| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
| |
And make it also work without a gui manager
|
| |
|
| |
|
|
|
|
|
|
|
| |
seperated the functionality into different files, and made the
paintdevice belong to the windowsurface. Everything really belongs to
the screen, since everything is full screen but, this makes it more on
par with the other plugins.
|
| |
|
| |
|
|
|
|
|
|
|
|
| |
This adds a getter to QPlatformGLContext for the "default shared
context". The setter is protected as it is the plugins responsibillity
to create it or not. QPlatformWindowFormat has also gotten a pointer to
a QPlatformGLContext for a non default shared context. Only implemented
shared contexts in testlite for now.
|
|
|
|
|
|
| |
Dont post the resize event and move event seperatly on the
windowsystemintegration event queue, but as one event which will be
processed in one go.
|
| |
|
| |
|
| |
|
|
|
|
|
|
|
|
|
|
| |
and moved it onto QBlittablePixmap making the class abstract. The
creator function was seen as to spesialised to be part of the
PlatformIntegration class, and belonged to the QBlittablePixmap class
anyway since that was the PixmapData type you would need to return to
use the blitter api. Also removed the cross platform examples as they
are more confusing than anything else. For usage of QBlittable look in
the directfb platform integration plugin.
|
|
|
|
| |
This cleans up after commit bc5f5d99078d9ca1f4fc80fc7d0fe753d5ef396a
|
|
|
|
| |
and cleaned up the headerfile
|
|
|
|
| |
This is exprimental and does not work with the glib eventloop.
|
|\ |
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
When setting focus on an item, subfocus chain was previously updated
only for items with a common ancestor. We now make sure that this chain
is updated also for the previously focused item (possibly not sharing
any common ancestor with the newly focused item).
Autotest included.
Task-number: QTBUG-12112
Reviewed-by: Alexis Ménard <alexis.menard@nokia.com>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Crashed because tabFocusFirst could end up being a dangling pointer when
removing an item from the scene before deleting it.
When setting tabFocusFirst in fixFocusChainBeforeReparenting, we now
check that the item is in the scene. If it is not, tabFocusFirst is set
to 0.
Autotest included.
Task-number: QTBUG-12056
Reviewed-by: Alexis Ménard <alexis.menard@nokia.com>
|
|\ \
| |/
| |
| |
| |
| |
| | |
Conflicts:
src/gui/image/image.pri
src/gui/image/qpixmapdatafactory.cpp
src/gui/painting/qgraphicssystem.cpp
|
| |\
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
4.7-integration
* '4.7' of scm.dev.nokia.troll.no:qt/qt-s60-public:
Minor modifications and optimizations to the Symbian HW glyph cache.
Convert to new EGL API.
Don't invert glyphs for the default VG font cache on Symbian.
Move QVGFontGlyphCache from source file into a separate header
Ensure glyphs are upright instead of upside-down.
Enable cross-process glyph cache for OpenVG graphics system on Symbian.
Factor out RSgImage -> VGImage conversion to separate function.
Add new file for Symbian specific VG stuff and add Symbian glyph cache.
Move QVGFontGlyphCache from source file into a separate header
Make cacheGlyphs() function (and the destructor) virtual.
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
This introduces a specialized OpenVG font cache for the Symbian
platform. By using RGlyphDataIterator we can create a VGImage to store
inside a VGFont without having to upload our own glyph image. This
works by utilizing RSgImage which can be used to share handles to
graphics memory across processes thus allowing glyph images to be
shared by multiple processes and reducing graphics memory usage.
Reviewed-by: Alessandro Portale
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Factor out the Symbian specific bits of QVGPixmapData into a separate
file and introduce our subclassed implementation of the VG font cache.
Also make our font cache a friend of the S60 font engine because the
cache needs to access the CFont member to create the glyph image.
Reviewed-by: Alessandro Portale
|
| |\ \
| | |/
| |/|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
* '4.7' of scm.dev.nokia.troll.no:qt/qt-qml: (65 commits)
Fix QTextDocument::markContentsDirty()
Replace 4.6 in all .qdocconf files
Resetting bindings through debugger interface
Fix crash with invalid role indexes
Make test pass and fix docs following removal of SpringFollow
Remove autotests of depracted element SmoothedFollow
doc improvements
Clean up at the end of each test.
Improve test reliability.
Fix crash
Follow -> Behavior
Added some documentation to spring animation
Fix spring animation
Update QtDeclarative def files
Fix exponential behavior of QTextCursor::removeSelectedText
Optimization: change signal/slot to function call
Don't show warning for attempts to load pixmaps asynchronously
Remove Image::pixmap property. QML applications should use
Update QmlChanges.txt
Don't double delete cancelled pixmap cache requests
...
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
The function markContentsDirty(from,length) is documented
to inform the document that it needs to lay out again
the part from from to from + length.
Trouble was that the function was implemented using
beginEditBlock()/endEditBlock(), which has the negative
side effect that it does increase the document's revision
number and emits all sorts of contentChanged() signals.
This did not matter from the one case where Qt uses the
method itself in QSyntaxHighlighter, because here a private
flag inContentsChange avoided the issue.
Without the change, we cannot implement semantic syntax
highlighting in creator without expensive extra selections
or triggering a full rehighlight via QSyntaxHighlighter.
Done-with: Roberto Raggi
Reviewed-by: Roberto Raggi
|
| | |
| | |
| | |
| | |
| | |
| | | |
Got introduced with change f62f6effab8d1551d8e5e5843dc478addee96de1
Reviewed-by: Roberto Raggi
|
| | |\
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
* '4.7' of scm.dev.nokia.troll.no:qt/qt-qml: (52 commits)
Update QtDeclarative def files
Fix exponential behavior of QTextCursor::removeSelectedText
Optimization: change signal/slot to function call
Don't show warning for attempts to load pixmaps asynchronously
Remove Image::pixmap property. QML applications should use
Update QmlChanges.txt
Don't double delete cancelled pixmap cache requests
Extend QDeclarativeImageProvider to support QPixmap loading and
Allow the debugger to modify method bodies
Micro optimization in QML date/time formatting functions.
Fix Symbian build
QML: Let the debugger now the name of embedded functions within a QML function
Added QDeclarativeSpringAnimation
docs
Add Symbian support for runtime.orientation property
Don't crash if drag.target has no parentItem
QDeclarativeText optimization.
Make Text, TextInput, and TextEdit all have the same size for the same text.
More generated images in anticipation of QT-3574
Remove deprecated Flickable.flickDirection
...
|
| | | |
| | | |
| | | |
| | | | |
This is an improved version of 62db6c18c7f1f60819783ed5e1340e9fc09e072e
|
| | | |\
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
* '4.7' of scm.dev.nokia.troll.no:qt/qt-qml: (35 commits)
Make Text, TextInput, and TextEdit all have the same size for the same text.
More generated images in anticipation of QT-3574
Remove deprecated Flickable.flickDirection
Revert "Fix exponential behavior of QTextCursor::removeSelectedText"
Prevent a recursive debug output loop when writing to the logger widget.
Fix performance of QTextDocumentPrivate::adjustDocumentChangesAndCursors
Convert QtDeclarative def files to use LF line endings
Update QtDeclarative def files
doc improvements
Changing currentIndex shouldn't cancel a flick unnecessarily.
Fix input methods for TextInput elements with key handlers
Document the QML enumeration basic type
Fix TextEdit with no color property defined is drawn with wrong color
Export QDeclarativePixmap
doc
Fix inconsistent reporting of module import errors when using versions.
Make declarative pixmap cache easier to use
Prepare for QTest persistent store for visual tests.
Loosen font-sensitive test.
Set correct license header.
...
|
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
This reverts commit 62db6c18c7f1f60819783ed5e1340e9fc09e072e, which
introduced a regression in tst_qtextpiecetable::checkDocumentChanged
|
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
As the changedCursors list grew large, the function used to spend an
extraordinate amount of time in QList::contains.
The fix removes the changedCursors list outright and replaces it with a
flag in QTextCursorPrivate.
Done-with: mae
|
| | | | |\
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
* '4.7' of scm.dev.nokia.troll.no:qt/qt-qml:
Make declarative pixmap cache easier to use
Prepare for QTest persistent store for visual tests.
Loosen font-sensitive test.
Set correct license header.
Work around QTBUG-11929
<br/> shouldn't trigger a new format range in QDeclarativeStyledText.
Add styled text layout benchmark.
Optimize QDeclarativeStyledText.
Don't write to the logger widget while the application is closing down.
Fix exponential behavior of QTextCursor::removeSelectedText
Fix test for N900 (ARM-specific SVG results).
docs - fix doc links, minor improvements
Fix TextEdit bitmap tests testing the wrong element
Make autotests work with qreal == float (in addition to double).
Fix TextInput text getting improperly clipped
fix doc links
Fix docs to link to the new QML Viewer page instead of the Runtime page
qmlviewer: Update the proxy factory when the user changes the proxy.
Don't accept *files* that exist when *directory* is required.
|
| | | | | |\
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
* '4.7' of scm.dev.nokia.troll.no:qt/qt-qml:
Make declarative pixmap cache easier to use
Prepare for QTest persistent store for visual tests.
Loosen font-sensitive test.
Set correct license header.
Work around QTBUG-11929
<br/> shouldn't trigger a new format range in QDeclarativeStyledText.
Add styled text layout benchmark.
Optimize QDeclarativeStyledText.
Don't write to the logger widget while the application is closing down.
Fix exponential behavior of QTextCursor::removeSelectedText
Fix test for N900 (ARM-specific SVG results).
docs - fix doc links, minor improvements
Fix TextEdit bitmap tests testing the wrong element
Make autotests work with qreal == float (in addition to double).
Fix TextInput text getting improperly clipped
fix doc links
Fix docs to link to the new QML Viewer page instead of the Runtime page
qmlviewer: Update the proxy factory when the user changes the proxy.
Don't accept *files* that exist when *directory* is required.
|
| | | | | | |\
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | | |
* '4.7' of scm.dev.nokia.troll.no:qt/qt-qml:
Don't write to the logger widget while the application is closing down.
Fix exponential behavior of QTextCursor::removeSelectedText
Fix test for N900 (ARM-specific SVG results).
docs - fix doc links, minor improvements
Fix TextEdit bitmap tests testing the wrong element
Make autotests work with qreal == float (in addition to double).
Fix TextInput text getting improperly clipped
fix doc links
Fix docs to link to the new QML Viewer page instead of the Runtime page
qmlviewer: Update the proxy factory when the user changes the proxy.
Don't accept *files* that exist when *directory* is required.
|
| | | | | | | |\
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | | |
* '4.7' of scm.dev.nokia.troll.no:qt/qt-qml:
Don't write to the logger widget while the application is closing down.
Fix exponential behavior of QTextCursor::removeSelectedText
Fix test for N900 (ARM-specific SVG results).
docs - fix doc links, minor improvements
Fix TextEdit bitmap tests testing the wrong element
Make autotests work with qreal == float (in addition to double).
Fix TextInput text getting improperly clipped
fix doc links
Fix docs to link to the new QML Viewer page instead of the Runtime page
qmlviewer: Update the proxy factory when the user changes the proxy.
Don't accept *files* that exist when *directory* is required.
|
| | | | | | | | |\
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | | |
* '4.7' of scm.dev.nokia.troll.no:qt/qt-qml:
Don't write to the logger widget while the application is closing down.
Fix exponential behavior of QTextCursor::removeSelectedText
Fix test for N900 (ARM-specific SVG results).
docs - fix doc links, minor improvements
Fix TextEdit bitmap tests testing the wrong element
Make autotests work with qreal == float (in addition to double).
Fix TextInput text getting improperly clipped
fix doc links
Fix docs to link to the new QML Viewer page instead of the Runtime page
qmlviewer: Update the proxy factory when the user changes the proxy.
Don't accept *files* that exist when *directory* is required.
|
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | | |
removeSelectedText adjusts all other cursors for every fragment
and block which gets removed. This becomes (for reasons yet
to be understood) exponential with loads of cursors (something
creator has for the semantic highlighting).
Done-with: Roberto Raggi
Reviewed-by: Simon Hausmann
|
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | | |
When reparenting an item into another, QGraphicsItem will fix the
subfocus chain by first removing subfocus from the previous ancestors,
and then optionally restoring the subfocus chain to the new ancestors
of the reparented item. This allows item trees to be created and
assigned a specific focus item before even added to a scene.
The QGraphicsItemPrivate::setSubFocus function did not take into
account that the root item might be in a different panel than the
reparented item, which is the case when reparenting one panel into
another. If so, setSubFocus must not mess up the new ancestor panel's
sub focus chain. Otherwise (former behavior), the new ancestor
panel's focusItem will point to the reparented panel's focus item. This
in turn breaks an invariant; one panel's focusItem can never point to
an item in another panel.
The bundled test case shows this point: Create two panels that each have
one child that sets focus. The first panel is added to the scene just
to prepare, and verify that it's activated as expected, and its focus
child immediately gains input focus. Then we create a second panel,
which also has a focus child (but it doesn't have focus as its panel is
not active). When we reparent the new panel onto the old one, we expect
the new panel to be activated, and the child focus item to gain input
focus. However, we also expect the parent panel's focus item to remain
intact, so that when we hide the child panel, and the parent panel is
reactivated, its original focus item regains focus.
Former behavior was that the parent's focus item still pointed to the
child panel focus item.
This change only has effect when reparenting to or from a panel.
Merge-request: 735
Reviewed-by: Yoann Lopes <yoann.lopes@nokia.com>
|
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | | |
The defaults were chosen to match major web browsers. See:
http://xhva.net/log/?p=160
Reviewed-by: Simon Hausmann
|
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | | |
Reviewed-by: Olivier Goffart
|
| |\ \ \ \ \ \ \ \ \
| | |/ / / / / / / /
| |/| | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | | |
4.7-integration
* '4.7' of scm.dev.nokia.troll.no:qt/oslo-staging-2:
Fixes the sqlite driver bug QTBUG-11904 (pointer aliasing)
Add test and fix style for the SSE2 implementation of ARGB32 conversion
Build fix, the header of QImageData did not declare QImageWriter.
SSE2 implementation of convert_ARGB_to_ARGB_PM_inplace for QImage
Moved primitive SSE2 painting utilities to qdrawingprimitive_sse2_p.h
Move logic for building SIMD extensions to gui.pro
Fix text drawing into alpha pixmap with opengl engine
|
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | | |
The commit beba018814b35c4bd032e6b9fa948e4bac34c59a introduce conversion
from ARGB32 to ARGB32_PM with SSE2.
This patch add a benchmark for this type of usage, and change the
name to use lowercase for SSE2 (style convention).
|
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | | |
Including this header in the SSE2 file broke the build because
QImageWriter was not declared.
|
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | | |
Merge-request: 725
Reviewed-by: Benjamin Poulain <benjamin.poulain@nokia.com>
|
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | | |
Merge-request: 725
Reviewed-by: Benjamin Poulain <benjamin.poulain@nokia.com>
|
| | |/ / / / / / /
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | | |
Enables SIMD files to be built outside of painting.pri by appending
files to SSE_SOURCES etc.
Merge-request: 725
Reviewed-by: Benjamin Poulain <benjamin.poulain@nokia.com>
|
| |/ / / / / / /
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | | |
Softkeys are updated via a compressable event that is posted via the
event loop. Since these events are not delivered immediately, there is
a chance that a call to handleCommand() could happen before the
softkeys have been updated which can lead to a crash if the previous
QAction's have been deleted already since the data structure used by
QSoftKeyManager is outdated. The likeliness of this is increased by the
fact that S60 commands are normally sent from the WSERV active object
which has a higher priority than the active object used by Qt's event
loop which means commands will preempt the event loop.
The fix is to introduce a flag that keeps track of pending update
requests and if a command is received while there are outstanding
requests, force the softkeys to be updated before handling the command.
Task-number: QT-3571
Reviewed-by: axis
|
| |/ / / / / /
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
The QTapAndHoldGestureRecognizer now can trigger also on mouse and
GraphicsSceneMouse events and all of them will cause the gesture start
event to be sent after the touch interval timed out.
Reviewed-by: Denis
|
| |\ \ \ \ \ \
| | |/ / / / /
| |/| | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
4.7-integration
* '4.7' of scm.dev.nokia.troll.no:qt/oslo-staging-2:
Revert "Character spacing when drawing a QPicture to a high DPI device."
|
| | |/ / / /
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
This reverts commit d0fb8557f3fc3e7c9305662d118228ceca9df72b.
This change breaks justified text drawing in QPrintPreview, so it's
reverted for now.
|
| |/ / / /
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
When drawPixmapFragments() is called with fragments that has invalid
source rects in it, then usually it causes the pixmap drawn on screen
to appear corrupted. However it has been reported that a crash can
occur (not reproducable locally) so by adding a warning in debug mode
only means that this can hopefully be caught at development time.
Reviewed-by: Trond
|