On Wed, Jan 05, 2005 at 05:39:20PM -0800, Bill Moseley wrote:
> On Wed, Jan 05, 2005 at 05:33:26PM -0800, Scott Hepler wrote:
> > No, I was just wanting to log requests in their own file.  I was looking
> > for info in the archive and happened to find this post, and didn't want to
> > patch each new version if the 'stock' code had the capability.
> 
> The web logs have the requests, of course.  But if you want to log
> separately then there is a callback called "request_fixup" you can
> set.  It passes the request object (which is a CGI.pm object, IIRC)
> and the configuration hash.  So you could use that for logging.

'request_fixup' sounds great -- I'll give it a try.  My main problem with
parsing the Apache logs is determining what's been done already, especially
in the case of failed runs.

> I think that message you quoted was about also reporting on the
> outcome of the search, too.  So that would mean you would need another
> hook later on in the script.  That doesn't exist.

Yes, that message was about logging the search result, which I don't think
is necessary in this case.

Thanks again,
-s.

-- 
Scott Hepler
System Administration and Security
Web Development
802.457.3796