updates to issues.guix.gnu.org

updates to issues.guix.gnu.org

[Ricardo Wurmus]

Hi Guix,

I’ve just updated mumi on issues.guix.gnu.org.  The new version is quite
a bit faster and features a more accurate search.

This is accomplished by storing essential bug information (such as
submitter, owner, status, and tags) locally in a sqlite database in
addition to indexing all original emails that are downloaded from
Debbugs.  The search no longer issues a complicated request to Debbugs.
Instead it is performed on the local message index, and any filtering
based on information not contained in the email messages (like
submitter, owner, status, and tags) is perfomed using the sqlite
database.

This is not perfect and a few slow Debbugs operations still remain, but
in general issues.guix.gnu.org should be much more usable now.

The code is here and I appreciate patches to improve it:

    https://git.elephly.net/software/mumi.git

--
Ricardo

Re: updates to issues.guix.gnu.org

[Pierre Neidhardt]

[-- Attachment #1: Type: text/plain, Size: 84 bytes --]

It is indeed, thank you very much!

-- 
Pierre Neidhardt
https://ambrevar.xyz/

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 487 bytes --]

Re: updates to issues.guix.gnu.org

[Ludovic Courtès]

Hi,

Ricardo Wurmus <rekado@elephly.net> skribis:

> I’ve just updated mumi on issues.guix.gnu.org.  The new version is quite
> a bit faster and features a more accurate search.

Very nice, thank you!

The “Recent Activity” thing on the front page displays much faster and
also looks more relevant.  Thumbs up!

Ludo’.

Re: updates to issues.guix.gnu.org

[Ricardo Wurmus]

Ludovic Courtès <ludo@gnu.org> writes:

> Hi,
>
> Ricardo Wurmus <rekado@elephly.net> skribis:
>
>> I’ve just updated mumi on issues.guix.gnu.org.  The new version is quite
>> a bit faster and features a more accurate search.
>
> Very nice, thank you!
>
> The “Recent Activity” thing on the front page displays much faster and
> also looks more relevant.  Thumbs up!

Yeah, previously this was all using Debbugs search queries, which are
terribly slow and which needed a lot of filtering.  The requested
maximum number of items returned from Debbugs needed to be reduced to
get a response at all, and that response set was highly redundant.
The response would usually not include all matching messages, so the
result would be unpredictable and incomplete.

--
Ricardo