| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
| |
For the QFontMetrics test, the different way of computing the with of a
text returs different value (1 to 2 pixel difference).
So we need to make sure to use the same way to compute the with that the
one used to ellide.
|
|
|
|
|
|
|
| |
The 'color' attribute is now explicitly parsed for gradient nodes.
Task-number: 260921
Reviewed-by: Trond
|
|\ |
|
| |\ |
|
| | |
| | |
| | |
| | |
| | | |
On embedded, the size of the screen is too small to leave enough space
for the dock widgets.
|
| | |
| | |
| | |
| | | |
Reviewed-by:jan-arve
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
The QTestLib XML output system throws exceptions when the system is
out of memory, which is normally quite reasonable. However when it is
used to report warnings during a catch block, this terminates the
program. So this change temporarily disables allocation failures while
the warning is being recorded.
Reviewed-by: Jason Barron
|
| | |
| | |
| | |
| | | |
Reviewed-by: Jeremy Katz
|
| | |\ |
|
| | |\ \ |
|
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
QGLPaintDevice::format() was returning the context's format, not
the format of the window surface's FBO. This caused the OpenGL1
paint engine to think that the window didn't have depth and
stencil buffers, even though the FBO most certainly did.
This change makes QGLPaintDevice::format() virtual and overrides it
in QGLFBOGLPaintDevice to return an updated format that includes
the context parameters plus the extra features that the FBO supports.
Reviewed-by: Tom Cooksey
|
| | |_|/
| |/| |
| | | |
| | | |
| | | | |
Add -all_load to make Objective-C categories work.
Enable hide_symbols to remove visibility issues.
|
| | |/
| |/|
| | |
| | | |
This has one known use case: compiling with the 10.4u SDKi on 10.6.
|
| |/
| |
| |
| |
| |
| |
| |
| | |
The problem was the QGLContextResource destructor which was called
when the QtOpenGL dll was unloaded. At this point in time, the gl
driver had already been unloaded so any gl calls at this point in time
would crash. It is simply wrong for the destructor to try to clean up,
so we instead output a warning if resources are leaked.
|
| |
| |
| |
| |
| | |
It's acceptable for the exceptionthrow test to output:
terminate called after throwing an instance of `int'
|
| |
| |
| |
| |
| |
| |
| |
| | |
Partially revert a974ce81d6324d8d85ade0153b5ce7c757ba4fdf. It seems odd
to just ignore the actual data in the input event but when the layer and
the event differs the layer is correct.
Reviewed-by: TrustMe
|
| |
| |
| |
| | |
Reviewed-by: TrustMe
|
| |
| |
| |
| |
| |
| | |
This will make DirectFB version tests work.
Reviewed-by: TrustMe
|
| |\ |
|
| | | |
|
| |\ \
| | |/ |
|
| | |
| | |
| | |
| | |
| | |
| | | |
Avoid warning when building DirectFB as part of Qt.
Reviewed-by: TrustMe
|
| |/
| |
| |
| |
| |
| |
| |
| |
| | |
Text drawing on OpenGL/ES 1.1 systems using QPF was displaying filled
boxes in place of the character glyphs. This is due to the QPF
implementation of alphaMapForGlyph() returning a different
color table than that expected by QGLGlyphCache::cacheGlyphs().
Reviewed-by: Sarah Smith
|
| |\ |
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
-waveOut callback was calling into class during destruction.
Changed to wait until close of waveOut/waveIn before returning from
destructor.
-Reduced default buffer on wince implementation.
-Added error checking on buffer allocation.
Task-number:261049, 261053
Reviewed-by:Bill King
|
| | |
| | |
| | |
| | | |
Reviewed-by: trustme
|
| |/
| |
| |
| | |
Reviewed-by: trustme
|
| |
| |
| |
| | |
Reviewed-by: trustme
|
| |
| |
| |
| | |
Reviewed-by: trustme
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
This item was inheriting from QObject and QGraphicsItem so there is
no point to not make it a QGraphicsObject. I have added some properties
that was specific to this class, i.e. elementId.
Reviewed-by: ogoffart
Reviewed-by: andreas
Reviewed-by: bnilsen
|
| |
| |
| |
| |
| |
| |
| |
| |
| | |
When using themed icons, we incorrectly used the systemTheme
as the fallback. This is not correct as the idea of the fallback
was to use the generic "gnome" or "oxygen" themes if the
proper theme was not available.
Reviewed-by: joao
|
| |
| |
| |
| |
| |
| |
| |
| | |
Partially revert 6715f2f29ae413b328243d8c3dbe15cfba51d33f
struct QWSServerCleaner { ~QWSServerCleaner(); } is not a function.
Reviewed-by: Noam Rosenthal <noam.rosenthal@nokia.com>
|
| |
| |
| |
| | |
Reviewed-by: Trust Me
|
| |
| |
| |
| |
| |
| |
| | |
Print a message when doing so and don't re-define qmake functions that
come from JavaScriptCore.pri.
Reviewed-by: Trust me
|
| |
| |
| |
| | |
Reviewed-by: thartman
|
| |
| |
| |
| | |
Reviewed-by: thartman
|
| |
| |
| |
| | |
Reviewed-by: thartman
|
| |
| |
| |
| | |
Reviewed-by: thartman
|
| |
| |
| |
| |
| | |
Reviewed-by: Trust Me
Really-reviewed-by: Legal
|
| |
| |
| |
| |
| | |
Reviewed-by: Takumi Asaki
Will-also-be-applied-to: 4.5
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
When loading binary files as text, QTextCursor did not check for
beginning and end of frame (QChar(0xfdd0) and QChar(0xfdd1)), while
QTextDocument asserted that those characters were absent.
The patch makes QTextCursor::insertText() treat frame bounderies like
paragraph separators.
Reviewed-by: ck
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
QFontMetrics::boundingRect() that takes a string needs to account for
the right bearing of the last glyph, as it is documented to be the
rectangle that contains the pixels of the text. I've added a test for
this, and fixed tst_QFontMetrics::elidedText() to use boundingRect() to
find the actual width of the text drawn (width() will return the advance
of the text, which is larger than the actual width of the pixels.)
I've also fixed a small typo in the "len" -> "ilen".
Reviewed-by: Simon Hausmann
|
| |
| |
| |
| |
| | |
Task-number: 260920
Reviewed-by: Trustme
|
| |
| |
| |
| |
| |
| |
| |
| | |
Removes dangling subFocusItem pointers when changing focus after
reparenting. This change also includes a mini-optimization when adding
focusable items to an inactive scene.
Reviewed-by: brad
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
This change ensures that only active scenes can have active input
focus items. If you try to set input focus on items that are in an
inactive scene, these items will gain input focus only when the scene
is activated.
For scenes attached to a view, this change should only fix the bug that
you could have a blinking line edit in a scene when the view is inactive,
meaning you couldn't type into the line edit.
For scenes that have no view, you now must activate the scene in order
to give the items active input focus. This will affect those who use
QGraphicsScene with custom key and focus handling.
Reviewed-by: brad
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
This change partially reverts f68fed388dcdba6ab6dad3af4933bcd3aa123cf8,
which was an attempt at getting FocusRealms working. Turns out this
approach is wrong. The flag behaves badly when reparenting, and the
feature conflicts with plain old setting focus.
Discussed with Aaron; this change was already reverted in the
kinetic-declarativeui branch.
Reviewed-by: TrustMe
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Previously, QtOpenGL assumed that by linking against QtGui, it would
automatically also be linked against EGL. However, this is no longer the
case after 83940f25dba51a9942ab55ed8475fc7fc8a8da84 which makes sure
only QtGui links against EGL and not other libs/apps linking against
QtGui.
Reviewed-by: Rhys Weatherley
|
| | |
|
| |
| |
| |
| | |
Reviewed-by: TrustMe
|
| |
| |
| |
| |
| |
| | |
Added begin/endNativePainting() calls.
Reviewed-by: Trond
|