[Fwd: pidgin: c0a4c265: As darkrain pointed out this isnt caused...]

[Paul Aurich]

Does anyone else have preferences on how purple_normalize() should behave
if it's passed a NULL string?

I personally prefer crashy behavior (perhaps an explicit g_assert?) because
it tends to clearly reveal bugs (in the case of #10115, the gevolution
plugin passing a NULL pointer as the PurpleBuddy* argument of
purple_account_add_buddy).

~Paul

-------- Original Message --------
Subject: pidgin: c0a4c265: As darkrain pointed out this isnt caused...
Date: Fri, 28 Aug 2009 16:55:21 -0400 (EDT)
From: sulabh at soc.pidgin.im
Reply-To: devel at pidgin.im
To: commits at pidgin.im

-----------------------------------------------------------------
Revision: c0a4c2650c0117a4b956199f56d5b990e0c3fb0d
Ancestor: 2c2401820569b102feeb7ce574b90abdc6406f99
Author: sulabh at soc.pidgin.im
Date: 2009-08-28T20:51:48
Branch: im.pidgin.pidgin
URL:
http://d.pidgin.im/viewmtn/revision/info/c0a4c2650c0117a4b956199f56d5b990e0c3fb0d

Modified files:
        libpurple/util.c

ChangeLog:

As darkrain pointed out this isnt caused by that certain bug in glib. My
mistake. Recomment, and change how I test str to be NULL and return NULL.


-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: Attached Message Part
URL: <http://pidgin.im/pipermail/devel/attachments/20090828/92097b05/attachment.ksh>
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: Attached Message Part
URL: <http://pidgin.im/pipermail/devel/attachments/20090828/92097b05/attachment-0001.ksh>