| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
| |
Reviewed-by: TrustMe
|
|
|
|
|
|
| |
Fixes a memleak
Reviewed-by: Peter Hartmann
|
|
|
|
| |
RevBy: Shane Kearns
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The test was reported failing by pulse on many platform, but noone of
us was able to reproduce it locally. The problem was that pulse runs the
tests in a hidden dirrectory. And this exposed a bug in the QDirModel.
We decided anyway that the bug was not worth fixing considered that
QDirModel is kind of deprecated in favor of QFileSystemModel.
(The bug is the one reported in task 255066)
Reviewed-by: mbm
Reviewed-by: Thierry
|
|
|
|
|
|
|
|
|
|
|
|
| |
When a text in QTextLayout contained tabs, these would be counted as
single glyphs when breaking the text on a fixed number of columns,
rather than the number of characters they span.
The patch calculates the number of characters represented by a tab by
using the average character width of the font engine.
Task-number: QTBUG-4468
Reviewed-by: Simon Hausmann
|
|
|
|
|
|
|
|
|
| |
If for some reason the theme name was not detectable, Qt would
incorrectly print "QGtkStyle cannot be used together with the
GTK_QT engine." I now added a separate "QGtkSTyle was unable to
detect the current GTK+ theme." warning instead.
Reviewed-by: Trust Me
|
|
|
|
|
|
|
| |
We were not respecting the rect offset on combo box arrows.
This was preventing the autotest from happily completing with GTK+ style.
Reviewed-by: joao
|
| |
|
|
|
|
| |
Reviewed-by: Kim
|
|
|
|
|
|
|
|
|
|
| |
1. The mono format was not handled at all.
2. We really, really wanted to use sub-pixel hinted glyphs even
when they were not available.
3. The glyphFormat type in the FT font engine wasn't updated to
reflect the correct system glyph type.
Reviewed-by: Kim
|
|
|
|
| |
Function is only used from within Q_OS_WIN or Q_WS_WIN-blocks.
|
|
|
|
|
|
|
| |
By default, waitForSignal() will block the entire test if it fails,
adding 20 false autotest failures.
Reviewed-by: TrustMe
|
| |
|
|
|
|
|
|
|
|
|
| |
such as
margins.h:90: warning: declaration of ???bottom??? shadows a member of _this_"
'a' stands for argument
Reviewed-by: jbache
|
|
|
|
|
|
|
|
|
|
| |
Get rid of the pages of warning you get when compiling code that uses
QtConcurrent::map for instance.
qmap.h:607: warning: dereferencing pointer 'y' does break strict-aliasing rules
qmap.h:605: note: initialized from here
Reviewed-by: Peter Hartmann
|
|\ |
|
| |
| |
| |
| |
| |
| |
| |
| | |
strictlyEquals has to continue to work if one of the values has an engine
and the other one doesn't. Use the other value's engine to construct a
JSValue if possible and use JSC::JSValue::strictEqual.
Reviewed-by: Kent Hansen
|
| |
| |
| |
| |
| |
| |
| | |
Don't create JSC types as the conversion back to a string now
requires an ExecState.
Reviewed-by: Kent Hansen
|
| |\ |
|
| | |
| | |
| | |
| | |
| | |
| | | |
Based on mkdist-webkit
Reviewed-by: Kent Hansen
|
| |\ \
| | |/
| |/|
| | |
| | |
| | | |
Conflicts:
src/corelib/kernel/qcoreapplication.cpp
tests/auto/windowsmobile/test/tst_windowsmobile.cpp
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Fix QGraphicsItem to clear clip path for items with an ancestor that
clips to shape. Added autotest to demonstrate clipping path problem.
Merge-request: 810
Reviewed-by: Alexis
Reviewed-by: Andreas
|
| | |
| | |
| | |
| | | |
Reviewed-by: TrustMe
|
| | | |
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
On Windows CE the plugin path is likely to be the same as the
application directory. That's why we must not compare app_location
with the plugins path.
The check app_libpaths->contains(app_location) should be enough.
Autotest: tst_QApplication::libraryPaths
Reviewed-by: thartman
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Previously a different version of this file was added to each source
package. As the file is now the same for all source packages, it can be
added to the repo.
Reviewed-by: Trust Me
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Replace convertHIconToPixmap() by QPixmap::fromWinHICON()
Move loadIconFromShell32() to qwindowsstyle.cpp
Merge-request: 1570
Reviewed-by: Eskil Abrahamsen Blomfeldt <eskil.abrahamsen-blomfeldt@nokia.com>
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Duplicate QPixmap <-> HICON conversion code removed from
qwidget_win.cpp & qsystemtrayicon_win.cpp.
Task-number: 218533
Merge-request: 1570
Reviewed-by: Eskil Abrahamsen Blomfeldt <eskil.abrahamsen-blomfeldt@nokia.com>
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
In S60 softkeys and menu items do not support key accelerators (i.e.
CTRL+X). Therefore, removing the accelerator characters from both softkey
and menu item texts.
Related to task QT-1505 "Make Qt dialogs look native"
Reviewed by Jason Barron
|
| | |
| | |
| | |
| | | |
Reviewed-by: mstormo
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
The problem is that the interpreter did not check for exception while
running the "while(true){}" loop, as it deduced that none of the generated
opcode would possibly generate an exception.
The solution is to force a check right after we come from a timeout.
Reviewed-by: Kent Hansen
|
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Added GCCE as a compiler in SymbianMakefileGenerator::writeMmpFileCompilerOptionPart
Task-number: QT-1163
Reviewed-by: axis
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
The target path for binaries (.path configuration) in the DEPLOYMENT
section of a .pro file now works for binaries as well as for resources.
Path on device is still forced to /sys/bin
Default path is still !:/sys/bin
This can be used to force a binary to be installed on system drive by
setting the path to "c:/sys/bin" or "$:/sys/bin" (for S60 3.2+)
Reviewed-by: Iain
|
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
The antialiasing is currently not gamma corrected and is disabled on
OpenGL ES 2.0.
Reviewed-by: Samuel
|
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Now that we don't make platform-specific source packages, we only have
one version of the docs. These docs need to include all platforms.
Reviewed-by: Trust Me
|
|/ /
| |
| |
| | |
Reviewed-by: Trust Me
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
These fonts used to be added to the qt-embedded-linux source package by
the packaging scripts. As we no longer make platform-specific source
packages, these files now belong in the Qt repository from which they
will be added to all source packages.
Reviewed-by: Trust Me
Acked-by: Thiago Macieira
|
| |
| |
| |
| | |
Reviewed-by: Trust Me
|
| |
| |
| |
| |
| |
| |
| | |
We now generate only one set of docs, for the qt-everywhere package.
Thus we no longer need to differentiate the different editions.
Reviewed-by: Trust Me
|
|\ \ |
|
| | |
| | |
| | |
| | |
| | |
| | | |
adds MYSQL_OPT_RECONNECT to the mysql driver options.
Task-number: QTBUG-4510
|
| | | |
|
|/ /
| |
| |
| |
| |
| |
| |
| | |
It is useful to be able to map direction vectors by the top-left
3x3 component of a 4x4 matrix, ignoring the translation and
projection components.
Reviewed-by: Sarah Smith
|
| |
| |
| |
| | |
This reverts commit 1b34feacef7a2d3ac005449a7cfbcb08a6bbf947.
|
| |
| |
| |
| | |
Reviewed-by: trustme
|
| |
| |
| |
| |
| |
| |
| | |
when cross compiling, freetype2 include pathes were not used when
compiling the fontconfig config.test. Fixed.
Reviewed-by: Simon Hausmann
|
| |
| |
| |
| |
| |
| |
| | |
Previously, the firstTimeObserver was released by the event loop. This
do not work if the event loop is never run.
This commit amend b0016ea9a6b225757e3ee06b50e8f7d05463ddf7.
|
| | |
|
| |
| |
| |
| | |
For more dependable test results.
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Since QXmlSchemaValidator resolves relative paths against the executable
location, we use QUrl::fromLocalFile before passing them along.
This is a more typical behavior for command-line applications and fixes
the autotest.
Reviewed-by: Frans Englich
|