Archive of RubyForge sup-talk mailing list
 help / color / mirror / Atom feed
From: Olly Betts <olly@survex.com>
To: sup-talk@rubyforge.org
Subject: Re: [sup-talk] [PATCH] xapian: do less work for update_message_state
Date: Thu, 4 Feb 2010 03:42:09 +0000 (UTC)	[thread overview]
Message-ID: <loom.20100204T042813-620@post.gmane.org> (raw)
In-Reply-To: <slrnhcbck3.5ah.olly@msgid.survex.com>

Olly Betts writes:
> On 2009-10-01, Rich Lane <rlane@club.cc.cmu.edu> wrote:
> > Excerpts from William Morgan's message of Thu Oct 01 09:46:20 -0400 2009:
> >> Reformatted excerpts from Rich Lane's message of 2009-09-30:
> >> > They're about 3 times faster on my machine with this patch. An
> >> > optimization the Xapian devs have been planning to make (and that this
> >> > patch is necessary to take advantage of) should increase performance
> >> > much more.
> >> 
> >> Awesome. Out of curiousity, what's the optimization?
> >
> > replace_document currently deletes all the old postings and inserts new
> > ones. It can be optimized to make the minimal set of modifications.
> 
> This is the ticket for it:
> 
> http://trac.xapian.org/ticket/250

This has now been fixed in Xapian SVN trunk, and I've backported the
patches to Xapian's 1.0 branch.  You can find snapshot tarballs including
these changes here:

http://oligarchy.co.uk/xapian/branches/1.0/

Xapian's testsuite passes (including the additional test coverage which I
also backported), and I looked over each change carefully, but I would be
interested to see some real world testing, particularly in the situation
which these changes are intended to improve (i.e. speed of adding/removing
tag terms from existing documents).  I'm not expecting problems, but this
is a larger change than we usually backport, so wider testing would be a
good thing.

If you are using Xapian 1.0's default backend (flint), you'll need this
patch (which was merged before sup 0.10) to get full benefit from the
changes:

http://article.gmane.org/gmane.mail.sup.devel/177

I've uploaded a snapshot of 1.0 with this patch to Debian experimental.
It hasn't built for all Debian architectures yet, but is available for at
least amd64 and x86, which are probably the most popular two.

If you aren't sure how to pull in packages from experimental, see:

http://wiki.debian.org/DebianExperimental

I've also put it in a Launchpad PPA for all currently supported Ubuntu
releases, and it has built for all of them already:

https://launchpad.net/~ojwb/+archive/experimental/

Cheers,
    Olly

_______________________________________________
sup-talk mailing list
sup-talk@rubyforge.org
http://rubyforge.org/mailman/listinfo/sup-talk


  reply	other threads:[~2010-02-04  3:58 UTC|newest]

Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2009-09-13 18:44 Rich Lane
2009-09-30 19:40 ` William Morgan
2009-09-30 20:16   ` Rich Lane
2009-10-01 13:46     ` William Morgan
2009-10-01 17:02       ` Rich Lane
2009-10-02  7:57         ` Olly Betts
2010-02-04  3:42           ` Olly Betts [this message]
2010-02-18  2:27             ` Olly Betts

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=loom.20100204T042813-620@post.gmane.org \
    --to=olly@survex.com \
    --cc=sup-talk@rubyforge.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