On Wed, 28 May 2003, Tom Meunier wrote: > I think I know what he means, because I've considered the same thing > myself, for a moment or two. > > He wants to parse his maillog file to throw away everything but the > "processing message <message_id> and the "identified spam" lines. Then > he can go back and look at <message_id> for troubleshooting purposes. > He has a feeling that he can't count on the "identified spam" line > always immediately following the "processing message" line. He'd like > to throw away everything but those two lines. Basically, I guess he's > asking for a report in the format: > > "Identified spam (29.6 points) in message <[EMAIL PROTECTED]>" > ... ... ... (more of same) > > Fair disclosure: Idunno how to do that.
I've considered doing something like this as well. To be honest I never did it either and I would have a lot of trouble getting it done since I'm not a programmer (hacker at best). However I know of some code that could be useful. http://www.analog.cx/helpers/maillog2commonlog The helper program "maillog2commonlog" on the website for Analog (HTTP stats generator FYI) has code that keys off of the Sendmail queue ID or MessageID, I forget which. I always planned on taking that code and attempting to write something that stuffed all the pieces of info into a database. I never figured out how to do it though. For the record I'm a MIMEDefang user so my log lines will be different that those written by spamd (at least I'd expect them to be). Justin ------------------------------------------------------- This SF.net email is sponsored by: ObjectStore. If flattening out C++ or Java code to make your application fit in a relational database is painful, don't do it! Check out ObjectStore. Now part of Progress Software. http://www.objectstore.net/sourceforge _______________________________________________ Spamassassin-talk mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/spamassassin-talk