ChangeLog for lccorr =============================== Open SPRs: ---------- 3403, 3404, 3406 Version 2.22.2 - 2005-08-12 (EO) -------------- + No code changes. Modified DISTRIBUTION, replacing "world" by "ssc" to avoid public distribution of this package in inmediate release of SAS 6.5.0. Version 2.22.1 - 2005-04-06 (IMS/LUX) ---------------------- - Upgraded DEPEND of attcalc2 to 1.0 to fix build problem. Version 2.22 - 2005-02-01 (IMS/LUX) ---------------------- - Fixed SPR 3435 (a typo in subroutine lccorr_mod.f90:writeWeights). Version 2.21 - 2004-12-01 (IMS/LUX) ---------------------- - Source and background CCD weight values are now written as keywords to the RATE extension. - eexpchipmap now deals with the situation in which readGtisFromDss returns an empty vector. - Call to exposure_aux:readGtis changed to readGtisFromDss in accordance with the change in name in ssclib-3.21. ssclib DEPENDency has been upgraded accordingly. - Ditto re lccorr_corrector. - Test harness for eexpchipmap temporarily disabled - the task is seg faulting on linux. Version 2.20 - 2004-11-16 (IMS/LUX) ---------------------- - lccorr and eexpchipmap now call the same version of exposure_map:getEnergyBounds, which has been rewritten as a function. - In evigmap, changed a print of image number to 'call message'. Version 2.19 - 2004-11-04 (IMS/LUX) ---------------------- - SPR 3405 on lccorr fixed: TIMEZERO now written to gti extensions, regardless of whether present in RATE extension or not. Version 2.18 - 2004-10-08 (IMS/LUX) ---------------------- - Removed a lot of obsolete code files. - Task now copies MJDREF and TIMEZERO keywords from the RATE table to the src and bkg gti tables (closes SPR 3400). Version 2.17 - 2004-09-20 (IMS/LUX) ---------------------- - Modified exposure_aux.f90 so as to fix SPR 3387. - Deleted some obsolete routines from exposure_map.f90. - Fixed SPR 3384 by enforcing normalization of the ccd weights in lccorr under all circumstances. - Fixed SPR 3383 - lccorr now renames the temp outset rather than cloning it. - SPR 3347: lccorr was subtracting centre times of adjacent frames to estimate the frame length on a frame-per-frame basis. However, in PN, frames may be missing, which buggered up the calculation. Task now uses modeParams%frameTime from CAL_getModeParameters. - Messages altered slightly to provide better indication of the start of lengthy stages. - Fixed an error in lccorr in which CAL_setState was being called with ccdChipId=0 when the CHANTYPE keyword ='PHA'. - Fixed SPR 3353 on lccorr: the task was counting only those frames whose centre times fell within a bin, where it should have been counting all frames which overlapped the bin. - Fixed SPR 3386 on lccorr. The task had assumed that all bins would be full of frames (yes I know this was stupid): bins which overlapped the start and end of the frame sequence were not being handled correctly. - Added new parameter --treatalias. When 'no', time series made from event lists in which the events were not dithered within the frames are left with the resulting aliasing uncorrected. - Resolved SPR 3385 by replacing --srcweights by three parameters: --srcweightsnode0, --withsrcnode1 and --srcweightsnode1. Same also for bkg weights. - Ratio of bkg to src collection area is now printed (at 'verbose' verbosity). - Added new task to the package: 'etimedither'. Version 2.16 - 2004-09-16 (IMS/LUX) ---------------------- - Fixed error in evigmap where theta was being sampled twice, in different ways. Version 2.15 - 2004-09-15 (IMS/LUX) ---------------------- - exposure_map.f90 can now deal with older-format PN offset extensions. - Included .par, .lyt and .info for evigmap. Version 2.14 - 2004-09-03 (IMS/LUX) ---------------------- - ssclib module exposure.f90 copied to lccorr as exposure_aux.f90 and 'use' statement lines in all other lccorr modules modified accordingly. Since lccorr and eexpchipmap seem to be the only tasks which use this module it seems to make more sense to have it in the lccorr package. Later I'll delete it from ssclib. - Fixed a bug in eexpchipmap in which chip RAWX/Y arrays were not being initialized. - Fixed a bug in exposure_aux:interpolateVigMap in which the same value was being given to every element of the output. - Brought into line with ssclib-3.16, attcal2-0.3 and eimchip2sky-1.6 and DEPEND updated accordingly. - Added new task evigmap. - Added parameters --withvigexpmapset and --withvigexpcubeset to eexpchipmap. Version 2.13 - 2004-08-17 (IMS/LUX) ---------------------- - eexpchipmap was dumping a test image to disk. This has now been stopped. - eexpchipmap now writes the cube with a 'key' table and without any blank image planes (dependency on eimchip2sky also increased to 1.5). - Output of vignetted exposure map now implemented in eexpchipmap. Version 2.12 - 2004-08-10 (IMS/LUX) ---------------------- - In DEPEND, cal version reduced to 3.172, so the task will build in reltrack. Version 2.11 - 2004-08-02 (IMS/LUX) ---------------------- - Task eexpchipmap added (per SCR 168). Libraries binnedatt and chipcube have been added to the Makefile:USEDLIBS line, and thus packages attcalc2 and eimchip2sky have been added to DEPEND. - testprods version in DEPEND upgraded to 1.3. Version 2.10 - 2004-07-26 (IMS/LUX) ---------------------- - Non-devtrack library names inadvertently left in USEDLIBS. These have now been removed. - Output TS only had a HDUCLAS2 keyword if a background TS was also supplied. This is now rectified (input source kwd is retained if no bkg). - Added 'epic' and 'timing' to config/lccorr.info. - Test harness has been temporarily weakened (it now only checks that the task has been built) until the compiler issues which seem to be causing seg faulting have been sorted out. - Documentation has been brought up to date. Version 2.9 - 2004-06-29 (IMS/LUX) ---------------------- - Overloaded exposure_map:readCcdExposureInfo so it will take either the name of the event list or the dataset pointer as arguments. - Fixed SPR 3348 by making the calculation of average TIMEDEL in a sensible fashion which retains all the precision available. - Fixed SPR 3349 by only asking the cal for ccdNodeId if instrument != EPN. - Fixed SPR 3350 by changing the constraints on --srcweights and --bkgweights from >0 to >=0. - Now prints a list of calculated background weights when --bkgweightstyle=flat. - Fixed SPR 3352 by adding new parameters --bkgarea and --srcarea for when, respectively, --bkgweightstyle and --srcweightstyle = 'user'; also these areas are now set to 1 if 'none' is selected for these styles. - The total-source-region and total-bkg-region gtis which were introduced in version 2.7 have been modified a bit. Tables SRC_GTIS and BKG_GTIS now contain gtis which are set to exclude only and all null values of, respectively, the source and background count rate columns. Version 2.8 - 2004-06-15 (IMS/LUX) ---------------------- - Changed the name of parameter --weights to --srcweights; changed --weightstyle to srcweightstyle; introduced analogous parameters --bkgweightstyle and --bkgweights. - Fixed SPR 3340. Version 2.7 - 2004-06-10 (IMS/LUX) ---------------------- - Now again calling emsaplib/badpixutils_mod:readBadOffsets since it was fixed. The fixed version however only has a minor version no. increment so I can't ensure the fixed version is used via dependencies. - Some rearrangement of code to (a) reduce the number of global variables, (b) to allow other tasks access to some of the algorithms, and (c) to speed things up a bit. - GTIs for all the CCDs which intersect the extraction region are now ORed and written to extensions of the output (SCR 162). - Now supports CHANTYPE='PHA'. - Removed prints of test output which were inadvertently left in v-2.6. - Fixed SPR 3306 (ssclib-3.10/exposure:constructQeArray now takes rawX/rawY arguments). - 'use polygon' changed to 'use regridders'. - Pointer piRanges in exposure_map.f90:calcPnCcdExpMap now deallocated (fixes SPR 3314). - Added parameters --weightstyle and --weights. This should fix SPR 3111 (see also SOC SCR 162). - Removed redundant calls to checkRangeOverlap in exposure_map:getEnergyBounds. - Included oal, selectlib in DEPEND and brought all versions up to date. - lccorr_mod:getCcdNumFromDssComp, lccorr_mod:attemptToGetSrcPosFromDss and lccorr_mod:getSrcPositions slightly modified to lessen their use of global variables. - Removed redundant 'use array_utils' from lccorr_mod header. - Removed energy-dependent factors from BKGRATIO calculation (fixes SPR 3310). Version 2.6 - 2004-04-19 (IMS/LUX) ---------------------- - Interfaces to ssclib brought up to 3.9 standard. - Removed dumpBounds from exposure_map.f90 (now in ssclib/dss_ranges_aux). - Fixed SPR 3302. This was an incorrectly sized array in exposure_map.f90:calcMosCcdExpMap. Version 2.5 - 2004-04-06 (IMS/LUX) ---------------------- - Added to the test harness an explicit test on the presence of the link to the input data in package testprods. Version 2.4 - 2004-04-05 (IMS/LUX) ---------------------- - Now uses test data in package testprods. (Added testprods to DEPEND.) Version 2.3 - 2004-03-25 (IMS/LUX) ---------------------- - Copy of patches to devtrack. Version 2.2.8 - 2004-03-17 (IMS/LUX) ------------------------- - Fixed SPR 3280. - Put in a temporary trap (with explanatory message) to work around SPR 3281 on ssclib. Version 2.2.7 - 2004-03-16 (IMS/LUX) ------------------------- - 'Slight change to error doco' appears to have been made to the wrong file: it doesn't seem to have appeared in the uploaded 2.2.6. Let's try again! Version 2.2.6 - 2004-03-15 (IMS/LUX) ------------------------- - Fixed SPR 3272. - Slight change to error documentation and also to the .lyt file, to be more user-friendly. - Added work-around for an error discovered in emsaplib. - Lccorr now rescales the background to the source collection area whether the background is subsequently subtracted or not. This was not really an error so much as an ugliness. Version 2.2.5 - 2004-03-04 (IMS/LUX) ------------------------- - The text for some warnings and errors had been missing; these have now been filled out. - The documentation has been brought up to date. Version 2.2.4 - 2004-01-29 (IMS/LUX) ------------------------- - Upgraded to handle PN. New module exposure_map contains stuff that used to be in ssclib/exposure. Version 2.2.3 - 2004-01-29 (IMS/LUX) ------------------------- - Changed call to instToRaw to cope with changes in ssclib-3.6.4 (see SSC SPR 3217). - ...however this introduced another bug which caused the test harness to fail (SPR 3237). This has now been fixed. It necessitated some changes to ssclib/exposure.f90. Version 2.2.2 - 2003-12-18 (IMS/LUX) ------------------------- - Changed subroutine interpolatePsfCubeToCcd so that ssclib regridding routine used instead (fixes SPR 3189). Version 2.2.1 - 2003-12-17 (IMS/LUX) ------------------------- - Added oal, dss and sasdal to the USEDLIBS line. Task won't compile statically without them. Fixes SPR 3179. Version 2.2 - 2003-11-24 (IMS/LUX) ------------------------- - Temp_ssclib modules now merged into ssclib-3.0. - (from 2.0.4): Documentation update. - (from 2.0.3): Fixed SPR 3028. - (from 2.0.3): Added some more messages and utility routines. - (from 2.0.3): Added parameters --srcposstyle and --srcposset to allow the source position to be obtained either from keywords SRC_RA, SRC_SET in the designated dataset or from the centre of the selection region as stored in the DSS (if this is possible). - (from 2.0.3): Allowed sources on CCD other than 1 to be processed. - (from 2.0.3): Slightly altered lccorr_test. - (from 2.0.3): Fixed SPR 3043. - (from 2.0.1): Now depends explicitly on sas. Version 2.1 - 2002-11-07 (IMS/LUX) ------------------------- - Same as 2.0 but uses param>2.0 for devtrack upload. Version 2.0 - 2002-10-17 (IMS/LUX) ------------------------- - Major rewrite. Also now needs dsslib-4.0. Version 1.7.4 - 12 Mar 2002 (IMS/LUX) ------------------------- - Distribution changed temporarily from 'world' to 'ssc' until some of the bugs are weeded out. Version 1.7.3 - 10 Mar 2002 (IMS/LUX) ------------------------- - Closed SPR 2703. Version 1.7.2 - 08 Mar 2002 (IMS/LUX) ------------------------- - Updated DEPEND. Version 1.7.1 - 07 Mar 2002 (IMS/LUX) ------------------------- - Rearranged some expressions to avoid a rounding error that made the testset fail on some platforms. Version 1.7 - 24 Jan 2002 (IMS/LUX) ------------------------- - Updated DEPEND. Version 1.6 - 30 Oct 2001 (IMS/LUX) ------------------------- - Updated dependencies. - Changed cal coord variables to type double to conform with cal-3.121. - Fixed test harness. Version 1.5 - 14 Sep 2001 (IMS/LUX) ------------------------- - Fixed another depend error. Version 1.4.5 - 17 May 2001 (IMS/LUX) ------------------------- - Fixed depend error. Version 1.4.4 - 16 May 2001 (IMS/LUX) ------------------------- - Fixed some bugs. There are plenty more. - Updated DEPEND. Version 1.4.3 - 1 Dec 2000 (IMS/LUX) ------------------------- - Added .lyt to config/Makefile. Version 1.4.2 - 23 Nov 2000 (IMS/LUX) ------------------------- - Updated documentation. - Broken html links fixed. Version 1.4.1 - 9 Nov 2000 (IMS/LUX) ------------------------- - Fixed SPRs 2019 and 2027. Version 1.4 - 10 Oct 2000 (IMS/LUX) ------------------------- - Updated dependencies. - Reduced size of test files. Version 1.3 - 9 Oct 2000 (IMS/LUX) ------------------------- - Changed mode etc names to accord with caloalutils v 1.34. Version 1.2.1 - 5 Jul 2000 (GV/SOC) ---------------------- - add taskmain to DEPEND. - SPR deprecating use of caps in taskname macro fixed. Version 1.2 - 27 May 2000 (GV/SOC) ---------------------- - updated DEPEND Version 1.1 - 26 May 2000 (IMS/LUX) ---------------------- - Now using /usr/local/bin/perl in lccorr_test Version 1.0 - 15 May 2000 (IMS/LUX) ---------------------- - Initial version for SAS v5.