[Difx-users] baseline-selective merge of two DiFX runs?

[Walter Brisken]

Hi Jan,

To be sure I understand -- you are recorrelating the same job, but with a 
different subset of baselines, and just want a visibility set that has as 
many unique baselines as possible?

I think this would be a valuable addition to the toolset and know of no 
other tools that can do this.  I suspect as long as the visibilities are 
in weak time order (e.g., doesn't matter within a timestamp the order of 
baselines, polarizations, or IFs, but timestamps should not go backwards), 
that difx2fits should work on it.

 	-Walter

On Mon, 8 Jan 2018, Jan Wagner wrote:

> Hi,
>
> we're in a situation where we have two DiFX output data sets:
>
> set 1 all stations present but one station (Eb) with VEX $FREQ incorrect
>
> set 2 some stations (Kt Ku Ky Mh) lost as their data were erased,
> but Eb VEX $FREQ corrected
>
> The correlator settings v2d and the .input files are identical
> between these two sets.
>
> I'd like to copy the visibilities from these two sets into a new
> "merged" output set. Essentially, take set 1, ignore all
> visibilities from baselines to Eb, and if set 2 has data on that
> particular baseline to Eb then copy those visibilities from set 2.
> This would produce a new correlator output dataset where the maximum
> number of baselines have good visibility data, except of course of
> the irrecoverable baselines (Eb) x (Kt Ku Ky Mh).
>
> Presumably this would be a trivial merge operation based upon the
> baseline tag of each visibility data record in pairs of .difx/DIFX_*
> files.
>
> But before I rush off and create a merging script in Python:
>
> Does anyone know of (or have :-)) a "mergeDiFX.py" or similar script
> that does this trick?
>
> kind regards,
> Jan
>
> _______________________________________________
> Difx-users mailing list
> Difx-users at listmgr.nrao.edu
> https://listmgr.nrao.edu/mailman/listinfo/difx-users
>

-- 
-------------------------
Walter Brisken
Director
Long Baseline Observatory
(575)-835-7133 (office)
(505)-234-5912 (cell)