[Pidgin] #5302: Pidgin uses a number of deprecated gtk interfaces

Pidgin trac at pidgin.im
Mon Mar 24 17:31:35 EDT 2008


#5302: Pidgin uses a number of deprecated gtk interfaces
---------------------------+------------------------------------------------
  Reporter:  stewb         |       Owner:            
      Type:  enhancement   |      Status:  new       
  Priority:  minor         |   Milestone:            
 Component:  pidgin (gtk)  |     Version:  2.4.0     
Resolution:                |    Keywords:  deprecated
   Pending:  1             |  
---------------------------+------------------------------------------------
Changes (by stewb):

  * pending:  1 => 0

Comment:

 res_query isn't in the LSB.  I realize that's an ugly hack, and didn't
 bother to mention that in the bug report as I don't have an alternative at
 the moment. The big piece of the patch is all the gtk_item_factory stuff.

 G_BEGIN_DECLS/G_END_DECLS do not seem to be defined in LSB. I'm not sure
 why.
 Seeing as these macros are being used outside of GTK itself. I think LSB
 should probably reconsider adding these macros. I've opened a bug at our
 end for them:

 http://bugs.linuxbase.org/show_bug.cgi?id=2008

 Some of the includes changes are due to the way the LSB headers are built
 up. Things that would be in discreet headers in the native headers end up
 in gtk.h.  I wasn't really proposing you make these changes, unless we
 were to come up with some LSB_MODE conditional that would be acceptable to
 you. Otherwise I'll just need to keep patching these out at our end. This
 (the monolithic gtk.h) was a decision made by the team that did the
 initial LSB GTK work.

 LSB is based on gtk-2.6.2, and we're looking at uplifting to 2.8 for LSB
 4.0.
 Any interfaces which were considered deprecated in gtk-2.6.2 were not
 included.

 I would love to keep the N_() macro, but I was having trouble getting it
 to work with the new menu structure. Any hints how to fix this would be
 appreciated.

 I'm not sure what's going on with the pidgin_xfer_dialog_show, I know my
 test build was segfaulting trying show the xfer dialog that had not yet
 been created before I made the change.  I'm sure this is probably fallout
 from other bits of the patch, but I wasn't finding where dialog is
 initialized to NULL, and the test was falling through with a segfault
 trying to display a dialog that did not exist on the first pass.

 I'll look at the unused GLists and look into monotone.

 Thanks

-- 
Ticket URL: <http://developer.pidgin.im/ticket/5302#comment:3>
Pidgin <http://pidgin.im>
Pidgin


More information about the Tracker mailing list