;----------------------------------------------------------------------- ; Copyright (C) 2003-2005 ; Associated Universities, Inc. Washington DC, USA. ; ; This program is free software; you can redistribute it and/or ; modify it under the terms of the GNU General Public License as ; published by the Free Software Foundation; either version 2 of ; the License, or (at your option) any later version. ; ; This program is distributed in the hope that it will be useful, ; but WITHOUT ANY WARRANTY; without even the implied warranty of ; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ; GNU General Public License for more details. ; ; You should have received a copy of the GNU General Public ; License along with this program; if not, write to the Free ; Software Foundation, Inc., 675 Massachusetts Ave, Cambridge, ; MA 02139, USA. ; ; Correspondence concerning AIPS should be addressed as follows: ; Internet email: aipsmail@nrao.edu. ; Postal address: AIPS Project Office ; National Radio Astronomy Observatory ; 520 Edgemont Road ; Charlottesville, VA 22903-2475 USA ----------------------------------------------------------------------- This file must be used to record any changes made to the Master version of AIPS kept in SOCORRO. An entry is: line 1: Date Program name Your name line 2: What was done (may span multiple lines) line n-1: Where has this change been moved (e.g. 15JUL94, nowhere) line n: Blank The details of line 1 are important since programs are used to find specified strings and do conversions on them. It MUST be typed as follows: 1 blank, 4 digits, a period, two blanks, then the Month, day, comma, year, some space, a title, some more space, your name. *********************************************************************** *********************************************************************** Changes while 31DEC04 is TST *********************************************************************** *********************************************************************** 11456. December 1, 2003 several Eric Changed 31DEC03 to 31DEC04 in install.pl, update,pl, INSTEP1, SETPAR, GETRLS, AIPSPATH.*SH, UPDCONTROL. Moved nowhere. 11457. December 2, 2003 FILLM Eric Fixed time advances while skipping tests to handle day roll over. There was a problem only when no data had yet been selected. Moved to 31DEC03 this date. 11458. December 2, 2003 $YPGNOT/DELZN Leonia Creating output file DELZN has carried out calculation of the values for the times given at the CL table rows. Now DELZN can calculate the values of the output file for the times even distributed over the time range of the input data. The new option works under control of APARM(8). The new version is useful when the output CL table (target sources) is attached to the different UV data. Moved nowhere. 11459. December 3, 2003 MAKE.MNJ Eric The part in do_daily that defines AIPS_ROOT should come ahead of the line that uses it. Moved to 31DEC03 this date. 11460. December 4, 2003 install.pl Eric Grammar error caused special hack for HPs not to work. Corrected $arch to $ARCH. Moved to 31DEC03 this date. 11461. December 8, 2003 SNSMO Eric There was no default for SMOTYPE despite what the help said. The clipping routines had a variety of troubles. Put in counters for what was deleted in CLPAPH, CLPRAT, CLPDEL. The amp/phase clip assumed that both polarizations needed to be present and so showed both bad if one was. This caused odd behaviors and failures to clip. Dropped this both pol nonsense. Fixed the setting of the BAD flag so that fewer records would be updated - only those actually changed. Moved to 31DEC03 this date. 11462. December 8, 2003 ZDCHIN, FILAIP Eric Set the default AIPSmark to 20 not 1 in ZDCHIN. Have FILAIP not re-init the SP and IC files if they pre-exist. The number of records in a user-independent catalog was not set (no longer used much anyway). Moved nowhere. 11463. December 8, 2003 GRBOXS Eric Modified messages to tell users what to expect a bit better. Moved nowhere. 11464. December 9, 2003 UVFIX Eric Two major errors in processing multi-subarray data: 1. The counter used to show start of data only worked for the subarray of the first sample. All others used parameters based on that one. 2. When bug 1 was fixed, several parameters were found to be set to the output values by the first subarray. This included EPOCH and the coordinates and so made handling the second one problematical. Made a bunch of parameters be picked up and saved at the beginning and then reused when needed. Moved to 31DEC03. Cannot patch 31DEC02 since UVFIX differs in other ways too. 11465. December 10, 2003 UVPLT Eric Fixed to do self-scaling on the binned values where appropriate. Corrected which values end up in the bins. Moved nowhere. 11466. December 11, 2003 BLOAT Eric The alternate reference pixel was not reset when the frequency reference pixel was. Moved to 31DEC03 this date. 11467. December 11, 2003 VPFLG Eric Add option to flag all IFs if 1 is bad. Moved to 31DEC03 this date. 11468. December 11, 2003 UVFLG Eric Expand error message when trying to flag ampl or weight on a multi-source (FG table) data set. Moved to 31DEC03 this date. 11469. December 12, 2003 CLIPM Eric Changed meaning of APARMS - APARM(5) = min allowed weight, 6-8 were 5,6,8 before, APARM(9) controls all IF flagging. Moved nowhere. 11470. December 16, 2003 PLOTR Eric Corrected symbol size by multiply or divide by SQRT(XYRATIO) to keep the symbols square when XYRATIO is not 1. Moved to 31DEC03 this date. 11471. December 18, 2003 CookBook Eric Corrected DELZN remarks in LaCook9.tex and COOK9.PS and COOK9.PDF. Moved to 31DEC03 this date. 11472. December 22, 2003 UVPRT Eric Changed displayed remark: u, v, w are scaled to selected frequency, not the reference frequency. Moved to 31DEC03 this date. 11473. December 23, 2003 ALGSUB, ALGSTB Eric Added field number to messages to show progress when more than one field is involved. Moved nowhere. 11474. December 29, 2003 ELINT Leonia ELINT has forced the imag part of the output CL table amplitude to zero. As a result ELINT has given the right output CL table only if the input CL table has zero for imag part of amplitudes. It fixed. Moved nowhere. 11475. January 1, 2004 install.pl Eric Changed 31DEC03 version of install.pl to myver NEW (stops the MNJ). Found that it had called itself 31DEC04 since 4 Dec! Moved nowhere (only 31DEC03). 11476. January 3, 2004 CLCAL et al. Eric Changed CLSMO Fix spelling of comment SNSMO I swear I did this once already: dropped an initial insistence on having both polarizations at each time. This got in bad trouble in some cases. CLCAL The smoothing of real and imaginary is not ideal and was changed back to smoothing phases keeping track of the last good phase to try to avoid wrap troubles. Moved to 31DEC03 - opening it just for this. 11477. January 3, 2004 AIPS, ZDCHIN Eric Modified the Copyleft statement, adding 2004 in the former, adding 1995- to the latter. Moved nowhere. 11478. January 4, 2004 SNPLT, WETHR Eric Changed these to do a fixed X scale when a start and stop time are given. Fixed the comments in the help files - PIXRANGE was and is fully a fixed scale plot. Moved nowhere. 11479. January 5, 2004 DBCON Eric The buffers for copying and concatanating tables were awfully small. Fixed to 3*MAXCIF which will cover BP tables. The basic VLBI tables were not concatenated - IM, MC, PC, AT, CT, OB, and GC tables were added to the concatenation list. NOTE: CL tables > 1 are copied from the first file but any in the 2nd file are ignored. SN amd BP tables are copied from the first file and any from the second file are ignored. Comments to this effect were added to the help file. Moved to patch 19 January. 11480. January 5, 2004 DSORC Eric New task - renumbers sources. Moved nowhere. 11481. January 6, 2004 PLOTR Eric The step line function did not start well. Corrected it. Moved nowhere. 11482. January 7, 2004 POSSM Eric Used LUN 39 to be SU ISTAB call - this is marked as a tape device here - we have to fix tape handling!!!! Moved nowhere. 11483. January 8, 2004 FITLD Eric It regarded the number of rows in an FQ table as the same as the max FQ number in the table. Fixed to get the max FQ and use it some places. VLBAUTIL will need fixing. Moved to patch January 19. 11484. January 11, 2004 Tape LUNs Eric We have used for tapes LUNs = 31 -> 31+NTAPED. This has caused trouble (see POSSM above). The default DEVTAB has 31-39 set aside. It is now time to change this. I expect that no one has used LUNs > 100 so I will for tapes use LUN = 129 - NTAPE Simple minded changes to the following just to change this formula: GRITP GR2TEX BAKLD BAKTP FILLR FITAB FITLD FITTP GSCAT IMLOD M3TAR MK3IN MK3TX UVLOD Other changes ZDCHIN Remove old DEVTAB(30+i) = 2, set DEVTAB(129-i) to 2. ZTAPE Changed several tests on LUN. ZTPOPN Changed tests on LUN, changed MTLUN (lock disk), changed FORMATs ZTPMIO Changed tests on LUN and FORMATs (I2 -> I3) ZTPWAT Changed tests on LUN and FORMATs (I2 -> I3) ZTPCLS Changed tests on LUN and FORMATs (I2 -> I3) AU4 Changed LUN formula TAPIO Changed test on LUN LSERCH Changed odd usage of LUN tests TPMON Changed setting of tape LUN and tests/comments FILLM Changed formula and tests on the LUN in the disk reading routine. Moved nowhere. 11485. January 12, 2004 WXSEL Eric The WX table needs to be selected too - in UVCOP, UJOIN, SPLAT, and FLGIT. Moved nowhere. 11486. January 12, 2004 STARTPMON Eric Now that tape units > 9 actually might work, it is time to fix STARTPMON - it looked for and made, e.g. TPMONa and then ordered TPMON10 to start! Moved nowhere. 11487. January 14, 2004 CALIB, FRING Eric 1. The subroutine that divides the vis data by the model is UVDOUT. The data weights it put out were those that came in times the amplitude of the model. Since weights are 1/sigma**2, it should have been times the amplitude squared of the model. 2. Created adverbs SOLSUB, SOLMIN for solution subinterval and minimum size. Changed POPSDAT.HLP, NEWPARMS.001, and new help files. 3. Changed CALIB - added these two adverbs. Changed to use dynamic memory for big solution data arrays which now are also dimensioned by SOLSUB. Dropped apparent loop over Stokes which could never have worked (did not back up the reads) but which was probably never done. Add up the data in sub-intervals, do a solution, then shift sub-intervals. 4. Changed FRING - added these two adverbs. Changed big dynamic arrays to include both polarizations since we have to retain correlator 1 data for subintervals 2-N while doing the 2nd correlator. Changed then to read in both correlators at once and solve properly. Shifts the arrays by sub-interval, adding data at the end for the next sub-interval. Changed also meaning of PRTLEV. 5. Changed KRING - added these two adverbs. Changed it to do the subintervals. Moved nowhere. 11488. January 14, 2004 OHGEO Eric The function PSNCVT in POSITION.FOR assumed that the X coordinate in the input image went with the X coordinate version in the out image without checking. It can handle simple transposes of the Ra-Dec. Fixed to check and blow awy some of the bad ones and to handle simple transposes. Moved to patch January 19. 11489. January 15, 2004 CLCOR Leonia CLCOR has used the different sign for the correction of phase (-) and delay (+). The sign has to be identical (+). It is fixed. Error was only in option CLOC. Moved nowhere. 11490. January 15, 2004 VLACLCAL Amy Updates VLACLCAL.HLP to take CLCAL smoothing changes into account. Moved nowhere. 11491. January 16, 2004 UVFLG Eric Fixed FLAGVER default remark - the code already used the highest but the help file said 1. Moved nowhere. 11492. January 18, 2004 remote tape Eric Fixed ZTPOPR Tested LUN to be 31-31+NTAPED for real tape. ZMOUNR Uses an lUN - should be the right kind even though it does not open ZVTPOPR Limited LUNs to <= 50! TPMON Assumed user has message file on remote machine Moved nowhere. 11493. January 19, 2004 IMFIT, JMFIT Eric Reordered the adverbs and changed task to allow the answers to be returned in GMAX, GPOS, and GWIDTH and the uncertainties in DOMAX, DOPOS, and DOWIDTH. Changed POPSDAT.HLP to add adverb RADIUS and changed default of DOBLANK to 0.0. Added RADIUS to NEWPARMS.001. Changed AU2A to DEFAULT DOBLANK to 0. Added RADIUS to IMFIT and JMFIT. If RADIUS > 0, the rms will be found in a circle of that radius in pixels, if = 0 the rms will be taken from the header or the full plane as previously, if RADIUS < 0 then abs(RADIUS) will be taken as the rms. Moved nowhere. 11494. January 19, 2004 TGIN, SGIN, VGIN Eric Changed AU2A and the help files to allow for sort options in the listings. Moved nowhere. 11495. January 19, 2004 Read-only file system Eric DVDs can be mounted on read-only systems but are still usable if the catalog routines do not insist on updating the last access time. Changes: DDRC.INC Include for parameter - to avoid DDCH recompile everything ZDRCHK Prepares parameters for ZDRCH2 ZDRCH2.C Opens read-write, if fail opend read only, sets parameters based on success. AIPS Call ZDRCHK after user number known and CATCR GTPARM Call ZDRCHK after user number known ZOPEN Set EXCL to -1 if read-only file system ZDAOPN.C Test for read-only as well as no access errors Handle excl==-1 as read-only open. ZCREAT Forbid any creation on a read-only file system CATDIR Forbid OPEN and CLOS on read-only disks, do not change status on read only disks CATIO Update status only on writable disks CATOPN Update date only on writable disks MCREAT Test near start for DSKLOK=1 and loop or die SCREAT Test near start for DSKLOK=1 and loop or die UVCREA Test near start for DSKLOK=1 and loop or die Moved nowhere. 11496. January 20, 2004 XAS Eric Rearranged the code in init.c to allow the screen size to be over-ridden by user Xdefaults options (AIPStv*xPixels and AIPStv*yPixels). Smaller than the visible screen works well, bigger may not play well with your screen manager. Also updated XAS.HLP. Moved nowhere. 11497. January 21, 2004 XTRAN Eric The initial read of the text file damaged the error return so that if an error was found in reading the file, it was not passed on to the following code and the program kept going. Also added the ability to skip lines in the text file beginning with $ - to allow comments. Moved nowhere. 11498. January 21, 2004 SAD Eric Fixed the sorting of the solution display to do S, X, Y, R, and D correctly taking into account real RA and Dec even when the image is rotated. Moved nowhere. 11499. January 21, 2004 IMFIT, JMFIT Eric Created new adverbs FMAX, FPOS, FWIDTH for the output of iMFIT and JMFIT - over-writing GMAX, GPOS, GWIDTH messes up some procs. Changed IMFIT and JMFIT help files, POPSDAT, NEWPARMS. Made new help files. Corrected Fortran files to reset the DOMAX et al parameters only if they are > 0. Moved nowhere. 11500. January 21, 2004 SETFC Eric Changed to return CELLSIZE and IMSIZE used - they change values if they start = 0. Moved nowhere. 11501. January 22, 2004 DELZN, CLCOR Leonia DELZN created the outfile with one column data corresponded to the atmosphere zenith delay and CLCOR (option 'atmo') read such a file format interpreting the only data column as a atmosphere zenith delay. If an user has wanted to make correction of atmosphere zenith delay and clock delay, CLCOR has carried out the wrong calculation of both delay and rate correction based on the only data column at the infile. Now DELZN creates the outfile with four data column: atmosphere zenith delay, rate of the atmosphere zenith delay, clock delay and clock drift. The format of the infile in CLCOR has been changed relevantly. The help files are updated. Moved nowhere. 11502. January 22, 2004 AIPS Eric Recent changes pointed out that certain subroutines did not always close all files - this left them open blocking later reuse. Fixed AU2, AU2A, and AUC. Moved nowhere. 11503. January 25, 2004 FRING Eric The weight array was local and so had a much larger dimension than required - but the zeroing thought it was just right. Solutions for the second IF were badly damaged when the number of working antennas increased in the middle of a data set. Moved nowhere. 11504. January 25, 2004 VLARUN Lorant Implemented minor changes to robust, inseq and refant to accomodate for Michaels requests - demonstration at xraydio Reserving major upgrade to make it infinitely robust to later 11505. January 27, 2004 install.pl Eric The test to see if a machine was in a pre-existing HOSTS.LIST used a numeric compare (==) which apparently is always true when comparing e.g. BOB and LOCALHOST. It should and now does use a string compare (eq) which works. Also added a comment to the user about the password in FILAIP. Moved eq to the 31DEC03 ftp area. 11506. January 28, 2004 tapes Eric ZTPOPR had a GT where a GE was needed - so a tape number = your highest one used as a remote tape would fail to open. MOved nowhere. 11507. January 29, 2004 SGLAST Eric Added explanatory message about lASTEXIT on error. Moved nowhere. 11508. January 29, 2004 TI2HA Eric Added advice about DBCON and PUTHEAD to be used when running TI2HA. Moved nowhere. 11509. February 9, 2004 Character in Z Eric I noticed that CHARACTER variables are sometimes being sent directly to C routines. This is a serious no-no. Changed ZABORT $APLGEN: changed to send HOLLERITH to ZABOR2 Changed ZABOR2 stub ($APLUNIX) comments ZACTV8 $APLUNIX: use HOLLERITH to call ZACTV9 ZERROR $APLGEN: changed to use HOLLERITH from ZERRO2 - changed ZERRO2 ($APLGEN and $APLVMS) accordingly ZFIO $APLGEN: changed to use HOLLERITH to ZFI2 and ZTKFI2 except on VMS. Changed comments and some code in the $APLGEN stubs for ZFI2 and ZTKFI2. ZFREE $APLGEN: changed big name buffer to HOLLERITH, revised $APLGEN and $APLVMS versions of ZFRE2 to support this. ZFULLN $APLGEN: changed to call ZPATH and ZMKTMP with HOLLERITH, revised $APLGEN and $APLVMS versions of ZPATH and ZMKTMP to support this. ZLASCL $APLGEN: use HOLLERITH to call ZLASC2 except for VMS, changed comments in $APLGEN version ZLASC2. ZLPCLS $APLGEN: use HOLLERITH to call ZLPCL2 except for VMS, changed comments in $APLGEN version ZLPCL2. ZMIO $APLGEN: do VMS and Unix differently incl single buffer Unix and hollerith to ZMI2 - drop $APLUNIX version of ZMIO.FOR. ZMI2 $APLGEN: improve comment re OPER ZMOUNT $APLGEN: use HOLLERITH in calling ZCRLOG and ZMOUN2, improved comments in $APLGEN ZMOUN2 and ZCRLOG. ZMSGDK $APLGEN: Use HOLLERITH for non-VMS in calling ZMI2. ZMSGER $APLGEN: Use HOLLERITH in calls to ZERRO2. ZPASWD $APLGEN: new calls ZPRPAS w or w/o HOLLERITH ZPRIO $APLGEN: calls ZPRI2 w or w/o HOLLERITH ZPROMP $APLGEN: new calls ZPRMPT w or w/o HOLLERITH ZROPEN $APLGEN: use HOLLERITH for ZROPN2 call, fix $APLGEN/ZROPN2 comments ZSHCMD $APLGEN: make new subroutine to call ZSHCM2 w or w/o HOLLERITH. Rename VMS, UNIX, IBM, CRI versions to ZSHCM2. ZSSSLK $APLGEN: use HOLLERITH in call to ZSSSL2, change $APLGEN/ZSSSL2 comments ZSSSOP $APLGEN: use HOLLERITH in call to ZSSSO2, change $APLGEN/ZSSSO2 comments ZTABOR $APLGEN: new calls ZTKILL w or w/o HOLLERITH ZTACTQ $APLGEN: change to call VMS ZTACT2 with CHARACTER, rest with HOLLERITH ZTACT2 $APLGEN: changed comment ZTAPE $APLGEN: use HOLLERITH in call to ZTAP2, fix error message, change comments on $APLGEN/ZTAP2 ZTKILL $APLGEN: change comments re task name ZTKOPN $APLGEN: use HOLLERITH to call ZTKOP2 on non-VMS and fix $APLGEN/ZTKOP2 comments ZTPMIO $APLGEN: change to do CHARACTER and old ZTPMID call sequence to VMS, HOLLERITH and new call sequence to the others. Delete $APLUNIX/ZTPMIO.FOR. ZTTOPN $APLUNIX: call ZTTOP2 with HOLLERITH ZTXMAT $APLGEN: call ZTXMA2 with filespec HOLLERITH except for VMS - the VMS routine uses I think the wrong length for file names which were already HOLLERITH. Changed comments in $APLGEN/ZTXMA2. ZVTVOP $APLGEN: Use HOLLERITH in call to ZVTVO2 and fix comments in $APLGEN/ZVTVO2. ZWHOMI $APLGEN: CHANGE CALL TO ZCRLOG TO HOLLERITH AU2 Changed to call ZTABOR AUC Changed ZPRPAS to ZPASWD PASWRD Changed ZPRPAS to ZPASWD PREAD Changed ZPRMPT to ZPROMP RDUSER Changed ZPRMPT to ZPROMP Moved nowhere. 11510. February 9, 2004 ZXHLP2 Eric Changed all cv references to aoc. Moved nowhere. 11511. February 9, 2004 CLSMO Eric Michael Bietenholz pointed out that the code swapped rate and delay columns and swapped the smoothing times of the two - so doing the right thing most confusingly. Rearranged. Moved nowhere. 11512. February 11, 2004 PLOTR Eric Allow comment lines starting with $ or # anywhere. Moved nowhere. 11513. February 11, 2004 XHELP Eric XHELP limited the length of a name to 8 characters - 10 is better as in explain etc al. Changed HELPS.FOR. Moved nowhere. 11514. February 11, 2004 BPASS Eric Added warning about data averaging across sources. Moved nowhere. 11515. February 11, 2004 TVPSEUDO Eric Changed TVPSUD and COLORC top add yet another color scheme suggested by Lincoln Greenhill. Moved nowhere. 11516. February 12, 2004 FCLEVEL Eric The default code seemed suspect - replaced the full case statement with a different grammar that appears to work everywhere. Moved nowhere. 11517. February 12, 2004 AIPS Eric We get lots of complaints about damaged LASTEXIT files when machines crash. I do not understand this - the abort handler does not reference LASTEXIT. Changed AIPS.FOR so that if SGLAST read does have an error, it is reported and INIT called. Moved nowhere. 11518. February 12, 2004 ZXHLP2 (XHELP) Eric Fixed up $SYSUNIX/ZXHLP2 and $SYSPERL/ZXHLP2.PL to work on the present NRAO web setup. Moved nowhere. 11519. February 15, 2004 CHANGES, AIPSPLIT.TAR Eric Removed $Id date crap that confuse the cvs system. Moved nowhere. 11520. February 15, 2004 START_TVSERVERS Eric The previous fix that set $DISPLAY to $TVDISP:0 is a bit dangerous and should be applied only for Inet sockets. To make this take effect, managers must move the file from $SYSUNIX to $AIPS_ROOT fixing the $AIPS_ROOT definition. The tv=local option did not work when $DISPLAY=server:12.0 - i.e. when the display is at the end of an ssh link. Moved nowhere. 11521. February 15, 2004 AIPS Eric Changed SGLOCA to return errnum 45 for no SAVE/GET directory when trying to read LASTEXIT. Changed AIPS to skip message in that case and improved the message about a new vocabulary. Tested that a bad LASTEXIT file is survived well. Moved nowhere. 11522. February 18, 2004 VLBAARCH.001/HLP Lorant New procedure to take care of archiving VLBA data from distribution tape or disk into single source and single frequency (-ID) files. Essentially a combination of VLBALOAD, VLBAFIX (taken from VLBAUTIL) and FITAB. Meant to be used by the analysts and myself 11523. February 20, 2004 CLIPM, UVMLN Eric Both of these messed up the relationship between source table row number and source number - they wrote flag entries for row number not source number. Moved to patch area today. 11524. February 23, 2004 SPLAT Leonia SPLAT has always copied PC tables even if DOCALIB=1 although HELP file has reported that "PC table is not copied considering PCAL calibration had been carried out". It is changed to not copy PC tables (assembling option) if the calibration has been applied (DOCALIB.GT.0). SPLAT.HLP is changed to make the statement more clear. Moved nowhere. 11525. February 23, 2004 TBOUT Eric For some reason local variables dimensioned only 50 were used when an include wa salso used that provided correct variable sizes (dimension 128 columns the max in aips). HF tables have more than 50 columns. Moved nowhere. 11526. February 24, 2004 DELZN Leonia OPTYPE='DELA' has been used when the multi band delay column of the input SN table is used. I have changed 'DELA' for 'MDEL' to exclude a possible confusion with the single band delay. Moved nowhere. 11527. February 24, 2004 REFREQ Amy Added variables and some explanation to help file. Made procedure more rational. Removed hidden adverb UNITS, I did not understand why anyone would want to use this adverb. Will see if anyone complains. Moved nowhere. 11528. February 24, 2004 CALIB, SETJY Eric Changed CALIB per Robert Braun to use the V polarization from the source table when calibrating the RR (I+V) and LL (I-V) correlators. This is particularly useful for WSRT with equatorial mounted linear feeds where XX=I-Q and YY=I+Q. If they are changed to Stokes -1 to -4 and the Q flux put in the V place then CALIB will work for the usually non-zero Q of calibrators. Changed SETJY to allow changing the Q,U,V fluxes while leaving the I (ZEROSP(1)) flux set by CALC alone. Moved nowhere. 11529. February 26, 2004 START_AIPS Eric Dieter Engels (University of Hamburg) suggested a much more comlex grammar for parsing the SSH_CLIENT variable which solves the new format used by some systems for its value: ::ffff:134.100.120.133 33151 22 Also put in SSH_CONNECTION since SSH_CLIENT may not be set when the other is. Moved nowhere. 11530. February 26, 2004 RLDIF Eric New task: does the LISTR operation to display RL and conjugate(LR) for calibrators to determine the R-L phase difference. It returns this in adverb CLCORPRM and optionally does the usual printing. Moved nowhere. 11531. February 26, 2004 LISTR Eric The "vector" sum line shown at the bottom of the matrix listings assumed that the real and imaginary parts in the two halves of the matrix are supposed to be the same. This is not always true - changed it to sum the two halves separately. Moved nowhere. 11532. February 26, 2004 SLBINI Eric It did not check that the rotation had nothing to do with the slice (AXTYP=4) before bumping type 3 (absolute) plots to type 4 (relative). Affect ISPEC the most, other tasks relinked: AIPS, SL2PL, PFPL2, PLCUB, SLFIT, XBASL, XGAUS, XPLOT. Moved nowhere. 11533. February 28, 2004 VLBAFIX/VLBASUBS Amy VLBAFIX/SUBS used to force everything into one subarray, I added an option to actually split the data into multiple subarrays using USUBA OPCODE='AUTO'. Moved nowhere. 11534. March 1, 2004 WX tables Eric Fixed SPLAT, FLGIT, UJOIN to avoid copying the WX table during the mass table copy, saving it for the WXSEL intelligent copy done later. Otherwise the latter tried to overwrite the former causing a variety of possible problems. Moved nowhere. 11535. March 3, 2004 QEDIUTIL Eric The UV editing portion did not update pointers properly when flagging a single baseline and either NEXT BASELINE or ENTER OTHER ANT functions were done. The correct antennas appeared on the screen but not in the flag table. Moved to patch area. 11536. March 8, 2004 ASTROMET.HLP Amy Updated, mostly adding comments about calibration transfer and referring to the VLBAUTIL scripts that might be useful. Also cleaned up the text a bit. Moved nowhere. 11537. March 12, 2004 CONFI Leonia There has been two options of finding maximum of the beam. One of them is two complicate and does not have any advantage. I have removed it. The default value for the minimum radius of optimization is added. This default is equal 1.2*lambda/ARRAYSIZE and correspond to the edge of the main lobe of the synthesized beam. Moved nowhere. 11538. March 13, 2004 TBOUT Eric The "simple" fix I made caused several problems. The length variable was replaced by a standard one in the the call to GETHUT but not where it was used making the length of character and bit outputs 0 rather than correct. The subroutine GETHUT declared the "known" lengths of the strings in its argument list. This is very bad when strings of some other (presumably larger) length are sent in as they were after the "simple fix". Corrected GETHUT. Moved nowhere. 11539. March 13, 2004 FILLM, PRTTP Eric The new archive system makes it very easy to download a very mixed bag of data and users expect FILLM to load it all in one run. Changed: a. It failed to recognize new data in which the number of streams increased, probably because it happened at a file boundary. Changed to not write more NX, CL et al records after the change has arisen and to force a close and reopen operation when the number of streams changes. b. It did not allow CL, TY et al intervals less than the integration time. Given that times are inaccurate when 200 days are added, this meant that a large number of CL records did not get written that should have been. c. Test for wheether the array center was moved from 0 to account for Pie Town at Q band used an uninitialized variable whan MXANT (29) should have been used. This variable became 29 after the routines were used once but Q band AN files may be funny! c. FILLM has happy to combine data from different configurations as if they were all from the first one encountered!! Changed it to read in pre-existing antenna files and to use the antenna station IDs to decide whether to concatenate or not. PRTTP was given the fix that FILLM has for reading short records at the end of disk files. It was still reporting a short record with no I/O error. Moved nowhere - should go to patch sometime. 11540. March 14, 2004 VLBAARCH.001/HLP Lorant The analysts didn't get enough disk space so far, so I modified VLBAARCH to be able to switch disks during operation, when new files are created, and make "zap" in the proc accept "indisk=0". Moved nowhere. 11541. March 13 & 14, 2004 ZXHLP2 Eric Fixed it so that it will do min match on symbols of 4 or more characters and exact match on 1-3 character symbols. Note that it takes the first symbol to match in the alphabetized list HLPIT.LIS. Cleaned up code to use functions and to do as many symbols as there are in a line except not the symbol for which we are doing the help. It is slwer now but really slick! Moved to cgi web location, nowhere else. 11542. March 16, 2004 adverbs help files Eric Corrected BATNLINE, GRADDRES help files - they are adverbs not verbs. Dropped from POPSDAT and NEWPARMS unused adverbs: PGAUSS, MGAUSS Changed POPSDAT.HLP to declare more adverbs properly. Revised AU2A to DEFAULT a couple more adverbs in special ways and to handle an init of characters > 4 in length. Created new help files for: AVOPTION BPASSPRM DEFER DELTAX DELTAY DOBTWEEN DOEBAR DOFIT PHASPRM SEARCH SNCORPRM SOUCODE SPECINDX TAU0 TRECVR VLBINPRM X XAXIS Y Moved nowhere. 11543. March 17, 2004 list, ABOUT files Eric For changes before today, changed HLPIT.LIS and ABOUT files ADVERB ANALYSIS ASTROMET CALIBRAT COORDINA MODELING OPTICAL POLARIZA RUN TASK UTILITY UV VLBI For the new help files changed ADVERB ANALYSIS CALIBRAT IMAGING PLOT POPS SPECTRAL TASK TV UV VERB VLBI Moved nowhere. 11544. March 18, 2004 AU2A, POPSDAT Eric SNVER should default to 0 - it has been 1 in POPSDAT and I fixed AU2A to match. Fix both to be 0 now. Moved nowhere. 11545. March 25, 2004 DELZN Leonia The OPTYPE='PHAS' has been added. This option allows to pick up phase delay instead of multi band delay (OPTYPE='MDEL'). The ambiguity problem is waiting for solution!!!! Moved nowhere. 11546. March 25, 2004 $APGNOT/FIT2A Leonia Number of digits at the third column of the output file has been changed from 3 to 6. Moved nowhere. 11547. March 26, 2004 BLCAL Amy BLCAL was not reading the CMETHOD input correctly because CMETH was declared as only one character long. Moved nowhere. 11548. April 2, 2004 DELZN Leonia The new option not fitting clock has been added. Now there is no default for number of terms in the fitting polynomial for clocks as it was originally (APARM(3)=0 => 2) Now Aparm(3)=0 means no fitting for clocks. The help file has been modified in particular to make more clear what is plotted. Moved nowhere. 11549. April 3, 2004 CALIB Eric If DOFIT was set, then GCALC2 was called to do a fit for some but not all antennas. Unfortunately, it used all included antennas for the data but had only gain=(1,0) for those antennas it was told not to fit. This got rather in the way of fitting. Changed it to omit all data which contain an antenna that is not to be solved. The fits are now much better. So why is this option there - ANTENNAS would do? Moved nowhere. 11550. April 6, 2004 DELZN Leonia Some misprints (which do not lead to an error) found in the program. Mapping functions are stored now instead of relevant elevations. This simplifies the program and makes easier to modify DELZN if it is required. Moved nowhere. 11551. April 7, 2004 CookBook Eric Changed $DOCTXT/LaCOOK?.tex and $AIPSPUBL/COOK?.PS mostly to allow for making html and linked pdf files. This involved changes to all chapters, mostly in the nature of things invisible in the output PostScript files. The macro file LaCook.mac was revised to make every section self-labeling, to write 2 kinds of index outputs with section reference in one of them, to have new macros for the section-ref index (LaCookJ), and to choose between jpg and eps forms of the figures. All figures were worked on to be 990 points or so wide in eps and then jpeg versions were made of them. Real changes to content were also made: 4 - Changed to talk about RLDIF rather than LISTR for the polarization right-left phase. Fixed CL version info from CLCOR and elsewhere. 9 - Improved the DELZN section (per Amy) and added multi-band delay section; CLCOR makes new gain versions. A - Changed LISTR to RLDIF for polarization and fixed GAINUSE after CLCOR. 13 - Updated ABOUT sections. Also Index and Table of Contents. Dropped $AIPSPUBL/COOK*.PDF and made new full-book, cross-referencing $AIPSPUBL/COOKBOOK.PDF instead. Changed COOKNX.FOR to allow for an italic index entry and to limit the number of references in any one row. Created COOKNS.FOR to make the section-reference index from the other form of output file. Moved nowhere. 11552. April 7, 2004 ZCLOSE Eric Some tests on tape LUNs still remained - in $APLUNIX and $APLGEN versions of ZWAIT and $APLGEN/ZCLOSE. It would not close files or wait on them if the LUN was in the range 31-30+NTAPED which is bad if NTAPED >= 10. Fixed comments in ZTPOPN, ZTPOPR, TAPIO, FILLM, and PRTTP about the LUNs and dropped a stupid line of code from TPIOHD. Relinked everything! Moved nowhere. 11553. April 7, 2004 SYSTEM Eric Changed AU3B, POPSDAT, NEWPARMS to make a new verb SYSTEM to send a command to the operating system. New help files SYSTEM, SYSCOM, SYSOUT. Moved nowhere. 11554. April 8, 2004 FILLM Eric For some reason, FILLM would not check a source table for entries if the number of vis was < 1. Thus a first scan with no data would create a source table entry for that source but not acknowledge it when the next source showed up. So the two ended up with the same source number although they were known to be different otherwise and the third source would be called number 3. Now if the source table has data it looks into it. Moved nowhere. 11555. April 9, 2004 SETFC Eric Changed it to return NFIELD as well as CELLSIZE and IMSIZE. Moved nowhere. 11556. April 10, 2004 Calibrator models Eric Wrote CALWR and CALRD taks to write source models into FITS disk files and CALRD to read them. The former will be mostly used by the AIPS group to prepare models to be shipped with AIPS. The latter will be used to read in the models. Each FITS file contains a postage stamp image and the merged CC file with header information including the rms noise off source and the total CC flux. Added to $AIPSTARS: 3C138_K.MODEL 3C138_Q.MODEL 3C138_U.MODEL 3C147_K.MODEL 3C147_Q.MODEL 3C147_U.MODEL 3C286_AK.MODEL 3C286_AQ.MODEL 3C286_U.MODEL 3C48_ABQn.MODEL 3C48_AK.MODEL 3C48_U.MODEL Moved nowhere. 11557. April 12, 2004 FRING Amy Changed FRING so that it does not reset the number of IFs (NUMIF) when the SU table is opened. Also added a warning if the number of IFs in the SU table is different from the number of IFs in the header. Moved nowhere. 11558. April 12, 2004 CALRD, SETJY Eric Added the current 12 models to the help file and a remark about setting flux in the SU table for CALIB. Added code to SETJY to recognize the 3 bands and 4 sources for which models are available at present and added info to the SETJY help file about the models. Moved nowhere. 11559. April 12, 2004 CALDIR Eric New verb to list all calibration models available. Changed POPSDAT, NEWPARMS, and AU7B. Added MGAUSS and PGAUSS back to the former two - their deletion causes trouble with compress. Also new help file. Moved nowhere. 11560. April 12-13, 2004 CookBook Eric Updated the ABOUT files for ADVERB, CALIBRAT, FITS, GENERAL, IMAGE-UT, TASK, VERB, VLA. Updated LaCook13 for these changes. Changed LaCook4 to describe CALRD, CALDIR and the use of models. Updated the table of contents and index as well. Moved nowhere. 11561. April 13, 2004 VLBAARCH Lorant Modified the FPOL in VLBAARCH not to croak on data sets with only one frequency and one IF (apparently this is requested by some users). Also suggested to Amy to fix this in VLBAUTIL Moved nowhere. 11562. April 15, 2004 MWFLT Eric Changed max filter size to 201 from 51. Moved nowhere. 11563. April 15, 2004 POSSM Eric The divide by channel 0 option had some errors: 1. The counter of failures was no re-zeroed and so blocked all plotting after it exceeded the count of the number of samples. 2. The flux test compared a flux cutoff with the square of the flux. 3. It forced BCHAN to 1, ECHAN to Nchan, and DOCAL to false for no good reason and would not use a single channel as the channel zero. Fix help file for DOCAL - the rest were not mentioned. 4. Its count of Vis subtracted an omitted correlators count despite the fact that they are blanked suitably. Moved nowhere. 11564. April 15, 2004 SNFLG Eric The phase difference routine tried to fix phases > 180 or < -180 but they could actually be nearly 720 degrees in the worst case so more of the fix statements are needed. Moved nowhere. 11565. April 16, 2004 FILLM Eric The sign of CPARM(3) was lost before it was written to the history file. Fixed misleading wording in help file. Moved nowhere. 11566. April 16, 2004 POSTSCRIPT Eric General information help file about PostScript and AIPS including macros to draw arrows and labeled bars. Added mention of this in LWPLA.HLP. Moved nowhere. 11567. April 19, 2004 START_TVSERVERS Eric Apparently lots of computers do not run networking! So DISPLAY=localhost:0 or DISPLAY=:0 do not work. Added code to convert this to :0 when the $HOST, $TVHIST, and $TVDISP are all the same. I wonder what wierd setups will now be revealed. Fle must be moved to $AIPS_ROOT with editing to set the value of AIPS_ROOT appropriate to the local system for this change to take effect. Moved nowhere. 11568. April 19, 2004 SAD Eric Raised the number of components allowed from 4000 to 40000. Added DPARM(10) option to expand (or contract) the area of the image taken about each component for fitting it. Moved nowhere. 11569. April 20, 2004 FLATN Eric Make help clearer about not applying beam pattern unless doing mosaicing. Moved nowhere. 11570. April 20, 2004 SAD Eric Changed to allow BWSMEAR to be < 0 to imply use abs(BWSMEAR) but not scale peak flux by correction. Fixed formats to allow for 5 digits in component number. Moved nowhere. 11571. April 20, 2004 SAD, JMFIT, IMFIT Eric The initial values for the max deconvolved Gaussian in all 3 were the same and were wrong. Changed to gues max and min = formal solution and go from there. Typos were found in the IMFIT version of the array deconvolver. SAD did not return the correct position angle in SAD's units (90 degrees different from the world incl IMFIT/JMFIT) for the bandwidth smearing term. This was most visible at the deconvolution stage but affected the initial guesses as well. Moved nowhere. 11572. April 21, 2004 CLCAL Eric Changed SN2CL so that it does 2PT on INTMOD 0, a new SELF on INTMOD 1, and the old SELF (nearest) in INTMOD 2. The new SELF adds a loop over source ID (having first found all source IDs) and does 2PT but using the source IDs one at a time. Changed CLCAL Fortran and help to rename the old mode SELN and to describe the new SELF and set INTMOD correctly. Changed BLAPP for the required value of INTMOD now = 0 for the hard-coded 2PT it uses. Moved nowhere. 11573. April 22, 2004 VLBAFPOL/CRPL/SNPL Amy Fixed VLBAFPOL's handling of one IF, one freq.. Per a users request, VLBACRPL/SNPL only do a TVINIT if DOTV>=1. Moved nowhere. 11574. April 22, 2004 UPDUPDATE Eric Script to be used later to update a MNJ site for changes to XAS, POPSDAT, START_TVSERVERS etc etc It is not necessary for the AIPS Manager to do things by hand except in the rare cases of changes to READLINE, PP.FOR, F2PS.C, et al. These require at least knowledge of the local compilers and so need to be done by hand, but they should almost never change. This is not used yet - I am going on vacation and do not dare change the MNJ yet. Moved nowhere. 11575. April 22, 2004 UVFND Eric Previously it checked only channels CHANNEL to CHANNEL+n where n was determined by the width of the window, number of correlators, etc. Added BCHAN and ECHAN to the adverbs and all of those are tested. The same channels as before are printed. Found a bug in which the frequency and number of channels could be wrongly determined in the print routines. Moved nowhere. 11576. April 22, 2004 IMEAN, SAD, IMFIT, JMFIT Eric Changed to exclude pixel values which are exactly zero from the fits for the actual image rms. Remarked on this is the help file, citing the use of REMAG to set blanked pixels to zero. Moved nowhere. 11577. May 3, 2004 BPLOT Eric To get the plot file to have all the parameters set, the default PIXRANGE was filled in. This led to the next plot (e.g. next time or antena) to use the PIXRANGE of the first plot. Moved nowhere. 11578. May 3, 2004 STARS Eric Changed STARPL back to allow for plotting stars at the edges of a field. Changed PLOTWO to use local variables for the plot positions do the limiting of a line to the field will not change the calling variable. Moved nowhere. 11579. May 3, 2004 UPDUPDATE Eric Fixed Logfile to LogFile and got the putbck system to accept the file as executable. Sent out warning that this file needs to be ln -s in the user's update area. Moved nowhere. 11580. May 4, 2004 POSSM Eric Added code to FILLPL to prevent exactly constant amps and/or phases from causing the plot routine to lose its cookies. Moved nowhere. 11581. May 4, 2004 MNJ Eric Changed AIPSUPD to call UPDUPDATE before the UPDCOMRPL to make sure it gets done. This might make trouble if POPSGN.FOR has had to change for something in POPSDAT.HLP. However, in case of failure of COMRPL, knowledge of the changes to the text will disappear so one must run UPDUPDATE in any case. Changed UPDCONTROL which spotted changes in files and issued a special mail message. Changed it to notice (with corrected names) changes in things that UPDUPDATE cannot handle and to tell AIPS Managers some clue of what to do. Fixed the info re READLINE and dropped XAS and POPSDAT sections. Moved nowhere. 11582. May 5, 2004 OFMFILE web address Eric Kristy Dyer found a web address in Japan that has images of all the standard AIPS OFMFILEs. Added this address to the help files for GREYS, KNTR, LWPLA, OFMGET, PCNTR. Moved nowhere. 11583. May 5, 2004 SPLIT, SPLAT, AVSPC Eric SPLIT and SPLAT attempted to reset the reference frequency and rescale u,v,w when averaging all of the channels. This was fine for the case of no averaging across IFs, but ignored the fact that the IFs are at different frequencies (and potentially channel widths, channel selection, etc) when the IFs were averaged. Fixed the average IFs to compute the average frequency as accurately as possible. Added notes to the help files to warn that the average frequency is a straight average while the data average takes into account data weights and flags which often differ between IFs. AVSPC was changed to eliminate the AVIF option - it claims to work for multi-FQ data sets and the like so fixing the scaling for averaging IFs is too tricky. However, it is so simple that one would like to retain it rather than require SPLAT. Moved nowhere. 11584. May 5, 2004 UPDUPDATE Pat/Eric Added redirect of standard error to standard out for make XAS and RUN POPSGN. Moved nowhere. 11585. May 6, 2004 CookBook Eric Changed the wording regarding AVSPC in Chapter 9. Also edited Chapters 0, 4, 5, 10, and A to drop the doFIG is statements. Chnaged COOK9.PS and COOKBOOK.PDF for the first. No PS changes for the others. Moved nowhere. 11586. May 7, 2004 CONVL Eric Changed it to accept blanked images as the main image but not the image with which the main image is convolved. Blanks are converted to 0.0. On output the image is reblanked at the same pixels that were blanked previously if new adverb DOBLANK > 0. Moved nowhere. 11587. May 11, 2004 MFPRT Eric Added option to print in STARS format and to use sexagesimal format for coordinates (rather than do offsets) with the basic print format. Fixed it to omit spectral channel when there is only one frequency image plane. Fixed scaling of flux to show microJy when cutting off at milliJy is inappropriate. Several minor errors were also fixed. The STARS format has the option of using the fit or the deconvolved widths and can have the output width scale with source flux (or not) and to have the symbol type scale with source flux or not. Moved nowhere. 11588. May 11, 2004 install.pl Eric There were typos ($m was set to 28 or 29 rather than $d) in finding the date of yesterday. Moved nowhere. 11589. May 12, 2004 $APGNOT/ATMCA.FOR...HLP Leonia The current phase referencing procedure considers observing of one strong calibrator located near the target source. Solution found for the delay toward the calibrator applies to the target source. This procedure works well if the calibrator and the target source are very close to each other so atmosphere can be considered identical for the two directions. In practice this is not true. The new task ATMCA reconstructs the delay at atmosphere toward the target(reference) source using the observation of several calibrators located at some vicinity of the target(reference) source. This method does not require the identical atmosphere delay for the target(reference) source and the calibrators requiring instead linear representation of the delay along RA, DEC and time, which can be corrected for a larger area of the sky and time. The atmosphere delay at the vicinity of the target(reference) source is represented by the following linear combination of the two sky coordinates and time: DELAY = PH0 + SLRA*DRA + SLDEC*DDEC + SLT*DT where PHO is delay at the point DRA=0, DDEC=0, DT=0 SLRA is slope of the delay at RA direction SLDEC is slope of the delay at DEC direction SLT is slope of the delay at time direction DRA is deviation of the reference point at RA DDEC is deviation of the reference point at DEC DT is deviation of the reference point at time The deviations are considered relatively the target (reference) source in sky and in time. The data for at least four calibrators (three if time dependence is excluded) are used to fit the unknown PH0, SLRA, SLDEC, and SLT. PH0 is the solution for delay toward the target(reference) source. The task finds the solution for each given time interval (SOLINT) independently and apples it to the given CL table. Moved nowhere. 11590. May 12, 2004 CLCOR Amy Added option OPCODE='TROP' which works basically identically to OPCODE='ATMO' except in 2 ways. The INFILE should contain the total zenith delay (rather than just the correction as for ATMO) which is subtracted from the atmospheric model used by the correlator (in column ATMOS in the CL table). This difference is then treated almost identically as in OPCODE='ATMO' except that a more sophisticated mapping function is used. Moved nowhere. 11591. May 12, 2004 FITTP, FITAB Eric Changed FITTP and fITAB to invoke 1PDm.n formats for double precision keyword values rather thna 1PEm.n although the fITS format does not differentiate between them in any way. There may be some value in the implication of higher precision. Moved nowhere. 11592. May 13, 2004 $APGNOT/ATMCA.FOR...HLP Leonia Error found at the calculation of the RE/IM correction at CL table. Fixed. Some misprints at .HLP file fixed also. Moved nowhere. 11593. May 15, 2004 FILLM, PRTTP Eric Correct the code at the end of file - a short record may actually contain more than one short record. Issue message only when the short record is shorter than needed - previously it gave a message if the short record was shorter or longer than needed. Moved nowhere. 11594. May 17, 2004 FITTP, FITAB, CALWR Eric The table extension data might not have been zero filled if the data took fewer than 2880 bytes. Fixed this. Moved nowhere. 11595. May 17, 2004 models Eric The model files were messed up by putbck which did not handle them as binary but rather put them through a string handling business. Rceomputed the FITS files to zero-fill the CC table data record and fixed putbck to handle $AIPSTARS/*.MODEL and $AIPSPUBL/*.PDF (was present but mis-spelled) with straight copies. Moved nowhere. 11596. May 18, 2004 FRING Amy When the new feature SOLSUB/SOLMIN was used in FRING, it did not properly update the GOTANT variable when going to the next solution interval. The subroutine that writes the SN table uses GOTANT to determine which antennas have solution. This bug prevented good solutions from being written to the SN table. Moved nowhere. 11597. May 19, 2004 $APGNOT/ATMCA Leonia The new option creating the output SN table has been added. The output SN table includes the delay difference of the input SN table and fitted atmosphere represented by linear combination at the vicinity of the calibrators The output SN table can be useful to see the effect of the atmosphere correction. The output SN table is appeared at the next after the highest existing SN table. The atmosphere correction is presented at the output SN table for the CALIBRATOR list of the sources. Moved nowhere. 11598. May 20, 2004 LISTR Eric Changed MATX type 4 to do a proper phase rms (based on Reals and imaginaries averaged suitably) rather than an amplitude rms. Changed phase displays to do scaling by powers of 10 in degrees rather than simply integer degrees. Moved nowhere. 11599. May 24, 2004 RANDOM, GRANDOM Eric New verb RANDOM returns value for 0 to 1, GRANDOM returns Gaussian distributed random number with mean 0, rms 1. Changed NEWPARMS, POPSDAT, KWICK (call RANDUM), RANDIN (more initialization), and AIPS, DAIP, and AIPSB (to call RANDIN). New help files too for GG, RANDOM, GRANDOM. Moved nowhere. 11600. May 24, 2004 SNPLT Eric The help and code said that SUBARR=-1 would plot all subarrays, but the inputs required the vale to be >= 0. Changed the limit to -1 and added a comment in the inputs. Moved nowhere. 11601. May 24, 2004 LISTR Eric Changed help and code to allow gain listings to include all subarrays on subarray < 1. Moved nowhere. 11602. May 24, 2004 ISPEC Eric The computation of velocity/frequency for printing out in the case in which the axis is reversed was not correct. The Z pixel wanted to be corrected for the inversion, but internally the data are already inverted including the header because of the plot routines. Moved nowhere. 11603. May 24, 2004 POSSM Eric Reversed spectra had bad velocity axis labeling - the total number of channels was used which is the number of spectral channels times the number of polarizations times the number of IFs. The number of spectral channels is what is now needed. checked the text file output and it seems okay. Moved nowhere. 11604. May 25, 2004 ATMCA Leonia Add the option fitting only two space slopes at the atmosphere behavior using only two calibrators. This option can give a a good fitting only if the two calibrators and the target source are located near a straight line. RMS of the residuals is added to the printout. Moved nowhere. 11605. May 25, 2004 SN2CL Eric Several bugs were exposed by a data set with a bit of missing data: 1. The loop over source and subarray branched to the exit code when the end of the input SN table was hit without a usable calibration for the current record. This caused it to omit any further subarrays and (now with the new SELF) sources that followed the one where the condition was hit. 2. The second solution got marked as possibly good before the check on record number that would show that there was no second solution. 3. The looping to find a second solution treated the last record as wanted even after finding out that it was not wanted. Blanking of the gains in this case probably caused it to avoid error but it did complicate the logic. Moved nowhere. 11606. May 26, 2004 MBDLY Amy Fixed annoying feature, where if FREQID was -1 (which is the default) then MBDLY would solve for nothing. Now if FREQID or SUBARRAY are negative then they are set to 0, which then treated as the default by MBDLY. Moved nowhere. 11606. May 27, 2004 WTMOD Eric Added SUBARRAY to limit the operation to the requested subarray. Moved nowhere. 11608. June 3, 2004 Line printing Eric Wanted to change DOCRT to allow suppression of form feeds and extra info when printing to text files. LPOPEN reused DOCRT to convey the line length making this impossible. So changed LPOPEN Added NACROS as output from LPOPEN, so changes DOCRT only for batch. Dropped unused code. Fixed call sequence in AU1A, AU7, OFMLST, PRTMSG, and PRINTER subroutines and in tasks IMFIT, JMFIT, LISTR, MFPRT, PRTAN, PRTSD, PRTUV, SHOUV, TXPL, UVFND, UVHOL, UVPRT, RLDIF, CORER, OTFIN, PRTAB, PRTAC, PRTCC, PRTTP, SAD, UVDIF, COHER, DTSUM, HFPRT, TBDIF, BLSUM, ISPEC. PRTLIN Changed to suppress the form feed character when going to text files and DOCRT < -1.5. Suppress all start-of-page stuff when DOCRT < -2.5. PRTALN As PRTLIN. INPUT The test on DOCRT ignored the fact that its value is forced to <= 0 on batch jobs. Changed task and help files to implement these options fully: BLSUM COHER CORER DTSUM HFPRT IMFIT ISPEC JMFIT LISTR MFPRT OTFIN PRTAB PRTAC PRTAN PRTCC PRTIM PRTSD PRTTP PRTUV RLDIF SAD SHOUV TXPL UVDIF UVFND UVHOL UVPRT Some fixing for batch jobs was also required. Fixed COHER - it used an improper format to read incorrectly the date string. Changed it to use DATEST. Fixed CORER - it worked on compressed data only if those data had 1 spectral channel and 1 IF. Changed TBDIF to implement these options and to allow for printing only those lines that differ. Changed it to do min match at some level on the column names since PRTAB does not show the full length of long column names. Changed verb subroutines and help files to implement: EXPLAIN, HELP (s.r. AU1A) OFMLIST (s.r. OFMLST) PRTHI (s.r. AU7) PRTMSG (s.r. PRTMSG) Moved nowhere. 11609. June 4, 2004 FITTP, FITAB Eric Some FITS readers (major ones!) cannot read D format numbers. Changed formats of double precision numbers back to E format. Moved nowhere. 11610. June 4, 2004 IMDIST Eric The position angle needed to be done with an ATAN2 not an ASIN. Fixed AU9. Moved nowhere. 11611. June 9, 2004 CALIB Eric Last Oct 1 change limited APARM(1) to > 2. Some users make use of pure baseline at a time solutions and require Min antenna to be allowed to be 2. Moved nowhere. 11612. June 12, 2004 VPLOT Eric The ratio operations (POLPLOT = e.g. RL/RR) did not work right. The averaging routine (over frequency) did not know about the offset to the data and the later routines assumed there was an offset when the averaging routine could not do one. The time-averaging section ignored the POLPLOT option entirely. Fixed these and the labeling. Moved nowhere. 11613. June 14, 2004 IMAGR Eric The error handling for a field that had no components could lead to that field showing up as "second best" but better than any other field but the field would not be taken again. The IGNORE flag has to be in the 2nd best section as well as the first best. Moved nowhere. 11614. June 16, 2004 PRINTER, RFI Eric Changed the print object to know the keyword NACROS after the printer has been opened. Changed RFI to call it - DOCRT is no longer the size of the print line. Moved nowhere. 11615. June 16, 2004 AIPS Eric Changed SGLAST so that when it reads a lastexit file it checks that it can see INSEQ and INNAME adverbs. If the file reads okay but is trashed this should catch it. Moved nowhere. 11616. June 18, 2004 PCNTR Eric Had two errors - the polarization vector routine assumed that the grey file was being read if grey was plotted but it was only initialized if it was going to be used to blacken vectors. The test for 3-D coloring in polarization used an uninitialized (0) variable. Moved nowhere. 11617. June 22, 2004 TVFLG Eric Changed code to ask for subarray on FLAG ANTENNA-DT only when there is actually more than one subarray. Moved nowhere. 11618. June 22, 2004 PCNTR Eric Changed it to allow any of the 4 images to be the contour image and any to be the grey image. Image 1 functions as Ipol for clipping and may be a Stokes cube, image 2 in Ppol or Q, image 3 is Pang or U, and image 4 is given if needed. Found an error in the background grey image that would cause the dodark not to work at the bottom of the image for contouring. (CONDRW uses buffers 1 and 2 for contour and background - it does not matter what was DBINITed.) Moved nowhere. 11619. June 23, 2004 HELP Eric Fixed a "bug" in AU1A - DOCRT=-1 caused HELP to print on the line printer when it was supposed to be < -1. Also changed EXPLAIN so the DOCRT=0 will be the terminal not printer. Fixed the latter help file, the former said what was intended. Moved nowhere. 11620. June 24, 2004 PCAL, SPCAL Eric PCAL: found that the no reference antenna technique produced somewhat peculiar results and erroneous error estimates. Changed all methods to choose a reference antenna if none is specified. Corrected the error computation for the ORI- method, adding BPARM(2) to suppress it should it give trouble. SPCAL: Made the same reference antenna changes as PCAL. Also corrected the error computation in the way we did in January 2001 following Steve Meyer's suggestions in PCAL. Moved nowhere. 11620. June 24, 2004 VLBAARCH Lorant There is no reason not to run USUBA in the archiver if there is (n)one subarray. Before, SUBARRAY was left zero, and if it needed a SUBARRAY=1 in VLBAFIX it was not always caught. Here we're not trying to save time but to get the best data, so better run USUBA to force data in one subarray (if not more). Moved nowhere. 11622. June 24, 2004 CALIB Eric Users are setting the I = Q + U + V flux in SETJY and then expecting CALIB to work. Put a test in for I=V and have it use only I in that case with a message on each source and echa IF. Moved nowhere. 11623. June 27, 2004 FLAGR Eric New task - to edit data in pipelines and the like. The two algorithms at present do a baseline-based and an antenna-based flagging on RMSes that are too large within a specified averaging time. This is definitely a test program, putbck now to allow some testing. Algorithms to compare one time with another are needed next. Was called FLRMS for 1 day. Moved nowhere. 11624. June 28, 2004 CLCOR, FITLD Eric Again, FITLD and CLCOR disagreed on coordinates. Fixed FITLD to use the reference date in the output header for precession to apparent position. It was using the REFDATE of the current tape file instead. Added another OPCODE to CLCOR: 'ANTC' is 'ANTP' except that the coordinate shift is applied to the epoch (mean) coordinates not the apparent ones. Added messages to show the two shifts and, on ANTC, fix the shifts returned from SOUMOD to be the apparent coord shifts for the CL table parts. Moved nowhere. 11625. June 28, 2004 model files Eric Renamed the model files that had array codes in the names. Claire meant nothing serious with those codes and they make life messy - no the names are simply 3C286_K and Q and 3C48_K and Q. Changed SETJY and CALRD help files accordingly and added a warning about coordinate epochs. Moved nowhere. 11626. June 29, 2004 Epoch test Eric Changed DUVH.INC Add parameter DEQUIN for equinox (epch). DBCON Made corresponding change in /CATHDR/ and NWORDS for the tasks local copies of the include. The epoch test was already done with the headers. UVPGET Pick up epoch. SETGDS Test epoch DEQUIN versus epoch in model headers. Error message and quit. Moved nowhere. 11627. June 29, 2004 DELZN Amy DELZN ignored flagging in the input SN table, i.e. both flagged and unflagged records were used for the fitting. Fixed. Moved nowhere. 11628. June 30, 2004 AIPSLetter Eric Put back the new aipsletter. It took new programs to run on the MNJ.FTP summary lists. Moved to web/ftp site. 11629. Jule 1, 2004 ATMCA Leonia The new option (APARM(1)=1) is added. This option allows make a correction for atmosphere when there is only one additional (to the main calibrator) calibrator. This option does not reconstruct the two dimensional atmosphere behavior. Instead it reconstructs the atmosphere along the line reference - the second calibrator. If the target source is located along (or close) to this line, then the phase correction for the target source can be done. The codes for other options are modified significantly. The help file is completely rewritten by Ed Fomalont, describing specifically the observation schedule and data reduction to prepare the ATMCA input SN table. Moved nowhere. 11630. July 1, 2004 AIPS Eric DSKLOK(1) > 0 is a disaster - changed error handling and added check and quit on this condition. Moved nowhere. 11631. July 1, 2004 CLCOR Eric Walter Brisken found something I forgot in CLCOR ANTC fix. Moved to AOC this date. 11632. July 1, 2004 SETFC Eric Changed help and code to allow user control over points per beam (3 default) and max phase error (20 degrees default). Since Wmax is known, fix the formulas to use that rather than Bmax to set field of view. Moved nowhere. 11633. July 2, 2004 ATMCA Leonia The phase of the reference calibrator (APARM(1)=1) has been added to the estimated correction. The warning about maximum angle Calibrator to Reference to Target is added. Moved nowhere. 11634. July 2, 2004 WENSS Eric The WENSS and WISH catalogs contain multiple-component sources twice, once as the various components and again as a sum. I did not realize that and included both in previous WE files. The files are now corrected to leave out the sum "sources" and to be sorted with a proper numeric sort. The stronger source (> 100 mJy) files are in $AIPSTARS and in the ftp area while the all source files are only in the ftp area. Moved nowhere. 11635. July 6, 2004 RECAT Eric Error messages from ZOPEN were blocked to stop the no such file message. Fixed it to show all other ZOPEN messages. Moved nowhere. 11636. July 6, 2004 FILLM Eric The velocity frame code was not examined properly to check for user fixed frequencies and the like. The result was that a frequency appeared in the source table as a velocity. Added some tests to stop this although it probably caused no harm. Moved nowhere. 11637. July 7, 2004 FILLM Eric Changed $APLUNIX/ZONLIN.C to add an error routine that all the other ones call to make a message telling the user that an online function has been called from a non-online version. The previous error messages were less than obvious. Moved nowhere. 11638. July 7, 2004 TVFLG Eric Added two more display options - the vector amplitude rms and rms/average. The prevous rms and rms/mean were scalar. Fixed the TVFLG.HLP and HLPTVFLG.HLP to differentiate and show the new commands. This will be helpful in preparing inputs for FLAGR. Moved nowhere. 11639. July 8, 2004 ATMCA Leonia Several changes following Ed's recommendations done. 1. Weight proportional to the distance CAL-REF added when averaging different calibrators (APARM(1).EQ.1) This makes less noise output when the secondary calibrator is too close to the reference calibrator. 2. The SN table has copied even when APARM(1).EQ.1 This error removed. 3. The last group of points at the SN table have skipped because a logical error at the codes. Fixed. 4. Warning about to big angle CAL-REF-TAR includes now specific information about CAL, TAR and the angle itself. Moved nowhere. 11640. July 8, 2004 ZSBIND Eric This unused Z routine has been causing problems. It declares gethostname to set types but some newer OSs have a different declaration. Moved to $APLCONTR to get it out of the way. Moved nowhere. 11641. July 12, 2004 ATMCA.HLP Leonia A lot of changes have been done by Ed Fomalont following the modification of ATMCA (11639). Moved nowhere. 11642. July 12, 2004 LISTR Amy If SUBARRAY = 0, OPTYPE = 'GAIN' then LISTR assumed that all subarrays were wanted (even if there was only one AN table). If a user then tryed to deselect antennas, LISTR did not handle this correctly. Fixed it so that if SUBARRAY = 0; OPTYPE = 'GAIN' and antennas are deselected LISTR now assumes SUBARRAY = 1. Moved nowhere. 11643. July 13, 2004 DBCON Eric The task did not check to see if there are FQID random parameters before deciding to make more than 1, renumbering the 2nd data set's FQID (usually 1 but of unlike frequency to that of data set 1). Added better tests to avoid this. Moved nowhere. 11644. July 13, 2004 START_TVSERVERS Eric The switch to DISPLAY=":0" when the TV display and host are the same overrides a user-set display such as "1.0". Fixed this to only override our "host:0.0" Note that special settings still require TV=local as well. Moved nowhere. 11645. July 13, 2004 PRTTP Eric Check OUTPRINT=' ' on print levels -3, -4 - fill in a default value and warn the user. Moved nowhere. 11646. July 13, 2004 FLAGR Eric Changed it a lot. The ANTE one has been corrected to get correct antenna-based solutions. All functions now support a clip on flux (after averaging and when appropriate after conversion to antenna based). A new function working over time has been added. An EXPLAIN file was written to I hope deal with the confusion over what the task does. Moved nowhere. 11647. July 13, 2004 AIPS Eric Moved the Copyright statement - it came before the user's message file is created and so had error messages. Looks lousy that way. Moved nowhere. 11648. July 13, 2004 DRAWBOX Eric New verb - draws Clean boxes on graphics plane and returns. Changed AU5C, POPSDAT.HLP, NEWPARMS.001, and new help file. Moved nowhere. 11649. July 14, 2004 VLBAARCH Lorant Analysts requested to put subarray variable back after last vget. Also took the opportunity to explain more in detail the step-by-step data handling done to the files, per request/suggestion of a user. Moved nowhere. 11650. July 18, 2004 VECTOR Eric Changed POPSDAT.HLP, NEWPARMS.001, AU2A.FOR and created VECTOR.HLP to make a new adverb to control whether vector or scalar averaging is done. Moved nowhere. 11651. July 18, 2004 FLAGR Eric Added flagging on weights and optype GAIN which combines all sources in the TIME operation dividing the amplitudes by the apparent source amplitude first. Added VECTOR as adverb. Moved nowhere. 11652. July 19, 2004 perl Eric Added versions to check_version.pl in $SYSPERL. MOved nowhere. 11653. July 19, 2004 TECOR.HLP Amy Added alternate ftp site to get ionex files. Moved nowhere. 11654. July 20, 2004 INPUT Eric Suprisingly, OOP tasks end with a direct call to an AIPS mid level routine of the old style (DIE). Created subroutine OUT2AV as a public function to return adverbs from an object to aips (if needed) and then shut down the current task. It uses a private function OUSTOP also added to INPUT.FOR. The former can replace the direct calls to DIE in order to add the usual level of obfuscation present in OOP. Moved nowhere. 11655. July 20, 2004 FINDR Eric New task - like FLAGR OPTYPE = 'TIME' and 'GAIN' but designed to find the averages and print them and return a summary in adverbs to pipelines. Moved nowhere. 11656. July 20, 2004 FLAGR Eric Change PRTLEV to match FINDR and fix minor typos and a weight solution on singletons. Improve grammar in explain file. Moved nowhere. 11657. July 20, 2004 DDT, Y2K Eric Recomputed masters since CALIB and VTESS have changed. Moved to ftp site. 11658. July 21, 2004 LISTR Eric Put the scaling of phases under user control. DPARM(7) must be > 0 or phases and angles are in integer degrees. Minor other corrections made. Moved nowhere. 11659. July 21, 2004 EHEX, REHEX Eric Make adverbs DENUMB, EHNUMB in POPSDAT.HLP, NEWPARMS.001, EHEX.HLP, REHEX.HLP, and AU8.FOR. These function as in/out adverbs for verbs EHEX and REHEX. Moved nowhere. 11660. July 21, 2004 TVLOD Eric Changed TVWIND subroutine to honor TXINC < -1, TYINC < -1 even if the window has to be squeezed, Moved nowhere. 11661. July 21, 2004 VLBAARCH Lorant Since USUBA is run by default, the NX table was left out. Fixed Moved nowhere. 11662. July 22, 2004 modeling Eric The source models made from subimages have pointed out the dependece AIPS model routines have on the image size of the model image even when using the CCs. Changed UVMTYP and SETGDS to call new routine FORCED to check the image dimensions. If they are not a power of 2 or are less than 128 DFT will be forced on CMETHOD = ' ' and will be subject to a warning message when 'GRID' has been forced (in modeling tasks that use SETGDS). Moved nowhere. 11663. July 23, 2004 CALIB Amy Fixed CALIB so that it writes the SOLSUB and SOLMIN in the history file if they are set. Moved nowhere. 11664. July 27, 2004 APCAL Amy Added the Trec and Tau0 that were used in the opacity correction to the history file. Previously only the input Trec and Tau0 were in the history file. Moved nowhere. 11665. July 27, 2004 Weights in solutions Eric The correct data weights when divided by a model or converted to only phase should be multiplied by AMP ** 2 where AMP is the amplitude being divided into the data sample. When we made this change to UVDOUT however more solutions began to fail. So we are adding a weight modification done after conversion to "correct" 1/sigma**2 weights and data averaging but before the data are solved for antenna-based gains or fringe fits. (1) Changed POPSDAT, NEWPARMS to make adverb WEIGHTIT. (2) Changed GCALC, GCALC1, GCALC2, NCALC to do correct weighting when mode is 2 (amplitude independent) and then to call REWAIT, a new routine to apply WEIGHTIT to the weights. (3) Changed CALIB.FOR and .HLP to offer WEIGHTIT. (4) Changed UVUTIL s.r. SLFCAL to pick up WEIGHTIT and pass it in the MODE parameter to SLFPA which calls the GCALC routines. (5) Changed SCMAP and SCIMG Fortran and Help files to add the WEIGHTIT adverb and to pass it on in as needed including TELL. Fixed declarations in SCMAP to avoid unused variable messages. (6) Changed FRING and KRING to correct the data weights by amplitude squared first and then WEIGHTIT before they are "averaged" (usually essentially no time smoothing, just placing in the arrays for fitting). Also fixed history to record SOLSUB and SOLMIN if they are used. Moved nowhere. 11666. July 27, 2004 GREYS Eric In true color plotting the pix ranges set by APARM were ignored when plotting the images and the wedges. Now they are properly used. Step wedge labels are for the red image - a fact added to the help file too. Moved nowhere. 11667. July 28, 2004 FLAGR Eric The task did not handle the choice of which FG table to copy right and neglected to change the table sort order to unknown. Moved nowhere. 11668. August 3, 2004 FILLM Eric An out antenna is VLA:OUT not just OUT so the test to ignore out antennas was ignored. Moved nowhere. 11669. August 5, 2004 install.pl Eric The $INSTALL_FROM_CD was treated as a logical one place where the compiled tools such as F2PS are supposed to be copied. Changed that test to looking at the string value like all the rest. I do not see how the CD install works anyway! Moevd nowhere. 11670. August 8, 2004 SCANLENG Eric New adverb in POPSDAT.HLP, NEWPARMS.001. Help files for WEIGHTIT and SCANLENG. Moved nowhere. 11671. August 8, 2004 FINDR, FLAGR Eric Added VDIF OPTYPE to do antenna-based amplitude of the vector difference between a SOLINT averaged data point and the SCANLENG averaged data surrounding ot. FINDR tells one about it and FLAGR does flagging. Also added more messages to FLAGR on PRTLEV to show individual flags and questionable points too. Moved nowhere. 11672. August 9, 2004 Y2KLOAD Eric Changed procs: 1. The checker of uv data now takes an argument of the flux level and L and H use 0.5 rather than 0.05. 2. The L and H VTESS now set FLUX = -SCALR3/10 which seems to get convergence where a FLUX constarined vtess does not. New data to anonymous ftp site. Moved nowhere. 11673. August 9, 2004 FINDR, FLAGR Eric Found a gross addressing error - how did it work? Moved to AOC this date. 11674. August 11, 2004 IMAGR, models Eric Changed: QUVUTIL Added calls to set FLDSZ before model subtraction or division to make sure it is set. FORCED Changed call sequence to look at all fields at once and to base decisions on more global image size parameters. If the minimum is < 64 or the average <= 128 or any < 768 are not power of 2, then FORCED is true. UVMTYP Change to call FORCED for all fields at once SETGDS Change to call FORCED for all fields at once Moved nowhere. 11675. August 11, 2004 FINDR, FLAGR Eric Added OPTYPE VRFI to do baseline-based amplitude of the vector difference. Kept working on explain sections. Moved nowhere. 11676. August 12, 2004 TVRGB Eric The formats used for certain parameters did not allow for really large images to be put out in PostScript correctly. Moved nowhere. 11677. August 12, 2004 FINDR, FLAGR Eric Pulled out subroutine ROBUST to dfo robust averages in an array. Changed FINDR and FLAGR to use ROBUST everywhere. This caused me to find wrong RMSes being printed by FINDR (they were 3 times too large), blanked solutions being handled and printed by FLAGR. FLAGR TIME OPTYPE now has the possibility of flagging correlators for all times. Moved nowhere. 11678. August 14, 2004 IMAGR Eric Changed: 1. QCLEAN to control SDI Clean differently. It should be based on the fraction of pixels in the Clean windows having residual values toward the upper part of the histogram. So IMAGRPRM(4) now selects SDI Clean when the fraction of pixels having residual value greater than half the peak residual exceeds IMAGRPRM(4). The lower residual value to include is set by IMAGRPRM(19) or 0.333, whichever is greater, times the RESMAX. The default for IMAGRPRM(19) is set by the shape of the dirty beam and can have values like 0.06 up to 0.5. Previously the lower residual value was set by RESMAX divided by (1 + GAIN * IMAGRPRM(4)) which could be rather too low. Worse still, the lower limit might have been applied to a BGC Clean rather than the limit carefully found to put MAXPIXEL pixels into the Clean inner loop. 2. IMAGR.HLP to change the SDI parameters. 3. HLPCLEAN.HLP to describe several TV options that were omitted including the ones involved with the SDI options. 4. QCLEAN to add header keywords CCFLUX (total Clean flux in this field - not just the total of this field's CC file) and CCTOTAL (total Clean flux over all fields). Moved nowhere. 11679. August 14, 2004 Y2K Huge Eric Use WEIGHTIT=1 for Huge. Compute new masters and make them available since the new approach to VTESS seems to work. Moved nowhere. 11680. August 15, 2004 TVFLG SPFLG Eric Changed to display the current pixrange in the status line(s). Moved nowhere. 11681. August 17, 2004 DFQID Eric NEW TASK: Renumbers FQ IDs inside a data set. Moved nowhere. 11682. August 17, 2004 VLBAARCH+HLP Lorant Changed VLBAARCH to only flag weights below 0.1 (versus 0.7), because this is the archive and that should be as complete as possible. VLBAPIPE (or the user) should flag what (s)he thinks is bad data with UVFLG, not the archiver. <0.1 is defenitely bad in uncompressed data (i.e., weight is per IF). NB I still need to test this added flagging after putting it in VLBAPIPE. Also, in some experiments the PI uses Doppler tracking per each source, which may result in freq-ID's that differ only in the 5th decimal of the frequency. As CHAR(freq) uses a F9.4 float, the name of the files may not be unique in the 4th decimal. I changed the naming convention a bit to make sure that for each frequency-ID we have an unique name - by appending a freq-ID#. Moved nowhere. 11683. August 18, 2004 WHATSNEW Eric Updated the web page dec04.shtml and took that list to the WHATSNEW help file (throwing away all the obsolete info it contained). Moved nowhere. 11684. August 25, 2004 UVFND Eric The VCLP option was not really functional - there was elaborate code to set the selection of which APARM to use and then the parameters that were set were never used. Made the code simpler. Moved nowhere. 11685. August 25, 2004 PBCOR Eric Changed help file (explain) to reflect the new adverb names and to show the correct cutoff for beam value. Moved nowhere. 11686. September 1, 2004 DELZN.HLP AIPSMEM110.PS Amy AIPSMEM110.PS (Strategy for Removing Tropospheric and Clock Errors using DELZN) was put in the system and it is now mentioned in DELZN.HLP. Moved nowhere. 11687. September 1, 2004 $APLNOT/JPLEPH.FOR Leonia The new software package includes JPL software to calculate position of the solar planets. Moved nowhere. 11688. September 1, 2004 FIXBX Eric New task - will convert Clean boxes from old boxfile to the facets of a new boxfile. Moved nowhere. 11689. September 1, 2004 PLOTR Eric Fixed a test on line type - an absolute value was needed to get the legitimate test to see that the symbol was desired. Moved nowhere. 11690. September 7, 2004 CLCOR Leonia The new opcode 'SUND' has been added. This opcode corrects the extra time delay caused by the bending of the light ray passing through the gravitational field of the solar system planets. The VLBA correlator (before September 2004) carries out this correction considering the source locates at the infinity. This option removed the correlator correction and put the correction taking into account the actual position of the source inside of the solar system. The correction is carried out only for the one source (located at the solar system) given at the adverb SOURCE. 11691. September 8, 2004 GAL Eric Raised buffer size to 1024x1024. Moved nowhere. 11692. September 16, 2004 SHADO Amy SHADO would croak if the UVCON antenna files in AIPSTARS were used as the INFILE, even though the file format was the same. This is because SHADO did not ignore comment lines. Fixed. Moved nowhere. 11693. September 21, 2004 CLCAL.HLP Amy Clarify what happens when multiple SN tables are used, either for in OPCODE='MERG' or to apply to a CL table. The SN tables are simply concatenated, so this feature is not intended for multiple SN tables with incremental calibration of the same source(s). Moved nowhere. 11694. September 22, 2004 ACOUNT Eric Replaced some GO TO's with IF-THEN-ELSE to stimulate the putbck check sum code. Moved nowhere. 11695. September 22, 2004 JPLEPH.FOR Leonia The new version with Bryan's fix of the bug in annual aberation for J2000. Moved nowhere. 11695. September 22, 2004 install.pl Eric Walter Brisken suggests changing the start line to #!/bin/env perl rather than a specific perl version. Moved nowhere. 11696. September 23, 2004 GAL Eric Changed it to: (1) handle 1536x1536 images, (2) scale the plot symbol by FACTOR and use SOLCON for the convergence criterion, and (3) return the APARM adverbs converted to fit answers from the initial guesses. Moved nowhere. 11697. September 23, 2004 CLIPM Eric Added flag on max weight, moving APARM(6-9) to APARM(7-10). Moved nowhere. 11698. September 23, 2004 SPFLG Eric Added option for vector rms and rms/mean as well as scalar which was already there. Fixed help file and HLPSPFLG for this. Fixed a units thing in TVFLG. Moved nowhere. 11699. September 24, 2004 CLCOR Leonia OPCODE = 'SUND'. The error in JPLEPH (see 11695) is taken care. The error reading the input file is fixed. Moved nowhere. 11700. September 27, 2004 POSSM Amy POSSM has the ability to divide by 'channel 0', the user can define what channels to use with BPARM(2-3). POSSM set BPARM(3)=BPARM(2). Fixed. Moved nowhere. 11701. September 28, 2004 CDi_j Eric Changed PCHDR which interprets the CDi_j and PCi_j keywords. They can contain skew not only in the two possible "rotation" angles but also in funny increments. Changed it to average the increments obtained by dividing by cos(rot) and those obtained by dividing by sin(rot) and to report any serious differences in the results. Affects AIPS (TPHEAD), IMLOD, UVLOD, FITLD. Moved nowhere 11702. September 28, 2004 TVALL Eric This proc had a bug - it cleared graphics channels 2, 4, and 8 although only GRCH and 8 are used. Fixed that clearing and moved the TVON command to later in the proc - all in POPSDAT. Moved nowhere. 11703. September 30, 2004 FITTP, FITAB Eric Clarified and improved the defaults for CATNO in help and code. Moved nowhere. 11704. September 30, 2004 FLAGR, FINDR Eric Pulled ASOLVE and VSOLVE from the two tasks and added robust computation of the antenna-based solutions. Added code to both to flag or report on closure error statistics. Moved nowhere. 11705. October 4, 2004 POPSDAT Eric My revised TVALL turned on the TV after TVWED and it must be on before TVWED. Moved to AOC this date. 11706. October 4, 2004 PCNTR Eric Fixed a MAPCLS loop to do all 5 possible on's. I added one to allow all images to be used for all things and did not increase this loop. Moved nowhere. 11707. October 6, 2004 SMOOTH Eric Changed: CALIB Wrote user adverb not SMOOTH (with defaults completed) to HI SPLAT as CALIB SPLIT as CALIB ACFIT as CALIB; also used DOSMTH as logical SPCAL Changed init of DOSMTH to 0 from FALSE ISPEC Allow SMOOTH(1)=5-8 to mean 1-4. WETHRC Allow SMOOTH(1)=5-8 to mean 1-4. DSEL.INC DOSMTH becomes an integer DATGET Test DOSMTH=1 to do before bandpass application and =2 to do after SELINI Init DOSMTH=1 UVGET On init call SETSM always, not testing DOSMTH. SETSM Set DOSMTH = 0, 1, or 2 and make functions 5-8 the same as 1-4 but applied after BP CALEDIT Change DOSMTH to integer and set init to 0. Changed HELP files re SMOOTH: SMOOTH ACFIT BLING BPASS CALIB CLPLT CPASS DECOR EDITR FRING FRPLT HORUS IBLED IMAGR ISPEC KRING LISTR POSSM RLDIF SHOUV SPFLG SPLAT SPLIT TVFLG UV2TB UVFIT UVHOL UVIMG UVPLT UVPRM UVPRT VPLOT WETHR WIPER Moved nowhere. 11708. October 9, 2004 SPFLG Eric Added option LOAD LAST BASELINE to go backwards through baseline numbers. Otherwise a mistaken LOAD NEXT BASELINE leads to a serious hastle to get backwards. Moved to AOC this date. 11709. October 11, 2004 ATMCA Leonia Interpolation along the elevation has been added as a new option (APARM(1)=5). SN table has been created now for APARM(1)=1,5. Moved nowhere. 11710. October 15, 2002 ATMCA.HLP Leonia Modify the help file in accordance of the code correction (11709). Clarified list of the source corrected at output SN table Moved nowhere. 11711. October 15, 2004 Y2K Eric Updated the help and explain for huge and for the move to the aoc from cv. Update ddt.html on the web site. Moved nowhere. 11712. October 17, 2004 WIPER Eric Added a line around the plot and at zero in each axis. The line is in graphics channel 5 and so is overridden by the data if needed. Added comment on this to help file. Moved nowhere. 11713. October 18, 2004 head/tail Eric/Walter Briskin Walter pointed out that the grammar of the unix utilities head and tail has changed to "head/tail -n L" from "head/tail -L" to show L lines. Fixed $INSUNIX files CDSETUP, INSTEP1, and REGISTER and $SYSUNIX files AIPSPATH.CSH, AIPSPATH.SH, LIBR, PRDEVS.SH, START_AIPS, TVDEVS.SH, UNIXSERVERS, XASERVERS, ZLASCL, COMLNK, COMRPL, and SYSETUP. This should test the new capabilities of the MNJ nicely. Moved nowhere. 11714. October 28, 2004 $APLNOT/SNAMBG.FOR Leonia The new subroutine SNAMBG reads the input SN table, calculates phase for each row IF and stoke and then solves 2 PI ambiguity for each antenna and source. The value of the ambiguity solved phase (in degrees) is recorded into the REAL column of the output SN table. IMAG column of the output SN table is equal zero. Having used the output SN table should be deleted either inside of the main routine (DELEXT) or outside using the verb EXTDEST. Moved nowhere. 11715. November 1, 2004 $APLNOT/SNAMBG.FOR Leonia The sorting information was lost at the output SN table. Fixed. Moved nowhere. 11716. November 1, 2004 Plotting beams Eric It can be annoying if the plot interferes with the plot of the beam. Changed: PLTBEM To set integer corners rather than floating to match PLTBSZ PLTBSZ Version of PLTBEM to return the pixels used to plot the beam so that the plot programs may avoid them. CONDRW Changed call seq to take the beam plot corners and blank all data at those pixels so that no contours will be drawn. PCNTR Changed to blank grey scale and polarization plots in the beam plot area if requested and pass beam corners to CONDRW. It calls PLTBSZ if requested to set the corners first. PCNTR HLP: changed CBPLOT meanings described CBPLOT HLP: changed CBPLOT meanings described CNTR Fixed call seq to CONDRW to do no blanking. CCNTR Fixed call seq to CONDRW to do no blanking. KNTR Like PCNTR plus a messier thing in the contouring since that is doen in panels by KNTR KNTR HLP: changed CBPLOT meanings described GREYS Add CBPLOT adverb and interpret it, call PLTBSZ, blank grey scale appropriately, change call to CONDRW, and finally call PLTBEM. GREYS HLP: added adverb CBPLOT, all meanings described Moved nowhere. 11717. November 1, 2004 YCURSE Eric YCURSE never checked that the TV coordinates were reasonable (i.e. 1 - MAXXTV). There is a suspicion that wierd values can come back from XAS on some systems and the software does not defend itself, causing mysterious aborts. Moved nowhere. 11718. November 2, 2004 ATMCA Leonia Two pi ambiguity is solved using the subroutine SNAMBG. The relevant correction of the codes for output SN table has been added. Help file is updated but is needed (probably) improvement. Moved nowhere. 11719. November 5, 2004 SETGDS Eric Some tasks wrongly initialized BITER to 0 rather than 1. Changed SETGDS to interpret BITER < 1 as = 1. Moved nowhere. 11720. November 7, 2004 Robust gains Eric Changed things to allow a choice of robust gain solutions instead of a single-pass solution using all the data: POPSDAT New adverb DOFLAG (putbck 21-Oct) NEWPARMS New adverb DOFLAG (putbck 21-Oct) GCALC Rearrange code to do robust looping or to omit it (i.e. do 1 to 10 or 10 to 10 for robust iteration) Call sequence returns fraction data that would heve been omitted from the next solution had there been one. Call sequence addition also of INIT to say whther gains are init to (1,0) or GLAST and to set the clip level of the last iteration. GCALC1 As GCALC. GCALC2 As GCALC, return DOFIT to original method which had been obliterated April 3. NCALC As GCALC - required much modernizing of excessive GO TO's - and the previous change to allow for weighting was not done correctly. NEVAL Used by NCALC - tested WT not equal 0, should test WT greater than 0 CLBPA Add FRAC to call sequence of above, add to histogram which is added to call sequence. Pass back the modified weights. Clip level of last iteration added to call sequence to pass into GCALC et al. Take the weights returned by GCAL et al and pass them back to calling program. UVUTIL Fix call sequences of above and interpret additional codes in SOLTYPE. Added FFLAST keyword fetch with 2.5 default and added that to calls of GCALC et al. CALIB Interpret additional SOLMODE values, add histogram init and display. Added adverb DOFLAG interpreting it and passing it along to CLBPA. Add code flagging data when rejected it on the last pass (including non-robust fits). BPASS Add SOLTYPE adverb. Drop local GCALC and offer GCALC, GCLAC1, NCALC methods including robust. Allow divide by channel 0 when autocorrelation data only. SCIMG Interpret additional SOLTYPE values SCMAP Interpret additional SOLTYPE values CALIB HLP : Add robust forms of SOLMODE, amplify DOFIT discussion. SCIMG HLP : Add robust forms of SOLTYPE SCMAP HLP : Add robust forms of SOLTYPE BPASS HLP : Add SOLTYPE adverb. Upgrade discussion of normalization and BPASSPRMs DOFIT HLP : Amplify discussion considerably. SOLTYPE HLP : Add robust solution values and discussion of how robust method is done. FRING HLP : Amplify DOFIT discussion KRING HLP : Amplify DOFIT discussion DOFLAG HLP : new Moved nowhere. 11721. November 8, 2004 ZLOCK.C Eric Changed $ALPUNIX and $APLBERK versions of ZLOCK.C to issue and error message and set an error return only when it is not a read-only file system. Moved nowhere. 11722. November 8, 2004 SETGDS Eric The epoch test mentioned ALTSW rather than EPOSW. Moved nowhere. 11723. November 8, 2004 CALIB Eric The second polarization did not count the solutions with too little data. Moved nowhere. 11724. November 8, 2004 PATGN Eric Fixed the default reference pixel and changed the CPARM adverbs so that they are all in arc seconds where appropriate. Some were in pixels and some in arc sec. Moved nowhere. 11725. November 8, 2004 EXTLIST Eric Changed AU8A (EXTLIST verb): to account for the addition of the CBPLOT adverb in GREYS. Corrected code for POSSM and FRMAP and FRPLT. All three used the wrong sign for the offset parameter and the latter 2 had the version entries placed wrongly. Moved nowhere. 11726. November 9, 2004 KNTR Eric Fixed an error done with the last change (CBCORN negative). It took 0 as a request for a beam in a separate panel. Moved nowhere. 11727. November 9, 2004 ATMCA Leonia The deleting of the intermediate SN table has had the mistake. The table was deleted from the catblk header but was not destroyed. As the result the task did not work at the sequence run. It fixed. Delete of the output SN table (if OPTYPE='PHAS') has been added. The help file updated. Moved nowhere. 11728. November 10, 2004 CALIB Eric A bug in counting in CLBPA caused aborts and failed solutions. Moved nowhere. 11729. November 11, 2004 Doc files Eric Updated LSAPROPO.HLP and ABOUT files for ADVERB, CALIBRAT, EDITING, GENERAL, HARDCOPY, IMAGING, INFORMAT, OOP, PLOT, POPS, PROCEDUR, TASK, TV, UV, VERB, and VLBI. Moved nowhere. 11730. November 12, 2004 BPASS Eric Added mode 4 for normalizing (BPASSPRM(10)) so that ICHANSEL can be a few chnnels if needed and all VLBI channels then used for normalizing. Moved nowhere. 11731. November 13, 2004 CookBook Eric Updated CookBook chapters: 0: Update 4: SMOOTH revised and warning added, BPASS SOLTYPE added, fix file name for CALRD example 5: IMAGR: SDI parameter meaning revised, mention FIXBX, CALIB: SOLTYPE, DOFLAG, WEIGHTIT; FLAGR & FINDR mentioned 6: DOCRT=-2,-3; POSTSCRIPT.HLP, PCNTR meaning of DOGREY changed, CBPLOT enhanced 7: MFPRT added option 9: CLCAL 'SELF' changed meaning; RLDIF rather than LISTR for polarization; mention FLAGR, FINDR, WIPER; drop use of SMOOTH and add warning; downgrade BLING; BPASS discussion enhanced and new section on A&P added. WEIGHTIT in FRING. 10: POSTSCRIPT.HLP; mark SPFLG figure as dated 13: Brought up to date. C: SMOOTH dropped; BPASS changed; CVEL added I: Update J: Update Recipe: new ones and update Also did COOKBOOK.PDF. Moved nowhere. 11732. November 16, 2004 PP.FOR Eric Changed $SYSUNIX/PP.FOR to translate 'LONGINT ' to 'INTEGER*8 ' on all machines except HP and HP2. They get plain 'INTEGER '. This is anticipating a use in the 31DEC05 release, but we need to get the change in now so install.pl on that release will build it correctly. Moved nowhere. 11733. November 17, 2004 CCOPTS.SH Eric The Solaris Ultra CCOPTS.SH had a case statement on Fortran compilers not on C compilers. Moved nowhere. 11734. November 18, 2004 UPDUPDATE Eric Added updating the $UPDUNIX files in $TST/$ARCH/UPDATE. Moved nowhere. 11735. November 18, 2004 LaCook.mac Eric The LaTeX people helped me debug my macros and found why the start of chapters typeset differently when doing one or many chapters. Putbck this fix to preserve it. Now I should look at the macros a bit more... Moved nowhere. 11736. November 18, 2004 UPDSRTUNQ.SED Eric This sed script file had a begin: lable that was unused and generated error messages on Macs. Dropped the unused label. Moved nowhere. 11737. November 18, 2004 CALIB Eric Adjusted the DOFLAG default: if APARM(6) = 0, the default is -2.5 so that the DOFLAG marking will generate some report. But if APARM(6) = 1 or more, the default is -99 which means that the gain solver will not mark any data bad so that the closure checker will receive all data that entered the gain solver. Moved nowhere. 11738. November 18, 2004 VLBALOAD.HLP Lorant Changed the upper limit of NFILES from 999 to 9999 to accomodate job numbers over 999 to be used in VLBAARCH. Moved nowhere. 11739. November 21, 2004 FILLM Eric Apparently if a run includes Pie Town and Q band at all, the antenna positions will be shifted at other bands too. Fixed it to react to > 25 OUT antennas and try the center moved fix. Changed the fix to leave a (0,0,0) antenna position unchanged. Changed task to print the first 10 bad nominal sensitivity messages and then a summary at the end. moved nowhere. 11740. November 22, 2004 SNPLT Leonia The new OPTYPE to plot out REAL or IMAG has been added. moved nowhere. 11741. November 22, 2004 EDITA Eric Changed QEDIUTIL - it chose badly for which IFs and polarizations to process when undoing and redoing flags. Moved nowhere. 11742. November 22, 2004 binary MNJ/install Wes/Eric In preparation for the binary install and MNJ putbck the MNJ procs: AIPSUPD Call UPDRSYNC on $UpdRsync else UPDCOMRPL and UPDCOMLNK UPDCONFIG Set $UpdRsync on existence of file $SYSLOCAL/RSYNC.ME MAKE.MNJ $TST/$ARCH/UPDATE to have $UPDUNIX files copied rather than symlinked MAKE.BMNJ New MAKE.MNJ without compiles, copies - just cvs, do_daily, UPDCONFIG UPDRSYNC New - does the rsync for binary UPDUPDATE Copy UPDUNIX files rather than link AIPSPATH.SH Set LD_LINRARY_PATH for binary on the RSYNC.ME file's existence. AIPSPATH.CSH As above. Moved nowhere. 11743. November 25, 2004 CALIB Eric When there are ony a few baselines the robust method does not make much sense nor does the marking of data as questionable. Changed GCALC, GCALC1, GCALC2, and NCALC to suppress the robust and flagging when the number of baselines is small. Added ECALC to do exact gain computations when the number of baselines is 1 or 2. Moved nowhere. 11744. November 26, 2004 CALIB Eric Fixed the scaling of gains in ECALC for two baselines. Moved nowhere. 11745. November 26, 2004 EDITA, EDITR, etc Eric Modified QEDIUTIL to 1. Allow UNDO of flags for IFs not in core. It just means dropping the flag so that it is not re-applied to the data when they are again loaded. 2. Change flag area, above, and below to do all points rather than just un-flagged points. Required a new OP in EDFLAG and a new subroutine EDGTAP (get all points not just unflagged). This allows a REDO to really redo the intent rather than just those points that were flagged the first time. Changed also:HLPEDICL, HLPEDITY, HLPEDIUV for this greatly improved REDO FLAGS. Changed EDITA.FOR and help to delete the new SN/CL table that is created for the editing process. Moved nowhere. 11746. November 27, 2004 UVFIX.HLP Eric Changed the help file to improve grammar and to mention option 20 in the inputs. Moved nowhere. 11747. November 27, 2004 Buffered TV Eric Fixed ZSSSO2.C - it reset buffered to 0 every time but only checked its setting on the first call in a task. Moved nowhere. 11748. November 27-28, 2004 install.pl, UPDRSYNC Wes/Eric Significant overhall of install.pl to install using rsync for all the source code and binaries on option -n. UPDRSYNC is the part of the MNJ that does the binary rsync's to $ARCH/SYSTEM, LIBR, and LOAD. It now checks for a file named BUILD.INPROG and waits if it is present looping 20 waits of 10 minutes before giving up. We just need to get the aipssrc cron fully up to speed and we should have a full binary system for MACPPC and soon for SUL. Moved nowhere. 11749. November 28, 2004 COMB Eric Fixed history display of operation to show 4-character OPCODE rather than a number and to have a value for SUMM in the comment field. Altered help file to admit to SUMM in the inputs area. Moved nowhere. 11750. November 28, 2004 install.pl, PAPC.INC Eric Changed the main PAPC.INC to be 20 megabytes. Changed install.pl to have 20.0 as the default and to skip the detailed screen when doing binary and CD installs. Moved nowhere. 11751. November 29, 2004 misc Eric Changed install.pl to skip menu 10 for binary installs. Fixed grammar for AIPSPATH.CSH for binary systems. Dropped update.pl from CDSETUP. Moved nowhere. 11752. December 1, 2004 CDSETUP, HLPIT.LIS Eric Updated the readline completion file. Changed CDSETUP to have 4 modes - the test with -f cannot tell the difference between a link file to a real file and a simple real file. Thus, which direction one is converting an existing system had to be double checked. It is simpler to ask up front. Moved nowhere. *********************************************************************** *********************************************************************** Changes while 31DEC04 is NEW *********************************************************************** *********************************************************************** 11754. December 2, 2004 MNJ Eric Changed UPDRSYNC to drop the work-around for an error and AIPSUPD to fix the error, definig $logFile and the error file with full pathnames rather than relative ones. We have been leaving LogFile entries lying around. Moved from 31DEC05 this date. 11755. December 3, 2004 more MNJ Eric UPDCVS and UPDREMOVE also had this work-around. (I hate work-arounds, one should always find the general solution.) Moved from 31DEC05 this date. 11756. December 5-6, 2004 more MNJ Eric Changed AIPSUPD to set the LogFile and ErrorFile logicals with full pathnames inside the version loop. Otherwise they were set with the calling version which might not be the version(s) being done. Changed AIPSUPD, UPDCONTROL, and UPDCVS always to tee every message that goes to ErrorFile also to LogFile. It was inconsistent previously. Changed UPDUPDATE to do the $AIPS_ROOT portions only if UpdDoCopy is YES and to provide the $AIPS_ROOT value as a call arg to AIPSROOT.DEFINE. Changed UPDREPORT to use the logical LogFile rather than a locally relative name. Moved from 31DEC05 this date. 11756. December 14, 2004 SN2CL Amy The IBM compiler exposed a problem in SN2CL (a subroutine that does the interpolation for CLCAL) in which it relied on floating point arithmetic to give it an exact zero in order to not interpolate FBLANK entries. SN2CL now no longer relies on this nor does it do any arithmetic on FBLANKs themselves. Moved from 31DEC05 this date. 11757. December 6, 2004 $APLNOT/J2B, B2J Leonia The new subroutines J2B, B2J convert J2000 to B1950 and vice versa. Moved to 31DEC04 on 14 Dec. 11759. December 8, 2004 Misc Eric Changed: 1. FLATN.HLP There is no reason to limit IMSIZE to 8192 2 UPDCONFIG Wrong home area for MPI. Moved to 31DEC04 this date, #1 to Hawaii, #2 to Bonn. 11760. December 9, 2004 $APLNOT/JPRECS.FOR Leonia The new version of JPRECS can work with both J2000 and B1950 coordinate systems. Originally JPRECS could work only with J2000. The new version converts B1950 mean coordinates (if EQUIN=1950D0 and DIR=1) into J2000 using the new subroutine B2J (see #11757) and recalculates the found J2000 mean coordinates to the apparent coordinates. If DIR=-1, then JPRECS converts the given apparent coordinates into the mean J2000 and then recalculates J2000 in B1950 using the new subroutine J2B (see #11757). Moved to 31DEC04 on 14 Dec. 11763. December 14, 2004 SN2CL Amy The IBM compiler exposed a problem in SN2CL (a subroutine that does the interpolation for CLCAL) in which it relied on floating point arithmetic to give it an exact zero in order to not interpolate FBLANK entries. SN2CL now no longer relies on this nor does it do any arithmetic on FBLANKs themselves. Moved to 31DEC04 this date. 11764. December 14, 2004 AIPSASSN Eric Changed to add VLADATADIR for the VLA site and to add a call to LOCALASSN.CSH or .SH if present. Moved to 31DEC04 this date. 11765. December 14, 2004 FILLM Eric Changed to force the apparent position to be computed by Leonia's new JPRECS. This removes the diurnal aberation included by the VLA but it was only right for one time not the whole data set anyway. Moved to 31DEC04 this date. 11767. December 15, 2004 MAKE.*MNJ Eric Changed MAKE.MNJ to copy the MAKE.*MNJ files to the update work area along with the A*, U*. and R* files. Also fixed the message that comes out ahead of doing this from link to copy. MAKE.BMNJ still was not executable in the putbck version. Moved to 31DEC04 this date. 11768. December 17, 2004 FILLM Eric Users may use fixed frequency for calibrator observations followed by tracking frequency for target source line observations. Changed the setting of the velocity type keywords to catch tracking modes should they occur after the first source. Moved to 31DEC04 December 29. 11771. December 17, 2004 install.pl, MAKE.*MNJ, UPDRSYNC Eric The former left out HP2 architecture from the list of cvs forbidden things and the latter including a debug printout. Added SYSTEM areas for all architectures to install.pl even for rsync installs. Moved to 31DEC04 this date. 11773. December 20, 2004 X band models Amy/Eric Put back models for 3C286 and 3C48 at X band. Moved to 31DEC04 automatically. 11786. December 29, 2004 FILLM Eric Added check on gain codes with message and flagging of data. Moved to 31DEC04 this date. 11787. December 31, 2004 install.pl Eric Fixed back to 31DEC04 and made to be NEW. Moved nowhere - this is for 31DEC04 only. 11788. January 4, 2005 install.pl, CDSETUP Eric install.pl was given option -c to mean CDrom as well as the test on logical AIPSCDROM=YES. It was fixed to turn off network when CD and vice versa and to test for previous network or CD specification not made this time. CDSETUP was excessively complicated since it assumed that one might be redoing the size of the AP or the XAS or similar things even when keeping things on the CD. These things cannot be done. I suppose that one could fully offload the CD to disk and then do a normal install rather than a CD one, but not when keeping a light footprint. Moved to 31DEC04 (04 version of install.pl) this date.