[daip] Re: ATCA specific code in 31DEC03

[Eric Greisen]

Chris Phillips writes:
 > On Thu, 2 Oct 2003, Eric Greisen wrote:
 > 
 > > I have been looking at this in considering what UVFIX should do...
 > 
 > Almost certainly no-one would notice if the "dont recompute uvw" option
 > went. It was probably put in one for one specific reason for one person.
 > If you feel it is not worthwile, discard the change.

It is more complicated - should we compute -NCP or swicth to -SIN.  I
have done all but GREYS as of now.  The GREYS may get a new file
format for the input data to support our local holography guys as
well.

The reply generated so far is:

Eric

VKEY.INC - done - the code you sent had an error.  It put 34
                  characters in a 32-character variable so the last 2
                  were lost.

ZABOR2   - done - nice magic although it causes a compiler warning

UVFIT    - done - I made it it much larger and fixed help file

GETJY    - done - why does at want to avoid doing frequencies
                  correctly?   The BIF part was our error.

GETANT   - done - this might not be the only place needed for this!

GEOM.HLP - done - Deleted GEOM task entirely, put comments in LGEOM
                  help file

MCUBE.HLP - done - It is simplistic and dated but I added it.

BPASS    - not done - buffer has to do Number IFs * Number sources
                  and that is anyone's guess.  It is equivalence to a
                  large buffer.  Made the number 16384.
                  SOLINT test is in fact an error on AT's part.  The
                  limit for end time > scan end time just avoids minor
                  round-off troubles.

CALIB    - done - Changed limit to MAX (3, MIN (6, NUMANT/2)).  Put
                  the inputs in as comments WITHOUT TAB CHARACTERS.

UVLOD    - done? - Raised main buffer sizes (also FITLD)
                   Stokes test needed for linear (also FITLD)
                   Buffer change for tables wrong (512 is all it uses)

FITTP    - done - The AT code did not work correctly.  Put correct
                  code with new adverb in FITTP and FITAB.

CLCOR    - done - Added array name tests, changed comment, skip FQ
                  test on ATCA and ATLBA

Added ATLBA to PCCOR, UVCRS, ACCOR, SPLIT, SPLAT, GETJY, LOCIT, CVEL,
VPLOT, CLPLT and made last 3 friendlier rather than depending on the
user to tell us the array type.

UVFIX    - done - also added code to change -NCP to -SIN with a
                  warning.