summaryrefslogtreecommitdiffstats
path: root/src/corelib/kernel
Commit message (Collapse)AuthorAgeFilesLines
* Corrected alphabetic order in the docs for events.axis2010-01-201-1/+1
| | | | RevBy: Trust me
* QMetaObject::disconnectOne()Aaron Kennedy2010-01-203-7/+29
| | | | | | | | Adds a disconnectOne() method that allows a single signal/slot connection to be disconnected, as required by QTBUG-6781. Reviewed-by: Warwick Allison Reviewed-by: Olivier Goffart
* Merge branch '4.6' of scm.dev.nokia.troll.no:qt/oslo-staging-2 into ↵Qt Continuous Integration System2010-01-131-256/+256
|\ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | 4.6-integration * '4.6' of scm.dev.nokia.troll.no:qt/oslo-staging-2: (23 commits) My changes Significant digits were lost in QDoubleSpinBox range when changing precision My changelog for 4.6.1. Added my changes to the 4.6.1 change log. Put LinesHint into QVectorPath hints to enable further optimizations Update changes file for 4.6.1 Avoid coordinate limitations in the raster engine. Update changelog with my 4.6.1 changes. 4.6.1 changes Fixed bug where QGLPixmapData::toImage() returned too dark image. Update change log with Yoann Lopes work. Fix completion in QFileDialog. Fix point drawing on raster engine for flat and square caps Update changes file with Rhys Weatherley's work. warning fixes for platforms, where qreal == float fix release mode crash in qfont.cpp initFontSubst() on Windows mobile Fixed justify aligned text drawing with the GL1 engine. Add odf-writer benchmark. Fixed subpixel antialiased text drawing with the GL2 engine. Fixes warning in the QMacStyle ...
| * Merge remote branch 'qt/4.6' into oslo-staging-2/4.6Paul Olav Tvete2010-01-1286-89/+89
| |\ | | | | | | | | | | | | Conflicts: dist/changes-4.6.1
| * | warning fixes for platforms, where qreal == floatJoerg Bornemann2010-01-111-256/+256
| | | | | | | | | | | | Reviewed-by: thartman
* | | Make compile on HPUXPierre Rossi2010-01-121-0/+2
| |/ |/| | | | | Reviewed-by: axis
* | doc: Updated the descriptions of the connection types.Martin Smith2010-01-121-1/+1
| | | | | | | | | | | | | | | | Replaced the ambiguous "delivered" with the explanation that the blocking queued connection type blocks the emitter's thread until the slot finishes executing in the receiver's thread. Task-number: QTBUG-6392
* | Update copyright year to 2010Jason McDonald2010-01-0686-88/+88
|/ | | | Reviewed-by: Trust Me
* doc: Added a missing \sa command, plus a \l in the text.Martin Smith2009-12-181-7/+8
| | | | Task-number: QTBUG-6288
* Merge remote branch 'staging/4.6' into 4.6Simon Hausmann2009-12-172-0/+7
|\
| * Avoid timer starvation when handling many X11 eventsBradley T. Hughes2009-12-172-0/+7
| | | | | | | | | | | | | | | | | | | | | | | | | | After commit d0d0fdb8e46351b4ab8492de31e5363ef6662b57, timers are normally run at idle priority. This makes it possible for the X11 handler to starve timers indefinitely. Fix this by enforcing one normal priority pass of the timer source after we have processed all X11 events. This has the added benefit of keeping animation timers smooth and consistent, which is the intention of this change. Reviewed-by: Jens Bache-Wiig
| * Revert "Changes to support locale change event for symbian platform"Denis Dzyubenko2009-12-164-20/+1
| | | | | | | | | | | | Revert a change that is not yet complete and was pushed by mistake. This reverts commit 44f7c1e097582a704a06ccbbf516536b88ddcd3a.
| * Changes to support locale change event for symbian platformdka2009-12-154-1/+20
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Subscribing to the locale change notification to be able to update the system locale whenever the user changes the current system locale. Also changed the initialization of the system locale to make construction of the QLocale object as lightweight as possible. So now the default contructor just creates a QLocale and QSystemLocale objects, but doesn't try to fill the cache in the latter with data from the system and postpones it until it is actually requested (most applications create QLocale objects on the stack and might not even use the data from the system locale, so we don't need to initialize system locale right away). Modified-by: axis Modified-by: Denis Dzyubenko Reviewed-by: Denis Dzyubenko Reviewed-by: axis
* | Merge branch '4.6' of scm.dev.nokia.troll.no:qt/qt-s60-public into ↵Qt Continuous Integration System2009-12-161-45/+109
|\ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | 4.6-integration * '4.6' of scm.dev.nokia.troll.no:qt/qt-s60-public: (31 commits) Fixed qstylesheetstyle benchmark for Symbian Fixed qdiriterator benchmark for Symbian Fixed events benchmark postEvent function Fix for QTBUG-4908 SVG transparency rendering problem. Long informative texts causes messagebox to grow outside of screen area qreal-ization qreal-ization qreal-ization QS60Style: Remove layouts with mirrored information Fixed qgraphicsview autotest build for winscw. Enabled input method update code for all platforms. (ODBC) Fixes segfault when error string is larger than 256 chars. Fixed QGraphicsView benchmark for Symbian. FEP indicator shown in status pane when it should not Skipped the most memory intensive tests in QByteArray benchmark. qreal-ization qreal-ization QS60Style: Theme graphics for QSlider in 3.1 QS60Style: Groove changes caused build break of S60 3.1 Slow spinbox on N95 when using keys Up/Down ...
| * \ Merge branch '4.6' of git@scm.dev.nokia.troll.no:qt/qt into 4.6-s60axis2009-12-164-11/+13
| |\ \ | | | | | | | | | | | | | | | | Conflicts: src/s60installs/bwins/QtGuiu.def
| * | | qreal-izationAleksandar Sasha Babic2009-12-151-45/+109
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Using Symbian Math package. Benhcmarks are showing that some Symbian native math functions (trigonometric, exp, pow ...) are faster then their Open C counterparts. Task-number: QTBUG-4894 Reviewed-by: Iain
* | | | Merge commit 'origin/4.6' into 4.6Olivier Goffart2009-12-142-2/+16
|\ \ \ \ | | |/ / | |/| |
| * | | Merge commit oslo-staging-2/4.6 into upstream/4.6Olivier Goffart2009-12-143-9/+10
| |\ \ \ | | |_|/ | |/| |
| * | | Check if the timeout expired during the time update in qt_safe_selectThiago Macieira2009-12-111-2/+3
| | |/ | |/| | | | | | | | | | | | | | | | | | | | | | | | | It can happen that select(2) returns -1 with EINTR, but the time remaining was too small for us to update the time. Our own processing (plus the syscall to clock_gettime(2)) leads us past the timeout, so we get a negative timeval. Task-number: QTBUG-6755 Reviewed-by: Bradley T. Hughes
| * | Merge commit 'origin/4.6' into 4.6-upstreamBradley T. Hughes2009-12-091-0/+13
| |\ \
| | * | Compile on Windows CE.Bradley T. Hughes2009-12-081-0/+13
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Windows CE does not support Windows hook at all, so we need to call our hook function "manually" from process events. Task-number: QTBUG-6083 Reviewed-by: joerg
* | | | Fix QMetaObject::connect and disconnect with "dynamic signals"Olivier Goffart2009-12-142-20/+24
| |_|/ |/| | | | | | | | | | | | | | | | | | | | QML might pass index that are larger that the method cound. We must not call QMetaObjectPrivate::originalClone in that case as this would read invalid memory Reviewed-by: brad
* | | Use 64bit for the connectedSignalsOlivier Goffart2009-12-092-8/+9
| | | | | | | | | | | | | | | | | | | | | | | | | | | as QML objects may have lots of signals Do not use quint64 as it would produce lots of useless padding on MSVC 32bit Reviewed-by: Brad
* | | Fix regression in qVariantFromValue when converting from complex type to ↵Olivier Goffart2009-12-091-1/+1
|/ / | | | | | | | | | | | | | | | | | | | | | | | | simple type QVariant v = QColor(Qt::red); v.setValue(1000); Would produce a variant with garbage. the destructor of QColor would not be called, and the 1000 would be in the QVariant::PrivateShared, while most of the QVariant code assume that numbers are dirrectly in the QVariant::Data union Task-number: QTBUG-6602 Reviewed-by: Thierry
* | Merge commit 'origin/4.6' into 4.6Olivier Goffart2009-12-042-29/+40
|\ \ | |/
| * Merge commit 'c0b81480b2909b18ac15bdd124a562ae005c2f41' into origin-4.6Olivier Goffart2009-12-041-0/+57
| |\
| * \ Merge branch '4.6' of scm.dev.nokia.troll.no:qt/oslo-staging-1 into 4.6Bradley T. Hughes2009-11-303-43/+27
| |\ \
| * | | Fix performance regression of posted eventsBradley T. Hughes2009-11-272-29/+40
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Commit 31f1ff91028dd7f90925d5b3737e4d88b5fb07aa introduced a performance regression by using SetTimer() to delay the next call to sendPostedEvents(). SetTimer() has a minimum resolution, around 15-16ms, which is too slow for most uses. Fast timers and QWidget::update() use posted events to do their job, and suffered as a result. This commit goes away from using SetTimer() to using a GetMessage() hook that examines the input queue as soon as message are pulled from the queue. Now there is no large delay between calls to sendPostedEvents(), they are sent as quickly as possible (once all other input and timer messages have been delivered). Reviewed-by: Prasanth Ullattil Task-number: QTBUG-6083
* | | | Made qobject.h "syncqt compliant".Jan-Arve Sæther2009-12-041-3/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The previous patch to this file made syncqt confused (maybe because of the brackets), and caused it to not generate the QObject header file. This made of course a lot of stuff not compile. With this patch syncqt behaves. Reviewed-by: Olivier Goffart
* | | | Merge branch 'win32-icc' into 4.6Jan-Arve Sæther2009-12-042-12/+4
|\ \ \ \ | |_|_|/ |/| | |
| * | | Make sure that QtGui links with QtCore for the win32-icc mkspec.Jan-Arve Sæther2009-12-041-0/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Fix it by exporting QObjectData. Since this is a compiler bug we will only export it for the compiler that has this problem. (Intel Compiler on Windows) The Intel compiler seems to inline the destructor of QObject (or at least parts of it) even if it is declared virtual. Thus, it seems that the destruction of the QScopedPointer<QObjectData> member is inlined into QtGui, and that code will eventually call the destructor of QObjectData, and since QObjectData was not exported from QtCore it would crash. Reviewed-by: Olivier Goffart
| * | | Revert "Compile fix for win32-icc."Jan-Arve Sæther2009-11-301-12/+0
| | | | | | | | | | | | | | | | This reverts commit db5e4496229a776768464d1d3d2e1f8e81bd6ba0.
* | | | qreal-izationAleksandar Sasha Babic2009-12-021-0/+57
| |_|/ |/| | | | | | | | | | | | | | | | | | | | | | | | | | | | | Adding more wrapper math related functions. This should enable us to have more control over which calls are actually made. Don't worry too much for some "if" statements, the compiler is smart enough to make direct calls to specific math functions. Task-number: QTBUG-4894 Reviewed-by: axis
* | | Merge commit oslo-staging-2/4.6 into upstream/4.6Olivier Goffart2009-11-273-43/+27
|\ \ \ | |/ / | | / | |/ |/|
| * Don't crash via proxy model when moving QModelIndexesStephen Kelly2009-11-261-1/+1
| | | | | | | | | | | | | | | | | | | | | | The layoutAboutToBeChanged signal gives proxies a chance to persist QModelIndexes. The d->itemsAboutToBeMoved method is part of the process of moving the persistent indexes. d->itemsAboutToBeMoved must be called after layoutAboutToBeChanged. Otherwise persistent indexes will not be updated. Merge-request: 2219 Reviewed-by: Olivier Goffart <ogoffart@trolltech.com>
| * Optimize QObjectPrivate::isSignalConnectedOlivier Goffart2009-11-262-42/+26
| | | | | | | | | | | | | | | | | | | | | | | | | | | | It can be seen in the GraphicsView benchmark. This function has to be as fast as possible. Make the function inline. We do not need 64bit for the connectedSignals bit array because it is very unlikely that an object will have more than 32 signals. (In Qt 4.5 it could have hapenned as the slot were counting in the index) Reviewed-by: bnilsen Reviewed-by: Brad
| * Compile fix for win32-icc.Jan-Arve Sæther2009-11-231-0/+12
| | | | | | | | | | | | | | | | | | The Intel Compiler inlined the destructor of QObjectPrivate too agressively, causing it to generate a call to ~QObjectData for QtGui. ~QObjectData is not exported from QtCore, so it failed linking. Task-number: QTBUG-5145 Reviewed-by: Alexis Menard
* | On Windows Flash is played only while moving mouse inside the plugin.Prasanth Ullattil2009-11-251-1/+1
| | | | | | | | | | | | | | Qt is using the same window message as Flash plugin. We need to check the window handle before we treat this as Qt internal message. Reviewed-by: Bradley T. Hughes
* | Compile fix for win32-icc.Jan-Arve Sæther2009-11-241-0/+12
|/ | | | | | | | | | The Intel Compiler inlined the destructor of QObjectPrivate too agressively, causing it to generate a call to ~QObjectData for QtGui. ~QObjectData is not exported from QtCore, so it failed linking. Task-number: QTBUG-5145 Reviewed-by: Alexis Menard (cherry picked from commit db5e4496229a776768464d1d3d2e1f8e81bd6ba0)
* Fix the evaluation nag in binary installs with commercial licenses.Thiago Macieira2009-11-181-12/+14
| | | | | | | | | | It wasn't generating any message, but it did think it was an evaluation install. So all Qt console applications produced an extra newline and the GUI ones showed the Qt logo briefly. I noticed because of the extra newline when building Qt itself. Reviewed-by: Trust Me
* Merge oslo-staging-2/4.6 into upstream/4.6Olivier Goffart2009-11-171-0/+2
|\
| * Fix QT_NO_SHAREDMEMORY while not breaking the QNX buildJørgen Lind2009-11-161-0/+2
| | | | | | | | Reviewed-by: Harald Fernengel <harald.fernengel@nokia.com>
* | Merge remote branch 'staging/4.6' into 4.6Simon Hausmann2009-11-171-6/+6
|\ \ | |/ |/|
| * Don't sendPostedEvents() twice when calling processEvents() manuallyBradley T. Hughes2009-11-161-6/+6
| | | | | | | | | | | | | | | | | | Commit fe0f807e1f4e7510c6d8cddd848bcbc25e358651 could cause sendPostedEvents() to be called twice, which caused regressions in tests/auto/qtimer. Fix this by only calling sendPostedEvents() "manually" if we didn't see a WM_QT_SENDPOSTEDEVENTS message. Reviewed-by: Prasanth Ullattil
* | Early return for allowMove within a parent QModelIndexStephen Kelly2009-11-131-4/+2
|/ | | | | | | | | If this is not done, the climbing ancestors later in the method uses srcParent.row() as pos causing failure depending on which rows are being moved, and what the row of the parent is. Merge-request: 2072 Reviewed-by: Olivier Goffart <ogoffart@trolltech.com>
* Merge commit 'upstream/4.6' into 4.6Bradley T. Hughes2009-11-121-0/+4
|\
| * Merge branch '4.6-platform' into 4.6Denis Dzyubenko2009-11-111-3/+11
| |\
| * | make qfunctions_wince.h C compatible againJoerg Bornemann2009-11-101-0/+4
| | | | | | | | | | | | Reviewed-by: mauricek
| * | compile fix for tst_qfile on Windows CEJoerg Bornemann2009-11-091-1/+1
| | | | | | | | | | | | | | | | | | Changed the signature of QT_OPEN to be the same on all platforms. Reviewed-by: thartman
* | | Make sure posted events are always sent when calling processEvents() on Win32Bradley T. Hughes2009-11-111-0/+5
| |/ |/| | | | | | | | | | | | | | | After 31f1ff910, posted events could be delayed by a previous call to processEvents(). This causes some tests to randomly fail, so bring back the invariant that processEvents() will always call sendPostedEvents() when called "manually" (i.e. not from exec()). Reviewed-by: Prasanth Ullattil
generated by cgit v0.12 at 2025-01-16 01:49:10 (GMT)