Re: Multi server fix?

Shalon Wood <[email protected]> writes:

> Doug McLaren <[email protected]> writes:
>> There's a lot of processing involved in that step -- grep it against
>> the $OPT_I (in ubh2 anyways) stuff, then remove the $OPT_X stuff, then
>> see if you have every part of a given item ... it's slow when a group
>> has 100,000+ articles.  And the only way I know of to really speed it
>> up would be to write that part in C, which really isn't practical and
>> may not help that much anyways (I've looked at the code, and don't see
>> anything obvious to do in perl anyways.)
>
> Some of that's done by ubhasm, right? And as I think about it, the
> $OPT_I and $OPT_X parts _could_ be done ahead of time for the values
> set in .ubhrc (you'd have to redo that step if the user specifies a
> value on the command line, of course).

Actually, after some more thought, how about:

1) after ubhasm, run a speculative filter with the I and X specified
   by .ubhrc, storing the results in a table.

2) If ubh doesn't specify I or X, use that table instead of the full
   table.

3) ubh saves its filtered list in another table. If it's run again
   (with the same batchfile, if specified) before ubhasm is run and
   without changing the I or X options, it uses the pre-made table
   instead of refiltering.

Make sense?

Shalon Wood

-- 
Check out Strange Love -- an ezine of science fiction, fantasy, and
paranormal erotica. Only $2 for more than 20,000 words of fiction!

                 http://strangelove.pele.cx


 
Yahoo! Groups Links

<*> To visit your group on the web, go to:
    http://groups.yahoo.com/group/ubh/

<*> To unsubscribe from this group, send an email to:
    [email protected]

<*> Your use of Yahoo! Groups is subject to:
    http://docs.yahoo.com/info/terms/