.help revisions Jun88 noao.imred.ccdred
.nf
t_ccdgroups.x
t_ccdhedit.x
t_ccdinst.x
t_ccdlist.x
t_ccdproc.x
t_combine.x
t_mkfringe.x
t_mkillumcor.x
t_mkillumft.x
t_mkskycor.x
t_mkskyflat.x
    Added a check that the filename given to the hdmopen() procedure wasn't 
    empty.  This provides a more informative error message than the "floating
    invalid operation' one gets now when e.g. no 'instrument' file is 
    specified (10/12/13, MJF)

src/ccdcache.x
    The 'bufs' pointer was declared as TY_REAL instead of TY_SHORT (5/4/13)

t_cosmicrays.x
    A pointer to a an array of pointers was used in one place as a real.  This
    is an error when integer and real arrays are not of the same size; i.e.
    on 64-bit architectures.  (8/2/12, Valdes)

=======
V2.16.1
=======

various
    Separated the generic combine code to a subdirectory as is done
    for imcombine, mscred, etc.  This is only a partial step towards
    sharing the standard imcombine code.  Because this is really old,
    working code that has diverged significantly it will take some time
    to update/merge the new imcombine code.  (1/6/11, Valdes)

=====
V2.15
=====

src/icstat.gx
    Fixed type declarations for the asum() procedures (8/25/09, MJF)

doc/ccdproc.hlp
    Removed the statements that calibration images are not reprocessed
    if they have CCDPROC even if they lack the keywords for specific
    operations.  I looked at the code and did not see much dependence
    on CCDPROC though there could be something I'm missing.  For now,
    since a user reported this, I will assume the behavior reported by
    the user is correct and the documentation is wrong for some historical
    reason.  (5/27/08, Valdes)

x_ccdred.x
    Added the alias qccdproc for use in the quadred.quadproc task.
    (3/12/08, Valdes)

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

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

ccdred/ccdred.hd
    Hooked up help pages for ccdtest package.  (2/14/04, Valdes)

ccdred/ccdtest/t_mkimage.x
    Removed unused variable.  (8/8/02, Valdes)

ccdred/src/icscale.x
    Error dereferencing a string pointer.  (8/8/02, Valdes)

ccdred/src/t_mkfringe.x
ccdred/src/t_mkillumcor.x
ccdred/src/t_mkillumft.x
ccdred/src/t_mkskycor.x
ccdred/src/t_mkskyflat.x
    There was a confusion with the "output" parameter which is also in
    the ccdproc pset.  Each task now explicitly calls its own output
    parameter.  (7/31/02, Valdes)

=======
V2.12.1
=======

=====
V2.12
=====

ccdred/src/icsetout.x
    When computing offsets the registration point was the reference pixel
    returned by mw_gwterm for the first image.  The code then went on to
    assume this was a logical pixel when comparing with the other images,
    which is not true when there is a physical coordinate system.  The
    algorithm was fixed by converting the reference point to logical
    coordinates.  (4/18/02, Valdes)

ccdred/src/t_ccdmask.x
    Fixed bug where the if the last line or last column had a bad pixel
    without a neighboring interior pixel then the mask value would be
    some number corresponding to the number of pixels in that last line
    or column.  (2/28/02, Valdes)

ccdred/ccdred.cl
ccdred/ccdred.men
ccdred/ccdred.hd
ccdred/src/mkpkg
ccdred/x_ccdred.x
    Removed COSMICRAYS from package tasks.  The source is still not
    removed.  (8/22/01, Valdes)

ccdred/src/setdark.x
    Added a check for a zero divide in calculating the dark time scaling
    which results in an appropriate error message.  (7/5/01, Valdes)

========
V2.11.3b
========

t_combine.x
    Modified the conversion of pclip from a fraction to a number of images
    because for even number of images the number above/below the median
    is one too small.  (9/26/00, Valdes)

ccdred/src/icmedian.gx
    Replaced with faster Wirth algorithm.  (5/16/00, Valdes)

ccdred/src/icgdata.gx
ccdred/src/iclog.x
ccdred/src/icmask.x
ccdred/src/icombine.gx
ccdred/src/icscale.x
ccdred/src/icsetout.x
    Changed declarations for the array "out" to be ARB rather than 3 in
    some places (because it was not changed when another element was added)
    or 4.  This will insure that any future output elements added will
    no require changing these arguments for the sake of cosmetic correctness.
    (1/13/99, Valdes)

ccdred/src/t_combine.x
    Added workaround for error recovery problem that loses the error
    message.  (10/21/99, Valdes)

ccdred$doc/ccdproc.hlp
    The overscan type name was incorrectly given as "average" instead of
    "mean".  This was corrected in the documentation.  (10/15/99, Valdes)

ccdred$src/generic/mkpkg
ccdred$src/cosmic/mkpkg
ccdred$src/mkpkg
    Added missing dependencies.  (10/11/99, Valdes)

=======
V2.11.2
=======

ccdred$src/t_ccdlist.x
    Date accidentally changed.  File not modified.  (5/13/99, Valdes)

ccdred$doc/ccdproc.hlp
ccdred$doc/mkskyflat.hlp
    Fixed minor formating problems. (4/22/99, Valdes)

ccdred$src/imcombine/icsetout.x
    The updating of the WCS for offset images was not being done correctly.
    (10/6/98, Valdes)

ccdred$src/t_ccdmask.x
    The overlapping of groups of columns was not quite working because
    you can't overlap imp... calls.  (9/10/98, Valdes)

ccdred$src/t_ccdproc.x
ccdred$ccdproc.par
ccdred$doc/ccdproc.hlp
ccdred$darkcombine.cl
ccdred$flatcombine.cl
ccdred$zerocombine.cl
    1.  Added output image option to CCDPROC.
    2.  The combine scripts all still do in place processing.
    (6/19/98, Valdes)

ccdred$doc/ccdproc.hlp
    Fixed font change typo in Revisions section.  (6/16/98, Valdes)

ccdred$src/t_ccdmask.x
    The test for a bad pixel used && instead of ||.  (4/24/98, Valdes)

=======
V2.11.1
=======

ccdred$src/icscale.x
ccdred$doc/combine.hlp
    When zero offsets or weights are specified in a file the weights
    are not modified for zero offsets.  (10/3/97, Valdes)

ccdred$src/setoutput.x
    It is now allowed to go from ushort input to short output.
    (9/29/97, Valdes)

ccdred$src/t_combine.x
    Fixed a segmentation violation caused by attempting to close the
    mask data structures during error recovery when the error occurs
    before the data structures are defined.  (8/14/97, Valdes)

ccdred$src/cosmic/crfind.x
ccdred$src/cosmic/crlist.x
    Changed arguments with adjustable arrays to use ARB.  (8/6/97, Valdes)

ccdred$src/setsections.
    Generalized the LTERM update to work with arbitrary WCSDIM.
    (7/24/97, Valdes)

ccdred$src/ccdcheck.x
    No change except date modified.
    (7/17/97, Valdes)

=====
V2.11
=====

ccdred$src/setoverscan.x
ccdred$src/proc.gx
ccdred$src/ccdred.h
ccdred$doc/ccdproc.hlp
    The overscan fitting function now allows "average", "median", and "minmax"
    for line-by-line overscan determination.
    (2/21/97, Valdes)

ccdred$src/setfixpix.x
ccdred$src/setproc.x
ccdred$src/proc.gx
ccdred$src/setsections.x
ccdred$src/setheader.x
ccdred$src/ccdred.h
ccdred$src/corinput.gx		-
ccdred$src/generic/corinput.x	-
ccdred$src/mkpkg
ccdred$src/generic/mkpkg
ccdred$doc/ccdproc.hlp
    The bad pixel fixing is now done with the new fixpix routines from xtools.
    As part of this the physical coordinate system is set to be that of
    the CCD.
    (2/21/97, Valdes)

ccdred$src/t_ccdmask.x  +
ccdred$ccdmask.par      +
ccdred$doc/ccdmask.hlp  +
ccdred$src/mkpkg
ccdred$ccdred.cl
ccdred$ccdred.hd
ccdred$ccdred.men
ccdred$x_ccdred.x
    A new task, CCDMASK, has been added.  This task finds deviant pixels
    in CCD data and creates a pixel mask.  (2/21/97, Valdes)

ccdred$src/icscale.x
    The ccdmean keyword is now updated rather than deleted.  However
    the ccdmeant keyword is delete to force a later computation if needed.
    (1/7/97, Valdes)

ccdred$src/icsetout.x
ccdred$doc/combine.hlp
    A new option for computing offsets from the image WCS has been added.
    (1/7/97, Valdes)

ccdred$src/icmask.x
ccdred$src/iclog.x
ccdred$src/icombine.com
ccdred$src/icmask.h	+
ccdred$src/icmask.com	-
    Changed to use a mask structure.  (1/7/97, Valdes)

ccdred$src/t_combine.x
ccdred$src/icombine.gx
ccdred$src/icimstack.x +
ccdred$src/iclog.x
ccdred$src/mkpkg
ccdred$doc/combine.hlp
    The limit on the maximum number of images that can be combined, set by
    the maximum number of logical file descriptors, has been removed.  If
    the condition of too many files is detected the task now automatically
    stacks all the images in a temporary image and then combines them with
    the project option.

    The project option probably did not work previously.  May not still
    work.
    (1/7/97, Valdes)

ccdred$src/icsort.gx
    There was an error in the ic_2sort routine when there are exactly
    three images that one of the explicit cases did not properly keep
    the image identifications.  See buglog 344.  (1/17/97, Valdes)

ccdred$src/calimage.x
    The use of SZ_SUBSET-1 can cause problems because the names are
    unique to SZ_SUBSET but if unique part is the SZ_SUBSET character
    this causes problems.  (1/17/97, Valdes)

==========
V2.10.4-p2
==========

ccdred$src/icpclip.gx
    Fixed a bug where a variable was improperly used for two different
    purposes causing the algorithm to fail (bug 316).  (10/19/95, Valdes)

ccdred$src/cosmic/crlist.x
    The output bad pixel data accidentally included some extra fields
    making it incorrect to use the file directly with BADPIXIMAGE.
    The extra diagnostic fields were removed.  (9/25/95, Valdes)

ccdred$src/cosmic/t_cosmicrays.x
    Added a test for interactive mode before opening the graphics
    stream and whether to call the training routine.  This change
    was needed to allow the task to run non-interactively on
    dumb, non-graphics terminals.  (7/24/95, Valdes)

=======
V2.10.4
=======

ccdred$src/t_combine.x
    If an error occurs while opening an input image header the error
    recovery will close all open images and then propagate the error.
    For the case of running out of file descriptors with STF format
    images this will allow the error message to be printed rather
    than the error code.  (4/3/95, Valdes)

ccdred$src/icscale.x
ccdred$doc/combine.hlp
    The behavior of the weights when using both multiplicative and zero
    point scaling was incorrect; the zero levels have to account for
    the scaling.  (3/27/95, Valdes)

ccdred$src/cosmic/t_cosmicrays.x
    There was an error in setting the x,y coordinates of the window
    such that it left some of the coordinates undefined.  This causes
    an FPE on the Alpha.  (2/17/94, Valdes)

ctype.h
ccdred$src/ccdsubsets.x
    Change the test for non-filename characters to map all characters
    but alphabetic, numbers, and period to '_'.  (2/17/95, Valdes)

ccdred$src/proc.gx
    The asum$t function was not properly declared.  (9/13/94, Valdes)

ccdred$src/t_mkfringe.x
ccdred$src/t_mkillumcor.x
ccdred$src/t_mkillumft.x
ccdred$src/t_mkskycor.x
ccdred$src/t_mkskyflat.x
    Added calls to ccd_open/ccd_close in order to initialize the image
    caching even if images are not actually cached.  (9/13/94, Valdes)

ccdred$src/cosmic/t_cosmicrays.x
ccdred$src/cosmic/crexamine.x
ccdred$doc/cosmicrays.hlp
    1.  A new parameter was added to the crexamine subroutine in the
	previous modification for "training" the program.  In the
	subroutine the parameter was used as a modifyable parameter but it
	was being called with a fixed constant.  The effect was the costant
	value was no longer correct after the first execution and the
	program would act as if a 'q' was typed after the first interactive
	execution.  This was fixed to treat the input argument as input
	only.
    2.  The help page now emphasizes that the "answer" parameter is not
	to be used on the command line and if it is then the task will
	ignored the value and act as if the user always responds with
	"yes".
    (8/17/94, Valdes)

ccdred/src/cosmic/t_cosmicrays.x
ccdred/src/cosmic/crfind.x
ccdred/src/cosmic/crexamine.x
ccdred/src/cosmic/crlist.x
ccdred/src/cosmic/crlist.h
ccdred/cosmicrays.par
ccdred/doc/cosmicrays.hlp
noao$lib/scr/cosmicrays.key
    Added some new parameters and a new functionality to allow setting
    the flux ratio threshold by training with respect to a user supplied
    list of classifications.  Normally the list would be the image
    display cursor.  (6/29/94, Valdes)

ccdred/src/cosmic/t_cosmicrays.x
    Added an imflush() and imseti() after the initial copy of the input
    image to the output is done and before the random access to replace
    the detected cosmic rays.  The imseti sets the image I/O advice to
    RANDOM.  (6/24/94, Valdes)

ccdred/src/ccdcheck.x
ccdred/src/ccdmean.x
ccdred/src/setheader.x
ccdred/src/scancor.x
ccdred/src/setillum.x
ccdred/src/t_mkillumcor.x
ccdred/src/t_mkfringe.x
ccdred/src/t_mkskycor.x
ccdred/src/t_mkillumft.x
ccdred/src/t_mkskyflat.x
ccdred/doc/ccdproc.hlp
ccdred/doc/ccdinst.hlp
    Added a CCDMEANT keyword giving the time when the CCDMEAN value was
    calculated.  Routines that later access this keyword check this time
    against the image modify time to determine whether to invalidate
    the value and recompute it.  This solves the problem of people
    modifying the image outside the CCDRED package and possibly using
    an incorrect scaling value.  For backwards compatiblity if the
    new keyword is missing it is assumed to be same as the modify time;
    i.e. the CCDMEAN keyword is valid.  (6/22/94, Valdes)

ccdred/src/t_mkillumcor.x
ccdred/src/t_mkillumft.x
ccdred/src/t_mkskycor.x
ccdred/src/t_mkskyflat.x
    Added an extra argument to the millumination subroutine to specify
    whether to print log information.  This is because this procedure
    is used as an intermediate step in things like the fringe correction
    the message is confusing to users.  (6/21/94, Valdes)


ccdred/src/icaclip.gx
ccdred/src/iccclip.gx
ccdred/src/icpclip.gx
ccdred/src/icsclip.gx
    1.	The restoration of deleted pixels to satisfy the nkeep parameter
	was being done inside the iteration loop causing the possiblity
	of a non-terminating loop; i.e. pixels are rejected, they are
	restored, and the number left then does not statisfy the termination
	condition.  The restoration step was moved following the iterative
	rejection.
    2.	The restoration was also incorrectly when mclip=no and could
	lead to a segmentation violation.
    (6/13/94, Valdes)

ccdred/src/iccclip.gx
ccdred/src/icsclip.gx
    Found and fixed another typo bug.  (6/7/94, Valdes/Zhang)

ccdred/src/t_combine.x
    For some reason the clget for the nkeep parameter was deleted
    (it was in V2.10.2 but was gone in the version as of this date).
    It was added again.  (6/6/94, Valdes)

ccdred/src/icscale.x
    The sigma scaling flag, doscale1, would not be set in the case of
    a mean offset of zero though the scale factors could be different.
    (5/25/94, Valdes/Zhang)

ccdred/src/icsclip.gx
    There was a missing line: l = Memi[mp1].  (5/25/94, Valdes/Zhang)

pkg/images/imarith/icaclip.gx
ccdred/src/icaclip.gx
ccdred/src/iccclip.gx
ccdred/src/icpclip.gx
ccdred/src/icsclip.gx
    The reordering step when a central median is used during rejection
    but the final combining is average was incorrect if the number
    of rejected low pixels was greater than the number of pixel
    number of pixels not rejected.  (5/25/94, Valdes)

ccdred/src/t_combine.x
    Added a workaround for image header copy problem which leaves part
    of the TEMPNAME keyword in the output image headers.  For an output
    pixel list file this could cause the file to be screwed up.
    (5/6/94, Valdes)

ccdred/src/icscale.x
ccdred/src/t_combine.x
    1.  There is now a warning error if the scale, zero, or weight type
	is unknown.
    2.  An sfree was being called before the allocated memory was finished
	being used.
    (5/2/94, Valdes)

ccdred/src/iclog.x
    Changed the mean, median, mode, and zero formats from 6g to 7.5g to
    insure 5 significant digits regardless of signs and decimal points.
    (4/13/94, Valdes)

ccdred/src/icaclip.gx
ccdred/src/iccclip.gx
ccdred/src/icsclip.gx
    The image sigma was incorrectly computed when an offset scaling is used.
    (3/8/94, Valdes)

ccdred/src/setoverscan.x
ccdred/doc/ccdproc.hlp
    It is an error if no bias section is given or if the whole image is
    given.  (1/3/94, Valdes)

ccdred/src/t_ccdinst.x
    There was an error causing reentrant formats which was fixed.
    (12/16/93, Valdes)

ccdred/src/ccdnscan.x	+
ccdred/src/scancor.x
ccdred/src/setzero.x
ccdred/src/setdark.x
ccdred/src/setflat.x
ccdred/src/calimage.x
ccdred/src/proc.gx

ccdred/src/t_ccdinst.x
ccdred/src/t_mkskyflat.x
ccdred/src/t_ccdproc.x
ccdred/src/ccdproc.x
ccdred/src/setfringe.x
ccdred/src/setillum.x
ccdred/src/mkpkg

ccdred/doc/ccdproc.hlp
ccdred/doc/ccdinst.hlp
ccdred/doc/instruments.hlp
    For short scan data the task now looks for the number of scan lines
    in the image header.  Also when a calibration image is software
    scanned a new image is created.  This allows processing objects with
    different numbers of scan lines and preserving the unscanned
    calibration image.  (12/15/93, Valdes)

ccdred/src/setoutput.x
ccdred/doc/ccdproc.hlp
ccdred/doc/ccdred.hlp
    1.  The output datatypes were extended from just short and real to
	include ushort, integer, long, and double.  The calculation types
	are still only short or real.
    2.  The output datatype is no longer allowed to be of lower precision
	than the input datatype.
    (12/4/93, Valdes)

ccdred/src/t_combine.x
ccdred/combine.par
ccdred/doc/combine.hlp
ccdred/doc/darkcombine.hlp
ccdred/doc/flatcombine.hlp
ccdred/doc/zerocombine.hlp
    1.  The "outtype" parameter was being ignored and the package "pixeltype"
	parameter was used instead.  This was fixed to use the "outtype"
	parameter.
    2.  The output pixel datatypes now include unsigned short.
    3.  The DARKCOMBINE, FLATCOMBINE, and ZEROCOMBINE scripts specified
	that the output datatype be "real" because of the bug noted
	above the output type was being determined by the package
	"pixeltype" parameter.  The change above fixes this so that
	the combined output will always be real.  The help pages did
	not state that what the output datatype would be so a sentence
	was added specifying the output datatype is real.
    (12/4/93, Valdes)

ccdred/icgrow.gx
ccdred/icpclip.gx
ccdred/icsclip.gx
ccdred/icaclip.gx
ccdred/iccclip.gx
ccdred/t_combine.x
ccdred/doc/combine.hlp
    If there were fewer initial pixels than specified by nkeep then the
    task would attempt to add garbage data to achieve nkeep pixels.  This
    could occur when using offsets, bad pixel masks, or thresholds.  The
    code was changed to check against the initial number of pixels rather
    than the number of images.  Also a negative nkeep is no longer
    converted to a positive value based on the number of images.  Instead
    it specifies the maximum number of pixels to reject from the initial
    set of pixels.  (11/8/93, Valdes)

ccdred/doc/ccdproc.hlp
    Added a sentence explicitly saying the fixpix option provides
    the same algorithm as FIXPIX.  (11/1/93, Valdes)

ccdred/src/icscale.x
ccdred/doc/combine.hlp
    The help indicated that user input scale or zero level factors
    by an @file or keyword are multiplicative and additive while the
    task was using then as divisive and subtractive.  This was
    corrected to agree with the intend of the documentation.
    Also the factors are no longer normalized.  (9/24/93, Valdes)

ccdred/src/icsetout.x
    The case in which absolute offsets are specified but the offsets are
    all the same did not work correctly.  (9/24/93, Valdes)

ccdred/doc/geometry.hlp
ccdred/doc/ccdproc.hlp
ccdred/doc/guide.hlp
    The help was modified to say that the overscan region length is
    determine from trimsec and is ignored in biassec.  (9/23/93, Valdes)

ccdred/doc/instruments.hlp
ccdred/doc/subsets.hlp
    Added notes that comments are allowed.  Also if there is more than
    one translation for the same CCDRED parameter the last one takes
    effect.  (9/20/93, Valdes)

ccdred/doc/combine.hlp
    Clarified how bad pixel masks work with the "project" option.
    (9/13/93, Valdes)

ccdred/src/t_combine.x
    The algorithm for making sure there are enough file descriptors failed
    to account for the need to reopen the output image header for an
    update.  Thus when the number of input images + output images + logfile
    was exactly 60 the task would fail.  The update occurs when the output
    image is unmapped so the solution was to close the input images first
    except for the first image whose pointer is used in the new copy of the
    output image.  (8/4/93, Valdes)

============
V2.10.3 beta
============

ccdred/src/icgdata.gx
    There was an indexing error in setting up the ID array when using
    the grow option.  This caused the CRREJECT/CCDCLIP algorithm to
    fail with a floating divide by zero error when there were non-zero
    shifts.  (5/26/93, Valdes)

ccdred/src/icmedian.gx
    The median calculation is now done so that the original input data
    is not lost.  This slightly greater inefficiency is required so
    that an output sigma image may be computed if desired.  (5/10/93, Valdes)

ccdred/darkcombine.cl
ccdred/doc/darkcombine.hlp
ccdred/doc/flatcombine.hlp
ccddb/kpno/direct.cl
ccddb/kpno/coude.cl
ccddb/kpno/cryocam.cl
ccddb/kpno/echelle.cl
ccddb/kpno/foe.cl
ccddb/kpno/specphot.cl
ccddb/kpno/sunlink.cl
    1.	Updated FLATCOMBINE defaults for KPNO data.
    2.	Changed package defaults for DARKCOMBINE to use "minmax" rejection.
    (4/19/93, Valdes)

ccdred/src/icombine.gx
    There was no error checking when writing to the output image.  If
    an error occurred (the example being when an imaccessible imdir was
    set) obscure messages would result.  Errchks were added.
    (4/16/93, Valdes)

ccdred/src/setfpix.x
ccdred/src/ccdproc.x
ccdred/src/t_ccdproc.x
ccdred/doc/ccdproc.hlp
ccdred/doc/instrument.hlp
    If a specified bad pixel file is not found an abort now occurs.  Also
    the FIXPIX processing header flag is set even if there are no
    bad pixels.  The documentation was revised to stress that an "untrimmed"
    bad pixel file refers to the original CCD coordinates which is
    especially important with subraster readouts.  (2/23/93, Valdes)

ccdred/src/icaclip.gx
ccdred/src/iccclip.gx
ccdred/src/icpclip.gx
ccdred/src/icsclip.gx
    When using mclip=yes and when more pixels are rejected than allowed by
    the nkeep parameter there was a subtle bug in how the pixels are added
    back which can result in a segmentation violation.
	if (nh == n2)  ==>  if (nh == n[i])
    (1/20/93, Valdes)

ccdred/zerocombine.cl
ccdred/darkcombine.cl
ccdred/flatcombine.cl
    Explicitly set ccdproc.noproc to no.  (11/23/92, Valdes)

=======
V2.10.2
=======

ccdred/src/calimage.x
    Added test on the requested ccdtype when setting up the calibration images
    to avoid mapping a calibration type image which is not going to be
    used.  (11/17/92, Valdes)

ccdred/darkcombine.cl
    Fixed typo in output parameter prompt string refering to a flat field.
    (11/10/92, Valdes)

ccdred/src/ccdred.h
ccdred/src/t_ccdproc.x
ccdred/src/proc.gx
    Separated the minreplace operation from the findmean operation.  It
    is now a separate operation only applied to flat images.
    (10/26/92, Valdes)

ccdred/ccdtest/demo.dat
    Removed display commands.  Because DISPLAY is always loaded in V2.10
    there was no way to escape the displaying.
    (9/30/92, Valdes)

ccdred$darkcombine.cl
ccdred$flatcombine.cl
ccdred$zerocombine.cl
ccdred$doc/darkcombine.hlp
ccdred$doc/flatcombine.hlp
ccdred$doc/zerocombine.hlp
    Added "blank", "nkeep", and "snoise" parameters.
    (9/30/92, Valdes)

ccdred$src/t_combine.x
ccdred$src/icaclip.gx
ccdred$src/iccclip.gx
ccdred$src/icgrow.gx
ccdred$src/iclog.x
ccdred$src/icombine.com
ccdred$src/icombine.gx
ccdred$src/icombine.h
ccdred$src/icpclip.gx
ccdred$src/icscale.x
ccdred$src/icsclip.gx
ccdred$src/icsetout.x
ccdred$combine.par
ccdred$doc/combine.hlp
    The weighting was changed from using the square root of the exposure time
    or image statistics to using the values directly.  This corresponds
    to variance weighting.  Other options for specifying the scaling and
    weighting factors were added; namely from a file or from a different
    image header keyword.  The \fInkeep\fR parameter was added to allow
    controlling the maximum number of pixels to be rejected by the clipping
    algorithms.  The \fIsnoise\fR parameter was added to include a sensitivity
    or scale noise component to the noise model.  Errors will now delete
    the output image.
    (9/30/92, Valdes)

ccdred$src/t_combine.x
ccdred$src/iclog.x
    The log now prints the final image name rather than the temp name when
    using the clobber option.  (8/25/92, Valdes)

ccdred$src/icaclip.gx
ccdred$src/iccclip.gx
ccdred$src/icpclip.gx
ccdred$src/icsclip.gx
    There was a very unlikely possibility that if all the input pixels had
    exactly the same number of rejected pixels the weighted average would
    be done incorrectly because the dflag would not be set.  (8/11/92, Valdes)

ccdred$src/icmm.gx
    This procedure failed to set the dflag resulting in the weighted average
    being computed in correctly.  (8/11/92, Valdes)

ccdred$src/icscale.x
    When scaling and zero offseting the zero level factors were incorrectly
    computed.  (8/10/92, Valdes)

ccdred$src/ic[acs]clip.gx
ccdred$src/icstat.gx
    Corrected type mismatches in intrinsic functions.  (8/10/92, Valdes)

=======
V2.10.1
=======

=======
V2.10.0
=======

=====
V2.10
=====

ccdred$src/icombine.gx
    Needed to clear buffers returned by impl1 during the memory check
    to avoid possible invalid values.  (4/27/92, Valdes)

ccdred$src/t_ccdproc.x
ccdred$src/calimage.x
    Made it an error if an explicit calibration image is specified but cannot
    be opened.  Previously it would then look in the input list for the
    appropriate type.  (4/24/92, Valdes)

ccdred$ccdproc.x
ccdred$t_ccdproc.x
    Made the COMP type be processed like and OBJECT rather that the
    default case.  The only effect of this is to not have CCDMEAN
    calculated.  (4/8/92, Valdes)

ccdred$src/icalip.gx
ccdred$src/icclip.gx
ccdred$src/ipslip.gx
ccdred$src/icslip.gx
ccdred$src/icmedian.gx
    The median calculation with an even number of points for short data
    could overflow (addition of two short values) and be incorrect.
    (3/16/92, Valdes)

ccdred$src/iclog.x
    Added listing of read noise and gain.  (2/10/92, Valdes)

ccdred$src/icpclip.gx
    Reduced the minimum number of images allowed for PCLIP to 3.
    (1/7/92, Valdes)

ccdred$darkcombine.cl
ccdred$flatcombine.cl
    Set default parameters as requested by the support people.
    (12/12/91, Valdes)

ccdred$src/icgrow.gx
    The first pixel to be checked was incorrectly set to 0 instead of 1
    resulting in a segvio when using the grow option.  (12/6/91, Valdes)

ccdred$src/proc.gx
ccdred$src/icgdata.gx
ccdred$src/icscale.x
ccdred$src/setfixpix.x
ccdred$src/t_combine.x
    Fixed argument mismatch errors found by SPPLINT.  (11/22/91, Valdes)

ccdred$src
    Replaced COMBINE with new version.  (9/1/91, Valdes)

ccdred$ccdtest/observe.cl -> artobs.cl
ccdred$ccdtest/observe.hlp -> artobs.hlp
ccdred$ccdtest/subsection.cl
ccdred$ccdtest/subsection.hlp
ccdred$ccdtest/mkimage.hlp
ccdred$ccdtest/demo.dat
ccdred$ccdtest/ccdtest.men
ccdred$ccdtest/ccdtest.hd
ccdred$ccdtest/ccdtest.cl
ccdred$ccddb/kpno/demo.dat
    Renamed OBSERVE to ARTOBS to avoid conflict with the CCDACQ task of
    the same name.  (8/29/91, Valdes)

ccdred$src/setoutput.x
ccdred$src/setproc.x
ccdred$src/setdark.x
ccdred$src/setzero.x
ccdred$src/setflat.x
ccdred$src/setfringe.x
ccdred$doc/ccdred.hlp
    The default output pixel type and computation type are now real.  
    The computation type may be separately specified.  (5/29/91, Valdes)

ccdred$src/t_mkskycor.x
    The computation of CCDMEAN failed to accumlate the last few lines causing
    the mean to be underestimated.  (4/16/91, Valdes)

ccdred$src/t_ccdinst.x +
ccdred$src/ccdinst1.key +
ccdred$src/ccdinst2.key +
ccdred$src/ccdinst3.key +
ccdred$src/hdrmap.x
ccdred$src/mkpkg
ccdred$ccdinstrument.par +
ccdred$ccdred.cl
ccdred$ccdred.hd
ccdred$ccdred.men
ccdred$x_ccdred.x
    Added the new task CCDINSTRUMENT.  This also involved some changes to
    the header translation package hdrmap.x.  (10/23/90, Valdes)

ccdred$src/imcscales.x
ccdred$src/imcmode.gx
ccdred$src/mkpkg
    Added error check for incorrect mode section specification.
    (10/3/90, Valdes)

ccdred$src/ccdred.h
ccdred$src/proc.gx
ccdred$src/setproc.x
ccdred$ccdproc.par
    Added a minreplace parameter to replace flat field values less than this
    value by the value.  This provides zero division prevention without
    requiring specific flat field checking.
    (10/3/90, Valdes)

ccdred$src/t_ccdproc.x
ccdred$src/ccdproc.x
ccdred$src/scancor.x
    1.  The scan correction now computes the CCDMEAN to account for the
	ramp down.
    2.  Did a simple move of the ccdmean call from before scancor to
	after scancor.  Since CCDMEAN is now computed in SCANCOR this
	has no real affect and is just cosmetic.  If CCDMEAN were not
	computed in SCANCOR then the new placement would have computed
	the right value at the expense of another pass through the image.
    (9/21/90, Valdes)

ccdred$src/t_badpixim.x
    The template image cannot be closed immediately after opening the NEW_COPY
    mask image because the STF kernel doesn't make the header copy until
    pixel I/O occurs.  This only affects STF images.  (6/19/90, Valdes)

====
V2.9
====

ccdred$src/t_combine.x
    Changed:
	char images[SZ_FNAME-1,nimages] --> char images[SZ_FNAME,nimages-1]
    The incorrect declaration results in each successive image name have
    additional leading characters.  Apparently, since this has not be
    found previously, the leading characters have generally been blanks.
    (3/30/90, Valdes)

ccdred$doc/combine.hlp
    Clarified and documented definitions of the scale, offset, and weights.
    (11/30/89, Valdes)

ccdred$ccdproc.par
    1.  All parameters now have default values.  (10/31/89, Valdes)

ccdred$src/cosmic/mkpkg
ccdred$src/gtascale.x -
ccdred$t_cosmicrays.x
    1.  Removed duplicate of gtools procedure.
    2.  Fixed transfer out of IFERR block message when input image was wrong.
    3.  The badpixel file was not initialized to null if the user did not
	want a badpixel file output.  (9/21/89, Valdes)

====
V2.8
===

ccdred$src/imcmode.gx
    Fixed bug causing infinite loop when computing mode of constant value
    section.  (8/14/89, Valdes)

ccdred$src/ccdproc.x
ccdred$src/ccddelete.x
ccdred$src/t_ccdproc.x
ccdred$src/t_mkfringe.x
ccdred$src/t_mkskyflat.x
ccdred$src/t_mkskycor.x
ccdred$src/t_mkillumft.x
ccdred$src/t_mkillumcor.x
ccdred$src/t_combine.x
ccdred$src/scancor.x
ccdred$src/readcor.x
    1. Added error checking for procedure ccddelete.
    2. Made workaround for error handling problem with procedure imrename
       so that specifying a bad backup prefix would result in an abort
       with an error message.  (6/16/89, Valdes)

ccdred$src/imcombine.gx
   Made same changes made to image.imcombine to recover from too many VOS
   file description error.  (6/14/89, Valdes)

ccdred$setinstrument.cl
ccdred$setinstrument.hlp
   Incorrect instrument names are now reported to the user, a menu is
   printed if there is one, and a second opportunity is given.
   (6/14/89, Valdes)

ccdred$ccdred.par
   Added an ennumerated subset for the output datatype.  (5/12/89, Valdes)

ccdred$src/imcombine.gx
   Because a file descriptor was not reserved for string buffer operations
   and a call to stropen in cnvdate was not error checked the task would
   hang when more than 115 images were combined.  Better error checking
   was added and now an error message is printed when the maximum number
   of images that can be combined is exceeded.  (5/9/89, Valdes)

ccdred$src/sigma.gx
ccdred$src/imcaverage.gx
    1.  Weighted sigma was being computed incorrectely.
    2.  Added errchk to imcaverage.gx.
    (5/6/89, Valdes)

ccdred$src/setdark.x
ccdred$src/setflat.x
ccdred$src/setfringe.x
ccdred$src/setillum.x
ccdred$src/setoverscan.x
ccdred$src/settrim.x
ccdred$src/setzero.x
    Made the trimsec, biassec, datasec, and ccdsec error messages more
    informative.  (3/13/89, Valdes)
 
ccdred$src/imcmode.gx
    For short data a short variable was wraping around when there were
    a significant number of saturated pixels leading to an infinite loop.
    The variables were made real regardless of the image datatype.
    (3/1/89, Valdes)
 
ccdred$src/t_mkskyflat.x
ccdred$src/t_mkskycor.x
    1.  Added warning if images have not been flat fielded.
    2.  Allowed flat field image to be found even if flatcor=no.
    (2/24/89, Valdes)
 
ccdred$src/imcthresh.gx
ccdred$combine.par
ccdred$doc/combine.hlp
ccdred$src/imcscales.x
    1.  Added provision for blank value when all pixels are rejected by the
        threshold.
    2.  Fixed a bug that improperly scaled images in the threshold option.
    3.  The offset printed in the log now has the opposite sign so that it
	is the value "added" to bring images to a common level.
    (2/16/89, Valdes)

ccdred$src/proc.gx
    When the data section had fewer lines than the output image (which occurs
    when not trimming and the overscan being along lines) pixel out of
    bounds errors occured.  This bug was due to a sign error when reading
    the non-trimmed overscan lines.  (2/13/89, Valdes)

ccdred$src/setoverscan.gx
    The overscan buffer for readaxis=column was not initialized yielding
    unpredictable and incorrect overscan data.
    (3/13/89, Valdes)

ccdred$src/imcmode.gx
    Added test for nx=1.  (2/8/89, Valdes)
 
ccdred$darkcombine.cl
ccdred$flatcombine.cl
    Changed the default parameters to use "avsigclip" combining and
    no scaling or weighting.  (1/27/89, Valdes)
 
ccdred$src/ccdcheck.x
ccdred$src/setillum.x
ccdred$src/t_ccdproc.x
    1.  If the illumination image does not have CCDMEAN in its header
	it is calculated.
    2.  If an error occurs in setting up for illumination or fringe
	correction during processing a warning is issued and these
	processing steps are skipped.  They can be done later if
	desired.  Previously this caused an abort.
    (1/27/89, Valdes)
 
ccdred$ccdgroups.par
ccdred$src/t_ccdgroups.x
ccdred$doc/ccdgroups.hlp
    Added two new group types; ccdtype and subset.  (1/26/89, Valdes)
 
ccdred$src/t_ccdlist.x
ccdred$doc/ccdlist.hlp
    The exposure time and dark time are now printed in long format.  This
    is useful to allow verifying the header translation is working
    correctly.  (1/26/89, Valdes)
 
ccdred$src/setfixpix.x
ccdred$src/t_badpixim.x
    The magic word "untrimmed" no longer needs whitespace preceding it.
    (1/24/89, Valdes)
 
imred$ccdred/src/imcscales.x
    Valdes, Dec 8, 1988
    1.  COMBINE now prints the scale as a multiplicative quantity.
    2.  The combined exposure time was not being scaled by the scaling
	factors resulting in a final exposure time inconsistent with the
	data.

imred$ccdred/src/t_mkskyflat.x
imred$ccdred/src/t_mkillumft.x
imred$ccdred/src/t_mkskycor.x
imred$ccdred/src/t_mkskyflat.x
imred$ccdred/src/t_mkfringe.x
imred$ccdred/doc/mkillumcor.hlp
imred$ccdred/doc/mkillumflat.hlp
imred$ccdred/mkillumflat.par
imred$ccdred/mkillumflat.par
    1.  Minor typo in declaration (calimage.x) which had no effect.
    2.  Missing include file (t_mkskyflat.x) caused "Cannot open image"
	when using MKSKYFLAT.
    3.  Added checks for division by zero which are reported at the end as
	the number of divisions by zero and the replacement value.
	The replacement value was added as a parameter value in MKILLUMCOR
	and MKILLUMFLAT.
    4.	Updated the help pages to reflect the new division by zero parameter.
    5.	Modified the log strings to be more informative about what
	was done and which images were used.
	(10/20/88 Valdes)

imred$ccdred/src/imcombine.gx
    A vops clear routine was not called generically causing a crash with
    double images.  (10/19/88 Valdes)

imred$ccdred/src/t_mkskycor.x
    Replaced calls to recipricol vops procedure to one with zero checking.
    (10/13/88 Valdes)

imred$ccdred/src/imcscales.x
    It is now an error if the mode is not positive for mode scaling or
    weighting.  (9/28/88 Valdes)

imred$ccdred/ccdred.par
imred$ccdred/doc/ccdred.hlp
    The plotfile parameter was changed to reflect the "" character
    as the new default.  (9/23/88 jvb)

imred$ccdred/src/imcmedian.gx
    The median option was selecting the n/2 value instead of (n+1)/2.  Thus,
    for an odd number of images the wrong value was being determined for the
    median. (8/16/88 Valdes)

imred$ccdred/src/scancor.x
imred$ccdred/src/calimage.x
imred$ccdred/src/ccdcmp.x +
imred$ccdred/src/mkpkg
    1.  The shortscan correction was incorrectly writing to the input image
	rather than the output image causing a cannot write to file error.
    2.  It is now a trapped error if the input image is the same as a
	calibration image.  (4/18/88 Valdes)

imred$ccdred/src/imcmode.gx
    The use of a mode sections was handled incorrectly.  (4/11/88 Valdes)

noao$imred/ccdred/src/setoverscan.x
    Minor bug fix:
	gt_setr (gt, GTXMIN, 1.) -> gt_setr (gt, GTXMIN, x[1])
	gt_setr (gt, GTXMAX, real(npts)) -> gt_setr (gt, GTXMAX, x[npts])
    (2/11/88 Valdes)

noao$imred/ccdred/src/t_mkillumflat.x -> t_mkillumft.x
noao$imred/ccdred/src/t_mkfringecor.x -> t_mkfringe.x
noao$imred/ccdred/src/t_badpiximage.x -> t_badpixim.x
noao$imred/ccdred/src/imcthreshold.gx -> imcthresh.gx
noao$imred/ccdred/src/generic/imcthresh.x -> imcthresh.x
noao$imred/ccdred/src/mkpkg
noao$imred/ccdred/src/generic/mkpkg
    Shortened long names. (2/10/88 Valdes)

noao$imred/ccdred/src/t_mkskycor.x
noao$imred/ccdred/doc/mkskycor.hlp
noao$imred/ccdred/doc/mkillumcor.hlp
noao$imred/ccdred/doc/mkskyflat.hlp
noao$imred/ccdred/doc/mkillumflat.hlp
noao$imred/ccdred/doc/mkfringecor.hlp
    1. 	When not clipping the first 3 lines of the illumination were always
	zero.
    2.  The clipping algorithm had several errors.
    3.  It was unclear what a box size of 1. meant and whether one could
	specify the entire image as the size of the box.
    4.  The smoothing box has been generalize to let the user chose the minimum
	and maximum box size.  This lets the user do straight box smoothing
	and the growing box smoothing. (2/2/88 Valdes)

noao$imred/ccdred/src/ccdtypes.h
    Added the comparison CCD image type. (1/21/88 Valdes)

noao$imred/ccdred/src/t_mkskycor.x
noao$imred/ccdred/src/t_mkillumcor.x
noao$imred/ccdred/src/t_mkskyflat.x
noao$imred/ccdred/src/t_mkillumflat.x
noao$imred/ccdred/src/t_mkfringecor.x
    Calling sequences to the set_ procedures were wrong. (1/20/88 Valdes)

noao$imred/ccdred/src/imcscales.x
    The exposure time is now read as real. (1/15/88 Valdes)

noao$imred/ccdred/src/corinput.gx
    Discovered an initialization bug which caused the fixing of bad lines
    to fail after the first image.  (11/12/87 Valdes)

noao$imred/ccdred/ccdtest/observe.cl
noao$imred/ccdred/ccdtest/subsection.cl
noao$imred/ccdred/ccdtest/demo.dat
    Made modification to allow the demo to work with STF format images.
    The change was in being more explicit with image extensions; i.e.
    obs* --> obs*.??h.  (11/12/87 Valdes)

noao$imred/ccdred/src/mkpkg
noao$imred/ccdred/src/ccdmean.x		 +
noao$imred/ccdred/src/ccdcache.h	 +
noao$imred/ccdred/src/ccdcache.com
noao$imred/ccdred/src/ccdcache.x
noao$imred/ccdred/src/t_ccdproc.x
noao$imred/ccdred/src/ccdproc.x
noao$imred/ccdred/src/ccdcheck.x
noao$imred/ccdred/src/setflat.x
noao$imred/ccdred/src/setdark.x
noao$imred/ccdred/src/setzero.x
noao$imred/ccdred/src/setfixpix.x
noao$imred/ccdred/src/setillum.x
noao$imred/ccdred/src/setfringe.x
noao$imred/ccdred/src/t_ccdlist.x
    1.  There was a recursion problem caused by the absence of the CCDPROC
	flag in a zero level image which did not need any processing
	because there was no trimming, overscan subtraction, or bad
	pixel correction.  The procedure CCDPROC left the image
	unmodified (no CCDPROC flag) which meant that later another unprocessed
	calibration image would again try to process it leading to
	recursion.  Since I was uncomfortable with relying on the
	CCDPROC flag I added the routine CCDCHECK to actually check
	each processing flag against the defined operations.  This will
	also allow additional automatic processing of calibration
	images if the users sets new flags after an initial pass
	through the data.  The CCDPROC flag is still set in the data
	but it is not used.
    2.  It is possible in data which has no object types for the flat
	field image never to have its mean computed for later scaling.
	There were two modifications to address this problem.  If an
	image is processed without a ccdtype then the mean will be
	computed at a very small cost in time.  If the image is later
	used as a flat field this information will then be present.
	Second, if a flat field calibration image does not have the
	mean value, even if it has been processed, the mean value
	will still be calculated.
    3.  In looking at the recursion problem I realized that some of
	the calibration images could be opened more than once, though
	READ_ONLY, once for the image being processed and later if the
	task has to backtrack to process a another calibration frame.  I
	was surprise that this was not found on VMS until I realized
	that for OIF format images the image header is read and the
	file is then closed.  No file is actually left open until pixel
	I/O is done.  However, this should cause STF images to fail on
	VMS because VMS does not allow a file to be open more than once
	and the STF image header is kept open.  I rewrote the image
	caching interface to cache the IMIO pointer even if the pixel
	data was not cached.  This will insure any calibration image
	is only opened once even if it is accessed independently from
	different parts of the program.
    4.  The error message when using fringe and illumination correction
	images which have not been processed by MKFRINGECOR and
	MKILLUMCOR was misleading when refering to the absence of the
	MKFRINGE and MKILLUM flag.  A user thought that the missing
	flag was FRINGCOR which refers to an image being fringe corrected.
	The message was made a little more clear.
    5.  The CCDLIST listing for fringe correction in long format was wrong.
    (11/12/87 Valdes)

noao$imred/ccdred/src/t_combine.x
noao$imred/ccdred/src/t_ccdhedit.x
noao$imred/ccdred/src/setoverscan.x
noao$imred/ccdred/src/setinput.x
noao$imred/ccdred/src/imcscales.x
noao$imred/ccdred/src/imclogsum.x
noao$imred/ccdred/src/ccdlog.x
noao$imred/ccdred/src/ccddelete.x
    Added calls to XT_STRIPWHITE to allow null strings to be recognized
    with whitespace.  It should probably use NOWHITE but this would make
    it incompatible with V2.5.  (11/6/87 Valdes)
.endhelp