| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
| |
The native border width need to be taken into account in QRenderRule::boxRect()
Task-number: QTBUG-6855
Reviewed-by: Benjamin Poulain
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* Move most of drawPixmaps to private
* Move most of stroke to private
* Remove dead code: context()
* Make optimiseForBrushTransform use xform type
* Use GLuint for uniform location
* Rename lastTexture -> lastTextureUsed
* Move qopengl2paintengine_cleanup_vectorpath to private
* Re-ordered declarations in header
* Remove dead temporaryTransform
* Rename use_system_clip -> useSystemClip
Reviewed-By: Kim
|
|
|
|
|
|
|
|
|
|
|
| |
The old code set and unset a temporary matrix for every draw command
when doing aliased rendering. Instead, we just use a flag to indicate
that updateMatrix() should add the offset. This means the offset is
added only once. When doing lots of small rendering operations on a
(non-multisampled) QGLWidget, this gives up to 72% performance boost
on the SGX.
Reviewed-By: Kim
|
|\
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
4.6-integration
* '4.6' of scm.dev.nokia.troll.no:qt/oslo-staging-2:
revert f36fb8b2b63b3734cc2bd66b329ca4fef1204845
Read QWS_SIZE from layer rather than screen
Get rid of no_cast_from_ascii warning
Docs: Amendments to latest changes to QImageReader docs.
images: Document QImageReader::loopCount behaviour for infinite loops
gif: Add unit test for QImageReader::size calls
gif: 10% speedup on microbenchmark on QImageReader
png: Avoid calling QImage::scanLine(int) from within a loop
QImageReader: Avoid errorString creation in the normal case
Fix text rendering on GL2 paint engine
|
| |\
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
4.6-integration
* '4.6' of scm.dev.nokia.troll.no:qt/oslo-staging-2:
revert f36fb8b2b63b3734cc2bd66b329ca4fef1204845
Read QWS_SIZE from layer rather than screen
Get rid of no_cast_from_ascii warning
Docs: Amendments to latest changes to QImageReader docs.
images: Document QImageReader::loopCount behaviour for infinite loops
gif: Add unit test for QImageReader::size calls
gif: 10% speedup on microbenchmark on QImageReader
png: Avoid calling QImage::scanLine(int) from within a loop
QImageReader: Avoid errorString creation in the normal case
Fix text rendering on GL2 paint engine
|
| | |\
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
4.6-integration
* '4.6' of scm.dev.nokia.troll.no:qt/oslo-staging-2:
revert f36fb8b2b63b3734cc2bd66b329ca4fef1204845
Read QWS_SIZE from layer rather than screen
Get rid of no_cast_from_ascii warning
Docs: Amendments to latest changes to QImageReader docs.
images: Document QImageReader::loopCount behaviour for infinite loops
gif: Add unit test for QImageReader::size calls
gif: 10% speedup on microbenchmark on QImageReader
png: Avoid calling QImage::scanLine(int) from within a loop
QImageReader: Avoid errorString creation in the normal case
Fix text rendering on GL2 paint engine
|
| | | |\
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
4.6-integration
* '4.6' of scm.dev.nokia.troll.no:qt/oslo-staging-2:
revert f36fb8b2b63b3734cc2bd66b329ca4fef1204845
Read QWS_SIZE from layer rather than screen
Get rid of no_cast_from_ascii warning
Docs: Amendments to latest changes to QImageReader docs.
images: Document QImageReader::loopCount behaviour for infinite loops
gif: Add unit test for QImageReader::size calls
gif: 10% speedup on microbenchmark on QImageReader
png: Avoid calling QImage::scanLine(int) from within a loop
QImageReader: Avoid errorString creation in the normal case
Fix text rendering on GL2 paint engine
|
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
as discussed at #qt-labs special dirs should be hidden on unix
to preserve qt3-like behavior
unhiding those dirs was not a good idea and now i do revert of my changes
Merge-request: 1894
Reviewed-by: Joerg Bornemann <joerg.bornemann@nokia.com>
|
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
When running the X11 emulator of DirectFB IDirectFBScreen->GetSize
returns the size of the entire X11 Display, not the actual DirectFB
window.
We can read this from the primary layer instead and this value is
correct both on devices and on the desktop.
Reviewed-by: Donald Carr <donald.carr@nokia.com>
|
| | | | | |
|
| | | | |
| | | | |
| | | | |
| | | | | |
Reviewed-by: Holger Hans Peter Freyther <zecke@selfish.org>
|
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
The gif and mng reader will return -1 as QImageReader::loopCount()
for infinite images. Document that this -1 is clashing with the
error handling of this method.
The documentation was mixing the default and error cases. In some
cases "Otherwise -1" and "Otherwise 0" was used. Change this to
mention the default coming from QImageIOHandler in the "Otherwise"
clause and document the error case with -1.
Add a test case to check that loopCount() returns -1 for infinite
animations.
Reviewed-by: Andreas
|
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
Verify that QImageReader::size is not consuming a frame and
can be called before each frame and that the last frame will
return a QSize of (-1x-1).
Task-number: QTBUG-6696
Reviewed-by: Andreas
|
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
On the microbenchmark of what is executed during
pure pageloading in regarding to QImageReader this
is an improvement.
Do not use the non const QImage::scanLine as this will
go through the QImage::detach all the time making
QImage::scanLine show up in the profile.
Introduce a FAST_SCAN_LINE that is doing the scanLine
calculation on local variables.
Reviewed-by: Andreas
|
| | | | |
| | | | |
| | | | |
| | | | | |
Reviewed-by: Andreas
|
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
Deploy the same idiom found in QIODevice::errorString(). Return
"Unknown Error" in case of am empty string. This is a tiny improvement
for QtWebKit.
Reviewed-by: Andreas
|
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
If you draw with a brush, then draw a pixmap, then draw with the
same brush, the GL2 engine wouldn't update the fragment shader to
use the brush. This is because it detected the brush hadn't changed
and so didn't need updating. To fix this, we now set the current
brush to noBrush when drawing an image/pixmap so the engine knows
it needs to update the GL state for the brush.
Reviewed-By: Kim
|
|\ \ \ \ \
| |/ / / /
|/| | | |
| | | | |
| | | | |
| | | | |
| | | | | |
4.6-integration
* '4.6' of scm.dev.nokia.troll.no:qt/oslo-staging-1:
Fix D-Bus marshalling of booleans in optimised code.
|
| |\ \ \ \ |
|
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
C++ booleans are 1 byte in size, but D-Bus booleans (dbus_bool_t) are
4 bytes. That means a boolean with a zero in the LSB byte but non-zero
garbage in the high bytes is a valid "false" in C++, but libdbus-1
will turn that to true when sending.
Task-number: QTBUG-7041
Reviewed-By: Trust Me
BT: yes
|
|\ \ \ \ \ \
| |/ / / / /
| | | / / /
| |_|/ / /
|/| | | |
| | | | |
| | | | |
| | | | |
| | | | | |
4.6-integration
* '4.6' of scm.dev.nokia.troll.no:qt/oslo-staging-1:
Use Mono instead of MonoLSB when writing pbm files.
Fix memory leak of QGLGlyphCoord objects in the OpenGL1 paint engine
Add QHostInfo benchmark
|
| |\ \ \ \
| | |/ / /
| | | | |
| | | | |
| | | | | |
Conflicts:
tests/benchmarks/benchmarks.pro
|
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
PBM files are MSB, not LSB - the pbm reader was correct,
but not the pbm writer.
Task-number: QTBUG-6937
Reviewed-by: Sarah Smith
|
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
Task-number: QTBUG-6936
Reviewed-by: Julian de Bhal
|
| | | | |
| | | | |
| | | | |
| | | | | |
Reviewed-by: Peter Hartmann
|
|\ \ \ \ \
| |/ / / /
| | | / /
| |_|/ /
|/| | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
4.6-integration
* '4.6' of scm.dev.nokia.troll.no:qt/oslo-staging-1:
Don't include dlfcn.h unconditionally.
Fix warning on 64-bit build with MSVC.
Add missing license headers to the new solaris-cc-stlport mkspecs.
FTBFS on hurd-i386 because since clock_gettime is not available there
Fix build on ARMv6 when NEON is detected.
Add two new mkspecs for SunCC with stlport.
|
| | | |
| | | |
| | | |
| | | |
| | | | |
Task-number: QTBUG-3344
Reviewed-by: Trust Me
|
| | | |
| | | |
| | | |
| | | |
| | | | |
Task-number: QTBUG-3792
Reviewed-by: Trust Me
|
| | | |
| | | |
| | | |
| | | | |
Reviewed-by: Trust Me
|
| |\ \ \
|/ / / /
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
4.6-integration
* '4.6' of scm.dev.nokia.troll.no:qt/oslo-staging-1:
FTBFS on hurd-i386 because since clock_gettime is not available there
Fix build on ARMv6 when NEON is detected.
Add two new mkspecs for SunCC with stlport.
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
qt4-x11 reverts to using gettimeofday, but it doesn't include
<sys/time.h>
where the latter is defined.
Debian bug #533526
Task-number: QTBUG-6961
Reviewed-by: Thiago Macieira <thiago.macieira@nokia.com>
|
| | | |
| | | |
| | | |
| | | | |
The compiler requires -mfpu=neon to enable NEON instructions.
|
| | | |
| | | |
| | | |
| | | | |
Untested, but not used anywhere yet, so this cannot cause any problems.
|
|\ \ \ \
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
4.6-integration
* '4.6' of scm.dev.nokia.troll.no:qt/qt-s60-public:
Minor optimization for QS60Style
Correcting white space in QS60Style
QComboboxes are squeezed in QVGA screens
Revert "Daylight savings time for Symbian."
Changed autodetection logic for stlport version and sqlite in Symbian
Setting background color to a QDialog doesn't work
QWidget with the window flag Qt::Dialog is not decorated as a dialog
Device flickers badly when orientation change occurs
Changed make sis only require .make.cache if QT_SIS_TARGET is not set.
|
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
Ask selected state only once in QS60Style when handling element
CE_TabBarTabLabel.
Reviewed-by: TrustMe
|
| | | | |
| | | | |
| | | | |
| | | | | |
Reviewed-by: TrustMe
|
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
ComboBoxes and SpinBoxes are squeezed in QVGA screens, due to
dodgy style calculations.
This fix ensures that sub-widget buttons for these widgets are always
square.
Task-number: QT-2158 (internal)
Reviewed-by: axis
|
| |\ \ \ \ |
|
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
To support building Qt as part of Symbian SDK where epoc32 is not yet
populated, changed the autodetection logic in determining stlport
version. Now, we assume that we want the new version, unless only the
old version exists on SDK already.
Sqlite binaries export autodetection is now skipped if CONFIG value
symbian_no_export_sqlite exists, allowing clean builds to explicitly
suppress exporting.
Task-number: QTBUG-6971
Reviewed-by: axis
|
| | |\ \ \ \
| |/ / / / /
|/| | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
4.6-integration
* '4.6' of scm.dev.nokia.troll.no:qt/qt-s60-public:
Setting background color to a QDialog doesn't work
QWidget with the window flag Qt::Dialog is not decorated as a dialog
Device flickers badly when orientation change occurs
Changed make sis only require .make.cache if QT_SIS_TARGET is not set.
|
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
Since dialogs use their own theme graphics for background, regular
method of checking the palette fails (as generic background brush is
probably unchanged). Therefore to check if the dialog background is
using default (=theme graphic), it needs to check the cache key value
of background - so that it matches with the one in stored theme
palette.
Task-number: QTBUG-5898
Reviewed-by: Janne Koskinen
|
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
Previously style casted widgets to QDialog to see if it is able to
draw dialog background theme graphic.
As a fix, we now query window flag from the widget and if it is
equivalent of Qt::Dialog, draw dialog background.
Task-number: QTBUG-5930
Reviewed-by: Janne Koskinen
|
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
Currently what happens when orientation changes is:
1. Application gets notification that status pane size has changed.
2. Qt resizes S60Data to new size, and sends resize event.
3. Application redraws itself. But style is still using incorrect
screen size internally, so background brush is incorrect. Redrawing
thus might showup white rect on-screen.
4. Style gets notification that orientation is changed and style then
deletes existing background brush.
5. Style creates a new background brush and sends events to all toplevel
widgets that style has changed
6. Widgets might draw themselves with new style background brush.
What this fix changes is:
1. When application first tries to redraw itself, style notices that
the background brush size does not match to active screen size.
2. Style immediately re-creates background brush
3. Since cachekey for pixmaps won't match, new background is not drawn
until after background brush has been updated to application palette.
4. Due to #1 style needs to remove deletion of background brush from
clearCaches to avoid deleting (and re-creating) background twice.
Task-number: QTBUG-6428
Reviewed-by: Janne Koskinen
|
| | |\ \ \ \
| | | |/ / /
| | |/| | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
4.6-integration
* '4.6' of scm.dev.nokia.troll.no:qt/qt-s60-public:
Changed make sis only require .make.cache if QT_SIS_TARGET is not set.
|
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
The .make.cache file contents are ignored if QT_SIS_TARGET environment
variable is set, so there is no point in requiring its existence in
those cases.
Task-number: QTBUG-4617
Reviewed-by: Janne Anttila
|
|\ \ \ \ \ \
| |_|/ / / /
|/| | | | /
| | |_|_|/
| |/| | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
4.6-integration
* '4.6' of scm.dev.nokia.troll.no:qt/oslo-staging-2:
Implement ScaledClipRect in the jpeg image plugin
Add support for ClipRect to the jpeg image plugin
Use libjpeg's builtin scaler for implementing setScaledSize()
Use 3x3 PMV matrices rather than 4x4 in the GL2 engine
Handle broken shaders better in the GL2 engine's shader manager
Add GLfloat[2][2] & GLfloat[3][3] uniform setters to QGLShaderProgram
Fix build with neon instructions enabled but not set in mkspec
|
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
Where possible, we convert the post-scale clip rectangle into a
pre-scale clip rectangle because it is more efficient to clip first.
Task-number: QT-2023
Reviewed-by: Sarah Smith
Reviewed-by: Daniel Pope
|
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
The jpeg is read and scanlines discarded until the clip region
is found (libjpeg doesn't support direct seeking). This is
faster than the previous approach of reading the entire jpeg
and then clipping.
Task-number: QT-2023
Reviewed-by: Sarah Smith
Reviewed-by: Daniel Pope
|
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
The libjpeg library has builtin support for scaling to 1/2, 1/4, and 1/8
the original size very quickly. Use this in the implementation of
setScaledSize() to get close to the desired size and then scale
with QImageSmoothScaler the rest of the way.
Task-number: QT-2023
Reviewed-by: Daniel Pope
|
| |\ \ \ \
|/ / / / /
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
4.6-integration
* '4.6' of scm.dev.nokia.troll.no:qt/oslo-staging-2:
Use 3x3 PMV matrices rather than 4x4 in the GL2 engine
Handle broken shaders better in the GL2 engine's shader manager
Add GLfloat[2][2] & GLfloat[3][3] uniform setters to QGLShaderProgram
Fix build with neon instructions enabled but not set in mkspec
|