Open Patches

[John Bailey]

John Bailey wrote:
> I've been slowly trying to make my way through the open tickets in Trac, and
> I've found something that surprised me--we have *93* open patches.

As a status update, we now have 82 open patches.  This number is slightly skewed
because some patches have been opened since my last message on this topic.
However, these patches now break down like so:

Status:
  * Pending - 3
  * New - 79

Component:
  * unclassified - 2
  * AIM - 2
  * Finch - 1
  * Gadu-Gadu - 3
  * ICQ - 8
  * libpurple - 22
  * MSN - 3
  * pidgin - 25
  * plugins - 1
  * QQ - 1
  * SIMPLE - 2
  * WinPidgin - 2
  * XMPP - 8
  * Yahoo! - 2

Milestone (incomplete listing):
  * 3.0.0 - 3
  * 2.6.0 - 1
  * Patches Needing Improvement - 19
  * No Milestone - 57

This of course represents a marked improvement from the numbers from just 17.5
days ago.  A few notes I have though:

  * Some of the XMPP patches may be closable, as they relate to work Marcus has
committed to branches under im.pidgin.cpw.malu.*

  * I'm thinking at least one of the AIM patches should be rejected outright, as
it introduces over-the-wire behavior that the official AIM client doesn't have.

  * There is overlap in the root problem (encoding) in several of the ICQ
patches, but each of the patches seem to focus on different areas of the OSCAR
code.  It seems that it might make sense to figure out how to fix the issues
without hacking in encoding-related crap in multiple places.

  * One of the open MSN patches, #3319, is for the "Group Chat" support that
appeared where you add a bot to your list and the bot relays messages to
everyone who has that bot on their buddy lists.  These bots identify the person
who really sent the message by using the "P4-Context" header in the message.
The current prevailing developer opinion on the ticket seems to be that the
implementation is a dirty hack.  Obviously users don't care, but do we really
want this in libpurple?  I think we need either a definite "not until this can
be done without hackery" (set milestone to Patches Needing Improvement) or
"never" (set type to rejected_patch or just close as wontfix) on this patch.

Again, I'd really like to see more of these patches being properly reviewed and
committed or rejected as applicable for 2.5.3 and 2.6.0.  Specifically, I'd love
to see attention given to all the ones that aren't on the 3.0.0 or Patches
Needing Improvement milestone.

John

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 197 bytes
Desc: OpenPGP digital signature
URL: <http://pidgin.im/pipermail/devel/attachments/20081130/f0fba1fc/attachment.sig>