path: root/noao/digiphot/daophot/Revisions

.help revisions Jan90 noao.digiphot.daophot
.nf

daophot/psf/dpfitpsf.x
    The ":function <func>" command was not saving the new functon when
    refitting with the 'f' keystroke.  In dp_fitpsf() the parameters are
    reinitialized and the startup function was being reused. (4/13/10, MJF)

daophot/psf/dpfitpsf.x
    An amovi() call referenced a Memr[] array that was declared as
    int, changed to Memi[]  (8/28/09, MJF)

daophot/*.par
    Minor changes for readability suggested by Jason Quinn, i.e. made
    consistent use of space before question marks (12/23/08, MJF)

daophot/lib/warning.dat
    Added some blank lines for readability before the package menu 
    (12/11/08, MJF)

=======
V2.14.1
=======

daophot/psf/dppwselmer.x
    PSTSELECT was showing the IMAGE param as 'name' instead of 'imagename'
    (7/8/08)

daophot/daophot.hd
    The source directories for pstselect and allstar were wrong.
    (5/27/08, Valdes)

=======
V2.14
=======

daophot/allstar/dpcache.x
     Fixed some procedure calls being closed with a ']' instead of a ')'
     (2/17/08, MJF)

daophot/psf.par
    Fixed a type in the 'pstfile' prompt (11/24/07)

========
V2.12.3
========

daophot/allstar/dpalwrite.x
    In dl_alxaccum there are conditions where the loop skips a star which
    are not checked in the write routines when they compute the sharp
    parameter.  When the loop is not completed the denom value is not
    computed and defaults to zero which caused a divide by zero error
    in computing the sharpness.  (11/18/04, Valdes)

daophot/daolib/invers2.x
daophot/daolib/mkpkg
daophot/allstar/t_allstar.x
daophot/allstar/dpalphot.x
daophot/allstar/dpastar.x
daophot/allstar.par
    Made a small change to the matrix inversion code to avoid problems with
    very small numbers.  This was done by translating the original fortran
    version to SPP and then replacing small numbers by zero.  In order to
    allow users to reproduces earlier results or if there is a problem
    a version parameter was added to ALLSTAR.  Setting the version to 1
    will use the old version of the inversion routine and the default of
    2 will use the new version.  (6/18/04, Valdes)

=======
V2.12.2
=======

daophot/daoedit/dpeconfirm.x
daophot/daoedit/dperprofile.x
    Added some missing sfree statements to the daophot package routines.

    (19/02/02, Davis)

daophot/daolib/dpppcache.x
    Added a call setting IM_BUFFRAC to 0 to the memory caching code in the
    daophot package tasks in order to force the imio buffer to be the size of
    the input image.

daophot/substar/dprestars.x
    There were 3 missing arguments in the dp_tptread call which cause
    trouble for people trying to read an input exclude file in tables
    format.

    (19/09/01, Davis)

daophot/psf/dppwselmer.x
    Changed an incorrect smark call to salloc.  This bug was introduced by
    recent output file mods. Should not have been a problem in released code.

    (19/09/01, Davis)

daophot/psf/t_psf.x
    Changed the type declaration for the clgwrd function from bool to int.
    This bug was introduced by the wcs mods. Should not have been a problem
    in realeased code.

    (19/09/01, Davis)

daophot/addstar/t_addstar.x
    Modifed the addstar task to add a ".art" suffix to the output artificial
    star coordinate files when the user supplies the output root name. This
    avoids file and image name conflicts when the user includes the image
    suffix, e.g. ".imh" in the output root name.

    Davis, September 17, 2001

daophot/
    Modified all the apphot tasks to accept input coordinates in logical, tv,
    physical, or world coordinates and to write the output coordinates in
    logical, tv, or physical coordinates. One consequence of this is that
    the apphot tasks will now work correctly off image sections in interactive
    mode. Another is that users can work on image sections while preserving
    the coordinate system of the entire image.

    Davis, June 8, 2000

daophot/
    Modified all the daophot tasks to strip the directory information from
    the input image and coordinate file names written to the output files,
    to the terminal, and to the plot headers. The colon commands will still
    read and write full image and coordinate file path names. This change
    improves the likelyhood  that the full root image name will be written
    to the output. This root image name is used by the photometric calibration
    code to construct images sets.

    Davis, June 8, 2000

daophot/mkpkg
    Modified the make procedure to pick up user compile and link flags.

    Davis, June 17, 1999

daophot/daolib/dpotime.x
    Modified the code which reads the time of observation value from the image
    header to extract the time field from the date-obs keyword if it is
    present.

    Davis, May 11, 1999

daophot/daolib/dpdate.x
    Modified the date and time photometry file keyword encoding tasks to
    write the date and time in the proper fits format and in units of GMT.
    Note that GMT is deactivated in digiphotx because the necessary routine
    is not yet in the system.

    Davis, May 11, 1999


daophot/psf/dppwrtgrp.x
    The row number in the psf star group file writing code was not being
    correctly initialized results in an "invalid row number error" if
    the psf task was run with daophot.text = no.

    Davis May 10, 1999

daophot/psf/dpgetapert.x
daophot/allstar/dpalwrite.x
daophot/group/dpwrtgroup.x
daophot/doc/daopars.hlp
    Removed the restriction on the number of stars imposed by the maxnstars
    parameter and in the process fixed a segmentation violation that occurred
    when the number for stars in the files was > maxnstars. Maxnstars is
    now used only for setting the initial buffer size.

    Removed the code which writes MAXNSTARS to the output photometry files.

    Davis May 8, 1999

daophot/psf/t_pstselect.x
    Added a missing imtclose statement.

    Davis, May 4, 1999

daophot/psf/dpfitpsf.x
    Changed the test for the radial weighting function from wt >= 1.0 to
    wt >= 0.999998 to avoid floating point errors (division by a very small
    number) under Linux and possibly other systems as well.

    Davis, April 19, 1999

daophot/peak/dppkfit.x
daophot/nstar/dpnstarfit.x
daophot/allstar/dpalphot.x
    Added a check to make sure that the predicted error computation can
    never be <= 0.0 This can happen in rare circumstances if the model
    prediction is exactly 0, the rdnoise is 0.0 and flaterr and proferr are
    0.0.

    Davis, April 12, 1999

daophot/daolib/dpnames.x
    Modified the automatic input and output image and file naming code
    to behave more gracefully in the case where the input images have
    kernel sections and cluster indices and sizes.

    Davis, January 17, 1999

daophot/daolib/dpdate.x
daophot/addstar/dpnaddstar.x
daophot/addstar/dpalwrite.x
daophot/group/dpwrtgroup.x
daophot/nstar/dpntwrite.x
daophot/peak/dppkwrite.x
daophot/psf/dppwrtgrp.x
    Modified the daophot output file writing routines to write the DATE
    header keyword in the new Y2k format even though the output files
    are not FITS files.

    Davis, Dec 29, 1998

daophot/allstar/t_allstar.x
daophot/allstar/dpcache.x
daophot/allstar/dpalinit.x
    Modified the allstar task to ensure that all the output and scratch
    images and scratch space are 2D even if the input image is greater than
    2D to avoid a non-obvious memory error. 

    Davis, Aug 3, 1998


daophot/daophot.cl
    Modified the package cl script to check whether the tables package is
    already loaded before checking whether it is defined and then loading it.

    Davis, Aug 1, 1998

daophot/psf/t_psf.x
    The psf task psf image list reading code was using the fntgfnb routine
    instead of the imtgetim routine to read image names from the psf image list.
    Although the code functioned correctly, if the input psf image list was
    an @ file, a control character (^P) was being written into the PSFIMAGE
    keyword value in the output *.psg.* and *.pst.* files. This caused an error
    in the nstar, allstar, etc task photometry file reading code. 

    Davis, Apr 1, 1998

daophot/addstar/t_addstar.x
daophot/allstar/t_allstar.x
daophot/group/t_group.x
daophot/nstar/t_nstar.x
daophot/nstar/t_peak.x
daophot/psf/t_pstselect.x
daophot/psf/t_psf.x
daophot/substar/t_substar.x
    Modified the addstar, allstar, group, nstar, peak, pstselect, psf,
    and substar tasks to be able to read and write default images and files 
    from and to directories other than the current directory just as the
    daophot tasks daofind and phot do.

    Davis, Feb 14, 1998

daophot/seepsf/t_seepsf.x
daophot/seepsf/dpmkimage.x
    Modified the seepsf task so it explicity sets the pixel type of the
    output psf image to real to avoid a problem with stf image kernel
    and dataless stf images.

    Davis, Jan 20, 1998

daophot/allstar/dpalinit.x
    Fixed an uninitialized memory problem that was occurring on the Dec
    ALPHA if cache=no and readnoise=0.0. This part of memory was never
    used by the code, but was being written to a scratch image of type real,
    which could result in an FPE if the affected memory was not a valid
    FP number. A similar problem was fixed for the cache=yes situtation
    awhile back.

    Davis, Nov 12, 1997


daophot/psf/dpfitpsf.x
    Fixed a bug in the psf task that was causing the fits image kernel to
    crash when opening a new image, but apparently did not affect the oif
    or stf kernels in any way. 

    Davis, June 23, 1997

daophot/daotest.cl
    Modified the rfits calling sequence so that the code will work
    correctly with the new version of rfits.

    Davis, May 29, 1997

daophot$psf/mkpkg
daophot$psf/dprstars.x
daophot$psf/dpspstars.x
daophot$psf/dpispstars.x
    The pstselect and psf tasks were not reinitializing the psf star list
    correctly when more than one image was being processed,  resulting in
    psf stars on successive images which have the same ids as psf stars
    in previous images being incorrectly rejected or include in the psf star
    list.

    Davis, Feb 21, 1997

daophot$psf/dpmkpsf.x
    The id number and magnitude arguments were reversed in the 'f' keystroke
    command call to the routine dp_addstar. This should only cause trouble
    if the psf star list need to be reread.

    Davis, July 18, 1996

daophot$doc/phot.hlp
    Fixed a type in the ERRORS section of the phot help page.

    Davis, April 11, 1996

daophot$daolib/dpinit.x
daophot$psf/dpfitpsf.x
daophot$psf/dppsfutil.x
    The psf task was not restoring the fit for the moffat25 and moffat15
    functions correctly, in the event that they were chosen as the best
    fitting analytic function. This was occurring because the constant
    beta parameter was not being copied into and out of the save array
    resulting in a totally incorrect look-up table being computed.

    Davis, March 20, 1996

daophot$psf/dpfitpsf.x
    Modified the code which computes the analytic component of the psf
    model to start from the same initial state in auto mode as it does
    if a single function is specified. Peviously the initial state
    in the auto case was the state computed by the previous function. 
    Because the convergence criteria have a specified tolerance, the
    resulting computed values of the analytic component functions and the
    look-up table elements can be slightly different in the two cases.

    Davis, March 18, 1996

daophot$allstar/dpcache.x
daophot$allstar/dpalmemstar.x

    Modified the malloc calls to calloc calls in the dpcache routine to
    avoid an unititialized memory FPE error than can occur if the readout
    noise is exactly 0. As far as I am aware this has only caused a problem
    on the Dec Alpha, but 0 readout noise situation does occur in the
    daophot test script. Removed an extra fixmem call from the allstar
    cleanup routine.

    Davis, February 20, 1996

daophot$daolib/dpgetapert.x
daophot$nstar/dpggroup.x
    Fixed a bug in the code which parses the input photometry or group files.
    If the numerical field to be extracted is adjacent to another numerical
    field and there is no white space between them, then the number
    extracted may be a combination of the two fields, rather than the desired
    individual field. This is most likely to occur when the id numbers are
    large, e.g. 10003, and the image name is a number, e.g. 8649.imh.

    Davis, February 20, 1996

daophot$daopars.par
daophot$daotest.cl
daophot$lib/daophotdef.h
daophot$lib/nstardef.h
daophot$daolib/dpgppars.x
daophot$daolib/dppppars.x
daophot$daolib/dpset.x
daophot$daolib/dpstat.x
daophot$daolib/dpverify.x
daophot$group/dpgconfirm.x
daophot$group/dpmkgroup.x
daophot$group/dpwrtgroup.x
daophot$nstar/dpnstarfit.x
daophot$nstar/dpntwrite.x
daophot$allstar/dpastar.x
daophot$allstar/dpalwrite.x
daophot$doc/daopars.hlp
daophot$doc/group.hlp
    Added a new parameter mergerad to the daopars parameter set. Mergerad
    permits the user to control the merging process. If mergerad is 0
    object merging is turned off altogether. If mergerad is INDEF to default
    mergeing radius is used. Otherwise the user can set the merging radius
    to a specific value.

    Changed the name of the critoverlap parameter to critsnratio to avoid
    user confusion over the meaning to the parameter.

    Davis, February 15, 1996

daophot$nstar/dpnmemstar.x
daophot$nstar/dpnstar.x
daophot$allstar/dpalmemstar.x
daophot$allstar/dpalphot.x
    Modified the nstar and allstar tasks so that they allocate less memory for
    the fitting matrices and vectors if object recentering is turned off. This
    can make a big difference if the the maximum group size is large.

    Davis, February 14, 1996

daophot$nstar/dpnstarfit.x
    Added a singular matrix check to avoid floating point operand errors
    in the nstar task.

    Davis, January 5, 1996

daophot$nstar/dpggroup.x
    The integer code array size was not being reallocated correctly in the
    case that the size of a group is greater the the value of the maxgroup
    parameter, causing nstar to die with a memory allocation parameter,

    Davis, January 3, 1996

daophot$doc/centerpars.hlp
    Edited the centerpars help page to include a description of the new
    INDEF-valued cthreshold option.

    Davis, Sept 29, 1995

daophot$lib/psfdef.h
daophot$psf/dpfitpsf.x
daophot$psf/dpmempsf.x
daophot$psf/dppsfutil.x
daophot$psf/dprstars.x
daophot$psf/dpispstars.x
daophot$psf/dpspstars.x
daophot$psf/dpmkpsf.x
daophot$psf/dpaddstar.x
daophot$doc/psf.hlp
    Modifed the way the magnitude of the psf is set. If matchbyid is
    yes (the default) the magnitude of the first psf star in the psf
    star list if any will be used; other wise the magnitude of the first
    psf star in the input photometry file will be used as before. This
    should help minimize photometric drift problems derived from repeated
    iterations.

    Davis, Sept 21, 1995

daophot$psf/dpdelstar.x
daophot$psf/dpsubpsf.x
    Added a missing mfree to the dpsubpsf routine. This was not causing
    a problem but might in a large script.

    In dpdelstar the routine dp_psubrast was being called with an input
    output variable that was never used being set to a constant. This was
    causing a segvio on Solaris but not on SunOS.

    Davis, Aug 25, 1995

daophot$seepsf/dpmkimage.x
    Seepsf was dying with an FPE error on the Dec Alpha. This was caused
    by an array out-of-bounds error which was not detected on the Sun
    machines.

    Davis, July 18, 1995

daophot$seepsf/t_seepsf.x
daophot$seepsf/dpmkimage.x
    The fix made to cure the previous seepsf problem broke the code for
    pure analytic psf images.

    Davis, July 3, 1995

daophot$psf/dpfitpsf.x
    Changed the illegal construct "} until (redo == false)" to
    "} until (! redo)". This was causing problems on the IBM/RISC6000
    compilers but was not caught by the Sun compilers.

    Davis, November 16, 1994

daophot$seepsf/t_seepsf.x
daophot$seepsf/dpmkimage.x
    The code that was computing the default size of the output psf image
    was incorrectly using the size of psf requested by the user (default
    radius = 11.0 pixesl) instead of the actual size of the input psf image.

    Davis, October 3, 1994

daophot$daoedit/t_daoedit.x
daophot$daoedit/dpeconfirm.x
    Made some minor modifications to the daoedit task which change the
    image cursor mode and graphics cursor mode interact.

    Davis, June 21, 1994

daophot$addstar/dpnaddstar.x
    Corrected an error in the format string for the output file.

    Davis, June 20, 1994

daophot$psf.par
daophot$lib/psfdef.h
daophot$psf/t_psf.x
daophot$psf/t_dprstars.x
daophot$doc/psf.hlp
    Added a new parameter, matchbyid, to the psf task so that stars in the
    psf star list can be matched to stars in the input photometry file
    by id or by position.

    Davis, June 20, 1994

daophot$psf/dpspstars.
    In non-interactive mode the pstselect task was writing a minimum
    of two stars to the output file even if the requested number of
    psf stars was one.

    Davis, June 18, 1994

daophot$addstar/dpartstar.x
daophot$addstar/dpartstar.x
    If the nimages parameter was greater than 1 and the addimage parameter
    was set to other than the default value, the task would terminate 
    prematurely with a "Cannot close file error"  after the first output
    image was written. The problem was caused by addstar failing to append
    the appropriate sequence number to the output star list.

    Davis, June 13, 1994

daophot$allstar/dpalwrite.x
daophot$group/dpwrtgroup.x
daophot$nstar/dpntwrite.x
daophot$peak/dppkwrite.x
daophot$psf/dppwrtgrp.x
daophot$psf/dppwselmer.x
daophot$select/dpgwselect.x
    Modified all the daophot package tasks which output sky values (pstselect,
    psf, peak, group, grpselect, nstar, allstar, pfmerge) to write the
    values with format -%15.7g like the apphot package tasks do, instead
    of %-12.3f/%-14.3f, to avoid precision problems with images that
    have been "flux calibrated".

    Davis, May 27, 1994

daophot$nstar/dpntwrite.x
daophot$nstar/dpalwrite.x
    Changed the output GROUPSKY parameter name in the phot files to GRPSKY
    to avoid a name matching conflict with the GROUP column. This was causing
    problems in reading nstar output.

    Davis, May 5, 1994

daophot$daopars.par
daophot$lib/daophotdef.h
daophot$allstar/dpaconfirm.x
daophot$allstar/dpalphot.x
daophot$allstar/dpalwrite.x
daophot$daoedit/daoedit.h
daophot$daoedit/dpecolon.x
daophot$daolib/dpgppars.x
daophot$daolib/dpppars.x
daophot$daolib/dpinit.x
daophot$daolib/dpset.x
daophot$daolib/dpstat.x
daophot$daolib/dpverify.x
daophot$nstar/dpnconfirm.x
daophot$nstar/dpnstar.x
daophot$nstar/dpnstarfit.x
daophot$nstar/dpntwrite.x
daophot$doc/allstar.hlp
daophot$doc/daopars.hlp
daophot$doc/nstar.hlp
    Added a new boolean parameter, groupsky, to the daopars parameter set.
    Groupsky determines whether the sky value for each pixel used  in the fit
    is set to, the mean of ALL the individual sky values of the stars in the
    group (groupsky = yes), or,  to the mean of the individual sky values of
    only those stars for which  the pixel in question is inside the fitting
    radius.

    Davis, Dec 20, 1993

daophot II installed
    Davis, May 31, 1993

    Daophot II installed.

daophot$phot.par
    Added a leading quote to the prompt string for the daophot.phot task
    output parameter.

    Davis, Mar 24, 1993

daophot$daolib/dpppars.x
    The psfrad, fitrad, and matchrad were being written to the parameter
    set in pixel units instead of scale units when update was set to
    yes.

    Davis, Feb 16, 1993

daophot$allstar/dpalinit.x
    In crowded regions allstar would occasionally refuse to 1) fit a group
    of bright stars or 2) fail to converge to reasonable values for a group
    of bright stars by the time the number of iterations equaled maxiter,
    resulting in a group of stars with very poor subtractions.
    The problem was caused by a bug in the code which steps through the
    stellar groups subtracting off the current best fit to produce a residuals
    image.  Occasioanally stars which should have been subtracted from the
    residuals image were not being subtracted.  Since the residuals image is
    used to determine the relative errors and weights,
    which in turn control the bad data rejection algorithm, allstar sometimes
    refused to fit stars because the residuals were too big, or was unable
    to converge to a reasonable fit. The bug is data dependent but is
    more likely to be a problem if 1) the stellar detection threshold is low
    2) the fitting radius is high producing very large groups.

    Davis, Dec 23, 1992

daophot$allstar/dpalphot.x
    If 1) cache=no, or cache=yes and memory allocation failed for one of the
    three arrays scratch, weight, or data, 2) one or more the the groups has >
    maxgroup stars, 3) regrouping was performed and 4) the position of next
    non-regrouped group was just right, allstar could fail with an "attempt to
    access the scratch, weight or data pixels randomly" error. This error
    occurrs because the regrouping process could produce groups which were out
    of y-order with with succeeding groups which had not been regrouped, 
    forcing an illegal non-sequential image access. The solution was to buffer
    enough data to fit original large group.

    Davis, Sept 16, 1992

apphot$datapars.par
apphot$centerpars.par
apphot$doc/datapars.hlp
apphot$doc/centerpars.hlp
    Changed the units of the cthreshold parameter to sigma and moved it
    to the centerpars parameter set.

    Davis, July 7, 1992

daophot$daophot.par
    The verbose, verify, update, graphics, and display parameters were
    added to the package parameters and the corresponding individual
    task parameters were redirected there by default.

    Davis, June 20, 1992

daophot$daopars.par
daophot$lib/warning.dat
    Fixed errors in the definition of the psfrad and fitrad parameters in
    the daopars parameter set. These parameters are now defined in units
    of scale not pixels as before.

    Fixed some bugs in the message printed by the daophot package if the
    tables package is not present.

    Davis, May 29, 1992

daophot$daophot.men
    Changed the entries for append, convert, dump, renumber, select, and
    sort to entries for pappend, pconvert, pdump, prenumber, pselect,
    and psort.

    Davis, Feb 28, 1992

daophot$psf/dpmkpsf.x
    Added missing fset declaration to the mkpkg line for this routine.

    Davis, Nov 20, 1991

daophot$
    Renamed the append, convert, dump, renumber, select, and sort tasks to
    pappend, pconvert, pdump, prenumber, pselect, and psort.

    Davis, Nov 11, 1991

    *** Ran spplint on the daophot package.

daophot$allstar/dpastar.x
    The routines dp_gst, dp_gwt, and dp_gdc were being called as subroutines
    instead of functions inside dp_astar. This was a recent change which
    would not affect the old testphot.

daophot/psf/dpcontpsf.x
    Changed the last argument in the call to dp_map_viewport from NO to false
    to fix a type mismatch. This could affect the old testphot.

daophot/psf/dpnewpsf.x
    Removed extra status argument from the call to dp_gaussfit. This could
    have caused a problem in the old daophot.

daophot/peak/dppkfit.x
    Removed an extra argument from the mfree call. This would not cause a
    problem in the old testphot.

    Davis, Oct 3, 1991

daophot$daophot.cl
daophot$daophot.men
daophot$daophot.hd
daophot$daotest.cl
    At the autotmatic package test task daotest to the daophot package.

    Davis, Oct 3, 1991

daophot$psf/dpsurfpsf.x
    Removed non-required variables mode, xres, yres from this routine.

    Davis, Oct 1, 1991

daophot$test/fits3.fits
    Added the fits test image to the daophot test subdirectory in preparation
    for making a test script.

    Added a test cursor input file for the psf task in preparation for 
    making the test script.

    Davis, Aug 13, 1991

daophot$daolib/dpgetapert.x
daophot$nstar/dpggroup.x
    Modified the i/o routines which read aperture photometry and group
    photometry format text files in order to take account of new additions
    to the text database routines.

    Davis, Aug 13, 1991

daophot$
    1. All the DAOPHOT tasks except PSF have been modified to accept lists of
    input and output files.

    2. Moved the text parameter from DAOPARS to the DAOPHOT package parameter
    file.

    3. Modified all the DAOPHOT routines so that psfrad, fitrad and matchrad
    are defined in terms of scale.

    Davis, Aug 5, 1991

daophot$
    1. Added support for a time of observation parameter to all the appropriate
    daophot tasks.

    2. Changed all the daophot file header parameters to be 23 characters long
    instead of 15 characters.

    3. Modified all the daophot tasks to strip whitespace from the filter id
    keywords and the iraf version environment variable string.

    4. Wrote an spp version of the error function routine which was originally
    in fortran.

    Davis, Aug 2, 1991

daophot$allstar/
    1. Redid the i/o of the ALLSTAR task to make the cache=no option run in
    a finite period of time for large images.

    Davis, Jun 24, 1991

daophot$group/
    1. Changed GROUP so that the groups are output in y order instead of in
    order of the size of the group. This will help make i/o more efficient
    in NSTAR.

    Davis, Jun 18, 1991

daophot$allstar/
    1. Changed the boolean arrays inside ALLSTAR to integer arrays. These
    use the same amount of space and are probably safer.

    2. Added protection in the code for the case that the x and y position
    of a star is INDEF or the sky value is INDEF.

    3. Changed the format of the output of the verbose option in several places
    in the code.

    Davis, Jun 17, 1991

daophot$nstar/
    1. Changed the boolean arrays inside NSTAR to integer arrays. Thought
    this was probably safer.

    2. Fixed a potential problem in NSTAR wherein for groups greater than
    the maxgroup parameter in size the old_size variable was not being
    correctly set. This would mean that some group members could get
    truncated from the output file.

    3. Fixed a bug in verbose mode wherein stars would not be fit, their
    magnitude would be set to INDEF, but no error message would be generated.
    This was occurring when the new center of a star moved too close to the
    edge of the image.

    4. Also took the opportunity to do some code cleanup.

    Davis, Jun 5, 1991

daophot$psf/
    1. Modified the PSF task so that potential psf stars are rejected if their
    sky or magnitude values are INDEF.

    2. Added a check so that stars with INDEF valued positions are treated as
    stars that were not found.

    3. Added a check in the code so that the same star could not be added to
    the psf twice.

    4. Found a code construct in the dp_friends() routine that
    could trigger an optimizer bug and removed it.

    5. Finally changed the code which deletes an empty psf image and group
    file so that it would handle an output ST table correctly.

    6. Did some minor code cleanup.

    Davis, Jun 3, 1991

daophot$peak/
    1. Found that the main fitting task in PEAK, dp_pkfit(), was missing an
    sfree statement. This may account for the problems with TESTPHOT on the
    mountain. Took the opportunity afforded by this bug to do a little
    code cleanup in peak.

    2. Had to do a couple of modifications in PSF because of the mod to the
    fitting code in PEAK. The two tasks share code. Found an extra sfree
    statement in the main PSF loop and removed it.

    3. Changed the size of the extractions box in PEAK from psfrad to
    (psfrad + fitrad + 1). This removes any problems in the unlikely
    event that the fitting radius is bigger than the psf radius.

    Davis, Jun 1, 1991

daophot$group/
    Modified GROUP so that any stars with INDEF valued centers are not
    written to the output file. Various code modifications were made
    to clean up the logic of the task and make it more structured.

    Davis, May 31, 1991

daophot$peak/
    Modified PEAK so that stars with undefined sky values don't cause a
    floating operand error. Also modified PEAK so that any stars with
    INDEF centers in the input file are not written to the output file.

    Davis, May 30, 1991

daophot$addstar/
    Addstar was not incrementing the row numbers correctly when writing
    the output star list if the output file was an ST table, resulting
    in an output file that was missing some rows. The output image was being
    computed correctly.

    Davis, May 26, 1991

daophot$
    Added the PEXAMINE task to the daophot package.

    Davis, May 24, 1991

daophot$allstar/
daophot$addstar/
    Modified the peak and allstar tasks so that they were writing the
    keyword IMAGE instead of IMNAME in the header. IMNAME was causing
    problems for the preprocessors.

    Davis, Apr 5, 1991

daophot$
daophot$psf/
    1. Modified the dpppars() routine in daolib to the datamin and datamax
    in the datapars pset are updated when the daophot fitting parameters are
    updated.

    2. Update the psf task so that the default psf image header will hold more
    than 22 stars. The current default will be three times that. Psf will
    still pack up on the min_lenuserarea parameter if it is the default.

    Davis, Apr 1, 1991

daophot$
    1. The tasks allstar, group, nstar, peak, psf and substar were all modified
    to include datamin and datamax in their verify routines.
    As part of this a set of general utility routines were written and
    stored in the file daolib/dpverify.x.

    2. The io routines were consolidated into a single file for output and/or
    another for input as appropriate. This means several files in the 
    addstar, allstar, nstar, and peak routines have disappeared.

    3. The nstar text file input routine was made for efficient by changing
    the strmatch calls to strncmp.

    Davis, Mar 30, 1991

daophot$psf/
    The psf task was writing the incorrect value of xpsf and ypsf into the
    psf image header causing the variable psf to be evaluated at the
    wrong position. Any task which evaluated the variable psf including
    seepsf, group, peak, nstar, and allstar would be in error. The
    bug was located in the file dp_writepsf.x in testphot$daophot/psf/.

    Davis, Feb 1, 1991

daophot$psf/

    The celling for surface plots was being set to 20000 in the routine
    dp_psfsetup in file daolib/dpinit.x causing floating point divide
    errors in images with pixels not in the usual CCD range.

    Davis, Jan 21, 1991

daophot$allstar/
    A call to dp_talwrite() was missing the chigrp argument.

    Davis, March 1, 1989

daophot$allstar/
    A data dependent error could occur in allstar if the user tried to
    write to an output ST table and the computed magnitude error was
    less than or equal to zero.

    Davis, February 28, 1989

daophot$
    All task except DAOFIND and PHOT were affected by an error in the
    ptools$pttables/pthdrs.x file. The tasks sometimes crash on input files
    created with append task with a memory corruption error.
    See the ptools Revisions file for a description of the error.

    Davis, January 4, 1990

December 21, 1989 -- Beta Daophot Release
.endhelp