Gnus development mailing list
 help / color / mirror / Atom feed
From: lee <lee@yun.yagibdah.de>
To: ding@gnus.org
Subject: Re: automatic group compaction?
Date: Tue, 04 Sep 2012 13:32:30 +0200	[thread overview]
Message-ID: <87wr0am3ap.fsf@yun.yagibdah.de> (raw)
In-Reply-To: <muxoblm9jld.fsf@xemacs.org> (Didier Verna's message of "Tue, 04 Sep 2012 12:18:54 +0200")

Didier Verna <didier@xemacs.org> writes:

> lee <lee@yun.yagibdah.de> wrote:
>
>> is there something like automatic group compaction?
>
>   Depends on what you mean by compaction. I wrote the compaction feature
> for nnml a long time ago (e.g. 'z' in the server buffer). That means
> re-assigning article numbers to avoid gaps.

Yes, that's the feature I'm referring to, i. e. 'G z' to compact a
particular group.

What's behind it is that I have a group "mail.0-incoming" for new mail
that isn't being split into mailing lists etc., and I have set
gnus-large-newsgroup to 2000. Now the article numbers of
"mail.0-incoming" had gone up to over 4000 over time with about only 10
mails actually remaining in that group.

So every time I entered the group, gnus was asking me how many articles
to fetch with a default of 2000 because it's configured to "(display
. [not expire])" for this group and some others. Accepting the default
resulted in not displaying mail that was older than "some time" --- I
guess because the article numbers were too low.

To circumvent the problem, I'm using an entry in gnus-parameters that
sets gnus-large-newsgroup to 50000 for "mail.0-incoming". The
alternative/addition is to compact "mail.0-incoming" from time to time.

However, this is rather awkward. Over time, other groups will be
affected the same way as well, and it doesn't make sense to set
gnus-large-newsgroup to some high number for all or for many groups to
circumvent the problem. It also doesn't make sense to manually go
through some or many groups and compact them, and compacting the whole
server with about 300000 mails in nnml storage is very reasonable,
either.

All the mails are bzipped, and I'm not sure what effect on performance
that has for compacting --- it seemed to me that the mails in
"mail.0-incoming" were bunzipped and bzipped when I compacted that
group. Compacting 300k mails can take a while.

Also, I sometimes delete SPAM from groups that hold mailing lists, and
it would make sense to compact such groups like once a year or so.

So the idea is to have something like an entry in gnus-parameters like
"(gnus-autocompact 90d)", which would compact the group in question
automatically every 90 days: Plain and simple.

If there are users who care about the actual number of articles
displayed in the group buffer or have a similar scenario, an
autocompacting feature would be rather useful for many.


-- 
http://www.asciiribbon.org/
http://tools.ietf.org/html/rfc1855
http://www.caliburn.nl/topposting.html



  reply	other threads:[~2012-09-04 11:32 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-09-04  9:40 lee
2012-09-04 10:18 ` Didier Verna
2012-09-04 11:32   ` lee [this message]
2012-09-04 18:31     ` Lars Ingebrigtsen
2012-09-04 23:02       ` lee
2012-09-05 11:08         ` Lars Ingebrigtsen

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=87wr0am3ap.fsf@yun.yagibdah.de \
    --to=lee@yun.yagibdah.de \
    --cc=ding@gnus.org \
    /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).