After an evaluation, GNOME has moved from Bugzilla to GitLab. Learn more about GitLab.
No new issues can be reported in GNOME Bugzilla anymore.
To report an issue in a GNOME project, go to GNOME GitLab.
Do not go to GNOME Gitlab for: Bluefish, Doxygen, GnuCash, GStreamer, java-gnome, LDTP, NetworkManager, Tomboy.
Bug 629471 - make size_request and size_allocate just vfuncs, not signals
make size_request and size_allocate just vfuncs, not signals
Status: RESOLVED WONTFIX
Product: gtk+
Classification: Platform
Component: .General
3.5.x
Other All
: Normal normal
: 4.0
Assigned To: gtk-bugs
gtk-bugs
Depends on: 633324
Blocks:
 
 
Reported: 2010-09-13 01:08 UTC by Havoc Pennington
Modified: 2014-05-22 00:25 UTC
See Also:
GNOME target: ---
GNOME version: ---


Description Havoc Pennington 2010-09-13 01:08:49 UTC 
The methods on GtkSizeRequest are not signals, so having the deprecated size_request be a signal is kind of bloat.

It's easier to imagine a reason to connect to size_allocate, but a notify::allocation (i.e. make allocation a property) would probably be a better alternative. Right now there's no notify::allocation but there likely should be.
Comment 1 Matthias Clasen 2010-09-13 18:05:19 UTC 
There's a few places where we connect to 'size-request' inside gtk itself; those would have to be fixed:

[mclasen@planemask gtk+]$ grep "size-request" gtk/*.c  | grep connect
gtk/gtkpathbar.c:    g_signal_connect (label_alignment, "size-request",
gtk/gtktoolitemgroup.c:  g_signal_connect_after (alignment, "size-request",
Comment 2 Havoc Pennington 2010-09-13 18:17:59 UTC 
Looks like both of these could be replaced with a call to the proposed set_natural_size(), run from inside style_set().
Well, assuming the flavor of set_natural_size() was "increase only"
Comment 3 Javier Jardón (IRC: jjardon) 2010-10-29 15:27:10 UTC 
size-request will be removed completely in GTK+3.0 . See bug #633324
Comment 4 Javier Jardón (IRC: jjardon) 2011-01-04 13:54:15 UTC 
size-request signal was completely removed in GTK+3. Can we close this?
Comment 5 Tristan Van Berkom 2011-03-01 00:47:39 UTC 
Yes I think we should close this.

The original bug was concerning size-request and that's gone.

Now we've released 3.0 so it doesnt make sense to break api
to remove the size-allocate signal.