[daip] [Fwd: BB255CB and CD]

[Eric Greisen]

Andreas Brunthaler wrote:
> Hi Amy,
> 
> it is pass 4 (the only one with only 16 channels). The files are:
> 
> BB255CB/BB255CB_4_HPOL4IF8MHZ16CH_22.197GHZ00
> BB255CD/BB255CD_4_HPOL4IF8MHZ16CH_22.197GHZ00
> 
> Thanks,
> Andreas
> 
> 
> Amy Mioduszewski wrote:
>> Hi Andreas,
>>
>> I am going to try to download your data and reproduce your problem. 
>> Which pass were you doing?
>>
>> Thanks,
>>
>> Amy
>>
>> Andreas Brunthaler wrote:
>>
>>> Dear Eric,
>>>
>>> Eric Greisen wrote:
>>>
>>>>> Dear Analysts,
>>>>>
>>>>> I have downloaded the data from BB255CB and CD today from  the archive
>>>>> (the 16 channel pass). When I try to load the two datasets with fitld
>>>>> and doconcat 1 I get the following error:
>>>>>
>>>>> PC2004> FITLD1: Task FITLD  (release of 31DEC09) begins
>>>>> PC2004> FITLD1: MISMATCH           6           2  AXIS COUNT
>>>>> PC2004> FITLD1: ERROR      1 RETURNED FROM UVFHDR
>>>>> PC2004> FITLD1: Purports to die of UNNATURAL causes
>>>>> PC2004> FITLD1: pc20045      31DEC09 TST: Cpu=       0.0  Real=       0
>>>>>
>>>>> The headers are almost identical:
>>>>>
>>>>> AIPS 1: Image=MULTI     (UV)         Filename=BB255C      .FXPOL .   1
>>>>> AIPS 1: Telescope=VLBA               Receiver=
>>>>> AIPS 1: Observer=BB255               User #=  255
>>>>> AIPS 1: Observ. date=08-APR-2009     Map date=05-MAY-2009
>>>>> AIPS 1: # visibilities   1046761     Sort order  TB
>>>>> AIPS 1: Rand axes: UU-L-SIN  VV-L-SIN  WW-L-SIN  TIME1  BASELINE
>>>>> AIPS 1:            SOURCE  FREQSEL  INTTIM  GATEID  CORR-ID
>>>>> AIPS 1:
>>>>> ----------------------------------------------------------------
>>>>> AIPS 1: Type    Pixels   Coord value     at Pixel     Coord incr  
>>>>> Rotat
>>>>> AIPS 1: COMPLEX      3   1.0000000E+00       1.00  1.0000000E+00   
>>>>> 0.00
>>>>> AIPS 1: STOKES       2  -1.0000000E+00       1.00 -1.0000000E+00   
>>>>> 0.00
>>>>> AIPS 1: FREQ        16   2.2197040E+10       0.53  5.0000000E+05   
>>>>> 0.00
>>>>> AIPS 1: IF           4   1.0000000E+00       1.00  1.0000000E+00   
>>>>> 0.00
>>>>> AIPS 1: RA           1    00 00 00.000       1.00       0.000000   
>>>>> 0.00
>>>>> AIPS 1: DEC          1    00 00 00.000       1.00       0.000000   
>>>>> 0.00
>>>>> AIPS 1:
>>>>> ----------------------------------------------------------------
>>>>> AIPS 1: Coordinate equinox 2000.00
>>>>>
>>>>>
>>>>> and
>>>>>
>>>>> AIPS 1: Image=MULTI     (UV)         Filename=BB255C      .FXPOL .   2
>>>>> AIPS 1: Telescope=VLBA               Receiver=
>>>>> AIPS 1: Observer=BB255               User #=  255
>>>>> AIPS 1: Observ. date=09-APR-2009     Map date=05-MAY-2009
>>>>> AIPS 1: # visibilities    970096     Sort order  TB
>>>>> AIPS 1: Rand axes: UU-L-SIN  VV-L-SIN  WW-L-SIN  TIME1  BASELINE
>>>>> AIPS 1:            SOURCE  FREQSEL  INTTIM  GATEID  CORR-ID
>>>>> AIPS 1:
>>>>> ----------------------------------------------------------------
>>>>> AIPS 1: Type    Pixels   Coord value     at Pixel     Coord incr  
>>>>> Rotat
>>>>> AIPS 1: COMPLEX      3   1.0000000E+00       1.00  1.0000000E+00   
>>>>> 0.00
>>>>> AIPS 1: STOKES       2  -1.0000000E+00       1.00 -1.0000000E+00   
>>>>> 0.00
>>>>> AIPS 1: FREQ        16   2.2197040E+10       0.53  5.0000000E+05   
>>>>> 0.00
>>>>> AIPS 1: IF           4   1.0000000E+00       1.00  1.0000000E+00   
>>>>> 0.00
>>>>> AIPS 1: RA           1    00 00 00.000       1.00       0.000000   
>>>>> 0.00
>>>>> AIPS 1: DEC          1    00 00 00.000       1.00       0.000000   
>>>>> 0.00
>>>>> AIPS 1:
>>>>> ----------------------------------------------------------------
>>>>> AIPS 1: Coordinate equinox 2000.00
>>>>>
>>>>> Do you have any idea why I get this error?
>>>>>
>>>>> Cheers,
>>>>> Andreas
>>>>
>>>> This is a bizarre looking error - the number of axes being 2 would apply
>>>> to some simple images perhaps although aips normally makes more axes
>>>> than that.  There was a bug in FITLD that I fixed in 31DEC09 on April 10
>>>> which conceivably could be behind this problem although the testing that
>>>> I did showed other issues mostly not serious.  Try bringing your version
>>>> up to date and re-running FITLD.
>>>
>>> I tried it both with our MPIfR installation and on my laptop (after
>>> bringing it up to date). Still the same problem. However, if I load both
>>> files separately and use DBCON, it seems to work.
>>>
>>> Andreas
>>>
>>>
> 
> 
I have looked at this issue and now understand it.  When the data being 
read are from FITTP or FITAB, DOCONCAT cannot work.  Actually I do not 
understand why we don't allow it for regular FITTP output - although it 
would be a lot of work to disentangle issues of renumbering antennas, 
sources, FQIDs, etc.  But for FITAB output - after FITAB was corrected 
to be a proper rather than cheating implementation of FITS tables - 
nothing is known about the structure of the UV data until the UV table 
itself has had its header parsed.  Thus we cannot decide in advance 
whether concatenation is allowed or not and we have to write the other 
tables somewhere before we get to the uv data table.

I have changed the fortran to ignore DOCONCAT for files other than those 
directly from the correlator and the help to make it clear that DOCONCAT 
applies only the FITS-IDI files from the correlator.

DBCON may be run on the files after they are loaded - although they may 
require MATCH first since DBCON will not renumber antennas (it does do 
sources and FQIDs).

Eric Greisen