From mboxrd@z Thu Jan 1 00:00:00 1970 X-Msuck: nntp://news.gmane.io/gmane.emacs.gnus.general/50588 Path: main.gmane.org!not-for-mail From: Simon Josefsson Newsgroups: gmane.emacs.gnus.general Subject: Re: group splitting question... Date: Wed, 05 Mar 2003 22:41:18 +0100 Sender: owner-ding@hpc.uh.edu Message-ID: References: <8765qxg1qz.fsf@sheffield.ac.uk> <87llztbnzb.fsf@sheffield.ac.uk> NNTP-Posting-Host: main.gmane.org Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Trace: main.gmane.org 1046900562 16170 80.91.224.249 (5 Mar 2003 21:42:42 GMT) X-Complaints-To: usenet@main.gmane.org NNTP-Posting-Date: Wed, 5 Mar 2003 21:42:42 +0000 (UTC) Cc: ding@hpc.uh.edu Original-X-From: owner-ding@hpc.uh.edu Wed Mar 05 22:42:40 2003 Return-path: Original-Received: from malifon.math.uh.edu ([129.7.128.13]) by main.gmane.org with esmtp (Exim 3.35 #1 (Debian)) id 18qgeu-0004CY-00 for ; Wed, 05 Mar 2003 22:42:40 +0100 Original-Received: from sina.hpc.uh.edu ([129.7.128.10] ident=lists) by malifon.math.uh.edu with esmtp (Exim 3.20 #1) id 18qgdv-00063H-00; Wed, 05 Mar 2003 15:41:39 -0600 Original-Received: by sina.hpc.uh.edu (TLB v0.09a (1.20 tibbs 1996/10/09 22:03:07)); Wed, 05 Mar 2003 15:42:39 -0600 (CST) Original-Received: from yxa.extundo.com (178.230.13.217.in-addr.dgcsystems.net [217.13.230.178]) by sina.hpc.uh.edu (8.9.3/8.9.3) with ESMTP id PAA04323 for ; Wed, 5 Mar 2003 15:42:26 -0600 (CST) Original-Received: from latte.josefsson.org (yxa.extundo.com [217.13.230.178]) (authenticated bits=0) by yxa.extundo.com (8.12.8/8.12.8) with ESMTP id h25LfIdD028070 (version=TLSv1/SSLv3 cipher=EDH-RSA-DES-CBC3-SHA bits=168 verify=OK); Wed, 5 Mar 2003 22:41:19 +0100 Original-To: Eric E Moore Mail-Copies-To: nobody X-Payment: hashcash 1.1 0:030305:e.e.moore@sheffield.ac.uk:87da7fa0ded82a9f X-Hashcash: 0:030305:e.e.moore@sheffield.ac.uk:87da7fa0ded82a9f X-Payment: hashcash 1.1 0:030305:ding@hpc.uh.edu:c30cb24dae49f43b X-Hashcash: 0:030305:ding@hpc.uh.edu:c30cb24dae49f43b User-Agent: Gnus/5.090016 (Oort Gnus v0.16) Emacs/21.3.50 (gnu/linux) Precedence: list X-Majordomo: 1.94.jlt7 Xref: main.gmane.org gmane.emacs.gnus.general:50588 X-Report-Spam: http://spam.gmane.org/gmane.emacs.gnus.general:50588 Eric E Moore writes: > Simon Josefsson writes: > >> Eric E Moore writes: >> >>> Why does splitting on group parameters remove the backend name? I >>> found this a little confusing, when I set up parameters to try and >>> migrate from nnml to nnimap (to make reading email from different >>> locations a bit easier), and instead of putting my mail in the nnimap >>> group that it was supposed to match, it put it in a new identically >>> named nnml group.... >> >> You can't split across backends, so the backend name wouldn't serve >> any purpose. > > Why not? The feature isn't implemented. >> What exactly is your splitting configuration? > > I have 2 machines. "work" and "home". I would like to be able, at > least temporarially, to get gnus to file emails read (from the mail > spool) on "work" to be filed in IMAP groups on an imap server running > on "home". If I understand correctly, I don't think there is any pre-defined functionality for doing what you want -- mail cannot be split into nnimap. Nnimap can only split mail from one mailbox on the server into other mailboxes on the same server. > I have nnimap pointing to home in secondary selects (and nnml in > primary). I have an nnimap group "INBOX.foo" with a split-regexp > group property that catches messages to a certain email address. > When I read mail from the spool at work, instead of being put in the > group with the property (e.g. nnimap+home:INBOX.foo), they're put in > an nnml group with the same name (nnml:INBOX.foo), which in addition > to being not what I want, is wrong to boot (nnml:INBOX.foo does not > have that split-spec as a group property). Yes. Sadly, this is probably how it is intended to work. Only the group name is used for splitting, and those names are assumed to apply to whatever backend that eventually receives mail. >> Nnimap and nnmail doesn't share splitting code, except for the fancy >> nnimap splitting variable that defaults to the nnmail spitting >> variable contents. > > but the group splitting code in gnus-group-split-fancy searches over > all groups, in all backends, to find groups that the message matches, > but then strips off the backend information. Surely it should either: > > a) only scan groups it can actually deliver to by default > or > b) be able to deliver to the groups it scans That function is hard coded for nnmail splitting, but nnimap's splitter isn't based on nnmail. That function doesn't really do any splitting, simplified, it seems to only iterates over all group and configures up nnmail-split-methods and then let the normal split code take over. To solve your problem, I think it is easiest to simply have all mail moved into a nnimap group and then use the nnimap mail splitter to split the mail the way you want. You need to write a macro or something that takes mail from your local spool (possibly via a nnml group) and move it into the nnimap group.