[daip] AIPS flag entry limitations

[Danielle Fenech]

Hello,

I'm currently working on flagging software to work on eMERLIN data via 
parseltongue, which writes out flag tables to be loaded into AIPS via 
TBIN. However, I have run into some problems with limitations of various 
tasks in handling a large number of flag entries.
TBIN itself declares a limitation of 1 million entries and won't write 
anything larger. Tasks such as SPFLG seem to have an even lower limitation 
of around 120,000 making assessing the flag entries difficult.

The datasets themsleves are reasonably large, up to 500G and possibly 
larger and typically the flag tables written by the flagging software are 
consistently a few to 15-20 million entries in size (sometimes more).

Is there any way to get the working limit for flag entries raised (or 
even removed) and would it be possible to have a consistent limitation as 
opposed to task-based?

I don't currently know where these limitations are set within the AIPS 
structure, but if you can point me in the right direction, I could have a 
look at making some changes locally. However, given the large volumes of 
data and the intended general use of such flagging software, ideally a 
more global fix for the problem would be better.


Many thanks,

Danielle
-- 




-------------------------------------------------
Dr. Danielle Fenech
Department of Physics and Astronomy,
University College London,
Gower Street,
London WC1E 6BT, UK


Tel: +44 (0)20 7679 7682
Fax: +44 (0)20 7679 7145
Email: dmf at star.ucl.ac.uk
-------------------------------------------------