Hi David,
re:
>I have a more basic question. Is it more efficient to
>run a single pqact with a large/long/complicated pqact.conf,
>or, is it better to run several pqact each with a more
>simple pqact.conf?
It is much better to run several pqacts as long as the feeds they
are working on are mutually exclusive from the others. The reason
for this is that pqact scans _ALL_ entries in a pattern-action file
before processing the next product.
>Each pqact.conf might access the same
>product and do something different with it, or each
>pqact.conf might handle particular groups of products.
I would suggest different groups of producs (e.g., actions for
different feeds, etc.).
>(for example you could have a pqact.conf for each software package,
>or, you could have a pqact.conf for every feedtype.)
How you mix and match the actions should be governed by the number
of products being processed and the number of actions being checked.
>If one were to split up and reorginise their pqact.conf
>file would there be a computational advantage or disadvantage
>to either method?
Yes, but a definitive statement as to what is the best approach
would depend on the specifics of the actions being taken.
>I'm just wondering is there a good reason for splitting
>up a pqact.conf into several smaller files,
There is.
>or would splitting it by inbeded comments serve just as well?
I don't understand this comment. pqact will compare every action
in its pattern-action file against the ID of the product being
processed to see if there is a match.
Cheers,
Tom
--
+----------------------------------------------------------------------------+
* Tom Yoksas UCAR Unidata Program *
* (303) 497-8642 (last resort) P.O. Box 3000 *
* yoksas@xxxxxxxxxxxxxxxx Boulder, CO 80307 *
* Unidata WWW Service http://www.unidata.ucar.edu/*
+----------------------------------------------------------------------------+
