summaryrefslogtreecommitdiffstats
path: root/src
Commit message (Collapse)AuthorAgeFilesLines
* Support rotated displays with 15 and 16 bit BGR pixel formats.Paul Olav Tvete2009-05-054-0/+51
| | | | | | | | To avoid increasing code size for an uncommon case, the code is #ifdef QT_QWS_ROTATE_BGR Task-number: 250299 Reviewed-by: Tom
* Merge branch '4.5' of git@scm.dev.nokia.troll.no:qt/qt into 4.5Morten Sørvig2009-05-053-45/+39
|\
| * Revert "Fix tabbar issues with Oxygen style"Jens Bache-Wiig2009-05-051-41/+32
| | | | | | | | | | | | | | | | | | | | The patch has some issues with text eliding which are not really possible to solve properly without providing the elidemode to the styleoption. Under these circumstances it seems better just to fix the problem in Oxygen itself which has already been patched when compiled with 4.5. This reverts commit d7072b5577e3c07e984885cf05d5f9b217c6f473.
| * QListView in icon mode could crash when removing the last elementThierry Bastian2009-05-051-3/+4
| | | | | | | | | | Task-number: 252292 Reviewed-by: ogoffart
| * On Windows loosing focus should result in clearing focus if widget is ↵Denis Dzyubenko2009-05-051-1/+3
| | | | | | | | | | | | | | | | | | | | | | embedded somewhere. If a widget is embedded to non-Qt window on Windows, then we should still clear the focus whenever we receive WM_KILLFOCUS since we won't get WM_ACTIVATEAPP in this case. This is an addition to 6ed196051d0f19bfe2d045eaf12f5f5ca30670d0 Task-number: 251259 Reviewed-by: Thierry
* | Fix memory leak in QFutureMorten Sørvig2009-05-051-0/+2
|/ | | | | | | | | The held results were not cleared by QFutureInterface::operator=(QFutureInterface) Add call to resultStore().clear(), similar to the destructor. This needs to be done in the header since we know the template type here. Task-number: 252208 Reviewed-by: brad
* Rename QDirectFBSurface to QDirectFBWindowSurfaceAnders Bakken2009-05-045-34/+34
| | | | | | | This helps avoid confusion between IDirectFBSurface and QDirectFBSurface. Reviewed-by: Donald <qt-info@nokia.com>
* Lock for read only when possibleAnders Bakken2009-05-049-25/+40
| | | | | | | | | | From benchmarking I've established that surface->Lock(DSLF_READ) is faster than surface->Lock(DSLF_WRITE) which is faster than surface->Lock(DSLF_READ|DSLF_WRITE). This patch will make us Lock for read only, when possible. Reviewed-by: Donald <qt-info@nokia.com>
* QNetworkCookieJar: do not allow cookies for domains like ".com"Peter Hartmann2009-05-041-0/+7
| | | | | | | | the domain attribute in cookies must always contain one embedded dot, according to RFC 2109 section 4.3.2 Reviewed-by: Thiago Task-number: 251467
* Added a check that X11 timestamp goes forward only.Denis Dzyubenko2009-05-041-15/+20
| | | | | | | | Got a case somehow with a timestamp of the mouse event that is less than the timestamp we already had, so we need to make sure time only goes forward. Reviewed-by: Brad
* Added an assert so that QColormap usage without a QApplication asserts.Trond Kjernaasen2009-05-041-1/+5
| | | | | Task-number: 252668 Reviewed-by: Samuel
* Ensure that we send the Apple Events even when Cocoa isn't ready.Norwegian Rock Cat2009-05-042-11/+23
| | | | | | | | | | | | | | In general, Cocoa handles the the Apple Events for us. However, this is time between creating the NSApplication and Cocoa has set everything up, usually after the event loop is running. This means that until that time, the events are dropped on the floor :-/. The workaround is to use the same handler that we use for Carbon, but to only have it enabled for until Cocoa is ready to handle things. This will result in not stepping on the toes when used in a plugin (if it does, we can conditionalize it). Task-number: 252795 Reviewed-by: Richard Moe Gustavsen
* Fixed D-Bus socket write notifications, broken since d47c05b1Thiago Macieira2009-05-041-1/+1
| | | | | | Shame on me: copy/paste from socketRead to socketWrite, I didn't change the DBUS_WATCH_READABLE to DBUS_WATCH_WRITABLE. Reviewed-by: Trust Me
* Greatly reduced the complexity of the boilerplate function.Thiago Macieira2009-05-041-90/+17
| | | | | | | | I found out that all I needed to load the proper libraries was to add a string to the ".interp" section of the ELF executable containing the path to ld.so Reviewed-By: Marius Storm-Olsen
* Added comment to clearify the use of indexes.Morten Engvoldsen2009-05-041-0/+8
| | | | | | | Added a comment about the use of negative indexes. Task-number: 249344 Rev-by: Marius Storm-Olsen
* Do not crash when passing wrong indexes to ↵Olivier Goffart2009-05-041-0/+9
| | | | | | | | | QSortFilterProxyModel::indexFomSource and *ToSource Show a warning instead Task-number: 252507 Reviewed-by: Marius Bugge Monsen
* Implemented an option to tune flippingAnders Bakken2009-05-044-3/+11
| | | | | | | | export QWS_DISPLAY=directfb:boundingrectflip to enable calling Flip on the bounding rect of the dirtied area rather than each dirty rectangle. This could be faster if you update many small rectangles. Reviewed-by: TrustMe
* Beautified codeAnders Bakken2009-05-022-17/+17
| | | | | | Qt's coding style => Object *ptr, not Object* ptr Reviewed-by: TrustMe
* Improve debug output (in debug mode only)Anders Bakken2009-05-011-2/+83
| | | | | | | Print out detailed information about acceleration mask, blitting flags and drawing flags when passing debug. Reviewed-by: Donald <qt-info@nokia.com>
* Extended surface capabilitiesAnders Bakken2009-05-012-6/+31
| | | | | | | Clean up code and make it possible to set more DFBSurfaceCapabilities on the primary surface. Also allow users to force systemonly for the surfaces.
* Cleaned up surface creation codeAnders Bakken2009-05-014-34/+28
| | | | | | | | Since I am taking a copy of the description anyway it makes sense to just pass this light-weight object in as a copy rather than a const pointer. Reviewed-by: Donald <qt-info@nokia.com>
* Improved readability of the flip codeAnders Bakken2009-05-011-7/+2
| | | | Reviewed-by: Donald <qt-info@nokia.com>
* Fixed possible crash in QDirectFBPaintEngine::clipAnders Bakken2009-05-011-1/+1
| | | | | | | d->clip() might return 0 at this point so make sure we check before accessing it. Reviewed-by: TrustMe
* Apparently this is the best way to determine which to useBill King2009-05-011-5/+4
| | | | Best as I can determine via trial and error. It should make vc6 compile again though.
* QNetworkCookieJar: allow cookies with wrong domain attributePeter Hartmann2009-04-301-5/+4
| | | | | | | | | | According to the (old) cookie RFC 2109, the domain attribute must always contain a leading dot. Some servers do not have that, but all browsers accept those cookies anyway, so we should do that as well. Reviewed-by: Olivier Reviewed-by: Denis Task-number: 228974
* Fixed busy indicator for a QProgressBar with a style sheet applied to it.jasplin2009-04-301-1/+1
| | | | | | | | | | | | | | | For a progress bar with a style sheet applied to it, this fix ensures that the timer event is passed to the event handler that updates the busy indicator animation state. Essentially, the bug was that the decision that the event was processed by the proxy style object (baseStyle()) was based only on the return value from the event() function. In this case it is necessary to check that the event was accepted as well. Reviewed-by: ogoffart Reviewed-by: brad Task-number: 252283
* Fixes a possible issue in itemviws where we would not scroll to theThierry Bastian2009-04-302-10/+17
| | | | | | | current item on show Task-number: 252534 Reviewed-by: ogoffart
* QApplication::setStyle() can cause a crashRichard Moe Gustavsen2009-04-301-1/+4
| | | | | | | | | | | QApplication::setStyle() caused a crash if called before constructing the QApplication instance for custom styles. The reason was that polish tried to create a pixmap (which is not allowed before qApp is running). This fix checks that qApp exists. Polish will anyway be called again when qApp gets constructed. Task-number: 243697 Reviewed-by: Bjørn Erik Nilsen
* QTransform::map(const QRegion&) cut-off for single rect regions.Bjoern Erik Nilsen2009-04-291-0/+4
| | | | | | Avoid QRegion<->QPainterPath conversion if possible. Reviewed-by: Samuel
* Speed-up QPainter::clipRegion().Bjoern Erik Nilsen2009-04-291-11/+20
| | | | | | | | | We can speed up the calculation by using rect intersections if possible, i.e. QRegion &= QRect instead of QRegion &= QRegion. Then we'll get rid of one QRegion construction and the intersection itself is slightly faster. Reviewed-by: Samuel
* Wrong clip in QWidget::render(QPainter *, ...) when using Qt::(Replace|No)Clip.Bjoern Erik Nilsen2009-04-291-2/+7
| | | | | | | | | | | | | | | The problem was that we didn't take the painter's clip into account when setting the system viewport ("hard clip"). We only used the system clip, but we have to use system clip + painter clip, which is the current engine clip. Unfortunately, we have to calculate it again since there's no cross-platform way of retrieving it. This was only a problem with Qt::(Replace|No)Clip, since we in all other cases combine the old clip with the new one. (Uber cool) auto test included. Task-number: 250482 Reviewed-by: Samuel
* Corrected description of the QLayout::takeAt() functionMorten Engvoldsen2009-04-291-1/+1
| | | | | | | An item is not deleted when removed from the index. The remaining items get a new index. I changed deleted to removed. Tasknumber: 252547 Rev-by: janarve
* Merge branch '4.5' of git@scm.dev.nokia.troll.no:qt/qt into 4.5Jarek Kobus2009-04-2921-244/+326
|\
| * Mac: Fixes a crash in tst_QWidget::updateOnDestroyedSignalBjoern Erik Nilsen2009-04-291-0/+3
| | | | | | | | | | | | | | | | | | | | Addendum to change 9b674ddcea7c0ff7912f185474b7d98bc83bd39f. The problem was that QWidget::paintEngine() (pure virtual) was called while being in the destructor. The use-case itself (see above change) is very very corner case, but still, it shouldn't crash. Reviewed-by: bnilsen
| * Get rid of an unnecessary image copy when grabbing a GL framebuffer.Trond Kjernåsen2009-04-291-2/+10
| | | | | | | | | | Task-number: Related to 241466 Reviewed-by: Samuel
| * QNetworkAccessManager: do not segfault when using a network cacheBenjamin C Meyer2009-04-291-1/+2
| | | | | | | | | | | | | | Don't setfault when setting 0 for the network cache such as when you want to disable it. Reviewed-by: Peter Hartmann
| * Try and clean up ODBC 64/32 bit type disparityBill King2009-04-291-6/+4
| | | | | | | | It seems that after approx version 2.7.0, microsoft changed the ODBC types of certain functions, to cater for 64bit architectures. This tries to be a little bit smarter about which types are used in which places.
| * Fixes performance degradation from precision fix in 4.5.1Bill King2009-04-291-20/+20
| | | | | | | | | | | | | | Alleviates the performance hit of querying the database except when necessary, try and use XSQLVAR as much as possible, except for floating point fields. Task-number: 252426
| * Remove superfluous call to base classAnders Bakken2009-04-291-2/+2
| | | | | | | | | | | | | | | | Make code more readable by not chaining to QScreenCursor for hide/show. Since hwaccel is true we don't need the exposeRegion call anyway. Reviewed-by: TrustMe
| * Cleaned up mouse/cursor codeAnders Bakken2009-04-293-33/+28
| | | | | | | | | | | | | | | | | | The enabling/disabling of the Cursor belongs in the QDirectFBMouseHandler class and not in the QDirectFBScreenCursor. Even if we use the Qt software cursor we need the events to come in from the mouse handler. Reviewed-by: TrustMe
| * Fixed cursors with QT_NO_DIRECTFB_PREALLOCATEDAnders Bakken2009-04-291-75/+75
| | | | | | | | | | | | | | | | | | | | Refactor cursor management code to do. No need to change the cooperative level four times for each show/hide. Simplify code a little. Also we used to disable the cursor when we wanted to hide it which meant that we no longer received mouse events. Reviewed-by: Donald <qt-info@nokia.com>
| * Fold boolean options into flagsAnders Bakken2009-04-284-20/+21
| | | | | | | | | | | | | | Instead of having multiple bool QDirectFBScreen::.* getters I added DirectFBFlags QDirectFBScreen::directFBFlags() function. Reviewed-by: TrustMe
| * Added an option for ignoring the system clipAnders Bakken2009-04-283-10/+23
| | | | | | | | | | | | | | | | | | | | | | | | This is currently the only way I can make QGraphicsView not have to fall back to the raster engine for all operations. It seems the QRegion passed to the paintEvent of QGraphicsView also is set as the systemClip which in the end makes the QRasterPaintEngine's clipRegion equal the systemClip. By exporting QWS_DISPLAY=directfb:ignoresystemclip you can now draw without having a complex clip and therefore take advantage of hw acceleration with DirectFB. Reviewed-By: Donald <qt-info@nokia.com>
| * Make the disable/warn on fallbacks more flexibleAnders Bakken2009-04-282-52/+82
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Allow programmers to select which operations they want to warn on/disable and which not to. To warn for DRAW_PIXMAP only you can do this: DEFINES += QT_DIRECTFB_WARN_ON_RASTERFALLBACKS=DRAW_PIXMAP To include fillRect you can do: DEFINES += \"QT_DIRECTFB_WARN_ON_RASTERFALLBACKS=DRAW_PIXMAP|FILL_RECT\" Reviewed-by: Donald <qt-info@nokia.com>
| * Cleaned up constness errorsAnders Bakken2009-04-281-14/+16
| | | | | | | | | | | | | | | | | | | | All these drawing functions really shouldn't be const even if some of them can be. They should never be called from const functions and when/if they change pen/brush stuff they cease being able to be const. Also make sure fillRegion() uses the right brush. Reviewed-by: Donald <qt-info@nokia.com>
| * Fix documentation about native dialogsAlexis Menard2009-04-281-4/+3
| | | | | | | | | | | | It basically says that on KDE and GNOME the native dialogs are called. It fix a mistake about a regular instantiation on QFileDialog object on Windows.
| * Fix regression in QSelectionModel::rowIntersectsSelectionOlivier Goffart2009-04-281-10/+10
| | | | | | | | | | | | | | | | | | | | and QSelectionModel::columnsIntersectsSelection The documentation says "if one is selected" inside the row/column, so we need to look over if we find one which is selected Task-number: 252069 Reviewed-by: Marius Bugge Monsen
| * Don't send extra events when using the scroll wheel in CocoaNorwegian Rock Cat2009-04-281-2/+2
| | | | | | | | | | | | | | | | Argh! Copy and paste is evil, not only was the test was wrong, We sent the event twice and the second time we sent the wrong value. Task-number: 250668 Reviewed-by: Morten Sørvig
| * Cocoa: qwidget auto test fails (windowMoveResize)Richard Moe Gustavsen2009-04-282-11/+31
| | | | | | | | | | | | | | | | | | | | | | | | The test fails because, in cocoa, when resizing a window to zero (either w or h), cocoa will also move the window up in the corner (!). So the fix is to issue an extra move back to it's true location after the resize. The faulty cocoa move is issued inside the resize callback, so we choose to not update the window location anymore from a pure resize callback. Task-number: 251895 Reviewed-by: Trenton Schulz
| * performance improvement of isUncRoot in qfsfileengine_win.cppJoerg Bornemann2009-04-281-2/+9
| | | | | | | | | | | | | | We always called QStringList::split in this function, which was just expensive. Reviewed-by: mauricek
generated by cgit v0.12 at 2025-02-04 21:18:35 (GMT)