My apologies for not searching the archives very thoroughly. I tried a few basic searches, but nothing seemed to leap out at me. If this problem has already been fixed, I'm glad to hear it, and please disregard this message. On with the bug report: I was using Graham Barr's Perl 5 module Date::Parse (a part of his TimeDate distribution). I was going through about 200MB of old mailing list messages, trying to file them by month. I found some really fascinating "Date:" lines, among them: | Date: Wed, 31 Dec 1997 21:10:28 +875400 | X-Mailer: Mutt 0.88e Looking further, I found another one: | Date: Thu, 31 Dec 1998 16:26:33 +875200 | X-Mailer: Mutt 0.91.1 The fact that they were both Mutt is the reason that I'm sending this bug report. As for the Subject: line... Here's what I sent to Graham: > See the pattern yet? And for my next trick: > > 2400*365 = 876000 > > They're trying to compute the offset by taking the gmtime and > comparing it to localtime, without accounting for the year. I'm > submitting a bug report to the Mutt site, since at least the first one > is Mutt ... yes, the other one is too: > > X-Mailer: Mutt 0.88e > > That's silly. Lasted at least a few versions, although they're up to > 0.95.something. Apologies for using the word "silly"; I meant it purely in a fun way, as we had seen some truly bizarre timezone / timestamp notations here. Although it's not an internet standard, there *is* an ISO standard for how to write timestamps, and I encourage people to use it wherever applicable: http://www.cl.cam.ac.uk/~mgk25/iso-time.html Thanks for your time. If you could keep me (<[EMAIL PROTECTED]>) on the CC list, I would appreciate it, as I'm not on the Mutt mailing list. Happy Hacking, T. p.s. Hopefully it's obvious, but while you're fixing this little buglet, you will have to make sure you take leap years into account! :)
