units of 10^-4-degrees.

The "lc_signif" signal-to-noise ratio (in units of centi-sigma) during the foreground interval
over the background interval(s).  The intrinsically floating point quantity
has been multiplied by 100 and then integerized.

The "fore_dur" the amount of time (in units of centi-sec) the defines the triggering interval.
There are only 4 foreground/triggering intervals 0.25s, 0.5s, 1s and 4s defined
in the on-board triggering logic.  This intrinsically floating point quantity
has been multiplied by 100 and then integerized.

The "bkg_dur1" is the amount of time (units centi-sec) used to determine the background countrate
when used for the lc_significance.  For the (on-board) Flight triggers, there is only
one background interval used, and it occurs before the foreground interval
(with a separation of X.X sec between the end of the background interval and
the beginning of the foreground interval).  This intrinsically floating point quantity
has been multiplied by 100 and then integerized.

The "bkg_dur2" is the amount of time (units centi-sec) used to determine the background countrate
when used for the lc_significance.  It is used only for the Ground-porcessed  triggers.
It comes after the fore_dur interval.  The separation between the end of the fore_dur
and the beginning of the bkg_dur2 interval is X.X sec.  This intrinsically floating point quantity
has been multiplied by 100 and then integerized.

The "lon_lat" is the Longitude and Latitude of the CALET s/c at the time of the CALET-GBM trigger.
This 4-byte quantity is composed of two 2-byte integers.  The high-order 2-byte word
is the latitude multiplied by 100 and integerized; the low-order 2-byte word
is the longitude multiplied by 100 and integerized; ie both values are in centi-degrees.

The "lo_e_hi_e" is the Lo_energy band limit and Hi_energy band limit of the trigger criteria.
This 4-byte quantity is composed of two 2-byte integers.  The high-order 2-byte word
is the hi_energy multiplied by 100 and integerized; the low-order 2-byte word
is the lo_energy multiplied by 100 and integerized; ie both values are in centi-keV.

The "trig_id" field contains (a) flag bits on the type of solution
that was found by the CALET pipeline processing and
(b) flag bits that are assigned in the GCN ground-processing.
The bit packing of the "trig_id" entry in the packet is:
BitNum  F-v-G  Item_name      Description
------  ----   ---------      -----------
0-4      -     spare          spare for future use
5        G     retraction     0=No or 1=Yes, it is definitely not real, a Retraction
6-28     -     spare          spare for future use
29       G     temporal_coinc 0=No or 1=Yes, there was a temporal coincidence with another event
30             test_submit    0=No or 1=Yes, this is a test submission (internal use only)
31       -     spare          spare for future use
The columns are:  "BitNum" is the 2^N location within the 4-byte word;
"Item_name" is the short-description name of the quantity;  and
"Description" is the longer description of the quantity.

The "misc" field contains various bit_flags.
The bit packing is:
Bit_Number     Item_name      Description
----------     ---------      -----------
0              HMX1_det       HMX1 detector participated in the trigger
1              HMX2_det       HMX2 detector participated in the trigger
2              SGM_det        SGM detector participated in the trigger
3-31           spare          spare

The URL for retrieving the lightcurve can be generated by appending 
the string stored at starting location url_c0_3 onto:
http://cgbm.calet.jp/cgbm_trigger/ground/
You can then use wget or curl (or similar) to retrieve
the lightcurve file from the CALET science data center.

There are a total of 13 "spare" items in the packet (usually filled
with zeros).  They are located at items 13-15, and 20-28.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.



================================================================================

TYPE=166 PACKET CONTENTS:

UNDER CONSTRUCTION!!!!!  UNDER CONSTRUCTION!!!!!  UNDER CONSTRUCTION!!!!!

The AMON_ICECUBE_CLUSTER Event packets consist of 40 four-byte quantities.
The order and contents are listed in the table below:

For Type = 166
Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=166)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100)
long         4       event_num    integer         Event Trigger/ID number
long         5       event_tjd    [days]          Truncated Julian Day
long         6       event_sod    [centi-sec]     (int)(sssss.sss *100)   
long         7       event_ra     [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         8       event_dec    [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         9       n_events     [evts]          Events [counts] (always 1)
long         10      spare        integer         4 bytes for the future
long         11      evt_error50  [0.0001-deg]    (int)(0.0 to 180.0 *10000) 50% Conf
long         12-15   spare[4]     integer         16 bytes for the future
long         16      stream       integer         Which instrument/data source
long         17      spare        integer         4 bytes for the future
long         18      trig_id      bits            Trigger identification flags
long         19      misc         bits            Misc stuff packed in here
long         20-38   spare[19]    integer         76 bytes for the future
long         39      pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 166, which tells that this packet
is of type AMON_ICECUBE_CLUSTER and therefore indicates the contents and format
for the other 39 fields.

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start at 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the trig_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "event_num" is a uniquely identifying number for each event/trigger.
(See also 'run_num' below.)

The "trig_tjd" is the Truncated Julian Day of the AMON event,
eg. TJD=14466 is 01 Jan 2008.

The "trig_sod" is the UT seconds-of-day (SOD) of the AMON event.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized). The trigger time can be off by TBD msec.

The "ra" and "dec" fields specify the RA,Dec location of the multiply detected
event (in J2000 epoch).  These intrinsically floating point quantities
have been multiplied by 10,000 and then integerized to make
an integer quantity with units of 10^-4-degrees.

The "n_events" field contains the number of events (ie neutrinos)
that occurred during the deltaT window.  For ICECUBE_EHE type,
this number will always be 1 (ie one PeV neutrino).   (Alway one, but this field
is still included in the EHE type to be symetric with the HESE and COINC types.)

The "event_error50" is the radius of the circle that will contain on average
50% of the sources.  The units are 0.0001-degrees (ie the fl.pt. error was multiplied
by 10000 and then integerized).
It contains the statistical error and the systematic error(TBD).

The "trig_id" field contains (a) flag bits on the type of solution
that was found by the AMON processing and
(b) flag bits that are assigned in the GCN ground-processing.
The bit packing of the "trig_id" entry in the packet is:
BitNum  F-v-G  Item_name      Description
------  ----   ---------      -----------
0        -     spare          spare for future use
1        G     role=test      Received with role="test"
2-4      -     spare          spare for future use
5        G     retraction     0=No or 1=Yes, it is definitely not real, a Retraction
6-27     -     spare          spare for future use
28       G     spatial_coinc  0=No or 1=Yes, there was a spatial coincidence with another event
29       G     temporal_coinc 0=No or 1=Yes, there was a temporal coincidence with another event
30-31    -     spare          spare for future use
The columns are:  "BitNum" is the 2^N location within the 4-byte word;
"Item_name" is the short-description name of the quantity;  and
"Description" is the longer description of the quantity.

The "misc" field contains various bit_flags.
The bit packing is:
Bit_Number     Item_name      Description
----------     ---------      -----------
0-7            spare          Spare bits within 'misc'.
8-9            cite_type      Citation type (0=undef,1=followup,2=supersedes,3=retraction)
10-23          spare          Spare bits within 'misc'.
24-31          sernum         8 bits containing the type of the EHE serial number(???)


There are a total of TBD "spare" items in the packet (usually filled
with zeros).  They are located at items TBD-TBD.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.






================================================================================

TYPE=168 PACKET CONTENTS:

UNDER CONSTRUCTION!!!!!  UNDER CONSTRUCTION!!!!!  UNDER CONSTRUCTION!!!!!

The GWHEN_COINC packet contains spatial/temporal coincidence
between LVC GW events and ICECUBE HEN neutrino events.

The GWHEN_COINC Event packets consist of 40 four-byte quantities.
The order and contents are listed in the table below:

For Type = 168:
Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=168)   
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      [centi-sec]     (int)(sssss.sss *100)
long         4       trig_num     integer         Trigger num
long         5       event_tjd    [days]          Truncated Julian Day
long         6       event_sod    [centi-sec]     (int)(sssss.sss *100)   
long         7       coinc_ra     [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         8       coinc_dec    [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         9       fluence      [J/m2]]         (int)(log10(fluence)*10000)
long         10      peak_freq    [centi-Hz]      (int)(Hz*100.0)
long         11      lvc_far      [Hz]            (int)(log10(lvc_far)*10000)
long         12      event_type   bit_fields      Search/Pipeline/Group index values
long         13      trig_subsec  [micro-sec]     (int)(0 to 999999)
long         14      duration     [centi-sec]     (int)(0 to NNN.NN *100)
long         15      eta          [milli-dn]      (int)(0.00 to 0.NN *1000)
long         16      chip_mass    [0.0001-Mo]     (int)(0.00 to 1.50 *10000)
long         17      max_distance [centi-Mpc]     (int)(0.00 to 600.00 *100)
long         18      trig_id      bits            Trigger identification flags
long         19      misc         bits            Misc stuff packed in here
long         20      hen_ra       [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         21      hen_dec      [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         22      hen_error    [0.0001-deg]    (int)(0.0 to +180.0 *10000)
long         23      coinc_far    [Hz]            (int)(log10(coinc_far)*10000)
long         24-28   spare[5]     integer         20 bytes for the future
quad_char    29-38   url[10]      chars           The URL for the skymap
long         39      pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS (for 168):

The "pkt_type" is an integer with a value of 168 for the GWHEN type.
It tells which type of packet this is and therefore what the contents are
(the contents change somewhat between the various types).

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start at 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the trig_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "trig_num" is a uniquely identifying number for each trigger.

The "trig_tjd" is the Truncated Julian Day of the LVC event,
eg. TJD=14466 is 01 Jan 2008.

The "trig_sod" is the UT seconds-of-day (SOD) of the LVC event.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized). The trigger time can be off by TBD msec.
LVC has higher precission for the actual trigger SOD, but this field is kept
for LVC to be uniform with all the other GCN Notice types.  See slot 13
for the full 6 digits of precision (ie micro-sec) for the LVC Trigger SOD.

The "coinc_ra" and "dec" fields specify the RA,Dec location of the convolution
of the two probability maps (the LVC_INITIAL map and the HEN probability map)
(in J2000 epoch).  These intrinsically floating point quantities
have been multiplied by 10,000 and then integerized to make
an 32-bit integer quantity with units of 10^-4-degrees.

The "fluence" is the estimated fluence of the GW burst signal.  The units are [J/m2].
It is defined (ie filled) only for the 'burst' class of trigger events.
The quantity stored in slot 9 is the log10() of the fleunce value from LVC
multiplied by 10,000 and then intergerized.  To retrieve the fluence value at the
receiving end, call fluence=pow(10.0,slot[9]/10000.0).

The "peak_freq" is peak frequency of the GW burst signal.  The units are [Hz].
It is defined (ie filled) only for the 'burst' class of trigger events.

The "lvc_far" is the false alarm rate.  The units are [Hz].
The quantity stored in slot 11 is the log10() of the FAR value from LVC
multiplied by 10,000 and then intergerized.  To retrieve the FAR value at the
receiving end, call far=pow(10.0,slot[11]/10000.0).

The "event_type" is a combination of the group-type, search-method and pipeline used.
The pipeline is packed in the bottom 8 bits (2^0 - 2^7), and the search-method is
packed into the next higher 8 bits (2^8 - 2^15), and group-type is packed
in the next higher 8 bits (2^16 - 2^23).
Pipeline: 
  0     undefined/illegal
  1     MTBAOnline
  2     CWB
  3     CWB2G
  4     GSTLAL
  5     GSTLAL_Spiir
  6     Hardware Injection
  7     X
  8     Q
  9     Omega
  10    Ringdown
  11    LIB
  12    Fermi
  13    Swift
  14    SNEWS
Search:
  0     undefined/illegal
  1     AllSky
  2     LowMass
  3     HighMass
  4     GRB
  5     Supernova
  6     MockDataChallenge
Group:
  0     undefined/illegal
  1     CBC
  2     Burst

The "trig_sod_usec" is the sub-seconds portion of the Trigger_SOD.
The units are [microsec].

The "duration" is duration of the GW burst signal.  The units are [sec].
It is defined (ie filled) only for the 'burst' class of trigger events.

The "Eta" is the ratio of the reduced mass to the total mass.  The units are [dn].
It is defined (ie filled) only for the 'LowMass/CBC/inspiral' class of trigger events.
The quantity stored in slot 15 is the eta value multiplied by 1000 and then intergerized.

The "chirp_mass" is ((m1*m2)^3/5)/(m1+m2)^1/5.  The units are [M_solar].
The chirp mass determines the leading-order amplitude and frequency evolution
of the gravitational-wave signal emitted by the binary during its inspiral.
It is defined (ie filled) only for the 'LowMass/CBC/inspiral' class of trigger events.
The units are 0.0001-solar_mass (ie the value was multiplied by 10,000 and then integerized).

The "max_distance" is the maximum distance this could be detected.  The units are [Mpc].
It is defined (ie filled) only for the 'LowMass/CBC/inspiral' class of trigger events.

The "trig_id" field contains (a) flag bits on the type of solution
that was found by the LVC pipeline processing and
(b) flag bits that are assigned in the GCN ground-processing.
The bit packing of the "trig_id" entry in the packet is:
BitNum  F-v-G  Item_name      Description
------  ----   ---------      -----------
0        G     internal_flag  LCV_internal distribution(=1), Full distribution(=0)
1        G     test_flag      Test-type: 0=Real, 1=Test
2-4      -     spare          spare for future use
5        G     retraction     0=No or 1=Yes, it is definitely not real, a Retraction
6-28     -     spare          spare for future use
29       G     temporal_coinc 0=No or 1=Yes, there was a temporal coincidence with another event
30             test_submit    0=No or 1=Yes, this is a test submission (internal use only)
31             spare          spare for furture use
The columns are:  "BitNum" is the 2^N location within the 4-byte word;
"Item_name" is the short-description name of the quantity;  and
"Description" is the longer description of the quantity.

The "misc" field contains various bit_flags.
The bit packing is:
Bit_Number     Item_name      Description
----------     ---------      -----------
0              LHO            LIGO-Handford Obs contribution flag (aka 'participated')
1              LLO            LIGO-Livingston Obs contribution flag (aka 'participated')
2              Virgo          Virgo Obs contribution flag (aka 'participated')
3              GEO600         GEO600 Obs contribution flag (aka 'participated')
4              KAGRA          KAGRA Obs contribution flag (aka 'participated')
5              LIO            LIGO-India Obs contribution flag (aka 'participated')
6              spare_inst1    TBD1 Obs contribution flag (aka 'participated')
7              spare_inst2    TBD2 Obs contribution flag (aka 'participated')
8-9            cite_type      Citation type (0=undef,1=followup,2=supersedes,3=retraction)
10             IceCube        IceCube contribution flag (aka 'participated')
11             Antares        Antares contribution flag (aka 'participated')
12             Auger          Auger contribution flag (aka 'participated')
13-18          spare          Spare bits within 'misc'.
19             Gnd_gen        When this bit is a 1, then this Notice was ground-generated; else flight-generated.
                              Always 1 for Test Notices.
20-23          Letter         The Letter prefix on the ID_number: 1='G', 2='T', 3='M', 4='Y', 5='H', 6='E', 7='K'.
24-31          serial         LVC-generated serial number for message sequence in this trig_num.

The "hen_ra" and "dec" fields specify the RA,Dec location of the HEN neutrino(s)
(in J2000 epoch).  These intrinsically floating point quantities
have been multiplied by 10,000 and then integerized to make
an 32-bit integer quantity with units of 10^-4-degrees.

The "hen_error" is the radius of the circle that will contain TBD%
of the correct direction of the HEN localizations.
The units are 0.0001-degrees (ie the fl.pt. error was multiplied
by 10000 and then integerized).

The "hen_far" is the false alarm rate of just the HEN event.  The units are [Hz].
The quantity stored in slot 11 is the log10() of the FAR value from LVC
multiplied by 10,000 and then intergerized.  To retrieve the FAR value at the
receiving end, call far=pow(10.0,slot[11]/10000.0).

The "url_c0-3" is the start of slot 29 for Types I/U/T).
It is not defined for the Preliminary or Counterpart Notice types.
It is a string that looks like: M131119/files/lalinference_nest.fits.gz
It needs to be combined with:
https://gracedb.ligo.org/apibasic/events
[Note that prior to 24jul15, it was recommended that  https://gracedb.ligo.org/events/
be used as the prefix string.  But this is the x509-based url string, and x509
requires that the recipient have an x509 certificate, and be able to launch a java script,
and this was just too complicated.  So LVC created a simplier web-based interface.]
to make the full URL:
https://gracedb.ligo.org/apibasic/events/M131119/files/lalinference_nest.fits.gz

For GWHEN there are a total of 11 "spare" items in the packet (usually filled
with zeros).  They are located at items 7-8 and 30-38.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.



================================================================================

TYPE=169 PACKET CONTENTS:

The AMON_ICECUBE_EHE Event packets consist of 40 four-byte quantities.
The order and contents are listed in the table below:

For Type = 169
Declaration  Index   Item         Units           Comments
Type                 Name
-----------  -----   ---------    ----------      ----------------
long         0       pkt_type     integer         Packet type number (=169)
long         1       pkt_sernum   integer         1 thru infinity
long         2       pkt_hop_cnt  integer         Incremented by each node
long         3       pkt_sod      
long         4       event_num    integer         Event Trigger/ID number
long         5       event_tjd    [days]          Truncated Julian Day
long         6       event_sod    [centi-sec]     (int)(sssss.sss *100)   
long         7       event_ra     [0.0001-deg]    (int)(0.0 to 359.9999 *10000)
long         8       event_dec    [0.0001-deg]    (int)(-90.0 to +90.0 *10000)
long         9       n_events     [evts]          Events [counts] (always 1)
long         10      spare        integer         4 bytes for the future
long         11      evt_error50  [0.0001-deg]    (int)(0.0 to 180.0 *10000) 50% Conf
long         12      spare        integer         4 bytes for the future
long         13      deltaT       [centi-sec]     (int)(0.0 to 0.999999 *100)
long         14      sigmaT       [centi?-sec]    (int)(log10(sigmaT)*10000)
long         15      energy       centi-TeV       (int)(0.0 to EEE.E *100
long         16      stream       integer         Which instrument/data source
long         17      rev          [dn]            0-N (revision/update serial number)
long         18      trig_id      bits            Trigger identification flags
long         19      misc         bits            Misc stuff packed in here
long         20      spare[2]     integer         8 bytes for the future
long         21      charge       [centi-pe]      (int)(0.0 to NNNN.N *100
long         22      signalness   [0.01-1.0]      (int)(log10(ssignalness)*10000)
long         23      run_num      [dn]            Run number in which the Event was found
long         24      spare        integer         spare
long         25      sub_sec      [micro-sec]     (int)(0.sssssss *1e6)
long         26-38   spare[12]    integer         52 bytes for the future
long         39      pkt_term     integer         Pkt Termination (always = \n)


PACKET ITEM DESCRIPTIONS:

The "pkt_type" is an integer with a value of 169, which tells that this packet
is of type AMON_ICECUBE_EHE and therefore indicates the contents and format
for the other 39 fields.

The "pkt_sernum" is the serial number for that packet type.  Each packet type
has its own sequential serial numbering.  They all start at 1
when the GCN program starts up.

The "pkt_hop_cnt" is a sequential numbering scheme that is incremented by 1
for each machine that the packet passes through (starting from 0 on the
machine of origin -- sometimes PK and sometimes capella2).  It is used
in checking the hand-shake processing through the PK, CC, and capella2 machines
within the GCN system.  Since the loss of GRO (May 2000), and therefore
the elimination of the PK and CC front-end processors, this hopcount item
is defunct -- all packets start on capella2 with a hopcount value of 0.

The "pkt_sod" is UT seconds-of-day when the packet was created/sent from
the GCN computer.  Do not confuse this with the trig_sod (see below).
The value is multiplied by 100 and integerized (i.e. the units are centi-sec; 10 msec).

The "event_num" is a uniquely identifying number for each event/trigger.
(See also 'run_num' below.)

The "trig_tjd" is the Truncated Julian Day of the AMON event,
eg. TJD=14466 is 01 Jan 2008.

The "trig_sod" is the UT seconds-of-day (SOD) of the AMON event.
The units are in centi-seconds (ie the fl.pt. seconds were multiplied by 100
and then integerized). The trigger time can be off by TBD msec.

The "ra" and "dec" fields specify the RA,Dec location of the multiply detected
event (in J2000 epoch).  These intrinsically floating point quantities
have been multiplied by 10,000 and then integerized to make
an integer quantity with units of 10^-4-degrees.

The "n_events" field contains the number of events (ie neutrinos)
that occurred during the deltaT window.  For ICECUBE_EHE type,
this number will always be 1 (ie one PeV neutrino).   (Alway one, but this field
is still included in the EHE type to be symetric with the HESE and COINC types.)

The "event_error50" is the radius of the circle that will contain on average
50% of the sources.  The units are 0.0001-degrees (ie the fl.pt. error was multiplied
by 10000 and then integerized).
It contains the statistical error and the systematic error(TBD).

The "deltaT" field contains the Time window of the burst.
The units are centi-sec (ie the deltaT was multiplied by 100 and then integerized).

The "sigmaT" field contains the Uncertainty of the time window.
The quantity stored in slot 14 is the log10() of the Sigma_T value from AMON
multiplied by 10,000 and then intergerized.  To retrieve the time_interval value at the
receiving end, call sigma_t=pow(10.0,slot[14]/10000.0).

The "energy" field contains the lower bound energy estimate of the neutrino (in units of TeV).
The quantity stored in slot 15 is the log10() of the Energy value (in TeV) 
multiplied by 10,000 and then intergerized.  To retrieve the time_interval value at the
receiving end, call energy=pow(10.0,slot[14]/10000.0).

The "stream" field contains the index number identifying
which analysis stream produced this trigger.  It ranges from 1 - NN.
The list of streams are: 0=undefined, 1=ICECUBE_HESE, 2=abc, 3=def, ...

The "rev" field contains the number of times this notice was revised (ie updated).
rev=0 means the first notice, 1 is the 1st revision, 2 is 2nd revision, etc.
For AMON_EHE type notices, the changes are most likely due to the RA,Dec,Error
values changing.

The "trig_id" field contains (a) flag bits on the type of solution
that was found by the AMON processing and
(b) flag bits that are assigned in the GCN ground-processing.
The bit packing of the "trig_id" entry in the packet is:
BitNum  F-v-G  Item_name      Description
------  ----   ---------      -----------
0        -     spare          spare for future use
1        G     role=test      Received with role="test"
2-4      -     spare          spare for future use
5        G     retraction     0=No or 1=Yes, it is definitely not real, a Retraction
6-27     -     spare          spare for future use
28       G     spatial_coinc  0=No or 1=Yes, there was a spatial coincidence with another event
29       G     temporal_coinc 0=No or 1=Yes, there was a temporal coincidence with another event
30-31    -     spare          spare for future use
The columns are:  "BitNum" is the 2^N location within the 4-byte word;
"Item_name" is the short-description name of the quantity;  and
"Description" is the longer description of the quantity.

The "misc" field contains various bit_flags.
The bit packing is:
Bit_Number     Item_name      Description
----------     ---------      -----------
0-7            spare          Spare bits within 'misc'.
8-9            cite_type      Citation type (0=undef,1=followup,2=supersedes,3=retraction)
10-23          spare          Spare bits within 'misc'.
24-31          sernum         8 bits containing the type of the EHE serial number(???)

The "charge" field contains the total neutrino event charge (in units of photo-electrons).
The quantity stored in slot 21 is the log10() of the charge value (photo-electronics)
multiplied by 10,000 and then intergerized.  To retrieve the Charge value at the
receiving end, call charge=pow(10.0,slot[21]/10000.0).
[Note that this is not the same packing method as 'charge' in the AMON_ICECUBE_HESE packet.]

The "signalness" field contains the probability that the neutrino event was signal and track-like.
The range is 0.00001 to 1.0.
The quantity stored in slot 22 is the log10() of the Signalness value from AMON
multiplied by 10,000 and then intergerized.  To retrieve the Prob_value value at the
receiving end, call signalness=pow(10.0,slot[22]/10000.0).

The 'run_num' is the 2nd of two parts ('event_num' is first part, see above)
that uniquely identifies this event.

The "sub_sec" field contains the subseconds portion of the trigger_sod.
It contains the fractional-seconds portion to 6 significant digits;
0.123456*1e6 and then integerized.

There are a total of TBD "spare" items in the packet (usually filled
with zeros).  They are located at items TBD-TBD.

The "pkt_term" is the packet termination character.  It is an ASCII
newline (decimal value 10) in the last byte of the last 4-byte long
in the packet.  This highly specialized termination character was required
by the firewall machine at LLNL in the very early days of operation.
It has since become a permanent fixture as multiple sites are now operating,
and it is not wise to play with format/content changes.  And it still
does function as a fixed-content termination value for validity checking.




================================================================================

COORDINATES EPOCH and QUANTIZATION COMPARISON:

The table below contains the epochs used for the various RA,Dec locations
given within the various packets within the entire GCN socket_packet process.
This table gives a rapid comparison, and allows the "pattern" of epoch-usage
to be seen (granted, there are two exceptions to this "pattern").  Current epoch
was used in the early days of operations when there was only the
BATSE-based Position Notices, and then when other sources came into existance
the Current epoch pattern was maintained for the "center locations".
But since some of these other sources also had other RA,Dec coordinate
information (notably the center of an IPN annulus), which were historically
reported in J2000 epoch, that precident was then followed.  With the addition
of the Swift Notices, the J2000 convention has been maintained.

The table also contains the quantization factors for each set of coordinates
locations in the various packets (centers and corners, etc).  The quantization
in the value in angular degrees in each integerized coordinate location
(i.e. the value in the packet is (int)(value/quant) ).


TYPE#    SOURCE_TYPE              CENTER_LOC  CORNERS_LOC    CENTER_LOC  CORNERS_LOC
NUMBER   TYPE                     EPOCH       EPOCH          QUANT       QUANT
 
 1       BATSE_ORIGINAL           Current     n/a            0.01        n/a
 2       Test                     Current     n/a            0.01        n/a
11       BATSE_MAXBC              Current     n/a            0.01        n/a
22       BATSE_FINAL              Current     n/a            0.01        n/a
24       BATSE_LOCBURST           Current     n/a            0.01        n/a
25       ALEXIS                   Current     n/a            0.01        n/a
27       XTE-PCA                  Current     n/a            0.0001      n/a
29       XTE-ASM                  Current     J2000          0.0001      0.0001
30       COMPTEL                  J2000 (1)   J2000          0.01        0.01
31       IPN_RAW                  J2000       J2000          0.0001      0.0001
32       IPN_SEGMENT              J2000 (2)   J2000 (3)      0.0001      0.0001
34       SAX-WFC                  Current     n/a            0.0001      n/a
36       SAX-NFI                  Current     n/a            0.0001      n/a
37       ASM_XTRANS               Current     J2000          0.0001      0.0001
38       testing                  Current     n/a            0.0001      n/a
39       IPN_POSITION             Current     J2000          0.0001      0.0001
41       HETE_S/C_UPDATE          Current     J2000          0.0001      0.0001
42       HETE_S/C_LAST            Current     J2000          0.0001      0.0001
43       HETE_GNDANA              Current     J2000          0.0001      0.0001
44       HETE_Test                Current     J2000          0.0001      0.0001
45       GRB_COUNTERPART          J2000 (4)   n/a            0.0001      n/a
46       SWIFT_TOO_FOM            J2000 (5)   n/a            0.0001      n/a
66       SWIFT_TOO_SC_SLEW        J2000 (5)   n/a            0.0001      n/a
51       INTEGRAL_POINTDIR        J2000 (8)   n/a            0.0001 (8)  n/a
53       INTEGRAL_WAKEUP          Current     n/a            0.0001      n/a
54       INTEGRAL_REFINED         Current     n/a            0.0001      n/a
55       INTEGRAL_OFFLINE         Current     n/a            0.0001      n/a
56       INTEGRAL_WEAK            Current     n/a            0.0001      n/a
57       AAVSO                    J2000 (4)   n/a            0.0001      n/a
58       MILAGRO                  J2000 (4)   n/a            0.0001      n/a
61       SWIFT_BAT_GRB_POS        J2000 (5)   n/a            0.0001      n/a
62       SWIFT_BAT_GRB_POS_NACK   n/a         n/a            n/a         n/a
63       SWIFT_BAT_GRB_LC         J2000 (5)   n/a            0.0001      n/a
64       SWIFT_BAT_SCALEDMAP      J2000 (5)   n/a            0.0001      n/a
65       SWIFT_FOM_OBS            J2000 (5)   n/a            0.0001      n/a
66       SWIFT_SC_SLEW            J2000 (5)   n/a            0.0001      n/a
67       SWIFT_XRT_POS            J2000 (5)   n/a            0.0001      n/a
68       SWIFT_XRT_SPEC           J2000 (5)   n/a            0.0001      n/a
69       SWIFT_XRT_IMAGE          J2000 (5)   n/a            0.0001      n/a
70       SWIFT_XRT_LC             J2000 (5)   n/a            0.0001      n/a
71       SWIFT_XRT_NACK_POS       J2000 (5)   n/a            0.0001      n/a
72       SWIFT_UVOT_IMAGE         J2000 (5)   n/a            0.0001      n/a
73       SWIFT_UVOT_SRCLIST       J2000 (5)   n/a            0.0001      n/a
76       SWIFT_BAT_GRB_PROC_LC    J2000 (5)   n/a            0.0001      n/a
77       SWIFT_XRT_PROC_SPEC      J2000 (5)   n/a            0.0001      n/a
78       SWIFT_XRT_PROC_IMAGE     J2000 (5)   n/a            0.0001      n/a
79       SWIFT_UVOT_PROC_IMAGE    J2000 (5)   n/a            0.0001      n/a
80       SWIFT_UVOT_PROC_SLIST    J2000 (5)   n/a            0.0001      n/a
81       SWIFT_UVOT_POS           J2000 (5)   n/a            0.0001      n/a
82       SWIFT_BAT_GRB_POS_TEST   J2000 (5)   n/a            0.0001      n/a
83       SWIFT_POINTDIR           J2000 (5)   n/a            0.0001      n/a
84       SWIFT_BAT_TRANS          J2000 (5)   n/a            0.0001      n/a
89       SWIFT_UVOT_POS_NACK      J2000 (5)   n/a            0.0001      n/a
98       SWIFT_BAT_QL_POS         J2000 (5)   n/a            0.0001      n/a
98       SWIFT_BAT_SUBTHRESH_POS  J2000 (5)   n/a            0.0001      n/a
99       SWIFT_BAT_SLEW_GRB_POS   J2000 (5)   n/a            0.0001      n/a
100      AGILE_GRB_POS_WAKEUP     J2000 (5)   n/a            0.0001      n/a
101      AGILE_GRB_POS_GROUND     J2000 (5)   n/a            0.0001      n/a
102      AGILE_GRB_POS_REFINED    J2000 (5)   n/a            0.0001      n/a
103      SWIFT_ACTUAL_POINTDIR    J2000 (5)   n/a            0.0001      n/a
107      AGILE_POINTDIR           J2000 (6)   n/a            0.0001 (9)  n/a
109      AGILE_GRB_POS_TEST       J2000 (5)   n/a            0.0001      n/a
110      FERMI_GBM_ALERT          J2000 (6)   n/a            0.0001      n/a
111      FERMI_GBM_FLT_POS        J2000 (6)   n/a            0.0001      n/a
112      FERMI_GBM_GND_POS        J2000 (6)   n/a            0.0001      n/a
113      FERMI_GBM_LC             J2000 (6)   n/a            0.0001      n/a
114      FERMI_GBM_INTERNAL       J2000 (6)   n/a            0.0001      n/a
115      FERMI_GBM_FIN_POS        J2000 (6)   n/a            0.0001      n/a
116      FERMI_GBM_ALERT_INTERNAL J2000 (6)   n/a            0.0001      n/a
117      FERMI_GBM_FLT_INTERNAL   J2000 (6)   n/a            0.0001      n/a
118      FERMI_GBM_TRANS          J2000 (6)   n/a            0.0001      n/a
119      FERMI_GBM_POS_TEST       J2000 (6)   n/a            0.0001      n/a
120      FERMI_LAT_GRB_POS_INI    J2000 (6)   n/a            0.0001      n/a
121      FERMI_LAT_GRB_POS_UPD    J2000 (6)   n/a            0.0001      n/a
122      FERMI_LAT_GRB_POS_DIAG   J2000 (6)   n/a            0.0001      n/a
123      FERMI_LAT_TRANS          J2000 (6)   n/a            0.0001      n/a
124      FERMI_LAT_GRB_POS_TEST   J2000 (6)   n/a            0.0001      n/a
125      FERMI_LAT_MONITOR        J2000 (6)   n/a            0.0001      n/a
126      FERMI_SC_SLEW            J2000 (6)   n/a            0.0001      n/a
127      FERMI_LAT_GND            J2000 (6)   n/a            0.0001      n/a
128      FERMI_LAT_OFFLINE        J2000 (6)   n/a            0.0001      n/a
129      FERMI_POINTDIR           J2000 (6)   n/a            0.0001 (7)  n/a
130      SIMBAD/NED_RESULTS       J2000       n/a            0.0001      n/a
131      FERMI_GBM_SUBTHRESHOLD   J2000 (6)   n/a            0.0001      n/a
133      SWIFT_BAT_MONITOR        J2000 (5)   n/a            0.0001      n/a
134      MAXI_UNKNOWN_SRC         J2000 (6)   n/a            0.0001      n/a
135      MAXI_KNOWN_SRC           J2000 (6)   n/a            0.0001      n/a
136      MAXI_TEST                J2000 (6)   n/a            0.0001      n/a
137      OGLE                     J2000 (6)   n/a            0.0001      n/a
139      MOA                      J2000 (6)   n/a            0.0001      n/a
140      SWIFT_BAT_SUB_SUB_THRESH J2000 (5)   n/a            0.0001      n/a
141      SWIFT_BAT_KNOWN_SRC_POS  J2000 (5)   n/a            0.0001      n/a
144      FERMI_SC_SLEW_INTERNAL   J2000 (6)   n/a            0.0001      n/a
145      COINCIDENCE              J2000 (6)   n/a            0.0001      n/a
146      FERMI_GBM_FIN_POS_INT    J2000 (6)   n/a            0.0001      n/a
149      SNEWS                    J2000       n/a            0.0001      n/a
150      LVC_PRELIMINARY          n/a         n/a            n/a         n/a
151      LVC_INITIAL              J2000       n/a            n/a         n/a
152      LVC_UPDATE               J2000       n/a            n/a         n/a
153      LVC_TEST                 J2000       n/a            n/a         n/a
154      LVC_COUNTERPART          J2000       n/a            0.0001      n/a
157      AMON_ICECUBE_COINC       J2000       n/a            0.0001      n/a
158      AMON_ICECUBE_HESE        J2000       n/a            0.0001      n/a
159      AMON_ICECUBE_TEST        J2000       n/a            0.0001      n/a
160      CALET_GBM_FLT_LC         J2000 (10)  n/a            0.0001      n/a Normal slots used
161      CALET_GBM_GND_LC         J2000 (10)  n/a            0.0001      n/a
166      AMON_ICECUBE_CLUSTER     J2000       n/a            0.0001      n/a
168      GWHEN                    J2000       n/a            0.0001      n/a
169      AMON_ICECUBE_EHE         J2000       n/a            0.0001      n/a


Notes:
(1) J2000 is used for the "center" location which is in contradiction
to the rest of the packets in the system.  This deviation from standard practice
was an oversight on my part.
(2) J2000 is used for both the "center of the IPN annulus" location
and for the "maximum probability" location.  This "center of the annulus"
is not really the same kind of entity as the "center of burst" locations
in the rest of the GCN packets.  However, the "max prob" is essentially
the same kind of entity as the "center of burst" locations
in the rest of the GCN packets.  This deviation from standard practice
was an oversight on my part.
(3) J2000 is used for the "1,2,3-sigma" locations along the IPN segment
(which is the closest matchup to the "corners of a box" locations
of the other Notices types).
(4) J2000 is used for all the GRB_COUNTERPART, AAVSO and MILAGRO RA,Dec positions
because that is the new standard epoch being used.  The switch from Current
to J2000 is being motivated by the Swift mission.
(5) J2000 is used for all the Swift and AGILE RA,Dec positions/pointings/etc
because that is the native epoch used by the mission, and if I converted it
back to Current epoch to be consistant with all the prior GCN packets contents
then the conversion would introduce errors comparable to the uncertainties
in the NFI-based positions (especially the UVOT positions which are 0.3arcsec).
(6) J2000 is used for all the AGILE, FERMI, MAXI, etc RA,Dec positions/pointings/etc,
because that is the new standard epoch being used within GCN, and it is the
native epoch used in those missions/instruments.
(7) The Initial/Starting RA,Dec Pointing Direction is quantized at 0.0001 deg,
but the 29 RA,Dec values into the future are quantized at 0.1 deg.
(8) The RA,Dec values for this s/c pointing direction are stored in slots 14 and 15;
not in the standard "center" locations of 7 and 8.  The values are also J2000 epoch
which is non-standard for the source-position-containing INTEGRAL notice types
(which was Current epoch of that beginning era of GCN).
(9) The RA,Dec values for the current AGILE s/c pointing direction.
(10) The RA,Dec values for the current CALET s/c pointing direction.


Return to GCN/TAN homepage.
This file was last modified on 22-Jul-17.