Patches to 31DEC13

You will find patches to known bugs in 31DEC13 here.

With previous releases, we did not change the frozen release tar ball; user sites were required to download the text files and do any compilations themselves. But, beginning with 31DEC04, we provide binary releases for which user sites are not required to have compilers. Therefore, we have reversed our policy. These patches have been applied to the official code of 31DEC13 and may be downloaded using a 31DEC13 "Midnight Job". Any tarball taken after the change date listed below will already have the patches applied. Of course, you may still download and compile the files as before.

The following is a list of pointers to the different patches in this document.

1. BPASS failed to apply a spectral index correction when SOLINT=-1 even when there was only one calibration source. 2014-01-14
2. LISTR failed to read source information while printing "gains". 2014-01-14
3. DOOSRO run file had a typo in its first line. 2014-01-27
4. UVFND in averaging channels used the real part as both the real and imaginary parts. 2014-02-07
5. DOBAND modes 2 and 4 had an initialization problem. 2014-02-09
6. DOOSRO had a POPS language error 2014-02-10
7. KNTR and PCNTR had minor issues which blocked display of true-color images. 2014-02-11
8. PCCOR did not handle blanked values from the PC table cable-cal measurements 2014-02-11
9. FITLD had trouble finding the correct records for MC and IM tables. 2014-03-11
10. DSKEW did bad things when the input image had a non-zero value of rotation. 2014-03-17
11. CLCAL failed ot re-reference the SN tables when requested. 2014-03-19
12. SNP2D needed clarification of BCHAN and BIF and to write reference channel phases as well as delays. 2014-03-27
13. TIORD had a bad format. 2014-02-31
14. XGAUS and RMFIT called the function routine incorrectly numerous times, mostly with remarkably benign results. 2014-04-02
15. PRTAB could abort when string data was very long. 2014-04-02
16. LISTR did not show the correct scaling for angles in the GAIN listing. 2014-04-11
17. TVFLG and SPFLG interpreted flagged rows in the flag command table as a serious error in a couple of places. 2014-05-29
18. DTSUM omitted the highest numbered antenna from its matrix list if it did not have autocorrelation records. 2014-06-13
19. CALIB in phase-only solutions could return amplitude gains other than 1.0 when only 2 baselines occurred in an interval. 2014-06-27
20. PCAL did not take data flagging into account when averaging data over time. 2014-08-08
21. TYSMO did not apply flagging correctly except if all flags applied to all Stokes. 2014-08-21
22. BPASS did not allocate quite enough dynamic memory causing bad messages when it closed. 2014-09-12
23. SPFLG, TVFLG, FTFLG made a mess of the header in memory when trying to load a fully flagged plane making many operations other than a good TV load fail. 2014-09-12
24. FITLD had trouble finding widely spaced records in the MC table. 2014-09-30
25. 2014-??-??

The remainder of this document is comprised of the patch descriptions pointed to by the above references.

Task:

BPASS

Summary: cannot apply a spectral index correction when all data are averaged together (SOLINT=-1) if there is more than one calibration source. However, it can apply the correction when all data come from one source. It failed to apply a correction and did not tell the user that it was failing.

Files:

/31DEC13/Q/PGM/NOTST/BPASS.FOR

/31DEC13/HELP/BPASS.HLP

Fix:

1. Fetch the files listed above and place them in the corresponding areas ($QPGNOT and $HLPFIL) on your system.
2. COMLNK $QPGNOT/BPASS

Documentor:

Eric W Greisen

Date:

2014-01-14

Task:

LISTR

Summary: in the gain scaling routine used an LUN that was already in use and so failed to read source information when the source number changed.

Files:

/31DEC13/APL/PGM/NOTST/LISTR.FOR

Fix:

1. Fetch the file listed above and place it in the corresponding area ($APGNOT) on your system.
2. COMLNK $APGNOT/LISTR

Documentor:

Eric W Greisen

Date:

2014-01-14

Task:

DOOSRO

Summary: RUN file had a comma at the start of the file causing trouble with RUNning it.

Files:

/31DEC13/RUN/DOOSRO.001

Fix:

1. Fetch the file listed above and place it in the corresponding area ($RUNSYS) on your system.

Documentor:

Eric W Greisen

Date:

2014-01-27

Task:

UVFND

Summary: now has the option to average channels. When doing that average it used the real part of the average for both the real and the imaginary parts. This always gave a phase of 45 degrees and an incorrect amplitude.

Files:

/31DEC13/APL/PGM/NOTST/UVFND.FOR

Fix:

1. Fetch the file listed above and place it in the corresponding area ($APGNOT) on your system.
2. COMLNK $APGNOT/UVFND

Documentor:

Eric W Greisen

Date:

2014-02-07

Task:

DOBAND

Summary: modes 2 and 4 apply the closest BP record in time to the current data. The array of numbers to apply was assumed to contain the values from the earlier time, but was not initialized for that nor kept current if there were major jumps in the times of data application (such as when viewing only one source).

Files:

/31DEC13/APL/SUB/NOTST/BPGET.FOR

/31DEC13/APL/SUB/NOTST/DOBAND.SRT

Fix:

1. Fetch the files listed above and place them in the corresponding area ($APLNOT and some work area of yours) on your system.
2. COMRPL $APLNOT/BPGET
3. source DOBAND.SRT

Documentor:

Eric W Greisen

Date:

2014-02-09

Task:

DOOSRO

Summary: is a procedure to do a pipeline-quality reduction of ordinary EVLA data sets. It contained a mis-use of the POPS command CLRTEMP which seemed to work for a while before it stopped working.

Files:

/31DEC13/DOOSRO.001

Fix:

1. Fetch the file listed above and place it in the corresponding area ($RUNSYS) on your system.

Documentor:

Eric W Greisen

Lorant Sjowerman

Date:

2014-02-10

Task:

KNTR

Summary: required the third axis to have multiple planes when its initialization did not indicate this. PCNTR had an error assigning LUNs which caused it to fail to open a file and thus quit. Both of these were revealed while trying to plot a true-color image plus a contour (single-plane) image.

Files:

/31DEC13/Y/PGM/KNTR.FOR

/31DEC13/Y/PGM/pcNTR.FOR

Fix:

1. Fetch the files listed above and place them in the corresponding area ($YPGM) on your system.
2. COMLNK $YPGM/{K,PC}NTR

Documentor:

Eric W Greisen

Date:

2014-02-11

Task:

PCCOR

Summary: did not handle magic-value blanked values from the PC table. These can occur for the cable-cal values and, when used blindly, had values of 10^98! Fixed the code to use the last valid value when the data are blanked.

Files:

/31DEC13/APL/PGM/NOTST/PCCOR.FOR

Fix:

1. Fetch the file listed above and place it in the corresponding area ($APGNOT) on your system.
2. COMLNK $APGNOT/PCCOR

Documentor:

Eric W Greisen

Date:

2014-02-11

Task:

FITLD

Summary: and INDXR use VLBI tables MC and IM to find the clock, atmosphere, geometric delay, and delay dispersion values used by the correlator. The routines finding the appropriate records had issues leading to occasionally failing to find a usable record even when it was present.

Files:

/31DEC13/APL/SUB/NOTST/GETMC.FOR

/31DEC13/APL/SUB/NOTST/GETDEL.FOR

Fix:

1. Fetch the files listed above and place them in the corresponding area ($APLNOT) on your system.
2. COMRPL $APLNOT/GET{MC,DEL}.FOR
3. COMLNK $APGNOT/FITLD
4. COMLNK $APLOOP/INDXR

Documentor:

Eric W Greisen

Date:

2014-03-11

Task:

DSKEW

Summary: did something rather strange when the input image had a non-zero rotation. The rotation angle was fed to sin and cos without conversion to radians.

Files:

/31DEC13/APL/SUB/OOP/INTERPLATE.FOR

Fix:

1. Fetch the file listed above and place it in the corresponding area ($APLOOP) on your system.
2. COMRPL $APLOOP/INTERPLATE
3. COMLNK $APGOOP/DSKEW

Documentor:

Eric W Greisen

Date:

2014-03-17

Task:

CLCAL

Summary: failed to re-reference the SN table when told to do so. A change meant to reduce the number of meaningless warnings about re-refencing failures caused it to not change the table at all.

Files:

/31DEC13/APL/SUB/NOTST/SNAPP.FOR

Fix:

1. Fetch the file listed above and place it in the corresponding area ($APLNOT) on your system.
2. COMRPL $APLNOT/SNAPP
3. COMLNK $APGNOT/CLCAL

Documentor:

Eric W Greisen

Date:

2014-02-19

Task:

SNP2D

Summary: had not been tested with real data until now. The help file and code were confused about which data set BCHAN and BIF applied to and about the need for a (large) phase to be written to the output SN table along with what should be a modest delay.

Files:

/31DEC13/HELP/SNP2D.HLP

/31DEC13/APL/PGM/NOTST/SNP2D.FOR

Fix:

1. Fetch the files listed above and place them in the corresponding areas ($HLPFIL and $APGNOT) on your system.
2. COMLNK $APGNOT/SNP2D

Documentor:

Eric W Greisen

Date:

2014-03-27

Task:

TIORD

Summary: failed to write the visibility number plus times correctly for samples out of time order.

Files:

/31DEC13/APL/PGM/NOTST/TIORD.FOR

Fix:

1. Fetch the file listed above and place it in the corresponding area ($APGNOT) on your system.
2. COMLNK $APGNOT/TIORD

Documentor:

Eric W Greisen

Date:

2014-03-31

Task:

XGAUS

Summary: and RMFIT rather frequently called the function routine with an argument that requested a derivative evaluation rather than a residual evaluation. Why this did not cause multiple failures is not clear. On Intel 64-bit optimized compilation, an I/O pointer got zeroed very mysteriously and so the XGAUS failed after one spectrum. RMFIT also had trouble with initial guesses since one place still acted as if there are 3 parameters per component when there are now 4.

Files:

/31DEC13/Y/PGM/RMFIT.FOR

/31DEC13/Y/PGM/XGAUS.FOR

Fix:

1. Fetch the fileAS listed above and place them in the corresponding area ($YPGM) on your system.
2. COMLNK $YPGM/{RMFIT,XGAUS}

Documentor:

Eric W Greisen

Date:

2014-04-02

Task:

PRTAB

Summary: could abort trying to handle strings longer than 128 characters. This will arise in NX tables with more than 32 IFs. It also used too few characters when switching to exponential notation for a column when floating notation is generally in use.

Files:

/31DEC13/APL/PGM/PRTAB.FOR

Fix:

1. Fetch the file listed above and place it in the corresponding area ($APLPGM) on your system.
2. COMLNK $APLPGM/PRTAB

Documentor:

Eric W Greisen

Date:

2014-04-02

Task:

LISTR

Summary: uses FACTOR to scale all gain listings when it is greater than zero. The display of the scaling for angles (phase, elevation, parallactic angle) displayed this scale only when it was self-scaled or had a forced scale. The verbose display form also did not test for the values overflowing the format. Now both forms show minuses and asterisks when the format has under- and over-flowed, respectively. The help file said things about FACTOR that were at best misleading.

Files:

/31DEC13/HELP/LISTR.HLP

/31DEC13/APL/PGM/NOTST/LISTR.FOR

Fix:

1. Fetch the files listed above and place them in the corresponding areas ($HLPFIL and $APGNOT) on your system.
2. COMLNK $APGNOT/LISTR

Documentor:

Eric W Greisen

Date:

2014-04-11

Task:

TVFLG

Summary: and SPFLG took flagged rows in the flag-command table as serious errors in a couple of places. Since the UNDO operation can flag rows in this table, the consequences of this could be severe, especially if the last row in the table was flagged.

Files:

/31DEC13/Y/PGM/NOTST/TVFLG.FOR

/31DEC13/Y/PGM/NOTST/SPFLG.FOR

Fix:

1. Fetch the files listed above and place them in the corresponding area ($YPGNOT) on your system.
2. COMLNK $YPGNOT/{SP,TV}FLG.FOR

Documentor:

Eric W Greisen

Date:

2014-05-29

Task:

DTSUM

Summary: omitted the highest numbered antenna from its matrix listing of visibility numbers when that antenna did not have autocorrelation records.

Files:

/31DEC13/APL/PGM/OOP/DTSUM.FOR

Fix:

1. Fetch the file listed above and place it in the corresponding area ($APGOOP) on your system.
2. COMLNK $APGOOP/DTSUM

Documentor:

Eric W Greisen

Date:

2014-06-13

Task:

CALIB

Summary: uses a special routine to deal with solution intervals containing one or two baselines. That routine did not insure that phase-only solutions shoould have amplitude 1.0.

Files:

/31DEC13/APL/SUB/NOTST/ECALC.FOR

Fix:

1. Fetch the file listed above and place it in the corresponding area ($APLNOT) on your system.
2. COMRPL $APLNOT/ECALC
3. COMLNK $QPGNOT/CALIB

Documentor:

Eric W Greisen

Date:

2014-06-27

Task:

PCAL

Summary: ignored data flagging when averaging data over time. This matters when all polarizations are not flagged identically. Changed to do a proper weighted average in each polarization. Also corrected an issue with data having non-standard axis order, and issues with failed solutions especially multiple channels in a row.

Files:

/31DEC13/Q/PGM/NOTST/PCAL.FOR

/31DEC13/HELP/PCAL.HLP

Fix:

1. Fetch the files listed above and place them in the corresponding areas ($QPGNOT and $HLPFIL) on your system.
2. COMLNK $QPGNOT/PCAL.FOR

Documentor:

Eric W Greisen

Date:

2014-08-08

Task:

TYSMO

Summary: did not handle flags correctly. It applied both right and left flags to right only and then dropped the record if right was all flagged. Furthermore, the polarization pattern of the flags was that of the last record read from the FG table, not the one that applied to the FG record attempting to be applied.

Files:

/31DEC13/APL/PGM/NOTST/TYSMO.FOR

Fix:

1. Fetch the file listed above and place it in the corresponding area ($APGNOT) on your system.
2. COMLNK $APGNOT/TYSMO

Documentor:

Eric W Greisen

Date:

2014-08-21

Task:

BPASS

Summary: could corrupt dynamic memory because it allocated almost, but apparently not quite, enough space. This led to bad messages and no history file addition.

Files:

/31DEC13/Q/PGM/NOTST/BPASS.FOR

Fix:

1. Fetch the file listed above and place it in the corresponding area ($QPGNOT) on your system.
2. COMLNK $QPGNOT/BPASS

Documentor:

Eric W Greisen

Date:

2014-09-12

Task:

SPFLG, TVFLG, FTFLG

Summary: corrupted the header array in memory when trying to load a fully flagged data plane. These tasks left things in an otherwise normal state to continue working on the good plane still displayed. But the corrupted header led to a variety of failures. Added protection to restore the previous header should there be no valid data to load.

Files:

/31DEC13/Y/PGM/NOTST/SPFLG.FOR

/31DEC13/Y/PGM/NOTST/TVFLG.FOR

/31DEC13/Y/PGM/NOTST/FTFLG.FOR

Fix:

1. Fetch the files listed above and place them in the corresponding area ($YPGNOT) on your system.
2. COMLNK $YPGNOT/{SP,TV,FT}FLG.FOR

Documentor:

Eric W Greisen

Date:

2014-09-12

Task:

FITLD

Summary: interpolates within the MC table to find the clock and atmosphere corrections. It was only willing to look within (initially) plus or minus 1 minute since all old MC tables were spaced every minute. It was changed to 2.0 minutes, but that is still not enough. Changed it to look within plus or minus 5.1 minutes.

Files:

/31DEC13/APL/SUB/NOTST/GETMC.FOR

Fix:

1. Fetch the file listed above and place it in the corresponding area ($APLNOT) on your system.
2. COMRPL $APLNOT/GETMC.FOR
3. COMLNK $APGNOT/FITLD
4. COMLNK $APLOOP/INDXR

Documentor:

Eric W Greisen

Date:

2014-09-30

Task:

Summary:

Files:

/31DEC13/

Fix:

1. Fetch the file listed above and place it in the corresponding area ($???) on your system.
2. COMLNK $

Documentor:

Eric W Greisen

Date:

2014-??-??