From mboxrd@z Thu Jan 1 00:00:00 1970 From: marianne.promberger+sup-talk@gmail.com (Marianne) Date: Mon, 29 Dec 2008 18:27:19 +0100 Subject: [sup-talk] a few sup newbie questions In-Reply-To: <1229552235-sup-1046@entry> References: <1229548441-sup-531@audrey> <1229552235-sup-1046@entry> Message-ID: <1230571384-sup-9910@audrey> Hi, Thanks for the quick reply; sorry for my slow one. Excerpts from William Morgan's message of Wed Dec 17 23:46:15 +0100 2008: > Reformatted excerpts from marianne.promberger+sup-talk's message of 2008-12-17: > > (1) Automatically apply labels to incoming mail defined on search > > patterns? > > Stuff like that is handled by Sup's hook system. Thanks! I found more examples on the wiki. Is there any more documentation on the hook system? I've found /var/lib/gems/1.8/doc/sup-0.6/rdoc/index.html but I have trouble getting anything out of it. Sorry if this is obvious to others, but I am not a programmer and haven't used Ruby before. I cobbled two together, but I'm less than sure that they're correct; they seem to work for me. If someone could give me the go-ahead, I'd post them to the wiki, since I think it would be nice to have some more recipes for others to copy. This one assigns labels according to X-Label header. I'd like to do this since I already add some labels with procmail before mail goes to the maildir that is a source for sup. (Not sure the regex matching the label is exhaustive enough for everyone; it is for me). if message.raw_header =~ /X-Label: / xlabelheader = message.raw_header[/X-Label:.*/] xlabelheader.scan(/ [a-z0-9\-_+]+/) { |x| message.add_label x.lstrip } end The second one assigns labels based on an external file containing a list of e-mail addresses. The external file is one I maintain from mutt; I have a macro set up to quickly add the sender of a message to my "private" group. privatfile = File.open("/home/mpromber/.mutt/privataddr","r") if ! privatfile.grep(/#{message.from.email}/).empty? message.add_label :privat end Another question: If I add a new rule like this to the before-add-message hook, what is the recommended way to get sup to rescan messages to apply this rule to all existing mail? I am currently using "sup-sync -a --discard -e -x -v", but this takes forever (and I can't use regular "sup" in the meantime). (I'm using the -e because "sup-sync" on its own marks all mail as unread, even just running "sup-sync -c".) Is it correct that there is no faster way? > > (2) Save search history across sup sessions, so that I can use > > up-arrow to access terms I searched for last time I read mail in sup? > > Not implemented, but a good idea, and probably pretty easy! I've been thinking about this some more, and maybe even better would be not to save all searches, but to have the possibility to explicitly save some searches. This would turn the saved searches into a list of quickly accessible virtual mail folders. Maybe a shortcut to do this could be added to the buffer-list-mode? To start, it could even just be an external text file that has to be edited manually. Sorry I can't write patches in ruby :) > You can search for things like "from:me" or "to:me". That should work > with :alternates, but unfortunately doesn't work if you use :regexen. Thanks. Another question: Is there a search term like "from:" and "to:" but that stands for "anywhere in "from" or "to" or "cc" (like "a:" in mairix)? In general, is there more documentation about the search capacities, especially on building search terms, and where could I find it? One more question: Is there some way to have more fine-grained control over what the "From" e-mail address is when replying? As far as I understand it, I can have ":accounts:" sections in config.yaml, with different "alternates". Sup will reply from the e-mail address that someone uses as "to:" when e-mailing me. But for example, I'd like to use my "myname+sup-talk at gmail.com" address when replying to this list, but obviously, mail to this list has "sup-talk..." as the "to:" address. And the point of the "accounts" in config.yaml is just picking the right signature, right? I haven't discovered any other effect so far. Thanks, Marianne