[daip] New Client Reply - [!QHQ-880861]: CL tables with no entries for a source

[Michael Bietenholz]

New Client Reply: CL tables with no entries for a source

I agree that if CLCAL is not writing out entries it should that should be fixied.  I also agree that the only one-CL table method is probably not for the general user :-)

However, I don't see how you can, in general, prevent the possibility of missing, or shall we say non-existent, CL records in the output of CLCAL.   Users might quite legitimately run CLCAL only w/ specific SOURCES selected.  The output CL table from such a run would be missing entries for any non-selected sources, which is all as it should be. 

The problem I just ran across is if the user then mistakenly sets GAINU=True and selects one of the 
previously de-selected sources, they get supposedly calibrated visibility data which is likely garbled instead of just being alerted to the fact that they're trying to apply a non-existent bit of calibration.

There *is* a related CLCAL issue, in that sometimes it *refuses* to interpolate over flagged (-ive weight) SN table entries, and flags the corresponding CL entries. I've occasionally had to resort to TACOP' to strip the flagged SN table entries before running CLCAL to get it to not created gaps in the output CL table. (Note this is with sufficiently long CUTOFF and DOBLANK=0 in CLCAL).  I'm not sure when this crops up, I don't think it happens whenever you have a flagged SN table entry.  I will send an example next time I run across it.

Ticket Details
===================
Ticket ID: QHQ-880861
Department: AIPS Data Processing
Priority: Default
Status: Open
Link:  https://help.nrao.edu/staff/index.php?_m=tickets&_a=viewticket&ticketid=2910