2.3exp: expire and tradspool

[Richard Michael Todd]

In message <m11IJYf-000FaLC at independence.ecn.ou.edu>I wrote:
>In message <19990821161431Y.kondou at inn.do.mms.mt.nec.co.jp>you write:
>>btw, expireover takes much longer time if most of articles
>>are spooled into timecaf.  I know why it does;
>>SMretrieve(RETR_STAT) needs to open caf file even if it
>>just checks the existense.  Any resolution?
>
>Not any easy one I know of.  Timehash is probably the same way.  We could  hac
>k
>timehash and timecaf to keep a database of the currently active tokens
>in spool, but that would slow down writes and add one more thing to get out
>of sync.  

I thought of a way around this that might work without making the timecaf
system too much more complex.  Basically, I'd have requests for 
SMretrieve(RETR_STAT) cache the CAF TOCs for the CAF files it runs across,
so that each file will have to be accessed at most once.  It'll make 
expireover take up more memory (about 12 bytes per article), but should 
otherwise be fairly efficient.