[daip] FITS IDI loading via fitld is changing telescope ordering on import

[Eric Greisen]

On 06/24/2014 03:50 PM, Paul Harrison wrote:
> Dear Eric,
>
> sorry to mail you directly but the help desk https://help.nrao.edu/ always complains that the department ID has not been set.
>
> we are exporting data from eMerlin as FITS IDI and then using FITLD in AIPS to input - The intention is to sort the data into time-baseline order, and we believe this is done properly within the IDI however on import it seems as if the ordering is swapped for some telescopes (marked with an S in the following table) - it is not clear to me if there is some subtlety in the ordering as defined in AIPS - i.e. it is not simply antenna number, or whether there is a bug in FITLD -
>
> FITS	 AIPS order	
> 2 2	 2- 2	
> 2 7	 2- 7	
> 5 2	 2- 5	 S
> 5 5	 5- 5	
> 5 7	 5- 7	
> 5 8	 5- 8	
> 6 2	 2- 6	 s
> 6 5	 5- 6	 s
> 6 6	 6- 6	
> 6 7	 6- 7	
> 6 8	 6- 8	
> 6 9	 6- 9	
> 7 7	 7- 7	
> 8 2	 2- 8	 s
> 8 7	 7- 8	 s
> 8 8	 8- 8	
> 9 2	 2- 9	s
> 9 5	 5- 9	s
> 9 7	 7- 9	s
> 9 8	 8- 9	s
>
> I can supply a smallish example IDI file (though still 100MB - so I have not attached it unsolicited)

The convention established long ago was that the smaller numbered 
antenna is listed first.  FITLD enforces this - and you should note that 
the sign of the phase and RL/LR switch with this switch also since we 
take the order of the two numbers to imply the direction of the baseline.

When AIPS says TB sort order all that really is assumed is the T since 
we often get data with the antennas sorted not by real antenna number by 
by DCS address or some other non-physical scheme.

Eric Greisen