Path: utzoo!attcan!utgpu!jarvis.csri.toronto.edu!rutgers!ucsd!orion.cf.uci.edu!uci-ics!nagel
From: nagel@ics.uci.edu (Mark Nagel)
Newsgroups: news.software.b
Subject: Re: inews for NNTP and C news (plus log report generator info)
Message-ID: <1989Sep1.212023.14592@paris.ics.uci.edu>
Date: 1 Sep 89 21:20:23 GMT
References: <1989Sep1.191449.1292@anise.acc.com>
Sender: news@paris.ics.uci.edu (Network News)
Organization: UC Irvine Department of ICS
Lines: 38

pst@anise.acc.com (Paul Traina) writes:

>I've noticed some problems with inews when using the POST command of NNTP
>with C news.  There are two flavors of inews,  one which handles postings,
>and one that closely resembles rnews (it doesn't do header processing).

>One of them does too much (tries to do local user lookups),  one of them not
>enough (won't create Message-ID's and other headers).

I had the same problems.  The way I fixed it was to add a flag to
the inews script called '-nosig.'  If used, the signature code is
ignored.  Other flags could be used to bypass other parts.  I hate
the idea of duplicating inews, one per purpose, since so much
important stuff happens in there that won't happen if the article
goes straight into rnews.  Whatever you do, don't pay attention to
the man page for inews that says you could probably just throw
formatted articles from posters (like Pnews) directly into
relaynews.  For one thing, relaynews simply barfs when it gets an
article in a moderated group if the inews preprocessing hasn't been
done.

On another note, I just finished a Perl script for generating a log
report slightly similar to those generated by B news report.awk.  I
made no changes to the logging behavior of C news, so the script
requires approximately 10 minutes to run on a medium loaded Sun
4/260.  That's because it has to retrieve information that wasn't
stored in the logs from the history file.  I used newshist to get
history lines, so I'm not sure how much faster the script can get
without duplicating newsgroup information in the log.

If anyone would like a copy of this script, please email me.  I'd
like to test it and get ideas for speeding it up and/or making it
more informative prior to posting it to the whole net.

--
Mark Nagel
UC Irvine, Department of Information and Computer Science
ARPA: nagel@ics.uci.edu         UUCP: ucbvax!ucivax!nagel