Subject: Re: SMA MIRIAD Bug Report
From: Ken Young <rtm@cfa.harvard.edu>
Date: 04/11/2013 04:28 PM
To: Jun-Hui Zhao <jzhao@cfa.harvard.edu>
CC: gpetitpas@cfa.harvard.edu, smamiriad@cfa.harvard.edu

On Thu, 11 Apr 2013, Jun-Hui Zhao wrote:


Hi Glen,
    I will look at the problem and let you know as soon as have a
solution.
    Jun-Hui

The problem was almost certainly caused by the dropping of antennas 1 and
4 from the project.    The antennas were dropped from the project at
UT 08:05:14, but the newFile command was not given until about 20 minutes
later.   So the 130411_01:33:45 data set has variable numbers of baselines
in the sch_read file.   MIR/IDL can still process the data set (although
very slowly) but it may confuse Miriad.   Had they just dropped antennas
from the project, things would have been ok.   But they re-dopplerTracked
and restarted the correlator without opening a new data set, and that
caused the number of visibilities written for each scan to complete.   The
files can still be processed, because the pointers to the data are all OK,
but any code that assumes each scan is layed out in the same way will
fail.

I don't see anything wrong with the other file you listed,
130411_08:23:16, and MIR/IDL will read both of them.

Taco