[Xcb] New libxcb and xcb-proto releases (code name: "Oh, shit")

Uli Schlachter psychon at znc.in
Sun Dec 8 06:34:57 PST 2013


On 08.12.2013 15:15, Julien Cristau wrote:
> On Mon, Nov 18, 2013 at 21:40:12 +0100, Uli Schlachter wrote:
>> On 18.11.2013 12:37, Arnaud Fontaine wrote:
>>> Thanks for the list. Just updating the steps after merging some patches
>>> and what we have been talking about lately.
>> [...]
>> I think I got all of this list either into patches on the list or into bugs on
>> bugzilla that block [0]. Just to start a new list, the patches are about the
>> soname bump, NEWS and xcb_ge_event_t.
>> Let's see what happens...
> Hi Uli,
> I'm afraid I lost track.  What's the blocker at this point?

Well, the bug and its dependencies had no real activity. So we have the tracker 
bug (71752 aka xcb-1.10 to make it easier to find). The dependencies in the bug 
tracker are:

https://bugs.freedesktop.org/show_bug.cgi?id=68387 Wrong code generated for XI2
https://bugs.freedesktop.org/show_bug.cgi?id=71502 broken API in xcb_ge_event_t
https://bugs.freedesktop.org/show_bug.cgi?id=71507 broken ABI in xkb
https://bugs.freedesktop.org/show_bug.cgi?id=72253 CMSG_FIRSTHDR() broken on mac

The first three bugs are old and AFAIK have patches laying around somewhere. The 
last one is new and has a patch, too. I only added it as a blocker to force 
someone else to decide "not a blocker!", but since it has a patch... ;-)

Then there is also patchwork:


I tried to keep this updated (mostly by marking patches as merged or 
superseded). I haven't followed closely, but from a quick look it seems like 
"[lib,5/5] c_client.py: No sizeofs for Events, Replies, Requests" breaks ABI/API 
(removing functions) and all the other ones aren't urgent.

So IMO the things in the bug tracker need to be handled and someone has to 
decide if the CMSG_FIRSTHDR change is needed. I decided to do as everyone else 
does and wait for someone else to do this (aka "a miracle").


More information about the Xcb mailing list