* Will Yardley ([EMAIL PROTECTED]) wrote: > you can do the exact same thing with a simple procmail recipe. it > can be argued (and as the gnus manual points out) that this might be > unreliable. i use it anyway since i get loads of dupes. you can save > them to a separate folder rather than deleting them outright if you're > paranoid.
I had this and quickly backed it out, because it deletes the message to the list, not to me, as that's what I usually get first. Because of how I filter ml's (and changing this would increase the ruleset complexity *truckloads*), this results in the crap still ending up in my Inbox, *and* the message in the ml folder being nuked, which is even more annoying. Unfortunately, short of shooting anyone who ignores Mail-Followup-To, I don't think there's a workable solution. About the best way I can think of is on every list delivery, scan Inbox for the message id (and same/similar content if I'm feeling paranoid) and if it's found, nuke it. That's rather expensive though :) Nuking [crap] in ml subjects is another thing; I could change it at procmail level, but I'm not too keen on munging incoming mails like that. I'd really rather mutt apply a subject_munge_regex to it or so to the message list when it's displayed. Since a regex is already applied for subject threading, maybe that's not too much to ask for... ? It's especially needed for lists like ruby-talk, where each subject line is munged horrendously with [ruby-talk:12345], which, of course, aside from wasting tonnes of space, also makes mutt re-display the subject for every mail (unless I make re_regex even more complex, which still doesn't fix it when a subject is displayed. aagh). -- Thomas 'Freaky' Hurst - [EMAIL PROTECTED] - http://www.aagh.net/
