# BUGS.V29 -- Known bugs in the frozen IRAF Version 2.9 (and previous # versions). Started 15JUN90. # # Record Format: # # NUMBER: record number, decimal, sequential. # MODULE: package.task or library.procedure or 'unknown' or ... # SYSTEM: versions of iraf in which bug was present # DATE: date bug logged, unix format date string # FROM: user login name # BUG: description of the bug # STATUS: 'fixed in V2.X', 'unresolved', etc. # # New records are added to the tail of the bugfile. Left justify field labels, # indent text to the first tab stop, one blank line between bug entries. # ---------------------------------------------------------------------------- NUMBER: 120 MODULE: apphot.daofind SYSTEM: V2.8 DATE: Fri Mar 30 08:02:12 MST 1990 FROM: davis BUG: In rare circumstances daofind may abort with a "pixel file truncation error" when trying to read back the convolved images it has just written. This only occurs on certain sized images and is due to the interaction of the read, write and boundary extension in image i/o. For example daofind works fine on a 640 by 1024 image but fails on one that is 641 by 1025 pixels. STATUS: The problem is fixed in 2.9. The solution was to add an imflush statement to flush the image i/o buffers after the write was complete and before initiating the read. There is no workaround. Contact the IRAF group for a fix. NUMBER: 121 MODULE: utilities.curfit SYSTEM: V2.9 DATE: Mon May 21 15:12:40 MST 1990 FROM: sjacoby BUG: The errors in the coefficients reported by CURFIT are incorrect when points have been rejected from the sample by the iterative rejection algorithm. User deleted points are handled properly, but points automatically rejected by sigma criteria are being incorrectly included in the coefficient error calculation. STATUS: The bug is fixed in V2.10. There is no workaround, unless the user can identify and delete by hand those points the algorithm would reject. NUMBER: 122 MODULE: cl SYSTEM: V2.9 DECstation/IRAF DATE: Wed May 23 20:58:47 MST 1990 FROM: tody BUG: If IRAFARCH is not defined in the user environment, an attempt to run the CL in DECstation/IRAF will fail with the message "... /bin.f68881/cl.e not found". This is caused by a malformed if-else (missing else) on line 41 of hlib$cl.csh. STATUS: A simple workaround is to define "setenv IRAFARCH mips" in the user environment before starting the CL. This bug is so easy to fix however, that a permanent fix is probably the best approach. Merely edit the file iraf/unix/hlib/cl.csh and change the "if" on line 41 to "else if". NUMBER: 123 MODULE: imtranspose SYSTEM: V2.9 DATE: Wed Jun 6 08:37:42 MST 1990 FROM: davis BUG: Imtranspose fails with an "unknown pixel type error" on image of type ushort. STATUS: The bug has been fixed in 2.10. There is no workaround except to change the pixel type of the image to int or long. NUMBER: 124 MODULE: proto.imedit SYSTEM: V2.9, V2.8 DATE: Wed Jun 6 09:22:54 MST 1990 FROM: valdes BUG: The fixpix format input interpolates across the longer dimension rather than the shorter dimension. If the region is a complete column/line then no correction is made. STATUS: Now fixed. For square or nearly square regions the interpolation across the longer dimension is probably adequate. Other workarounds are to use FIXPIX itself or convert the fixpix format file to regular cursor input to imedit (which then also allows use of any of the other options). NUMBER: 125 MODULE: apextract.apscatter SYSTEM: V2.9 DATE: Mon Jun 11 17:35:22 MST 1990 FROM: valdes BUG: When specifying a "line" to be plotted a check is made against the wrong image axis; i.e. if a line is specified then it forced to be in the range 1 to the number of image columns and vice-versa for the other axis. STATUS: This has been fixed. There is no workaround but the task will still function correctly even though it is not possible to examine some image lines or columns. The effect becomes obvious only in significantly nonsquare images. NUMBER: 126 MODULE: apphot package SYSTEM: V2.9 DATE: Wed Jun 27 16:47:34 MST 1990 FROM: davis BUG: Due to fractional pixel effects the sky fitting routines in apphot can fail to preallocate enough space to hold the sky pixels, resulting in a memory corruption error or a segmentation violation on exit from a task. This condition occurs rarely most often when the sky fitting annulus is very narrow. All the apphot tasks which do sky fitting are affected. STATUS: The bug is fixed in 2.10. There is no workaround although increasing the width of the sky fitting annulus slightly may bypass the problem. Contact the IRAF group for a fix. NUMBER: 127 MODULE: artdata.mkobjects SYSTEM: V2.9 DATE: Mon Jul 2 11:41:14 MST 1990 FROM: valdes BUG: Objects which fall partially off the lower edges of the image (objects with centers near the (1,1) image origin) are shifted by 1 pixel to larger coordinates. STATUS: This bug is caused by incorrect rounding of negative numbers. The bug has been fixed. The workarounds are to either compensate for the error in the analysis or offset the coordinates to larger values (say by using the offset parameters) and make a larger image which can then be trimmed with IMCOPY. The tricky thing is knowing when the objects go off the edge since the full object size depends on seeing and the specified dynamic range. NUMBER: 128 MODULE: IMFORT SYSTEM: V2.9 DATE: Tue Jul 10 10:33:02 MST 1990 FROM: tody BUG: In IRAF V2.9, IMFORT has a bug which prevents access to images not in the current working directory. If one tries to access an image using a pathname an error message such as "image not found (.imh)" is seen. STATUS: The workaround is to CD (SET DEF) to the directory before accessing the image. A patch (patch #2) is available in the IRAF network archive which can be installed to fix the bug. Installation instructions are given in the README files for each supported architecture (SOS4, VMS5, etc.). NUMBER: 129 MODULE: onedspec.splot, onedspec.calibrate SYSTEM: V2.9 DATE: Tue Jul 10 17:07:46 MST 1990 FROM: valdes BUG: In SPLOT the first pixel is ignored. In CALIBRATE the wavelength scale for extinction correction will be in error by one pixels. STATUS: These two bugs are related because the header parameter NP1, which is the offset to the first valid pixel, should be zero in most cases but a bug fix in Nov. 1989 put in a minimum value of 1. The symptom in SPLOT is to lose the first pixel but the wavelength scale is correctly compensated. In CALIBRATE the wavelength scale is not compensated though the error should be extremely small since the calibrations are interpolations from a smooth curve. There is no easy work around for SPLOT. If the one pixel wavelength error in CALIBRATE is a concern one could reset the wavelength zero point parameter W0 or CRVAL1 to be greater by WPC or CDELT1; i.e. make the starting wavelength parameter refer to pixel 2. Note that this will make the wavelengths wrong for SPLOT and other ONEDSPEC tasks. This bug is corrected in V2.10. NUMBER: 130 MODULE: artdata.mkobjects SYSTEM: V2.9 DATE: Thu Jul 19 15:14:14 MST 1990 FROM: valdes BUG: The radius parameter for the "gaussian" model PSF is being used as a full width instead of a radius as intended and described in the documentation. The effect is to give narrower stars than expected. STATUS: The workaround is to give a FWHM for the gaussian model instead of a radius at half maximum. NUMBER: 131 MODULE: artdata.mkobjects SYSTEM: V2.9 DATE: Mon Jul 23 10:53:43 MST 1990 FROM: valdes BUG: The moffat function size scaling (as set by the radius parameter) is incorrect. STATUS: The shape of the function is based on the correct beta but the size is scaled for a function with beta of: beta1 = 2 * beta - 2 Another way to look at it is that the flux level corresonding to the specified radius is different than the half intensity. The actual flux level is: F(radius) = 0.5 ** (beta / beta1) Finally, the radius which must be specified, rfudge, to get a desired radius at half intensity, rhalf, is: rfudge = rhalf * sqrt ((2**(1/beta1)-1) / (2**(1/beta)-1)) Note that there is no error for beta=2, the size is too large for beta>2, and the size is too small for beta<2. The workaround is to adjust the specified radius for a desired radius at half intensity using the above formula. NUMBER: 132 MODULE: utilities.curfit SYSTEM: V2.9 DATE: Mon Jul 30 17:37:48 MST 1990 FROM: davis BUG: Curfit was crashing with a segmentation violation when it tried to fit the second file of data or second image in a list of images. The pointer to the icfit data structure was being freed after the first file of data was fit and not reallocated before the next fit resulting in a reference to an undefined pointer. STATUS: The bug has been fixed in 2.10. There is no workaround except to use a script to loop over a large number of input data files. NUMBER: 133 MODULE: math$gsurfit/dgsder SYSTEM: V2.9 DATE: Wed Aug 1 16:27:43 MST 1990 FROM: davis BUG: Due to a typographical error in the file math$gsurfit/gsder.gx, dgsder (the routine which computes the derivative of a double precision surface) was inadvertantly calling salloc with a pointer address instead of an array size. If the assigned pointer value is very large this can cause an "out of memory error" in any task which calls dgsder. Currently the only task affected by this bug is TRANSFORM in the LONGSLIT package. STATUS: The bug is fixed in 2.10. There is no workaround. Contact NOAO for a fix. NUMBER: 134 MODULE: XC (SPP compiler) SYSTEM: Sun/IRAF V2.9 (Sun systems only) DATE: Wed Aug 8 10:04:34 MST 1990 FROM: tody BUG: Sun/IRAF V2.9 will not work with the SunOS 1.3 or 1.3.1 Fortran compiler (which Sun came out with after IRAF was released). The new Sun compiler locates important executables such as f77, cc, etc., in a nonstandard place, one of the Fortran libraries has been done away with, the command line arguments have changed, etc. (in short Sun has completely revamped their compiler). STATUS: A new version of XC is available from IRAF site support which will work with the new compiler. Note that although this allows the compiler to be used, and it appears to be possible to compile and link programs, the IRAF V2.9 libraries were compiled with the old Sun Fortran compiler and we cannot be certain that these libraries are fully compatible with the new compiler and its libraries. Also, the default host compiler flags supplied with Sun/IRAF MKPKG are not necessarily what is best for the new compiler. Most likely these are not serious problems, but Sun/IRAF will not fully support the new compiler until we have had time to recompile the full system for a future release. NUMBER: 135 MODULE: noao.artdata.mkobjects SYSTEM: V2.9 DATE: Thu Aug 9 09:24:41 MST 1990 FROM: valdes BUG: The general cumulative profile file input for both the PSF and a an object type does not work. The error is can't open image even though the file is not an image. STATUS: This bug is due to the inability of the image access checking routine to distinguish images from regular files. Because of the order of the logic the task always checks to see if the specified file is an image template first, decides it is an image, and then fails with the error can't open image. There is no work around. Those requiring this capability wil have to contact site support for help. NUMBER: 136 MODULE: wfits/rfits SYSTEM: V2.9 DATE: Mon Aug 27 07:37:57 MST 1990 FROM: davis BUG: Wfits sometimes crashed with a segmentation violation when files were written with a non-standard block size (block_factor > 10). The error was in the code which 0-filled the unused portion of the last output block, and occurred if the unused portion of that block was greater than 2880 bytes. Rfits sometimes reported read errors when it tried to read tapes with a non-standard block size (blocks not a multiple of 2880 bytes). Rfits was not always counting the number of characters read from the tape correctly when the read attempted to cross tape record boundaries. Tapes with small block sizes like 512 and 1024 were the most affected. STATUS: Both bugs have been fixed in version 2.10 and the 2.9.1 patch. NUMBER: 137 MODULE: IMDKERN SYSTEM: V2.9 DATE: Thu Sep 6 11:59:44 MST 1990 FROM: tody BUG: There is a bug in the V2.9 version of the IMDKERN graphics kernel (used to draw color graphics overlays on the image display) which can cause the kernel to die on a segmentation violation when run as a connected subkernel. STATUS: The workaround is to spool the graphics metacode for the plot in a file and then plot this file using the IMDKERN task in PLOT. The bug is fixed in the V2.9.1 patch, available from the IRAF network archive. NUMBER: 138 MODULE: images.convolve SYSTEM: V2.9 DATE: Wed Nov 28 13:38:17 MST 1990 FROM: davis BUG: CONVOLVE was terminating with the error "Kernel rows are different lengths" if the user supplied a 1D kernel without the terminating delimiter character ';'. For example the valid kernel "1.0 2.0 1.0;" would work but the equally valid "1.0 2.0 1.0" would not. 2D kernels did not have this problem. STATUS: The bug is fixed in IRAF 2.10. The workaround is to always append the delimiter character to the kernel. NUMBER: 139 MODULE: noao.artdata.mkobjects SYSTEM: V2.9-V2.9.1 DATE: Tue Dec 4 12:00:32 MST 1990 FROM: valdes BUG: Memory associated with the stellar, galaxy, and cosmic ray templates is not freed in the tasks MKOBJECTS and MKNOISE. Repeated executions (without flushing the process) eventually overflows the swap space or causes out of memory errors. STATUS: The bug is caused by a simple typo and has been fixed. Large amounts of memory are tied up only with a large number of repeated calls. Normally all objects are created in one step rather than repeated calls to add individual objects and so there is generally no difficulty. The work around when using these tasks in a loop is to add a "flprcache" call to flush the process after each call or set of calls. NUMBER: 140 MODULE: dataio.rfits SYSTEM: V2.9 DATE: Thu Dec 6 11:38:42 MST 1990 FROM: davis BUG: A command of the form "rfits mta 1-8 "" oldirafname+" will generate the message "ERROR: T_RFITS: Error reading output file name" because the code was not dealing properly with an empty output file list. STATUS: Rfits has been modified in 2.10 so that a temporary output file name is created if oldirafname is true or a clear error message is generated if oldirafname is false, and the user has set the output file name to "". The workaround is to avoid setting the output file name to "". NUMBER: 141 MODULE: identify SYSTEM: V2.9 DATE: Wed Dec 19 16:47:06 MST 1990 FROM: valdes BUG: The automatic line identification algorithm fails to find some lines in certain circumstances. The source of the problem is when multiple lines in the line list end up being centered in the same place in the data. For example if two lines in the list are 4888.1234 and 4889.1234, the second one is intrinsically weak, and the data is low resolution (say 3A per pixel) then as far as the data is concerned there is just one line. This line will be marked twice with the same position but different wavelengths. The complication is that valid weaker lines will be removed based on the maxfeatures criteria (for example only the 50 highest peak values are kept of which say 10 are multiple assignments to the same peaks in the data). Then after the lines are found the minsep criteria is applied to winnow out the multiple assignments to the same pixel leaving 40 lines found and 10 of the weaker valid lines not found. This is a complex behavior dependent on data resolution, the initial dispersion solution, the line list (the problem occurs with dense line lists used for high resolution such as the thorium list used with lower resolution data), and the maxfeatures and match parameters. STATUS: The bug fix is to winnow out the multiple identifications to the same pixel before selecting the maxfeatures strongest lines. The workaround most likely to work is to reduce the match parameter so that some of the multiple identifications are thrown out based on wavelength differences with the dispersion function estimates. Another thing is to increase maxfeatures but this will result in many undesired weak lines. NUMBER: 142 MODULE: testphot.daophot.psf SYSTEM: V2.9 DATE: Tue Feb 5 16:39:13 MST 1991 FROM: davis BUG: The task psf in testphot.daophot was writing incorrect values of the zero point position of the psf "XPSF" and "YPSF" into the psf image header. Although this was not a problem for the constant psf fitting code, the variable psf fitting code was interpolating in the wrong place in the look-up table, resulting in a very strange looking psf fit. In effect the coordinate system of the look-up table was shifted with respect to the image. STATUS: The bug has been fixed in the ftp archive version of testphot and on all our local nodes (orion, gemini, ursa). Users can either get a new version of testphot from the archive or contact the IRAF group for a patch since only one file is affected. NUMBER: 143 MODULE: images.imsurfit SYSTEM: V2.9 DATE: Mon Feb 25 10:03:39 MST 1991 FROM: davis BUG: There is a bug in the bad pixel rejection code in the IMSURFIT task which occurs when the parameter upper > 0.0 and lower = 0.0, or if lower > 0.0 and upper = 0.0. In the former case all points with negative residuals were rejected instead of none, and in the latter case all points with positive residuals were rejected instead of none. IMSURFIT was computing the rejection limits by multiplying the computed standard deviation by upper and lower respectively without checking for the zero case. STATUS: The big is fixed in IRAF 2.10. There is no workaround, except to set upper or lower to very large values if you do not want to reject pixels. NUMBER: 144 MODULE: artdata SYSTEM: V2.9.1 DATE: Mon Mar 4 09:43:28 MST 1991 FROM: valdes BUG: The tasks sometimes fail when the output image is in STF (.hhh) format. This is due to a problem in how image header comments are put in the image header affecting only the STF format. Note that the original version released with V2.9 does not provide header comments and so it does not have a problem. The newer version with this STF header problem came as part of V2.9.1. STATUS: This problem has been fixed in the next version of the package. The only workaround is to use OIF (.imh) format and then convert to STF format with IMCOPY. NUMBER: 145 MODULE: apextract SYSTEM: V2.9 DATE: Wed Mar 27 16:57:13 MST 1991 FROM: valdes BUG: For some default background functions and sample regions a singular solution error can occur. This is caused by defining the function range to be the entire image dimension while the sample region is only a small part of this range. It probably also depends on the function type and order and the degree of curvature in the fitted data. When apertures are read from the database or reset by the 'b' mode in APEDIT the fitting limits are set to the limits of the sample region. STATUS: The singular solution message should be harmless. Reading apertures from a database does not have this problem. Entering the 'b' mode in APEDIT (where the message might be seen) and exiting will fix the fitting function properly. Of course if background subtraction is not required this problem does not arise. The problem has been fixed for V2.10. NUMBER: 146 MODULE: images.geomap SYSTEM: V2.9 DATE: Tue Apr 9 14:08:45 MST 1991 FROM: davis BUG: Geomap would not permit the user to fit cross-terms (terms containing x*y) in the the x coordinate fit if xxorder=2 and xyorder=2, or in the y coordinate fit if yxorder==2 and yyorder=2. For higher order fits cross-term fitting was enabled. STATUS: This bug has been fixed in IRAF 2.10. There is no workaround. NUMBER: 147 MODULE: ccdred.mkskycor ccdred.mkillumcor SYSTEM: V2.9 DATE: Tue Apr 16 16:42:36 MST 1991 FROM: valdes BUG: The CCDMEAN parameter computed by the task MKSKYCOR and MKILLUMCOR is computed incorrectly in the sense that it is smaller than the correct value by a small amount. The last few lines of the image were not accumulated before dividing by the number of pixels in the output image. STATUS: This bug has been fixed for V2.10. One workaround is to use yboxmin of 0. The other workaround is to use IMSTATISTICS and HEDIT to compute and replace the correct value. Failing to do so and then correcting images with CCDPROC will slightly change the data level which may be acceptable. NUMBER: 148 MODULE: ecdispcor, msdispcor SYSTEM: V2.9 DATE: Mon Apr 22 13:23:09 MST 1991 FROM: valdes BUG: The combining options "sum", "average", "minmax" do not work correctly. The cause is failing to clear an array between each spectrum so that as subsequent spectra are added the preceding spectra are multiply added. STATUS: The workaround is to dispcor each order to the same dispersion with onedspec output format and then add or average the spectra with imcombine. The problem is fixed for V2.10. NUMBER: 149 MODULE: artdata.mkobjects SYSTEM: V2.9 DATE: Mon Jun 10 10:44:47 MST 1991 FROM: valdes BUG: The PSF position angle (parameter pa) is intended to be input as degrees and converted internally to radians. The conversion is not being done with the effect that the input position angle is being interpreted as radians. STATUS: The conversion has been added to the program; a simple use of the DEGTORAD macro. The workaround is to specify the position angle in radians. Note that the object position angles specified in the object list are correctly interpreted as degrees. NUMBER: 150 MODULE: dataio.wfits SYSTEM: V2.9 DATE: Tue Jun 11 13:04:44 MST 1991 FROM: davis BUG: Wfits was setting the IRAFNAME image header keyword to a blank string if the input image was a section. This was done originally to avoid rfits trying to rename the output image (if parameter oldirafname = yes) to an image section, but had the side effect of making IRAFNAME useless for book-keeping purposes. STATUS: Wfits has been modified in 2.10 to write the full image specification including the image section but minus the directory specification in the IRAFNAME keyword. Rfits has been modified to check for and remove any image section before renaming the output image to the orginal iraf name. If the renaming fails for any reason, then the name specified by the iraf_file parameter will be used as before. NUMBER: 151 MODULE: noao.twodspec.longslit.fitcoords SYSTEM: V2.6-V2.9 DATE: Fri Jul 12 08:38:03 MST 1991 FROM: valdes BUG: Fitting a single trace made along the x or horizontal axis does not work correctly. STATUS: The workaround is to transpose the data and trace the feature vertically. This is corrected in V2.10. NUMBER: 152 MODULE: images.imshift SYSTEM: V2.9 DATE: Mon Jul 29 11:42:28 MST 1991 FROM: davis BUG: Imshift was not correctly initializing the shifts file descriptor to NULL, when the shifts_file parameter was set to "", sometimes causing a later conditional test in the code to fail. This bug is triggered when a user sets shifts_file to a file name and then sets it back to "" without flushing the process cache, as may happen after repeated executions in a script. STATUS: The bug has been fixed in 2.10. There is no workaround except to flush the process cache between executions. NUMBER: 153 MODULE: digiphot.apphot.apselect SYSTEM: V2.9 DATE: Mon Aug 12 15:07:29 MST 1991 FROM: davis BUG: Apselect can sometimes fail with a segmentation violation if the input file has variable length records, (as can be the case for example if the user changes the number of apertures interactively, or if the user has defined polygons of various sizes) if the first record is not the longest record in the file, and if the size of the variable length record exceeds 20, the initial guess for buffer allocation. STATUS: The bug has been fixed in 2.10. There is no workaround except to ensure that the longest record comes first in the output file. NUMBER: 154 MODULE: proto.fixpix SYSTEM: V2.9 DATE: Tue Aug 13 12:27:17 MST 1991 FROM: valdes BUG: Interpolation across columns with type double images does not work. STATUS: Fixed in V2.10. There is no workaround other than to avoid this rare combination of datatype and direction of interpolation. NUMBER: 155 MODULE: apphot.radprof SYSTEM: V2.9 DATE: Mon Aug 26 11:25:36 MST 1991 FROM: davis BUG: 1. Radprof was computing the total intensity of the fitted radial profile by integrating (RP) instead of the physically meaningful quantity 2 * PI * (r * RP). 2. The computed total intensity (TINORM) was too small by a factor equal to the profile step size in pixels, although the shape of the curve was correct. STATUS: Both bugs have been fixed in 2.10. Bug 1 was fixed some time ago at the request of a user at ST, but was not previously logged. There is no workaround. Bug 2 was only recently discovered. The workaround is to multiply the quantity TINORM by the step size in pixels, and compare it to the aperture photometry results. The two numbers should be equal to within the expected errors. NUMBER: 156 MODULE: images.imstatistics SYSTEM: V2.9 DATE: Tue Sep 24 09:17:16 MST 1991 FROM: davis BUG: The mode statistic was not being computed correctly in imstatistics, because the parabolic interpolation correction for computing the histogram peak was being applied in the wrong direction. STATUS: The bug has been fixed in IRAF 2.10 and the help page which uses dev$pix as an example has been modified appropriately. There is no workaround. NUMBER: 157 MODULE: XC compiler SYSTEM: V2.9 Sun/IRAF, all versions DATE: Wed Oct 9 14:22:11 MST 1991 FROM: tody BUG: XC will not recognize the V1.4 Sun Fortran compiler. The bug is due to XC looking for V1.3 and finding V1.4 instead. STATUS: The workaround is to do a "ln -s f77-1.4 f77-1.3" in /usr/lang, or wherever the compiler is installed. NUMBER: 158 MODULE: dataio.rfits SYSTEM: V2.9 DATE: Wed Oct 9 15:50:36 MST 1991 FROM: davis BUG: IRAF 2.9 rfits was not reading fits files with short last logical records correctly, i.e. records not correctly padded out to 2880 bytes. For each output image line contained or partially contained in the final short record, an error message was generated, and incorrect data was written to the output image. STATUS: The bug has been fixed in 2.10 and was not a problem in IRAF 2.8 and earlier. The workaround is to use Unix dd command or the IRAF reblock task to reformat the file. NUMBER: 159 MODULE: images.imstatistics SYSTEM: V2.9 DATE: Fri Oct 11 13:54:19 MST 1991 FROM: davis BUG: The computed value of the kurtosis was 1.0 too small, i.e. a Gaussian distribution would have a computed kurtosis of -1.0 instead of 0.0. STATUS: The bug has been fixed in IRAF 2.10. The workaround is to add 1.0 to the computed value of the kurtosis. NUMBER: 160 MODULE: onedspec.standard SYSTEM: V2.9 DATE: Thu Oct 17 09:22:06 MST 1991 FROM: valdes BUG: If the calibration bandpasses are less than a pixel wide the flux summation is incorrectly done when both endpoints are within the same pixel; the code assumed that the two endpoint pixels were not the same. Generally the calibration data, in onedstds$, has much larger bandpasses than typical spectra and so this bug has not be seen except in one reported case where a private calibration file with very small bandpasses was used. STATUS: Fixed in V2.10. The workarounds are to revise the calibration file to larger bandpasses (if using a private version), resample the spectra to smaller dispersion, or, the best choice, specify the bandpass widths and separations in the task to override the calibration file bandpasses and interpolate the calibration file data. NUMBER: 161 MODULE: apextract.apsum SYSTEM: V2.9 DATE: Wed Oct 30 11:02:38 MST 1991 FROM: valdes BUG: If the background regions go off the image, due to tilts in the spectrum such as with echelle data, while the object apertures remains on the image a segmentation violation occurs. This will only happen with one-sided background regions as otherwise this condition could not occur. The error is caused by a failure to check for an error condition from background fitting. STATUS: This error is fixed in V2.10. The workaround is to insure at least part of the background regions are on the image while the object aperture is on the image. This would be true if background regions are defined on both sides of the aperture. Note that it is ok for the object aperture and background regions to both be off the the image. The source fix is simple and could be supplied if absolutely necessary. NUMBER: 162 MODULE: reidentify SYSTEM: V2.9 DATE: Mon Nov 18 10:27:25 MST 1991 FROM: valdes BUG: If reference features are identified in user coordinates other than pixels, usually Angstroms for arc lines, and the user forgets to type 'f' to fit a dispersion function, then REIDENTIFY will compute a very large coordinate shift and then fail to trace any features. STATUS: The workaround is to recognize this behavior and then go back and do a dispersion function fit for the reference. The logical bug in the task is fixed in V2.10. NUMBER: 163 MODULE: artdata.gallist SYSTEM: V2.9 DATE: Tue Nov 26 17:07:03 MST 1991 FROM: valdes BUG: The randomization of the galaxy size distribution when using the Schecter luminosity function is incorrect. Instead of a 20% range about the scaled eradius/sradius for a given magnitude a range of 100% is computed coupled with a reduction of the size by a factor of 2.5. STATUS: There is no workaround for the increased dispersion but the factor of 2.5 in the size can be compensated by adjusting the eradius parameter. This has been fixed in V2.10 NUMBER: 164 MODULE: artdata.gallist SYSTEM: V2.9 DATE: Tue Nov 26 17:11:37 MST 1991 FROM: valdes BUG: The axial ratios for elliptical galaxies when a mixture of spiral elliptical galaxies is computed are incorrect. In particular there will be axial ratio values greater than 1 in the output file. STATUS: The work around is to compute the elliptical and spiral galaxy data separately. This has been fixed in V2.10. NUMBER: 165 MODULE: plot.pvector, proto.imexamine SYSTEM: V2.9 DATE: Wed Dec 11 11:32:36 MST 1991 FROM: davis BUG: Profiles computed and plotted using the pvector task or the 'v' key in imexamine occasionally show glitches, where 1-3 pixels have bad values. This bug was traced to any error in the column limits computed by the code which reads the image pixels. As a result the image interpolator was requesting values off the ends of the pixel array producing bad output values. STATUS: The bug has been fixed in 2.10. There is no workaround. NUMBER: 166 MODULE: apphot.polyphot SYSTEM: V2.9 DATE: Tue Feb 25 16:37:41 MST 1992 FROM: davis BUG: The intersection points of an image line and a polygon could be incorrectly translated into a list of line segments if the polygon was concave and contained one or more sides exactly collinear with the image line. The symptom of this problem is a larger than expected difference (larger than can be accounted for by fractional pixel effects) in the computed area of the polygon as it is shifted to different positions in the image (shifting by floating point numbers tends to remove the condition of exact collinearity); or larger than expected differences in the area and flux of the same object measured with the same polygonal aperture at slightly different positions in different images (again shifting the polygon tends to remove the condition of exact collinearity). STATUS: The bug has been fixed in 2.10. There is no work around except to alter the shape of the polygon slightly. NUMBER: 167 MODULE: images.imstatistics SYSTEM: V2.9 DATE: Tue Mar 10 14:58:59 MST 1992 FROM: davis BUG: Precision was being lost unnecessarily in the computation of the standard deviation, skew, and kurtosis because two of the intermediate variables in the computation were real instead of double precision. STATUS: The bug is fixed in 2.10. There is no workaround. NUMBER: 168 MODULE: noao.twodspec.longslit.fluxcalib SYSTEM: V2.9 DATE: Thu Mar 19 11:27:18 MST 1992 FROM: valdes BUG: Flux calibrating a short pixel type image will produce a short pixel type output image which is generall all zeros since the typical flux values are of order 10e-14. STATUS: The workaround is to convert the images to real (with chpixtype for instance) before flux calibrating. For V2.10 the output image will be of type real pixels. NUMBER: 169 MODULE: imred.irs.batchred, imred.iids.batchred SYSTEM: V2.10 DATE: Mon Jul 6 10:01:35 MST 1992 FROM: valdes BUG: The PROCESS script generated by BATCHRED included parameters no longer part of STANDARD and CALIBRATE. Running this task produces the error ir> process ERROR on line 9: parameter `recformat' not found process () process () STATUS: To be fixed in V2.10.0. The workaround is to edit the process.cl script generated by BATCHRED and delete two cases of "recformat=yes," and one case of "apertures='',". NUMBER: 170 MODULE: apphot SYSTEM: V2.9 DATE: Wed Jul 8 17:43:45 MST 1992 FROM: davis BUG: The apphot tasks were failing to close up the coordinate files in the case that 1) the number of input images was greater than one and 2) the input coords parameter was set to "default". If the input image list was sufficiently long the task could run out of file descriptors producing a system error. STATUS: The bug is fixed in 2.10.0. The workaround is to break the image lists into smaller groups (<= ~ 60). NUMBER: 171 MODULE: apextract.apfit, apextract.apflatten SYSTEM: V2.10.0 DATE: Thu Jul 9 16:17:01 MST 1992 FROM: valdes BUG: In these tasks which output the difference, ratio, or fit based on a profile model the profile model is low by a factor of the gain. Thus if the gain is not one the results will not be correct. For example APFLATTEN will produce flat field values within the aperture which are multiplied by the gain except outside the apertures. The workarounds are either to use a gain of 1 or correct the output such as with IMARITH. The latter option may be difficult to apply for APFLATTEN since special steps need to be taken for the interaperture values. Note that using an gain of 1 even though the true gain is higher will often be unimportant, especially if cosmic ray cleaning is not done. STATUS: Fixed in V2.10.1. NUMBER: 172 MODULE: onedspec SYSTEM: V2.10.0 DATE: Mon Jul 20 14:24:07 MST 1992 FROM: valdes BUG: The ability to extract and operate on long slit images using the dispaxis/nsum parameters does not work. There is no workaround; which is particularly unfortunate since TOONEDSPEC is no longer available. One must resort to APEXTRACT to extract 1D spectra from long slit images. STATUS: Fixed in V2.10.1 NUMBER: 173 MODULE: images.gauss SYSTEM: V2.10 DATE: Thu Jul 23 08:17:27 MST 1992 FROM: davis BUG: An incorrect convolution kernel is computed in the case theta=90.0, 0.0 < ratio < 1.0, and bilinear=yes. The long axis of the kernel is aligned along the x axis instead of the y axis as intended. The workaround is to set bilinear=no. STATUS: Fixed in 2.10.1 NUMBER: 174 MODULE: onedspec.dopcor SYSTEM: V2.10.1 DATE: Thu Jul 30 16:29:31 MST 1992 FROM: valdes BUG: There is a sign error in the conversion from velocity to redshift. Thus one has to change the sign of velocities to get the effect described in the documentation. The documentation was also misleading in how to add a redshift in that one needs to complement the redshift (1/(1+z)-1) rather than reverse the sign of the redshift. The example is therefore incorrect. STATUS: The task has been modified to correct the sign error and to add consistency checks on the redshifts and velocities. The help has been improved and the erroneous example corrected. NUMBER: 175 MODULE: tv.imexamine SYSTEM: V2.9-V2.10.1 DATE: Fri Jul 31 12:06:34 MST 1992 FROM: valdes BUG: Attempting to use a single column or line at the top or left edge (high column/line numbers) of an image yields an out-of-bounds message. For some reason (typo?) the simple test for out-of-bounds had an greater than or equal to with the edge rather than a greater than. There is no workaround. STATUS: Fixed for V2.10.2 NUMBER: 176 MODULE: images.imsurfit SYSTEM: V2.9.2 and later DATE: Mon Aug 3 15:12:01 MST 1992 FROM: davis BUG: There is a typo in the imsurfit code which reads the sections file, which can result in some sections specified by the user being ignored in the fitting process. Whether or not the code executes correctly depends on section specifications themselves since the intended test "if (y2 < y1)" was actually coded as "if (y2 < x1)". STATUS: The bug is fixed in 2.10.1. There is no workaround. NUMBER: 177 MODULE: images.imcombine, ccdred.combine, onedspec.scombine SYSTEM: V2.10-V2.10.1 DATE: Wed Aug 12 10:30:19 MST 1992 FROM: valdes BUG: The minmax rejection option when used with a final average and with weights does not correct for the rejected pixels in the weights normalization. For example with four images having identical weights and rejecting the high and low the final output is the sum of the middle pixels divided by 4 instead of 2. There is no workaround. If the weights are equal the end result is just that the mean intensity levels are low by a constant factor. However, when the weights are not equal the final result will have pixel position dependent effects. STATUS: Fixed for V2.10.2 NUMBER: 178 MODULE: images.fmedian SYSTEM: V2.9 DATE: Tue Aug 18 11:31:49 MST 1992 FROM: davis BUG: The fmedian task can crash with a segmentation violation or a floating operand error if image pixel to integer mapping is turned off (hmin = zmin and hmax = zmax) and the input image contains data outside the range defined by zmin and zmax. The bug is not seen often because for most ccd images the data is between -32768 and 32767. STATUS: The bug is fixed in 2.10.1. The work-around is to either let zmin and zmax default to the image minimum and maximum or set hmin and hmax to numbers that are slightly different from hmin and hmax. NUMBER: 179 MODULE: daophot.allstar SYSTEM: V2.10 DATE: Wed Sep 16 14:37:46 MST 1992 FROM: davis BUG: If 1) cache=no, or cache=yes and memory preallocation fails for one or more of the data, scratch or weight arrays, 2) one or more of the groups has greater than maxgroup stars, 3) regrouping is performed on the too large group and 4) the first star in the next non-regrouped group has a y value less than the y value of the first star in one of the regrouped groups, allstar can fail with an "attempt to access the scratch, weight, or data pixels randomly" error. STATUS: The bug is fixed in 2.10.2. Possible work arounds include: 1) setting cache=yes instead of no if the machine has enough memory and swap space 2) setting the maxgroup parameter to a larger value than the default value of 60 3) decreasing the fitting radius somewhat since this affects the grouping process. NUMBER: 180 MODULE: identify/reidentify SYSTEM: V2.10.1 DATE: Wed Sep 16 16:24:05 MST 1992 FROM: valdes BUG: If a spectrum is shifted using IMSHIFT with a positive shift then the physical pixel coordinates are incorrectly computed. The workaround is to reset the WCS with WCSRESET. STATUS: Fixed in V2.10.2. NUMBER: 181 MODULE: ccdred.ccdproc SYSTEM: V2.9(?) V2.10.1 DATE: Tue Sep 22 08:40:14 MST 1992 FROM: valdes BUG: When using STF images with a flat field that has been processed but has lost the CCDMEAN keyword CCDPROC will produce an error of the form: ERROR: OPEN: File does not exist (tmp4294ha.hhd) A flat field in this state would most likely be produced by FLATCOMBINE with preprocessing before combining selected. The combined flat field will lack the CCDMEAN keyword. The workarounds are to either use OIF images or add the CCDMEAN keyword manually. This would be done by using IMSTAT to compute the mean and adding the CCDMEAN keyword with that value using HEDIT. STATUS: The error is ignored by CCDPROC in V2.10.2. The reason for the error when calling imunmap with an image which has been created but for which no pixel data has been written is under investigation. NUMBER: 182 MODULE: apnormalize SYSTEM: V2.10.1 DATE: Tue Sep 22 13:40:26 MST 1992 FROM: valdes BUG: The parameter t_nlost is missing from the parameter file apnorm1.par. When attempting to run APNORMALIZE there will be an error message concerning this parameter. It can be fixed if desired by adding the following line after line 63 in the file iraf$noao/twodspec/apextract/apnorm1.par: t_nlost,i,h,)aptrace.nlost,,,>aptrace.nlost After adding this parameter one must "unlearn apnorm1". There is no workaround other than to add the parameter. STATUS: Fixed in V2.10.2 NUMBER: 183 MODULE: doargus, dohydra, dofibers, skysub SYSTEM: Newimred, V2.10.0, V2.10.1 DATE: Wed Sep 23 15:57:40 MST 1992 FROM: valdes BUG: When using the skyedit option to review and eliminate bad sky apertures there is a limit to the number of sky apertures that can be retained. Exceeding this limit can cause the CL to crash. The limit is determined by how many aperture numbers can be fit into 160 characters in a string of the form 11,12,13,...; for 2 digit aperture numbers this would be a little over 50 sky apertures. The workarounds are to use less than about 50 sky apertures or to not use the sky edit option. STATUS: Planned to be fixed in V2.10.2. NUMBER: 184 MODULE: splot SYSTEM: V2.10.1 DATE: Thu Sep 24 08:48:57 MST 1992 FROM: valdes BUG: The '(' and ')' keys for scrolling through multiple spectra images does not work properly in cases where the aperture numbers are not the same as the line number. The only workaround is to use the explicit '#' key to access particular apertures. STATUS: Fixed in V2.10.2 NUMBER: 185 MODULE: imcombine, ccdred.combine, onedspec.scombine SYSTEM: V2.9 DATE: Thu Oct 1 17:06:23 MST 1992 FROM: valdes BUG: When using the CCDCLIP or CRRJECT rejection option with mclip=yes (which is the default) a minimum of 3 images/spectra is required. This means that two image/spectrum cosmic ray rejection doesn't work as intended. The workaround is to use mclip=no. This is equivalent for two images/spectra since the median and average are the same in this case. STATUS: Fixed in V2.10.2. NUMBER: 186 MODULE: ccdred.ccdproc SYSTEM: V2.9-V2.10.1 DATE: Mon Oct 26 11:04:21 MST 1992 FROM: valdes BUG: If an image is processed which has an unknown CCD type it will have the minimum replace operation applied to it. This operation is only supposed to apply to flat field types. The workarounds are to setup/define a translation allowing the CCD types to be determined or to set the minreplace parameter to a very large negative value to avoid limiting the data. Note that this will only apply to data which has low or negative values in the first place. STATUS: Fixed for V2.10.2. NUMBER: 187 MODULE: echelle.dofoe SYSTEM: V2.9 DATE: Thu Oct 29 10:32:12 MST 1992 FROM: valdes BUG: When there is a second arc it is incorrectly extracted with a .ms extension instead of a .ec extension causing downstream processing to fail. The workaround is to rename the file to .ec and then start again. STATUS: Fixed in V2.10.2. NUMBER: 188 MODULE: longslit.fitcoords SYSTEM: V2.6-2.10.1 DATE: Wed Nov 11 10:43:43 MST 1992 FROM: valdes BUG: FITCOORDS is supposed to ignore features which have an INDEF for the line identification. This was not happening but instead some random value was substituted for the INDEF. The workaround is to avoid creating features with INDEF in IDENTIFY. STATUS: In V2.10.2 any features with INDEF line ids will be ignored. NUMBER: 189 MODULE: onedspec.splot/specplot SYSTEM: V2.10-V2.10.2 DATE: Fri Dec 4 09:41:16 MST 1992 FROM: valdes BUG: The units conversion to millimeters and centimeters is off by a factor of 10 too large. The inverse function on these units will be off by a corresponding amount. STATUS: Fixed for the next release. NUMBER: 190 MODULE: longslit.fitcoords SYSTEM: V2.10-V2.10.2 DATE: Mon Dec 7 11:05:09 MST 1992 FROM: valdes BUG: If "combine=yes" and the input images do not have the same fit axis a segmentation error occurs. The task is supposed to ignore all images which don't agree with the first image in the fit axis but there is a bug where it attempts to clean up for a skipped image using a NULL pointer. The workaround is to check that all the images being combined have the same fit axis. Note that a likely cause of this error is to inadvertently have combine=yes though what is really desired is combine=no to fit all the inputs separately in which case both fit axes can be done in one execution. STATUS: It is fixed and a warning is now printed when mismatched fit axes are encountered indicating which input image(s) are ignored. Currently expected to be released with V2.10.3. NUMBER: 191 MODULE: apextract.apscatter (and in related packages) SYSTEM: V2.10-V2.10.2 DATE: Tue Dec 8 17:24:34 MST 1992 FROM: valdes BUG: If interactive=no then the smooth parameter is ignored and the scattered light subtraction will not apply the smoothing along the dispersion. There is no way around this. The only suggested partial workaround is to leave interactive=yes, set fitsmooth=no and fitscatter=no, and then answer NO for the queries. With a list of images this will then process all further images with no queries. STATUS: Fixed for the next release of the APEXTRACT package (expected to be V2.10.3). NUMBER: 192 MODULE: photcal.fitparams SYSTEM: V2.10 DATE: Thu Dec 17 17:22:04 MST 1992 FROM: davis BUG: The fitparams task could produce incorrect weight, chi, and fitted parameter error estimates in the case that weighting=photometric, nreject > 1, and low_reject > 3.0 || high_reject > 3.0. This problem was occuring because the weight array was not being reinitialized correctly after each iteration of the rejection cycle. The actual problem was in the inlfit package routine inlfit$inreject.gx. STATUS: The bug is fixed for the next release of IRAF. The work around is to use interactive bad point deletion with the graphics cursor instead of the automatic iterative rejection algorithm. NUMBER: 193 MODULE: sensfunc SYSTEM: V2.9-2.10.2 DATE: Mon Dec 21 13:01:36 MST 1992 FROM: valdes BUG: SENSFUNC will not work when the standards file has a starting wavelength greater than the ending wavelength. The workarounds are to linearize the spectra with an increasing wavelength, flip the spectra in STANDARD (i.e. standard spec[-*]), or edit the standards file to reverse the starting and ending wavelengths. In the latter two cases the sensitivity function will then have a reversed sense from the data but the calibration will still correctly match wavelengths. STATUS: It is the intention that no spectral task care which way the wavelength runs relative to the pixels. SENSFUNC has been modified to work in either case for the next release. NUMBER: 194 MODULE: splot SYSTEM: V2.10.2 DATE: Thu Dec 31 10:34:52 MST 1992 FROM: valdes BUG: The 'e' key in SPLOT either produces a floating operand error or a nonsensical equivalent width when the data values are less than about 1e-10 such as is the case with flux calibrated spectra. This problem was introduced due to a change in a system routine. The workaround is to scale the data, say with IMARITH or SARITH, to yield pixel values near unity. The equivalent width and wavelengths will be independent of the scaling and the other flux quantities will simply have the same scale factor. STATUS: Fixed for the next release. NUMBER: 195 MODULE: daophot.allstar SYSTEM: V2.10 DATE: Sun Jan 3 11:04:59 MST 1993 FROM: davis BUG: In crowded regions allstar could occasionally refuse to 1) fit a bright star or group of bright stars, or 2) fail to converge to reasonable x,y, and magnitude 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 and large chi values. The problem was caused by a bug in the code which steps through the stellar groups, subtracting off the current best fit for all the stars, to produce a residuals image. Due to this bug, on occasion 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 are used to control the bad data rejection algorithm, allstar sometimes refused to fit stars because the residuals were too big or could not converge to a reasonable value. This bug is data dependent but is most likely to be a problem if, 1) the stellar detection threshold is very low, 2) allstar has to do a lot of regrouping to get the groups below maxgroup in size 3) the fitting radius is large resulting in very large groups. STATUS: This bug is only a problem in 2.10 (not in the external testphot package) and has been fixed for the next release of IRAF. There is no reliable workaround for allstar, but the nstar task which fits fixed groups does not have this problem. NUMBER: 196 MODULE: splot SYSTEM: V2.10.2 DATE: Fri Jan 8 16:39:01 MST 1993 FROM: valdes BUG: Use of "zero" in the option parameter (or ymin=0.) or the 'b' key with flux calibrated data causes a plot with limits -0.01 and 0.01 making the data appear all zero. The workaround is to set the minimum value to a number much smaller than the data but not exactly zero. For example setting ymin=1e-30 when the data is around 1e-14. This plot limit behavior applies to all tasks (GRAPH, IMPLOT) but is most likely to appear in SPLOT with flux calibrated data. STATUS: A change to a system routine is required and has been requested. NUMBER: 197 MODULE: images.imcombine, ccdred.combine (and related scripts) SYSTEM: V2.10.2 DATE: Wed Jan 20 15:54:28 MST 1993 FROM: valdes BUG: When reject=[avsigclip|sigclip|ccdclip|crreject|pclip], combine=average, mclip=yes, and nkeep!=0 there is a rare possiblity of a segmentation error. This is more likely the fewer pixels are allowed to be rejected. This occurs in the step where the pixels with the lowest residuals are added back after too many are rejected and the pattern of rejections is such that only low pixels have been rejected. The sure workaround is to change the combine option to a median or mclip=no. However, changing other parameters such as nkeep, the sigma thresholds, or the combining algorithm is likely to avoid the problem. STATUS: Fixed for V2.10.3. NUMBER: 198 MODULE: images.imcombine SYSTEM: V2.10.2 DATE: Wed Jan 20 16:30:10 MST 1993 FROM: valdes BUG: The CCDCLIP and CRREJECT algorithms quietly fail to reject any pixels. The basic source code is actually correct but due to a dependency condition based on file dates the wrong derived code wes used resulting in a mismatched argument list. This mismatch generally results in the rejection algorithm seeing that there are zero images to process and so no rejection is done. The workarounds are to use a different rejection option or the CCDRED.COMBINE version which is the same except for the ccdtype, keyword mapping, and filter subset features. STATUS: Fixed in V2.10.3. NUMBER: 199 MODULE: tv.wcslab SYSTEM: V2.10 DATE: Thu Jan 28 11:10:51 MST 1993 FROM: davis BUG: Wcslab was failing with the error "error processing newline" for images which were sections of a higher dimensioned parent images, and whose coordinate system was "physical" or not recognized recognized by wcslab, e.g. glon,glat. The actual error message was coming from the cl and was due to a trailing \n in the string passed to the error trapping routine. The actual problem was due to an error in the wcslab axis mapping code which was failing to take proper account of an existing axis map. STATUS: The bug is fixed in 2.10.3. A possible workaround is to temporarily set the wcsdim parameter in the image header to be equal to the dimension of the image. NUMBER: 200 MODULE: astutil.rvcorrect SYSTEM: V2.10-V2.10.2 DATE: Mon Feb 1 10:31:22 MST 1993 FROM: valdes BUG: When using RVCORRECT with a list of input images that don't have the OBSERVAT keyword, the observatory name supplied by the user in the parameter file is truncated to two characters. An error message is printed and a prompt is issued for another name. Replying to this prompt allows the task to complete; i.e. the reply to the prompt is not truncated. This is caused by a typo in RVCORRECT which reads only the first two characters from the observatory parameter. The workarounds are to either enter the proper observatory when prompted or set the observatory in the image headers under the keyword OBSERVAT. STATUS: Fixed for the next release (currently expected to be V2.10.3). NUMBER: 201 MODULE: identify SYSTEM: V2.10-V2.10.2 DATE: Wed Feb 3 16:49:36 MST 1993 FROM: valdes BUG: The 'i' initialize key fails to initialize the wavelength scale. STATUS: Fixed in V2.10.3. NUMBER: 202 MODULE: apscatter, doecslit SYSTEM: V2.10-V2.10.2 DATE: Fri Feb 5 15:25:52 MST 1993 FROM: valdes BUG: When using the APSCATTER option (either directly or in the DOECSLIT script) without specifying an output name instead of a temporary image name being used for intermediate results the name ".imh" is used. This is a hidden file and if the task aborts or is aborted it will be left behind and prevent any further use of this task. The workaround is to recognize that this is why it fails to work and delete the image ".imh". STATUS: Fixed in V2.10.3. NUMBER: 203 MODULE: sarith, scombine, splot SYSTEM: V2.10-V2.10.2 DATE: Wed Feb 10 10:55:11 MST 1993 FROM: valdes BUG: In certain circumstance (listed below) when the dispersion coordinates are inverted to get pixel coordinates the wrong pixel coordinates are obtained. In practice this is important only when resampling spectra such as in arithmetic or combining between two spectra as might occur with SARITH, SCOMBINE, or the function mode in SPLOT. It also can cause a problem with the SPLOT equivalent width options (giving a floating divide by zero). This problem does NOT apply to DISPCOR! The conditions under which this problem occurs are when all the following apply: 1. Multispec coordinate system images with more than one spectrum. 2. The spectra have different coordinate systems; i.e. are not all at a common coordinate system. 3. The spectrum aperture numbers differ from the line numbers; for example line 1 is aperture 3 and line 3 is aperture 1. Note that if an image section is used this will change the line number so even if the apertures are the same as the line numbers in the original image this will not be true with an image section. In many applications DISPCOR is used to linearize all spectra to a common dispersion system. This problem does not apply to such data. It is likely to affect echelle spectra and users who use nonlinear dispersion coordinates. The workarounds are to assign aperture numbers sequentially increasing with pixel coordinate when extracting, renumber the apertures with the renumber option in SCOPY, dispersion correct to a common system with DISPCOR, separate the spectra into ONEDSPEC images before doing the arithmetic/combining, or avoid arithmetic/combining operations. What is happening is that the coordinate transformations between pixel and world coordinates are (x,l) <-> (w, a) where x is the pixel image coordinates along the dispersion, l is the line number, w is the wavelength, and a is the aperture number. Note that the image line maps to an aperture and vice-versa. The bug is that instead of using the aperture number the inverse transformation being used is (w,l) -> (x,?). So unless the line number happens to be the same as the aperture number or the dispersion functions are the same for all apertures the dispersion function from the wrong spectrum is used for the inversion. STATUS: This will be fixed in V2.10.3. NUMBER: 204 MODULE: daophot SYSTEM: V2.10 DATE: Tue Feb 16 09:08:48 MST 1993 FROM: davis BUG: The parameters psfrad, fitrad, and matchread were being written to the daopars parameter set in pixel units instead of scale units if the daophot task update parameter was on. STATUS: Fixed for the next release of iraf. There is no workaround except to turn off the update parameter or always check the input parameters using the verify parameter. NUMBER: 205 MODULE: imslice SYSTEM: V2.9 DATE: Tue Feb 16 13:52:34 MST 1993 FROM: davis BUG: Removed a check that was preventing imslice from being used to reduce the dimensionality of images where the length of the slice_dimension = 1. STATUS: Fixed for the next version of iraf. The work around is to use imcopy. NUMBER: 206 MODULE: photcal.obsfile SYSTEM: V2.10 DATE: Mon Mar 1 15:18:47 MST 1993 FROM: davis BUG: The obsfile task was not decoding the image names correctly from the obsparams file, when obsfile was being called directly from the cl rather than from the script tasks mknobsfile and mkobsfile. Since obsfile could not match the names of the images in obsparams with those in imsets it simply ignored the contents of the obsparams file. STATUS: The bug is fixed for the next release of iraf. There is no workaround. NUMBER: 207 MODULE: splot SYSTEM: V2.10-V2.10.2 DATE: Tue Mar 2 14:47:45 MST 1993 FROM: valdes BUG: When doing arithmetic between two spectra and an invalid image name is entered for the second spectrum , usually because of a typo, the correct error message is returned. However, attempting to repeat the operation, say with the correct image name, results in a segmentation error and the task aborts. This will continue to happen until the task is flushed from the process cache. The workaround is simple to flush the processes cache with the command "flpr splot". STATUS: Fixed for the next release (V2.10.3) NUMBER: 208 MODULE: adccdrom.catalog SYSTEM: adccdrom external package: archive prior to March 4, 1993 DATE: Thu Mar 4 16:03:16 MST 1993 FROM: valdes BUG: Use of the "evsptype" function results in a segmenation or bus error. This was caused by using a wrong pointer datatype. There is no workaround other than to not use this function. However the external package archive was updated. STATUS: Fixed in the archive version of March 4, 1993. NUMBER: 209 MODULE: doslit, doecslit SYSTEM: V2.10-V2.10.2 DATE: Wed Mar 17 11:10:30 MST 1993 FROM: valdes BUG: The input list of arcs is sorted alphabetically and first sorted arc is used as the arc reference rather than the first specified by the user. The workaround is to rename the desired reference arc to be first alphabetically. STATUS: Fixed in V2.10.3. NUMBER: 210 MODULE: photcal.mknobsfile,photcal.mkobsfile,photcal.obsfile SYSTEM: V2.10 DATE: Fri Apr 2 11:45:11 MST 1993 FROM: davis BUG: The mknobsfile,mkobsfile, and obsfile task were occasionally dropping stars from the output observations file in the case that: 1) the parameter allfilters=no, 2) there were multiple matches to a given star in 1 or more filters, and 3) the closet matching star was not the first candidate star found. This bug was occuring because the code was correctly matching to the nearest of the possible candidate stars, but was not resetting the matching index from the previous star, confusing the indexing scheme, and causing the incorrectly matched star to be dropped from the list. STATUS: The bug is fixed in 2.10.3. There is no fullproof workaround but accurately determining the shifts between frames and using small matching tolerances will minimize the problem. NUMBER: 211 MODULE: longslit.fitcoords SYSTEM: V2.10 - V2.10.2 DATE: Mon Apr 5 10:01:12 MST 1993 FROM: valdes BUG: When there are many samples to be fit an 'Out of memory" error can occur. This is caused by repeatedly opening an image section for each sample and failing to close it. If there are enough samples this can cause memory to run out. If this happens on other than the first invocation of FITCOORDS then the workaround is to flush the process cache between executions, "flpr fitcoords". If it happens on the first execution then the only workaround is to reduce the amount of data being fit by using a larger step size, and avoiding multiple runs of REIDENTIFY (since redoing an IDENTIFY appends to the database rather than replacing resulting in multiple entries). For marginal cases it might be possible to make sure the process has the maximum amount of memory available by minimizing memory use by other processes. STATUS: This is a rather severe bug which is fixed in V2.10.3. NUMBER: 212 MODULE: scombine SYSTEM: V2.10-V2.10.2 DATE: Thu Apr 22 09:54:07 MST 1993 FROM: valdes BUG: When combining spectra which have log sampling (DC-FLAG=1) and one chooses log=yes the dispersion per pixel parameter is incorrectly computed in the output. The workaround is to use the first=yes option. STATUS: Fixed in V2.10.3 NUMBER: 213 MODULE: images.imcombine ccdred.combine SYSTEM: V2.9-V2.10.2 DATE: Mon May 10 14:57:05 MST 1993 FROM: valdes BUG: When using a median combining operation, no rejection or with mclip=no, and outputing a sigma image, the sigma image is incorrect. This is due to the median computation destroying some of the input data values before the sigma is computed. The workaround when both output median and sigma images are desired is to used one the rejection options with mclip=yes. By choosing the rejection option sigclip with low and high sigma thresholds of 1000 no pixels will be rejected and the results are the same as not specifying a rejection algorithm. STATUS: Fixed in V2.10.3. NUMBER: 214 MODULE: scopy, sarith SYSTEM: V2.10-V2.10.2 DATE: Thu May 20 17:07:53 MST 1993 FROM: valdes BUG: When using SCOPY or SARITH with a specified wavelength range, w1 and w2 not equal to INDEF, and rebin=yes the spectra are correctly resampled to the requested wavelength interval but the coordinate system in the header is updated incorrectly. The coordinates may be off by a fraction of a pixel. For example if w1=6400 and w2=6600 the output data will have the first pixel at 6400 but the coordinate reported by LISPIXEL or any other task might be 6400.25 (where the dispersion is, say, 3 Angstroms per pixel). This error is caused during the conversion from logical to physical pixel coordinates when the physical coordinates are treated as integer; i.e. a wavelength of 6400 for the first logical pixel might be physical pixel 185.325 but it is then truncated to 185 and the coordinate system is set so that physical pixel 185 is 6400 and pixel 185.325 is 6400.25. This is a subtle error which may be discovered by looking at the coordinate of the first pixel with LISTPIX and comparing it with the requested wavelength of the first pixel. STATUS: Fixed in V2.10.3. NUMBER: 215 MODULE: imcombine, ccdred.combine SYSTEM: V2.10-V2.10.2 DATE: Wed May 26 14:06:15 MST 1993 FROM: valdes BUG: The following combination of parameters give a floating divide by zero due to an error in the code. Rejection is CRRJECT or CCDCLIP, grow is greater than zero, and the images are offset. The simplest workaround is to not use a growing factor (grow=0). STATUS: Fixed in V2.10.3 NUMBER: 216 MODULE: noao.astutil.asttimes SYSTEM: V2.9 DATE: Thu May 27 15:01:18 MST 1993 FROM: valdes BUG: The epoch printed is not clearly defined and the siderial time is incorrect by a few seconds. There is also a gross error at the new year for observatories with a negative time zone. The epoch used is the day of the year, including the fraction of the day, divided by a Julian year of 365.25 days. However the siderial time calculation assumes a J2000 Julian epoch which is not the same as the previous definition. The previous epoch definition is also no smoothly continuous at the new year since some years have 366 days and other 365 and no years have 365.25 days. Other tasks in the ASTUTIL package also use the astronomical time routines but it appears that generally the error cancels out (because the date->epoch->julian date transformations use the same epoch definition) or is insignificant (such as in precession or air mass calculations). STATUS: For V2.10.3 the routines have been modified to consistently use J2000 epochs. ASTTIMES will print these epochs and the sideral times will not have the errors of a few seconds. NUMBER: 217 MODULE: identify SYSTEM: V2.9 DATE: Fri Jun 4 10:00:06 MST 1993 FROM: valdes BUG: The labels associated with marked features are not correctly updated when a feature is deleted. The effect is that the wrong labels get associated with the features after deletion. There is no workaround though the labels are informational so the main job of determining dispersion functions is unaffected. STATUS: Fixed in V2.10.3. NUMBER: 218 MODULE: fitprofs SYSTEM: V2.9 DATE: Tue Jun 8 09:11:06 MST 1993 FROM: valdes BUG: When the dispersion axis is 2 and the number of lines is greater than the number of columns a floating operand error can occur. This is caused by dimensioning an array by the number of columns rather than the number of elements along the dispersion axis. The workaround is to either use an image section to make the number of lines equal to the number of columns or to transpose the image. Transposing the image also requires reseting some of the WCS keywords. STATUS: Fixed in V2.10.3. NUMBER: 219 MODULE: images.geomap SYSTEM: V2.9 DATE: Wed Jul 7 11:04:31 MST 1993 FROM: davis BUG: If the input coordinate file contains fewer than 3 stars geomap dies with a segmentation violation. The problem was caused by some missing errchk statements in the error handling code, with the result that geomap tried to evaluate a non-existent solution. STATUS: The problem has been fixed in 2.10.3. NUMBER: 220 MODULE: ptools.pconvert SYSTEM: V2.9 DATE: Wed Jul 28 08:34:39 MST 1993 FROM: davis BUG: In append mode the pconvert task was refusing to append data to columns derived from multi-valued quantities in the original apphot/daophot text output, e.g. MAG[1]. The problem was due to an extraneous escape character "\" infront of the "[" in the column specification (this is required for column template expansion, but not for column definition). STATUS: Fixed in 2.10.3. There is no work-around. NUMBER: 221 MODULE: images.imcombine, ccdred.combine SYSTEM: V2.9-V2.10.2 DATE: Wed Aug 4 15:48:03 MST 1993 FROM: valdes BUG: When the number of input images + number of output images + logfile (if used) is exactly 60 the task fails with error 757. For example if there are 59 input images, an output image, and no logfile this error occurs. The workaround is to add or eliminate the logfile or else change the number of input or output images. STATUS: Fixed in V2.10.3. NUMBER: 222 MODULE: images.fit1d, generic.background, longslit.background SYSTEM: V2.10-V2.10.2 DATE: Fri Aug 6 10:36:36 MST 1993 FROM: valdes BUG: When the input and output images are the same the last part of the image will not be modified. The workaround is to use a different output image and then, if desired, delete the original and rename the new image to the original image. STATUS: Fixed for V2.10.3. NUMBER: 223 MODULE: photcal.fitparams SYSTEM: V2.10 DATE: Sun Aug 8 15:05:17 MST 1993 FROM: davis BUG: Errors in the catalog and observations variables were not being added to the total error computed for an observation by the fitparams task if these variables were used in a set equation, the set equation was then used in one or more of the transformation equations, and the weighting parameter was set to photerrors. STATUS: The bug is fixed in iraf 2.10.3. Possible workarounds include use of the weighting=equations option and avoiding the use of the set equations in transformations equations. NUMBER: 224 MODULE: apnormalize SYSTEM: V2.10 - V2.10.2 DATE: Fri Aug 27 18:40:27 MST 1993 FROM: valdes BUG: When the dispersion axis is 1 and the "cenorm" option is selected the result is not correct. One workaround would be to transpose the image and the other is to not use the "cenorm" option. STATUS: Fixed in V2.10.3 NUMBER: 225 MODULE: apfit, apnormalize SYSTEM: V2.10 - V2.10.2 DATE: Fri Aug 27 18:42:14 MST 1993 FROM: valdes BUG: The results of the "fit" and "difference" in APFIT and of APNORMALIZE are incorrect by a factor of the gain. This is basically the same bug as 171 which was not correctly fixed. The workarounds are to use a gain of 1. STATUS: Fixed in V2.10.3 NUMBER: 226 MODULE: artdata.mkobjects SYSTEM: V2.9-V2.10.2 DATE: Fri Sep 17 14:13:24 MST 1993 FROM: valdes BUG: The scales of the "expdisk" and "devauc" images are incorrect due to an integer truncation error. The error is such that the actual output radius for a given input radius is given by the formulas: expdisk: rout = rin * int(ln(D)/1.6783) / ln(D)/1.6783 devauc: rout = rin * int(ln(D)/7.67)**4) / (ln(D)/7.67)**4 where D is the dynamic range parameter in the ARTDATA package parameters, ln() is the natural logarithm, int() is the integer truncation function, and ** is the exponentiation operator. From this one can, fudge the input radii to give the desired output radii, select a dynamic range where the correction factor is 1, or account for this as needed when the artificial image is analyzed. The simplest thing is to adjust the dynamic range; for example, a dynamic range of 126514 for the "expdisk" model is the closest to the default value of 10000 which will give correct output scales. STATUS: Fixed in V2.10.3 NUMBER: 227 MODULE: longslit.transform SYSTEM: V2.8-V2.10.2 DATE: Sun Sep 19 14:21:35 MST 1993 FROM: valdes BUG: If the requested output coordinates extend outside of the input data range, the interpolated output data very near the edge may suffer a small stretch distortion. The error is confined to the end 50th of the image and the magnitude of the stretch depends on the specifics of the requested output sampling. The workaround is to either creating an output image which extends beyond the input data range. STATUS: Fixed in V2.10.3 NUMBER: 228 MODULE: images.imsurfit SYSTEM: V2.9 DATE: Wed Sep 22 13:29:23 MST 1993 FROM: davis BUG: Bugs in the bad pixel rejection code were causing the imsurfit task to either 1) hang or crash with a segmentation violation or 2) display oscillating behavior (sigmas which would decrease then increase again after succesive iterations), if the number of detected bad pixels was very large. The first problem was caused by a failure of the code to allocate sufficient space for the bad pixel list. The second problem was caused by a failure of the code to correctly resolve overlapping regions in some circumstances, causing the same pixels to be alternately rejected and included in the fit. STATUS: Fixed in 2.10.3. NUMBER: 229 MODULE: images.imcombine, ccdred.combine SYSTEM: V2.9 DATE: Fri Sep 24 10:40:31 MST 1993 FROM: valdes BUG: If an offset file is specified as containing absolute offsets but with all the offsets equal then the combining acted like relative offsets except the output image size would be that of the absolute offsets and garbage appears in the final lines/columns. The workaround is to either not have identical offsets or combine the images without offsets and then paste the result into a bigger empty image at the desired offset with IMCOPY. STATUS: Fixed in V2.10.3. NUMBER: 230 MODULE: images.imcombine, ccdred.combine SYSTEM: V2.9 DATE: Fri Sep 24 10:45:00 MST 1993 FROM: valdes BUG: The documentation says that scale and zero offset factors specified by an @file of header keyword are multiplicative and additive. However they are used as divisive and subtractive. The workaround is to know the values should be input as divisive and subtractive. STATUS: In V2.10.3 this has been fixed so the input is as defined in the documentation. NUMBER: 231 MODULE: imcombine, ccdred.combine SYSTEM: V2.10.2 DATE: Tue Nov 9 09:17:12 MST 1993 FROM: valdes BUG: If there are fewer pixels at some point than that specified by nkeep parameter, such as caused by offsetting, thresholding, or masking, then the tasks try to add data from the nonexistent images resulting either in garbage or a segmentation violation. The workaround is to use nkeep or 1 or 0 rather a higher value unless there are no offsets or other factors that will result in a smaller number of points than the number of images at some points. STATUS: This has been fixed for V2.10.3. This also resulted in a change in how a negative nkeep is used. In the new version, if one specifies a maximum number to reject, that number applies to the initial set of pixels rather than to the total number of images. NUMBER: 232 MODULE: images.geomap SYSTEM: V2.9 DATE: Wed Dec 29 16:39:24 MST 1993 FROM: davis BUG: Geomap was computing an incorrect coordinate transformation if the functional form of the coordinate transformation was "chebyshev" or "legendre" x fit x or y order was > than 2 and the y fit was linear (x and y order equal to 2), or if the y fit x or y order was > 2 and the x fit was linear (x and y order equal to 2). This was occurring because geomap was incorrectly redefining the functional form of the linear fit to be "polynomial". STATUS: Fixed in 2.10.3. There is no direct workaround but setting the cross terms parameter of the linear fit to "yes" will avoid the problem at the expense of computing one extra term in the fit. NUMBER: 233 MODULE: apphot package tasks SYSTEM: V2.9 DATE: Fri Dec 31 14:46:23 MST 1993 FROM: davis BUG: If the number of input images is greater than one, the number of output files is exactly equal to one, and the input coordinate file(s) do not contain any decodable coordinates, the input coordinate file rather than the empty output file is deleted. The problem occurs because the output file name is not being correctly stored as the tasks looped through the input image list. STATUS: The bug is fixed in 2.10.3. The workaround is to make sure at least one of the coordinate files contains valid coordinates. NUMBER: 234 MODULE: onedspec.standard SYSTEM: V2.10-V2.10.2 DATE: Fri Jan 7 09:38:51 MST 1994 FROM: valdes BUG: When using STANDARD on a long slit spectrum with a dispersion axis of 2 (where each column is treated as a spectrum) the task uses the length of the second axis rather than the first axis to define the number of spectra. If the second axis length is greater than the first axis then the last column is processed repeatedly. One may limit this by using the apertures parameter to specify just the columns; i.e. 1-250 in a 250x800 images. If the number is smaller then some of the columns will not be processed. On workaround is to transpose the image but then some fudging of the WCS will be necessary. Another workaround would be to use SCOPY to extract all the spectra to a multispec format. STATUS: Fixed for the next release. NUMBER: 235 MODULE: digiphot.apphot.fitpsf SYSTEM: V2.9 DATE: Mon Jan 17 14:18:33 MST 1994 FROM: davis BUG: In cases where the fitting box was entirely off the image or there were too few pixels in the fitting box to fit the psf model (e.g. < 5 in the case of function=radgauss), the output fitted parameter and parameter error arrays were not being initialized correctly. Instead, although the fit was correctly tagged with the appropriate error code, the results of the previous fit were written to the output file. STATUS: Fixed in 2.10.3. There is no workaround except to make sure that the fitting box contains enough pixels and/or to check the error codes of the fitted objects. NUMBER: 236 MODULE: digiphot.photcal.invertfit SYSTEM: V2.10 DATE: Wed Jan 19 08:03:42 MST 1994 FROM: davis BUG: On VMS systems the invertfit task will crash with an adjustable array dimension error if the user has no defined set equations in his/her configuration file. In this case invertfit passes a 2D array with a zero length first dimension to a subroutine, triggering a run time error under VMS, even though the array is never accessed. STATUS: Fixed in 2.10.3. The workaround is to include a dummy set equation in the transformation section of the configuration file. NUMBER: 237 MODULE: IEEE NaN support SYSTEM: Sun/IRAF V2.10, V2.10.1, V2.10.2 DATE: Fri Jan 21 15:55:46 MST 1994 FROM: tody BUG: When converting native floating point to IEEE floating point (as when writing FITS files), V2.10 IRAF has a feature which allows pixels with a reserved floating point value to be replaced by IEEE NaN on output. In V2.10 Sun/IRAF, the code which does this was correct, but the wrong version of the affected object module was present in the system library (libvops.a) and in the IRAF shared image (S6.e). This would cause zero to be written as the output IEEE value instead of NaN. STATUS: Fixed in V2.10.3 or later. So far as I know the core IRAF system does not use this feature but it is used by some layered packages, e.g., STWFITS. The bug only comes into play when mapping of NaNs on output is enabled by the application being used. This is probably not a serious bug in most cases, but if the zeros are a problem the bug can be fixed by rebuilding the IRAF shared library S6.e. Contact the IRAF group for advice on how to do this. NUMBER: 238 MODULE: ccdred.combine SYSTEM: V2.10 DATE: Tue Feb 15 14:04:14 MST 1994 FROM: valdes BUG: When combining images with separate exposure and dark times the final dark time recorded in the header is actually the exposure time; so the final dark time and exposure time will always be the same. The correct dark time is computed but the wrong thing is being written to the image. This is likely to be of relevance for combining dark frames when the dark times of the individual frames are not the same as the exposure times. The workaround is to copy the dark time to the exposure time with HEDIT in the dark frames before combining. Note that this bug applies to the script DARKCOMBINE since it calls COMBINE. STATUS: This is a simple typo code fix which has been made for the V2.11 release. NUMBER: 239 MODULE: onedspec tasks SYSTEM: V2.10 - V2.10.2 DATE: Sat Mar 5 16:05:55 MST 1994 FROM: valdes BUG: Applying IMSHIFT with a positive shift can cause errors in the ONEDSPEC tasks. Negative shifts do not cause a problem. The same problem has been noted previous in bug log number 180 for IDENTIFY and REIDENTIFY. Errors occur when a wavelength is transformed to a pixel coordinate. This happens in a number of operations in SPLOT. There are various workarounds depending on what is desired. A WCSRESET of the physical coordinate system can be used though this will cause the dispersion coordinates to also shift; i.e. then the shift will move the pixel relative to the image data array and the wavelength coordinate system. As a general rule use of IMSHIFT with spectra, particularly dispersion calibrated spectra, should be avoided. STATUS: The problems have been fixed for V2.11. NUMBER: 240 MODULE: ptools.pexamine SYSTEM: V2.10 DATE: Tue Mar 15 08:43:20 MST 1994 FROM: davis BUG: If either or both of the columns specified by the pexamine task parameters xcolumn or ycolumn is undefined in the input image data, then pexamine will crash with a segmentation violation if the user types the o keystroke command in image cursor input mode. The problem occurs because the o keystroke command switches the task to graphics cursor mode, and tries to move the graphics cursor to a point in the default x-y plot which is undefined. STATUS: The bug is fixed in 2.11. The workaround is to set the xcolumn and ycolumn parameters to a column that is defined in the input data. NUMBER: 241 MODULE: sarith, scopy SYSTEM: V2.10 DATE: Tue Apr 12 13:56:49 MST 1994 FROM: valdes BUG: If a spectrum is "multispec" with more than 127 APID keywords the tasks will fail with a segmentation violation. This is due to a limit in a keyword template expansion step. The work around is to delete some of the keywords with: cl> hedit apid1?? del+ Note that you can't delete them all since HEDIT has the same limit. The APID strings can be restored from an aperture identification file using SAPERTURES at a later time. STATUS: In V2.11 the limit on the number of keywords in a template expansion has been increased to 1024. NUMBER: 242 MODULE: sarith SYSTEM: V2.10-2.10.2 DATE: Thu Apr 28 13:51:04 MST 1994 FROM: valdes BUG: Contrary to the help documentation, SARITH fails to allow a single second spectrum operand to be repeated when the first operand is a list. This is a bug. The workaround is to repeat the second operand as often as necessary even if it is the same name. STATUS: Fixed for V2.11. NUMBER: 243 MODULE: rotate/imlintran/geotran SYSTEM: V2.10 DATE: Wed May 11 10:36:18 MST 1994 FROM: davis BUG: Geotran was erroneously overriding the output image size requested by the user if: 1) no geomap database was defined, 2) the output size requested by the user was larger the size required to see the whole input image assuming no origin shift. The tasks rotate/imlintran were also affected by this problem. STATUS: The bug has been fixed for the next release. There is no workaround. NUMBER: 244 MODULE: imcombine, combine, scombine SYSTEM: V2.10 DATE: Wed May 25 11:10:50 MST 1994 FROM: valdes BUG: When combine=average, reject=(sigclip, avsigclip, ccdclip, or pclip), and mclip=yes (the value of mclip does not matter for pclip) the average will be incorrect if the number of low pixels rejected is greater than the number of unrejected pixels. For example if 3 low pixels are rejected and only 2 pixels remain then the final average will be incorrect for that pixel. There is no workaround other than to avoid this combination of parameters. STATUS: Fixed in V2.11. NUMBER: 211.1 MODULE: longslit.fitcoords SYSTEM: V2.10 - V2.10.2 DATE: Sat Jun 4 09:49:39 MST 1994 FROM: valdes BUG: When there are many samples to be fit an 'Out of memory" error can occur. This is caused by repeatedly opening an image section for each sample and failing to close it. If there are enough samples this can cause memory to run out. If this happens on other than the first invocation of FITCOORDS then the workaround is to flush the process cache between executions, "flpr fitcoords". If it happens on the first execution then the only workaround is to reduce the amount of data being fit by using a larger step size, and avoiding multiple runs of REIDENTIFY (since redoing an IDENTIFY appends to the database rather than replacing resulting in multiple entries). For marginal cases it might be possible to make sure the process has the maximum amount of memory available by minimizing memory use by other processes. A second manifestation of this problem occurs when using STF format (the SDAS group format). The error is ERROR: 827 where is the image section. This message means that the task has opened too many files and could not open the file which translates the error code. This behavior occurs because STF format image keep the image header open unlike the .imh format. The workarounds are the same as decribed previously. STATUS: This is a rather severe bug which is fixed in V2.11. NUMBER: 246 MODULE: images.imcombine, ccdred.combine, onedspec$scombine SYSTEM: V2.10.2 DATE: Mon Jun 6 17:03:15 MST 1994 FROM: valdes BUG: In rare circumstances when using one of the clipping algorithms nkeep > 0 the task will not complete because it has entered a nonterminating loop. This occurs when too many pixels are rejected and two or more rejected pixels have equal residuals the equal residual points are added back. However this is done within the iterative loop and after adding back the pixels the task finds there are now more than nkeep pixels and so repeats the loop with the same result. The work around is to change nkeep. A change is likely to fix the specific case. A value of zero for nkeep will avoid the problem entirely but with the possibility of rejecting all pixels. STATUS: The adding back of pixels should be done outside the iterative loop and this change has been made for the next release (V2.11). NUMBER: 247 MODULE: photcal.fitparams SYSTEM: V2.10 DATE: Mon Jun 13 16:12:33 MST 1994 FROM: davis BUG: If the fitparams task parameter weighting is set to "photometric" or "equations", the addscatter parameter is set to "yes", and either the number of parameters to be fit equals the number of data points, or the fit has no scatter, fitparams will crash with a floating operand error. This occurs because the expression used to estimate the additional scatter to be added to the weights in the next iteration of the fit goes to 0 in the denominator, causing a divide by zero error. STATUS: The bug is fixed in 2.11. The workaround is to set weighting to "uniform" or set the addscatter parameter to "no". NUMBER: 248 MODULE: image.imcombine,ccdred.combine,onedspec.scombine SYSTEM: V2.10.2 DATE: Mon Jun 13 18:50:46 MST 1994 FROM: valdes BUG: When the rejection algorithm is avsigclip, sigclip, avsigclip, or crreject and mclip=no and nkeep is not zero there is the rare possibility of a segmentation violation. This occurs when more than the maximum number of pixels is rejected and several of the rejected pixels have the same residual relative to the average. The workarounds are to reduce nkeep (possibly to zero), use mclip=yes, use another rejection algorithm, or adjust the sigma values. STATUS: This problem is fixed for V2.11. NUMBER: 249 MODULE: sarith SYSTEM: V2.10 DATE: Tue Jun 21 14:35:17 MST 1994 FROM: valdes BUG: When doing a binary operation with a constant second operand and "onedspec" output format the constant used is incorrect due to a bug in which the number is used as a real number but is actually a double precision number. The incorrect value can be seen in the verbose output. For example: cl> sarith specin - 48.5 specout format=onedspec verbose+ specin[1] - 3.128906 --> specout.0001 The workaround is to use multispec output format for the operation. A later SCOPY can then be used to convert formats if desired. STATUS: Fixed in V2.11. NUMBER: 250 MODULE: photcal.mknobsfile,mkobsfile,obsfile SYSTEM: V2.10 DATE: Wed Jun 22 17:16:03 MST 1994 FROM: davis BUG: The mknobsfile/mkobsfile/obsfile tasks were omitting stars from the output observations file for image sets where there was only a single image per image set, and that image was not the last image in the image set. STATUS: The bug has been fixed in 2.11. One possible work around is to set the tolerance parameter to 0, run mknobsfile/mkobsfile/obsfile on image sets with only one defined image separately from the other image sets, and concatenate the results from the two runs. NUMBER: 251 MODULE: sarith, scopy SYSTEM: V2.10.0 - V2.10.2 DATE: Wed Jun 29 17:32:20 MST 1994 FROM: valdes BUG: For multispec spectra which have a doppler factor in the coordinate system description (a non-zero value for the 7th field following the number of pixels) making a copy or doing arithmetic will cause the wavelength zero point to be off by a factor of 1/(1-z*z). The workarounds are to either avoid introducing a doppler factor, accounting for the error, or explicitly setting w1,w2 (but note bug #214). STATUS: Fixed 10/92 and included in V2.11. NUMBER: 252 MODULE: utilities.polyfit SYSTEM: V2.10 DATE: Mon Jul 11 09:17:27 MST 1994 FROM: davis BUG: The computation of chisqr and stdev are too large by a factor if yavg ** 2 if the weighting parameter is "instrumental" and abs(yavg) if the weighting parameter is "statistical". STATUS: Fixed in 2.11. There is no workaround except to apply the correction factors manually. NUMBER: 253 MODULE: transform, apall, apsum SYSTEM: V2.9 DATE: Mon Jul 25 14:14:57 MST 1994 FROM: valdes BUG: The extraction of dispersion calibrated long slit data, the result of TRANSFORM, does not properly pass on the dispersion information. One symptom of this is that if the dispersion axis is 2 then the final extracted spectrum will have a wavelength per pixel corresponding to the first axis in the CD1_1 keyword. Note that there is also a correct WPC and CDELT1 keyword. This does not cause a problem when using V2.9 ONEDSPEC routines because WPC has precedence over CD1_1. However in taking this data to V2.10 the precedence was changed and CD1_1 is used in preference to WPC or CDELT1. So a task like V2.10 SPLOT will plot the spectra with an incorrect dispersion. The work around to update the header for V2.10 is update the CD1_1 keywords or delete them: cl> hedit <1D spectra> cd1_1 '(wpc)' cl> hedit <1D spectra> cd1_1 del+ STATUS: The aperture extraction in V2.10 correctly passes on the dispersion coordinates produced by TRANSFORM during the long slit reductions. NUMBER: 254 MODULE: longslit.transform SYSTEM: V2.10-V2.10.2 DATE: Sat Aug 6 12:14:32 MST 1994 FROM: valdes BUG: If DISPAXIS=2 and one specifies log sampling (ylog=yes) then the specified sampling is done correctly but the header coordinate description is incorrect. Specifically the starting wavelength, CRVAL2, is given as a linear wavelength rather than a log wavelength. Display of this spectrum, say with SPLOT, produces a wavelength scale which is very compressed at the starting wavelength. The workaround is to fix the header of the transformed image with cl> hedit crval2 "(log10(crval2))" update+ STATUS: Fixed in versions after V2.10.2 NUMBER: 254 MODULE: longslit.transform SYSTEM: V2.10-V2.10.2 DATE: Sat Aug 6 12:54:42 MST 1994 FROM: valdes BUG: If DISPAXIS=2 and one specifies log sampling (ylog=yes) then the specified sampling is done correctly but the header coordinate description is incorrect. Specifically the starting wavelength, CRVAL2, is given as a linear wavelength rather than a log wavelength. Display of this spectrum, say with SPLOT, produces a wavelength scale which is very compressed at the starting wavelength. The workaround is to fix the header of the transformed image with cl> hedit crval2 "(log10(crval2))" update+ This was caused by a typo where parameter xlog was used where ylog should have been. Another thing to be aware of is that extraction of a 1D spectrum from the 2D spectrum with the APEXTRACT tasks will propagate the error. For such data SPLOT crashes because it tries to evaluate an impossibly large decimal exponent. STATUS: Fixed in versions after V2.10.2 NUMBER: 255 MODULE: apall, apsum SYSTEM: V2.10-V2.10.2 DATE: Tue Aug 9 10:47:21 MST 1994 FROM: valdes BUG: When extracting spectra with dispaxis=1 (spectra horizontal) using variance weighting and/or cleaning with a gain other than 1 then occasionally a "floating operand error" or some arithmetic exception can occur. This is caused by reference to unitialized memory during application of the gain. The uninitialized memory does not affect the results but if an invalid arithmetic value is encountered it will cause the extraction to abort. The workarounds in this case are to use a gain of 1 if it is reasonable otherwise to transpose the image and reset the wcs (with wcsrest wcs=physical) and the dispaxis value. STATUS: Fixed after V2.10.2. NUMBER: 256 MODULE: ccdproc SYSTEM: V2.8-V2.10.3 DATE: Mon Aug 29 13:57:03 MST 1994 FROM: valdes BUG: If the calculation datatype is set to short, calculation of the CCDMEAN value is incorrect and, if used during flat fielding, will give incorrect scaling for the flat fielding. The workarounds are to not use short for the calculation datatype though it is ok to use short output datatype to save diskspace. Note that if no calculation datatype is specified it defaults to real so it takes an explicit setting of pixeltype="short short" or "real short" to force this bug to appear. STATUS: Fixed after V2.10.3. NUMBER: 257 MODULE: mkfringe, mkillumcor, mkillumflat, mkskycor, mkskyflat SYSTEM: V2.8-V2.10.3 DATE: Mon Aug 29 16:25:11 MST 1994 FROM: valdes BUG: If the tasks are run more than once without flushing the process then a segmentation violation or error attempting to access an unopened file is likely to occur. The workaround is to flush the process cache with "flpr" before running any of these tasks. STATUS: Fixed future versions. NUMBER: 258 MODULE: reidentify SYSTEM: V2.10-V2.10.3 DATE: Thu Sep 15 10:29:50 MST 1994 FROM: valdes BUG: If a step of zero is specified the task will hang in an infinite loop. The way the step is used is to increment/decrement the starting line until the ends of the image are reached. With a step of zero this will never happen. However, if one is interested in only reidentifying a single line in a 2D image it seems reasonable to set the step to zero. The actually method is to set the step to a very large number. STATUS: For the next release the case of a zero step is checked and causes REIDENTIFY to only reidentify the reference line in other images. NUMBER: 259 MODULE: images.tv.tvmark SYSTEM: V2.10 DATE: Mon Oct 3 09:01:29 MST 1994 FROM: davis BUG: The tvmark append (a) keystroke command does not function correctly under Solaris IRAF 2.10.3 BETA. The objects to be appended to the coordinate list are marked correctly on the image display but are not appended to the output coordinate file. The problem is apparently caused by a failure to call the flush command before doing a seek to EOF. On SunOS systems the seek to EOF appears to force a flush, while on Solaris systems it does not, resulting in the appended objects never being written to the output file. STATUS: Fixed for the next release of IRAF. There is no workaround. NUMBER: 260 MODULE: digiphot.daophot.seepsf SYSTEM: V2.10 DATE: Mon Oct 3 13:08:46 MST 1994 FROM: davis BUG: The seepsf task was forcing the output psf image to be less than or equal 23 pixels square, corresponding to a psfrad less than or equal to 11 pixels, no matter what the actual psfrad in the input image psf was. The computed portion of the output psf image was correct. The problem was that seepsf was using the wrong quantity to compute the default size of the output psf image. STATUS: Fixed in 2.10.3. There is no workaround. NUMBER: 261 MODULE: apphot package tasks SYSTEM: V2.10 DATE: Tue Oct 4 09:21:04 MST 1994 FROM: davis BUG: The directory information was inadvertantly being stripped from the default input/output file specifications. For example the phot task would look for the default input coordinate file in the current directory even though, the file specification was dir$default. Similarly the default output file was being written to the current directory, even though the file specification was dir$default. Directory information was not being stripped from fully specified file names. STATUS: Fixed in the next release of iraf. NUMBER: 262 MODULE: echelle.doecslit SYSTEM: V2.10-V2.10.2 DATE: Thu Oct 6 16:21:57 MST 1994 FROM: valdes BUG: DOECSLIT fails to actually use the recentered and (optionally) retraced apertures determined for each object spectrum when it extracts the orders. Instead it ignores the object apertures and uses the reference spectrum apertures. This can be seen since the actual apertures used for extraction are saved in the database and they will be the same as the reference aperture. There are no workarounds other than to modify the script or to do the extractions separately with APALL and then use DOECSLIT for the remainder of the operations. STATUS: This fixed in versions after V2.10.2. NUMBER: 263 MODULE: photcal.mkapfile SYSTEM: V2.10 DATE: Thu Oct 13 11:37:48 MST 1994 FROM: davis BUG: If there was more than one photometry file in the input file list mkapfile was quitting with the error "ERROR on line 187: cannot open `file'" for writing. This was caused by the script trying to append to an existing file with "> file" instead of ">> file". STATUS: Fixed for the next release of IRAF. NUMBER: 264 MODULE: rv.fxcor SYSTEM: V2.10 DATE: Thu Oct 13 15:32:33 MST 1994 FROM: fitz BUG: The batch processing code was not initializing the task structure properly and a "Memory has been corrupted" error can occur when processing in batch and the template spectrum has to be rebinned. The bug is not seen if it's the object spectrum that is rebinned. Possible workarounds include rebinning the data to a common disp- ersion and starting wavelength before correlating, setting the 'rebin' parameter to force it to rebin to the object's dispersion, or else just process interactively. The code change is trivial for sites who wish to fix it. STATUS: Fixed in versions after V2.10.3beta. NUMBER: 265 MODULE: expanding file templates on Solaris SYSTEM: SunOS/IRAF V2.10.2 executing on Solaris DATE: Fri Oct 14 13:27:26 MST 1994 FROM: tody BUG: Various users have reported a subtle problem in IRAF which occurs when SunOS/IRAF version 2.10.2 is run on a Solaris system in compatibility mode. On occasion IRAF code which reads host directories may not "see" all the files in a directory. Hence file and image templates may not expand correctly, omitting some of the files in a directory. STATUS: This problem is not present in the native Solaris version of IRAF, Solaris/IRAF V2.10.3BETA. The solution is to update to this version. If V2.10.2 must be used, use host system facilities (e.g. ls, provided as a foreign task in unix/iraf systems) to expand file templates. NUMBER: 266 MODULE: apphot.daofind SYSTEM: V2.10 DATE: Thu Oct 20 13:57:31 MST 1994 FROM: davis BUG: If the sigma parameter was left at its default value of INDEF, a floating point overflow value would result when daofind was run from the apphot package. The problem was the result of a missing check for INDEF in the code. STATUS: This bug was not a problem for versions of IRAF prior to 2.10.3 BETA and has been fixed for the next release of IRAF. The workaround is to set the sigma parameter to some reasonable value. NUMBER: 267 MODULE: rv.rvidlines SYSTEM: V2.10.3beta DATE: Thu Oct 20 14:44:50 MST 1994 FROM: valdes BUG: The intent of this program is that if the header information does not exist to compute heliocentric velocities it should simply compute observed velocities. However, the trap for this conditions in not correct so instead the program quits. The only workaround is to add dummy keywords for those missing and ignore the heliocentric results. STATUS: Fixed after version V2.10.3beta. NUMBER: 268 MODULE: imexpr SYSTEM: V2.10.3 DATE: Tue Oct 25 11:42:52 MST 1994 FROM: tody BUG: The "imexpr @file" way of invoking imexpr doesn't work; imexpr reports a syntax error due to the newline at the end of the text file containing the expression. STATUS: Fixed in the V2.10.3 patch. To workaround the bug, either put the expression on the command line or enter it as a macro function in an expression database file. NUMBER: 269 MODULE: artdata.gallist SYSTEM: V2.9-V2.10.3beta DATE: Sat Oct 29 16:56:56 MST 1994 FROM: valdes BUG: The internal absorption model for disk galaxies is not being applied as described in the help. Regardless of the absorption parameter there is no absorption correction to the magnitudes of the disk galaxies at any inclination. There is no workaround. STATUS: Fixed after V2.10.3beta. NUMBER: 270 MODULE: imshift, magnify, xregister, geotran SYSTEM: V2.10 DATE: Fri Dec 9 10:48:34 MST 1994 FROM: davis BUG: Flux may not be conserved when the IMSHIFT, MAGNIFY, XREGISTER, or GEOTRAN tasks are run on undersampled data using the spline3 interpolant. In the worst case situation, completely undersampled data, fractional y shifts of 0.4-0.6, and placement near the top or bottom of the data buffer, the discrepancy is ~ +/- 0.3%. For well-sampled data the discrepancy is negligible. For y shifts which are exactly 0.0 there is no problem. The cause of the problem was that the number of edge lines being allocated for the spline3 buffers was too small for undersampled data and some of the power in the high frequency components of the interpolant was being lost. STATUS: Fixed for next release of iraf. NUMBER: 271 MODULE: photcal.mknobsfile,photcal.mkobsfile,photcal.mkapfile,ptools.pdump SYSTEM: V2.10.3BETA DATE: Tue Dec 20 10:47:32 MST 1994 FROM: davis BUG: A reference was being to a tables.ttools.tdump task parameter outside the loop for processing ST binary tables in the pdump and mkapfile scripts, resulting in an "unknown task tdump" error when the tables package was undefined or defined incorrectly. The pdump error could in turn produce the same error in the mknobsfile and mkobsfile tasks. STATUS: Fixed for the next release in iraf. The workaround is to install the tables package. If this is not feasible contact the iraf group for a fix. NUMBER: 272 MODULE: dispcor SYSTEM: V2.10.3BETA DATE: Mon Jan 16 13:01:54 MST 1995 FROM: valdes BUG: When the input spectra are in equispec format (all spectra with the same linear dispersion relation) and there are aperture numbers which are greater than the number of image lines then the error "WFMSPEC: Coordinate out of bounds" may occur. Such a situation will only occur when resampling an an already dispersion corrected set of spectra and not when using an IDENTIFY or ECIDENTIFY database. This is due to a logical error in the code concerning mapping of aperture numbers to line numbers. The workarounds are to use SARITH if simply extracting a region, extract the spectra to onedspec format with SCOPY, or change the WCS with: cl> hedit cd2_2,cdelt2 1000 update+ cl> hedit crpix2 1 add+ STATUS: Fixed for the next release. NUMBER: 273 MODULE: dopcor SYSTEM: V2.10.3BETA DATE: Wed Jan 18 17:22:38 MST 1995 FROM: valdes BUG: For multispec spectra which have independent wavelengths for each spectrum (identified by system=multispec in the header) there is a bug such that if there are more than 8 spectra the message: Warning: MWCS: too many coordinate transformation (ctran) descriptors ERROR: segmentation violation appears. Most commonly this will affect echelle data. The workaround is to apply DOPCOR with the aperture parameter in groups of 8 or less spectra. For example: cl> dopcor obj.ec obj1.ec -120 isvel+ add- ap=1-8 cl> dopcor obj1.ec obj1.ec -120 isvel+ add- ap=9-16 ... Note the second and subsequent steps work in place. STATUS: This bug, which is a typo in the code, is fixed for a release after V2.10.3BETA (possibly the Solaris patch) and V2.11. NUMBER: 274 MODULE: doslit (specred, kpnocoude, kpnoslit) SYSTEM: V2.10.3BETA DATE: Tue Jan 24 13:56:11 MST 1995 FROM: valdes BUG: If the first object spectrum has more than one object on the slit and apertures are defined for them then after doing the dispersion solution on the reference arc the following error occurs: ERROR ... - Missing reference for aperture 2 This is a bug, introduced in V2.10.3BETA, since the intent of the task is to allow multiple apertures per long slit spectrum. It is caused by using the apertures from the first object as the apertures for the arc reference. The workarounds are 1) make the first object image be one that has only one aperture (other objects can have multiple apertures) or 2) when in IDENTIFY doing the arc reference dispersion solutions do 'k' after finishing the dispersion solution to step to the other (unnecessary) apertures and do a solution with 'a' 'c' and 'f' (all center and fit). STATUS: Fixed in later versions. NUMBER: 275 MODULE: apnormalize SYSTEM: V2.10.3BETA DATE: Fri Jan 27 17:16:35 MST 1995 FROM: valdes BUG: In bug numbers 171 and 225 it was noted that when using variance weighting or cleaning in APNORMALIZE the result would end up scaled by the gain value. The fix was to divide the result by the gain. However, the default parameters for APNORMALIZE do not call for variance weighting or cleaning and so the gain and readnoise are not read. The uninitialized value of the gain used to divide the result of APNORMALIZE is zero causing a divide by zero error. The workaround is to set the weights paraemter to "variance". The actually weighting and gain and readnoise parameters are not important to this algorithm. STATUS: This only affects version V2.10.3BETA and will be fixed in later versions. NUMBER: 276 MODULE: imexamine SYSTEM: V2.8-V2.10.2 DATE: Tue Jan 31 15:06:41 MST 1995 FROM: valdes BUG: The standard deviation values returned by the 'm' key in IMEXAMINE can be inaccurate when the mean data value is large and the standard deviation is relatively small. This is due to calculation of the standard deviation formula using the difference in the sum of squares and the square of the sum with real precision. The workarounds are to use IMSTATISTICS with an image section or to first subtract something like the mean value from the image before using IMEXAMINE. STATUS: In V2.10.3 and after the calculation is done in double precision and the standard deviation is as accurate as IMSTATISTICS. NUMBER: 277 MODULE: photcal.mknobsfile,mkobsfile,obsfile SYSTEM: V2.10.3BETA DATE: Thu Feb 9 13:25:36 MST 1995 FROM: davis BUG: The time of observation and airmass columns are switched in the output observations file if position matching is turned off, resulting in the airmass being written under the otime column and vice versa. STATUS: Fixed for the next release of iraf. One possible workaround is to simply switch the column numbers in the configuration file since the number can still be read in correctly. NUMBER: 278 MODULE: splot, fitprofs SYSTEM: V2.10.3BETA DATE: Thu Feb 16 18:17:59 MST 1995 FROM: valdes BUG: The new error estimates for Gaussian profile fitting and deblending computed if a noise model is given are incorrect when there is more than one line being fit; though they are roughly correct. This is caused by an indexing error such that the error is not always the 68% (1 sigma) point and the error on a particular line is really a mixture of measurements from all the lines. There estimates are valid for single line Gaussian fits. There is no workaround. STATUS: Fixed in subsequent versions. NUMBER: 279 MODULE: doslit, doecslit SYSTEM: V2.10-V2.10.2 DATE: Fri Feb 17 10:06:09 MST 1995 FROM: valdes BUG: If no arc image or standard star image is specified an error such as ERROR on line 103: OPEN: File does not exist (tmp$iraf1665j) doecslit () will appear. This is true even if one is not dispersion correcting or standard star fluxing. Thus, even if dispcor=no or fluxcal=no you must put a placeholder image name in the arcs and standard parameters. STATUS: Fixed in V2.10.3 and later such that not specifying arc and standard star images is ok if not dispersion or flux calibrating. NUMBER: 280 MODULE: sensfunc SYSTEM: V2.6-V2.10.3 DATE: Wed Feb 22 11:18:52 MST 1995 FROM: valdes BUG: Deleted points and deleted stars are not excluded from the grey shift calculation. This is a bug in the sense that this is not what most users would expect. In particular, if a star is deleted and it is the one with the highest apparent sensitivity one would not expect the rest of the data to be shifted to match this deleted star. For the case of deleting a star and excluding it from the grey shift calculation the workaround is to delete the data from the input data file. STATUS: Deleted points are excluded from the grey shift calculation in future versions. NUMBER: 281 MODULE: splot SYSTEM: V2.10.3beta DATE: Thu Feb 23 16:10:43 MST 1995 FROM: valdes BUG: The 't', '/' combination for continuum flattening spectra fails for flux calibrated data and instead produces a spectrum which is identically 1. A check to avoid division by zero fails to discriminate flux calibrated data (any value less than about 1E-8) as different from zero. The resultant value of 1 is used when the divide by zero condition is detected. STATUS: Fixed in future releases. NUMBER: 282 MODULE: fit1d SYSTEM: V2.10.3beta DATE: Thu Feb 23 16:16:41 MST 1995 FROM: valdes BUG: The ratio output does not work in V2.10.3beta due to a typo error in the code. The workaround is to output the fit and use IMARITH to take the ratio. STATUS: Fixed in future releases. NUMBER: 283 MODULE: ptools.pexamine SYSTEM: V2.10 DATE: Thu Feb 23 16:23:12 MST 1995 FROM: davis BUG: On Dec Alpha systems pexamine will crash with an invalid floating point operation error if the user tries to plot the histogram of an array with the h keystroke command. STATUS: The problem was caused by a typo in a statement which resulted in a memory access of the end of an array. The bug is fixed for the next release of iraf. There is no workaround. NUMBER: 284 MODULE: photcal.mknobsfile,photcal.mkobsfile SYSTEM: V2.10 DATE: Wed Mar 22 07:56:37 MST 1995 FROM: davis BUG: A very serious problem has been found in the iraf 2.10.3 and digiphotx mknobsfile/mkobsfile scripts. Running mknobsfile/mkobsfile on files written by the allstar/nstar tasks (but not the phot task) may corrupt the input files. This is definitely not problem for mknobsfile/mkobsfile in 2.10.2 and earlier version of iraf or for digiphotx linked with versions 1.2.1 and earlier of the tables package. The problem crept in when the tables package was modified to support text tables. The error check in mknobsfile/mkobsfile previously used to detect whether an input file was a text file or binary table began to fail for allstar/nstar files. The code thought the text file was a binary table, opened it as such, and attempted to edit it. Although the code was smart enough to detect that the proper columns etc were not present and did not do any actual editing, simply opening and closing the file as a table in READ_WRITE mode caused to file to be rewritten and its format corrupted. STATUS: This problem has been fixed for the next release of iraf and in the forth-coming 2.10.3 patch as well as in the ftp archive version of digiphotx. If you are running later versions of tables there is no workaround except installing the patch and/or the new version of digiphotx. We have confirmed that there is a problem running mknobsfile/mkobsfile with tables 1.3.2 and 1.3.3 and that there was definitely not a problem with tables 1.2.1 and earlier but are not certain of the status of intermediate versions. NUMBER: 285 MODULE: echelle.ecidentify SYSTEM: V2.9-V2.10.3 DATE: Tue Apr 11 10:55:54 MST 1995 FROM: valdes BUG: The 'x' key to find a shift between a set of features and the peaks in the data fails with a memory access error (i.e. segmentation violation). There is no workaround to use this key though the 's' key can be used with a known line to set and apply a shift. STATUS: Fixed for the next release. NUMBER: 286 MODULE: onedspec.sbands SYSTEM: V2.10.3 DATE: Wed Apr 12 09:33:20 MST 1995 FROM: valdes BUG: The output formating for the flux column produces 0.00 when the spectra are flux calibrated; i.e. have very small values of order 1E-14. The workaround is to multiply the spectrum by an appropriate large number and remember the scaling used. STATUS: The formating will be changed to allow exponential notation in a future release. NUMBER: 287 MODULE: photcal.mkapfile,photcal.apfile SYSTEM: V2.10 DATE: Fri Apr 28 10:00:14 MST 1995 FROM: davis BUG: The filter id keyword read from the input photometry files was not being written correctly to the output magnitudes file, resulting in a filter id of INDEF for all images. STATUS: Fixed for the next release iraf release. There is no workaround. NUMBER: 288 MODULE: ptools.pexamine SYSTEM: V2.10 DATE: Mon May 1 09:16:17 MST 1995 FROM: davis BUG: The colon commands :rinner and :router were incorrectly defined as :rin and :rout in the pexamine definitions file. In addition the new router value was being written back into the rinner parameter, a bug that was masked by the previous problem. STATUS: Fixed for the next release of iraf. The workaround is to use the :epar radplot command and edit the parameter set as a whole. NUMBER: 289 MODULE: apscatter SYSTEM: V2.9-V2.10.3 DATE: Wed May 3 16:50:54 MST 1995 FROM: valdes BUG: When not smoothing the scattered light surface along the dispersion, smooth=no, and there are multiple apertures the task may abort with the error "Range descriptor undefined". This occurs when the number of scattered light points being fit across the dispersion changes as the spacing between apertures changes. This is probably a rare event. The only workaround is to add the smoothing along the dispersion. STATUS: Fixed in after V2.10.3. NUMBER: 290 MODULE: scopy, sarith SYSTEM: V2.10.3 DATE: Sat May 13 13:54:11 MST 1995 FROM: valdes BUG: When spectra are modified with SPLOT and then written out, the current display units are also recorded. SCOPY and SARITH incorrectly create new spectra with the dispersion units set to the display units. For units which are linearly related (such as any between wavelength units or between energy/frequency units) this causes no problem. However, for other pairs of units, such as dispersion units of angstroms but display units of EV, the new coordinate system will not match the spectrum pixels. The workarounds are to avoid this situation or change the "units" in the copied/output spectrum back to the original units with WCSEDIT. STATUS: Fixed in V2.10.4 NUMBER: 291 MODULE: argus.msresp1d, hydra.msresp1d SYSTEM: V2.10.2 DATE: Mon May 22 09:38:27 MST 1995 FROM: valdes BUG: The task MSRESP1D in the ARGUS and HYDRA packages will produce an error reporting that there is no param file. This is caused by an error in specifying a directory name. The workaround is either to interactively type the following after loading the desired package: ar> redefine msresp1d=specred$msresp1d.cl or change the system files argus$argus.cl and hydra$hydra.cl so that "msred" becomes "specred". STATUS: Fixed in V2.10.3 and later. NUMBER: 292 MODULE: imexamine SYSTEM: V2.10-V2.10.4 DATE: Wed May 31 10:08:47 MST 1995 FROM: valdes BUG: The output of the 'x' and 'y' keys is not written to the logfile if one is used. There is no workaround other than copying the terminal output in some way. Note the task RIMCUR can be used to get coordinates from an image in both pixel and world coordinates. STATUS: Fixed in subsequent releases. NUMBER: 293 MODULE: register, geotran SYSTEM: V2.10 DATE: Fri Jun 9 16:43:56 MST 1995 FROM: davis BUG: If the xsample or ysample parameters are greater than 1 and the nxblock or nyblock parameters are smaller than the image dimensions, register/geotran may fail with a memory corruption error. The error was due to a buffering error in the y coordinate interpolation surface. STATUS: Fixed for the next release of iraf. The workaround is to set xsample and ysample back to 1, which makes coordinate evaluation slower for high order surfaces but does not make much difference for simple surfaces, or set nxblock and nyblock to be larger than the input image size. NUMBER: 294 MODULE: photcal.mkapfile,photcal.apfile SYSTEM: V2.10 DATE: Tue Jun 13 17:25:01 MST 1995 FROM: davis BUG: Mkapfile/apfile may terminate prematurely or hang with a floating overflow exception if the input data is such that it is difficult to determine a good value for the Moffat function radius parameter RO. In this case RO can be driven towards and beyond 0 which produces the overflow. STATUS: A check was added to the code to trap this condition and report a fitting error. Possible work arounds including upping the default number of parameters to be fit from 3 to 4 and/or deleting the suspect data from the input file. NUMBER: 295 MODULE: dofibers, dohydra, doargus SYSTEM: V2.10-V2.10.4 DATE: Wed Jun 14 14:44:15 MST 1995 FROM: valdes BUG: The use of subapertures does not work and aborts with the error ERROR on line 28: Bad aperture/column/line list This is caused by an incorrect aperture range in the script code. There is no workaround other than not use subapertures or modifying the script code. STATUS: Fixed for the next release. NUMBER: 296 MODULE: digiphot.apphot.daofind SYSTEM: V2.10 DATE: Mon Jun 19 09:34:40 MST 1995 FROM: davis BUG: If the datamin parameter is set above the sky level of the image the code which computes the x and y positions of the detected objects can crash with a floating divide by zero error for the fainter objects. STATUS: Fixed for the next release of iraf. The workaround is to set the datamin parameter to a more reasonable value. NUMBER: 297 MODULE: longslit, longslit.identify, longslit.reidentify SYSTEM: V2.10.0 DATE: Mon Jun 19 14:08:46 MST 1995 FROM: valdes BUG: This is a tardy bug report since I noticed this major bug is not actually in the bug log. In the first release of V2.10 there was a serious bug in the ONEDSPEC tasks which apply to 2D images such as long slit data. The tools would not access dispersion lines beyond 2. The most common symtom is that IDENTIFY will only show lines 1 and 2 in long slit data. Normally one encounters this in the LONGSLIT package. If one uses a section parameter of "middle lines" IDENTIFY will show the image section [*,2]. REIDENTIFY is similarly limited to the first two dispersion lines. There is no workaround other than to upgrade V2.10 to any higher version. It was first fixed in V2.10.1 STATUS: Fixed in V2.10.1 and later. NUMBER: 298 MODULE: onedspec.sbands SYSTEM: V2.10-V2.10.4 DATE: Fri Jun 30 09:25:30 MST 1995 FROM: valdes BUG: When the number of input bandpasses is greater than 10 a segmentation violation error occurs. This is a bug in the memory allocation logic of the task. The workaround is to break up the bandpasses into groups of 10 or less (it might work up to 20 but is not guarenteed). STATUS: Fixed in later versions. NUMBER: 299 MODULE: daophot.seepsf SYSTEM: V2.10 DATE: Mon Jul 3 17:24:04 MST 1995 FROM: davis BUG: Seepsf was crashing with a floating operand error if the input psf image was a purely analytic function, keyword varorder=-1. This error was accidentally introduced when the previous seepsf error (bug log # 260) was fixed. STATUS: Fixed in 2.11 and in the external addon package digiphotx. One possible workaround is make the psf image with varorder=0, then make a copy of it setting all the pixel values to 0.0 with imarith, and inputing the new image to seepsf. NUMBER: 300 MODULE: dispcor SYSTEM: V2.10.4 DATE: Thu Jul 13 11:08:04 MST 1995 FROM: valdes BUG: A new capability was added in V2.10.4 to allow DISPCOR to work with IDENTIFY database files containing only a shift and no dispersion function. This would occur when using IDENTIFY to compute a shift to previously dispersion corrected spectra. This does not work correctly when there is more than one aperture in an image. The workaround is to separate multiaperture images into single aperture images with SCOPY and then do the IDENTIFY/REIDENTIFY and DISPCOR. STATUS: Fixed in later versions. NUMBER: 301 MODULE: daophot.seepsf (Dec Alpha) SYSTEM: V2.10 DATE: Tue Jul 18 10:14:50 MST 1995 FROM: davis BUG: Seepsf will terminate with an FPE error on the Dec Alpha machine if the the psf model is not analytic. The problem was caused by as missing test which resulted in an out-of-bounds array reference in some cases. On other systems the FPE error did not occur but pixels outside the psf radius were not being set to 0 as intended but were instead were interpolated in the noisy edges of the look-up table. Pixels inside the psf radius are correct. STATUS: Fixed for the next release of iraf and in the ftp archive version. There is no workaround. NUMBER: 302 MODULE: apscatter SYSTEM: V2.10-V2.10.4 DATE: Tue Jul 18 17:35:37 MST 1995 FROM: valdes BUG: If the aperture trace functions cross or cause the apertures to overlap then the algorithm for finding the scattered light pixels outside the apertures gets confused. This can have the consequence of either using data within an aperture or a memory error (memory corrupted or segmentation violation). The workaround is to insure the aperture definitions and traces do not behave in this way. One way to check this is with APEDIT after the apertures have been defined and traced. In APEDIT you can examine various lines with ":line #" and the data line and apertures will be shown. Look at the first line and last line in particular. The interactive option of APSCATTER selecting various lines can also be used though this may crash in the same way. STATUS: A different algorithm will be used in the next release to correctly find the scattered light points even in the unreasonable case where the apertures cross. NUMBER: 303 MODULE: rv.fxcor SYSTEM: V2.10-V2.10.4 DATE: Wed Jul 26 15:59:57 MST 1995 FROM: fitz BUG: In rare cases a Gaussian fit would converge to a nonsensical result (e.g. a sigma of 3e10) and would cause a "floating overflow" error when the results were written out. The fix requires a trivial code change, as a workaround select a different set of points for the initial fit by adjusting the 'width' parameter or interactively with the 'g' or 'y' keystrokes. STATUS: Fixed for the next release. NUMBER: 304 MODULE: proto.wcsreset SYSTEM: V2.10 DATE: Tue Aug 1 09:12:11 MST 1995 FROM: davis BUG: Wcsreset was refusing to reinitialize the world coordinate system of any image whose world coordinate system was not support by the iraf mwcs system, mostly images with unrecognized sky projection types. STATUS: Fixed in 2.11. There is no workaround except to edit the headers directly with hedit or hfix. NUMBER: 305 MODULE: apextract tasks SYSTEM: V2.10-V2.10.4 DATE: Wed Aug 2 13:57:09 MST 1995 FROM: valdes BUG: If the image header has an invalid DISPAXIS value, for example a value of -9999, there is no warning and bad things happen. If the value is less than 0 then the editing plot will show a flat line with an X range 1-2. If the value is 0 or greater than 2 then the aperture axis will be shown as either 1 (for even values) or 2 (for odd values) but tracing will fail and extraction will produce a floating operand error. The workaround is to either delete the invalid keyword and use the "dispaxis" parameter or fix the image header value. STATUS: In subsequent releases there will be a warning message and the invalid value will be ignored. The "dispaxis" value will then be used. NUMBER: 306 MODULE: identify SYSTEM: V2.10-V2.10.4 DATE: Thu Aug 3 13:12:43 MST 1995 FROM: valdes BUG: The rarely used :add command to add lines can cause a segmentation violation when exiting. The results are still valid and the error can be ignored. A workaround is that after adding the lines and doing whatever is desired, a stored past solution can be read or an intialization be done. STATUS: Fixed in later versions. NUMBER: 307 MODULE: splot SYSTEM: V2.10.3-V2.10.4 DATE: Fri Aug 11 18:54:55 MST 1995 FROM: valdes BUG: When writing out a spectrum with the 'i' key an error about the keyword WAT1_002 not being found or cannot be deleted may occur. Because this occurs in graphics mode the error may flash by too fast to see (with XGTERM) and you may be repeatedly prompted for an output image. This loop can only be killed by killing the CL. The cause of this is a bug in SPLOT which occurs if there are BANDIDn keywords as produced by extraction with the APEXTRACT package. This problem has been reported with DEC ALPHA but could potentially affect any V2.10.3-V2.10.4 system. The workaround is to delete the BANDID keywords, which are only used for documentation currently, using HEDIT. You can then use SPLOT and write out modified spectra with 'i'. STATUS: Fixed in future versions. NUMBER: 308 MODULE: artdata.mkobjects SYSTEM: V2.9-V2.10.4 DATE: Mon Aug 14 09:55:27 MST 1995 FROM: valdes BUG: When using an image either as a PSF model or as an object model the result is very subtly wrong in the fluxes. The problem showed up when using a symmetrical PSF model and the resulting star or galaxy profile was not symmetric at a low level. The measured centroid was off by 0-0.05 pixels. This is caused by a numerical precision error that occasionally adds an extra subpixel to the flux computed for an output pixel. There is no workaround though the effect can be minimized by using a larger number of subpixels. However, the effect is very small and simulations with added noise would likely swamp the error. STATUS: Fixed in later versions. NUMBER: 309 MODULE: sfit, continuum SYSTEM: V2.10.3-V2.104 DATE: Tue Aug 15 18:33:59 MST 1995 FROM: valdes BUG: When the capability to select on bands was added the logic to check whether all lines and bands have been fit to simple. The effect is that if a subset of lines is fit covering all bands then later attempts to fit other lines will fail because the task will see that all bands have been done but doesn't realize some lines have not been done. The workaround is to set the override parameter to yes. STATUS: In later versions the check on the bands is not done. The user is responsible for dealing with multiband data. NUMBER: 310 MODULE: daophot.psf SYSTEM: V2.10 DATE: Mon Aug 28 08:55:09 MST 1995 FROM: davis BUG: Using the 'd' or ':d' keystroke commands will cause a segmentation violation on Solaris systems and may cause problems on other systems, although apparently not on SunOS systems. The problem was caused by a subroutine which was being called with an unused output variable set to a constant STATUS: Fixed for the next release of iraf. One possible workaround is to fit the psf with the 'f' keystroke command and then use the 's' or ':s' keystroke command to examine and optinally delete the star. NUMBER: 311 MODULE: ccdred.cosmicrays SYSTEM: V2.10.3-V2.10.4 DATE: Mon Sep 25 16:14:16 MST 1995 FROM: valdes BUG: The bad pixel output file is no longer just the coordinates of the replaced pixels but it also includes the flux and ratio values. These extra columns were intended only for development of the interative training option but accidentally were included in the release. The problem this causes is that it is no longer correct to simply use this file with ccdred.badpiximage. The workaround is to use the FIELDS tasks to select the first and second columns and then use that in BADPIXIMAGE. STATUS: The extra output is eliminated in the next release. NUMBER: 312 MODULE: slist, bplot SYSTEM: V2.10.3-V2.10.4 DATE: Tue Sep 26 14:27:26 MST 1995 FROM: valdes BUG: A segmentation error may occur when running SLIST and BPLOT (the latter calls SLIST). This error has been reported with LINUX IRAF but could occur with other versions. The output of SLIST is correct and the error occurs when the task finishes and frees memory. When running SLIST separately you can just ignore the error. This is more of a problem with BPLOT. There is no workaround for BPLOT. Contact iraf@noao.edu for assistance with the BPLOT problem. STATUS: Fixed for the next patch or release of IRAF. NUMBER: 313 MODULE: badpiximage SYSTEM: V2.9-V2.10.2 DATE: Mon Oct 2 18:10:45 MST 1995 FROM: valdes BUG: If the output bad pixel image is specified as a pixel list format image, the .pl extension, the image is likely to contain bad values or cause a system exception (segmentation violation or bus error). This is due to a problem in the system code that writes the pl data to disk. The workaround is to output the bad pixel image in one of the other image formats (such as .imh) and then use IMCOPY to copy to the .pl format if desired. STATUS: This problem is fixed in V2.10.3 and later versions. For reference see the system notes for imwrpx.x on 3/7/94. NUMBER: 314 MODULE: apphot centering routines SYSTEM: V2.10.2 and earlier DATE: Thu Oct 5 11:15:21 MST 1995 FROM: davis BUG: A bug in the centering algorithms thresholding code resulted in the cthreshold parameter being interpreted as the threshold in counts above (below) which pixels were used for centering rather than the threshold in counts above (below) the local data minimum (maximum). The problem occurred because the local data minimum maximum was being computed but not entered into the appropriate data structure. STATUS: This bug is not a problem for 2.10.3 and later versions of apphot or for the version of apphot in digiphotx. NUMBER: 315 MODULE: photcal.mkapfile SYSTEM: V2.10 DATE: Sat Oct 7 12:04:01 MST 1995 FROM: davis BUG: Mkapfile could terminate prematuringly with a floating divide by zero error in interactive mode or a fit did not converge error in non- interactive mode, if the input photometry files were binart ST tables. The problem was caused by missing commas in the tables column template string which resulted in some of the input aperture radii not being correctly read in to mkapfile. STATUS: Fixed in 2.11 and in the external addon package digiphotx. There is no workaround but using input photometry files written in text format instead of ST tables will avoid the problem. NUMBER: 316 MODULE: images.imcombine, ccdred.combine, onedspec.scombine SYSTEM: V2.10 - V2.4 DATE: Thu Oct 19 12:24:43 MST 1995 FROM: valdes BUG: In the PCLIP rejection option, if more than the maximum number of pixels that are allowed to be rejected (set by the nkeep parameter) are rejected at some point along the output line then for all subsequent pi