pipeline cannot handle a bad beam properly

[teuben]

if one session is all good, and another sessions has a bad feed/beam, this is not done correctly, unless the bad feed file is manually removed. which is a manual job, example for last night:

g=UGC02239
mask_UGC02239_Havfield_v1.fits
rm -rf $g
./reduce.py -m $m -g 33 -f 2    $g
# i now have to manually remove feed2 files for this
rm -i $g/*feed2*
./reduce.py -m $m -g 1,2        $g

thus we need the script to remove (or never create) those feed files.

[teuben]

solution would be not to write the bad beam spectra