Johan Dahlin [Fri, 19 Feb 2010 16:53:17 +0000 (14:53 -0200)]
[annotations] Add allow-none
This commit was created using a script that searched for all docstrings
containing a parameter and the string 'or %NULL'.
Gdk backends and demos excluded as they are not part of a public API
Javier Jardón [Tue, 16 Feb 2010 04:44:22 +0000 (05:44 +0100)]
[gdk/x11] Change the name of the functions in the header file
_gdk_keymap_add_virtual_modifiers() changed to
_gdk_keymap_add_virtual_modifiers_compat() in
commit 13d69e552dffe681ee80c0600366a8c04fc205c6
This patch change the header name too.
Stefan Kost [Thu, 11 Feb 2010 08:38:44 +0000 (10:38 +0200)]
widget: don't check before setting to avoid costly type checks
There are no side effect of setting these flags and we don't fire g_object
notifies also. Thus set/unset right away without costly check of current state.
Matthias Clasen [Fri, 12 Feb 2010 20:45:16 +0000 (15:45 -0500)]
Fix a compatibility problem
It turns out that my attempt at handling Super, Hyper and Meta better
is causing problems, mostly because Alt and Meta are commonly colocated
in the modmap, and apps do a check for the Alt modifier regularly.
Matthias Clasen [Tue, 9 Feb 2010 01:08:22 +0000 (20:08 -0500)]
Fix a refcounting issues in GtkPlug
When the foreign window already has a GdkWindow, we forget to take
an extra references, leading to crashes later on. Patch by
Karl Tomlinson, bug 607061.
Steffen Macke [Sat, 6 Feb 2010 18:04:30 +0000 (20:04 +0200)]
Fix menu bar problem in ms-windows theme engine on Windows 7
Patch from bug #591186. It won't have any effect in this branch as the
whole XP-and-later theming code is broken and disabled, but when/if
that is fixed, then this patch should help.
On X11 we receive enter notify and motion notify events for a window
regardless of its focus state. On Mac OS X this is not the case. This
commit improves the semantics to overcome this difference. It improves
on my earlier patch that sent a motion notify event when a window became
main.
Instead of sending a motion notify when a window becomes main, we now
send one when a window becomes key, which comes closest to a window
getting focus in X11. This motion notify is needed because Mac OS X does
not send motion events when an application is inactive (none of its
windows have focus), these events are sent in X11. This dummy motion
notify event (with current coordinates of the mouse cursor) allows an
application to get its prelight and other state right when it gets focus
and thus user attention.
Another change is to send an enter notify event when updating the
tracking rectangle of a GdkQuartView and the mouse cursor is currently in
this rectangle. This rectangle is at least updated on window creation.
This enter notify event is important for the case where a new window
appears right below the mouse cursor. The window has to receive an enter
notify event for the subsequent events to be processed correctly. Mac
OS X does not send one in this case, so we generate it ourselves.
Both of these synthesized events have to go through
_gdk_windowing_got_event() for updating statekeeping, etc.
append_event() has a boolean flag now to make this convenient.
Michael Natterer [Fri, 22 Jan 2010 15:52:56 +0000 (16:52 +0100)]
Bug 607082 - Add accessors for sealed GtkRange members
Add accessors for range->range_rect, range->slider_range,
range->slider_size_fixed and range->min_slider_size. Didn't add
properties for any of them because thir purpose is mostly to enable
proper subclassing.
Claudio Saavedra [Fri, 22 Jan 2010 13:46:46 +0000 (15:46 +0200)]
GtkTreeView: plug a leaking GtkTreeRowReference
This is exposed by overshooting in a HildonPannableArea. Leaking
a GtkTreeRowReference also means leaking the whole model, since
the GtkTreeModel reference count is increased for each row
reference.
Fixes bug #607770 - Leak in GtkTreeView with HildonPannableArea
If e.g. the right edge of the leftmost rectangle is near MIN_INT, and
the left edge of the rightmost rectangle is large then subtracting these
can lead to an integer overflow, making the resultant "width" falsely
positive, thus returning a very wide result instead of the expected
no-intersection result.
We avoid the overflow by not doing the subtraction unless we know the
result will be positive. There are still risks for overflow if x + width
or y + width is larger than MAXINT, but we won't ever overflow for valid
rects now.