Re: [Evolution-hackers] time zone problem: extending API in stable branch?

2008-06-13 Thread Patrick Ohly
 Can you just point us to the right patch, that extends API ? I seem to
 hit the wrong patch. I dont see any .h changes also. 

Sorry for the confusion. The patch I pointed to 
(http://bugzilla.gnome.org/attachment.cgi?id=112638) explains the required API 
version change, not the API extension itself. As Chenthill said, the new 
functions are in e-cal-check-timezones.h.

 Is there a way, we can work around for stable branch alone?

The functions contained in e-cal-check-timezones could be compiled as static 
functions into each calendar backend. It wouldn't be nice, but doable. Clients 
which also need them (like SyncEvolution) would have to do the same (but 
SyncEvolution already does that anyway in order to also work with older 
Evolution releases).

Bye, Patrick
-- 
Pt! Schon vom neuen GMX MultiMessenger gehört?
Der kann`s mit allen: http://www.gmx.net/de/go/multimessenger
___
Evolution-hackers mailing list
Evolution-hackers@gnome.org
http://mail.gnome.org/mailman/listinfo/evolution-hackers


Re: [Evolution-hackers] time zone problem: extending API in stable branch?

2008-06-12 Thread Srinivasa Ragavan
Patrick,

Can you just point us to the right patch, that extends API ? I seem to
hit the wrong patch. I dont see any .h changes also. 

Is there a way, we can work around for stable branch alone? We may need
release team approval, if we have to update API (I guess so, though EDS
isn't part of platform, we may have to atleast check with them)

-Srini.

On Thu, 2008-06-12 at 22:25 +0200, Patrick Ohly wrote:
 Hello,
 
 the patch that I am suggesting to fix the time zone issues in Evolution
 (see http://bugzilla.gnome.org/show_bug.cgi?id=528902) adds new
 functions to libecal. Note that the extended library is backwards
 compatible, i.e., this is not an API change which requires recompiling
 other programs (http://bugzilla.gnome.org/attachment.cgi?id=112638).
 
 Is this an acceptable solution for the 2.22.x stable branch? Chentill
 reviewed the patch and agreed to committing it on trunk after some minor
 changes (which I have done in the meantime), but he is concerned about
 the API and rather would like to keep this out of the stable release.
 
 I'm writing here to also get the opinion of Evolution packagers who
 might not monitor the discussion taking place in the bug tracker. Is
 this issue important enough to extend the API in the next 2.22.x
 release? If Evolution upstream doesn't include it in 2.22.x, are
 distributions going to apply the patch (it is written against the stable
 branch and applies cleanly) anyway?
 
 I personally think that this is important because Evolution will
 continue to use out-dated time zone definitions for new meetings in
 perpetuity unless people wipe out their old calendars or the patch is
 applied. Quoting Paul Smith, who missed a meeting this spring due to
 this bug:
 I have to add my voice to those saying please, please, PLEASE
 someone fix this complete and utter disaster!
 

___
Evolution-hackers mailing list
Evolution-hackers@gnome.org
http://mail.gnome.org/mailman/listinfo/evolution-hackers