Gnus development mailing list
 help / color / mirror / Atom feed
From: Kevin Greiner <kgreiner@xpediantsolutions.com>
Subject: Re: Agent: delete agentized message?
Date: Wed, 04 Jun 2003 00:51:18 -0500	[thread overview]
Message-ID: <u3ciqgzl5.fsf@xpediantsolutions.com> (raw)
In-Reply-To: <84smqreey0.fsf@lucy.is.informatik.uni-duisburg.de>

kai.grossjohann@gmx.net (Kai Großjohann) writes:

> My boss has his nnimap:INBOX under agent control.  There are some
> very old articles known to the agent, but they are not on the server
> anymore.  Hitting `B DEL' on them says `cannot delete article 1054'.
>
> Is it normal for Gnus to behave like that?

It would seem that the behavior is entirely up to the backend.  The
command 'B DEL' is bound to gnus-summary-delete-article. It then calls
gnus-request-expire-articles to perform the actual work.  g-r-e-a
passes the request to the request-expire-articles method of the
backend.  That method performs the request then returns a list of
articles that could not be deleted by the backend.  Finally, that list
is used by g-r-e-a to compute the articles that should be removed from
the agent.

From your description, the backend is apparently failing to
distinguish between an error caused by an article that is no longer on
the server and an error caused by, for example, being unable to
connect to the server.

> I did M-x gnus-agent-regenerate-group RET, but that didn't help.  (It
> did move all those old articles from ancient status to unread status,
> though.)

How did you respond to the 'Reread?' prompt?  If you evaluate
(gnus-agent-regenerate-group "group.name" t), you'll get the behavior
that you just described.  On the other hand, if you evaluate
(gnus-agent-regenerate-group "group.name"), you should NOT see any
change in an article's status.

> I then marked some of the old spam as expirable and did M-x
> gnus-agent-expire-group RET.  That did the job.

Good.

> WIBNI the Agent would delete its copy of a message when people do B
> DEL on an article that isn't on the server?

Well, it would seem that that decision is controlled by the backend.
I took a look at nnimap-request-expire-articles and I'm not going to
be able to take this much further.  Someone comfortable with the
nnimap backend will need to see if the 'cannot delete article BECAUSE
IT DOESN'T EXIST ON THE SERVER' error can be handled differently than
every other error.

Kevin




      parent reply	other threads:[~2003-06-04  5:51 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2003-06-03  8:35 Kai Großjohann
2003-06-03  9:39 ` Niklas Morberg
2003-06-03 13:34   ` Kai Großjohann
2003-06-04  5:51 ` Kevin Greiner [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=u3ciqgzl5.fsf@xpediantsolutions.com \
    --to=kgreiner@xpediantsolutions.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).