| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This reverts commit fcfc19878a0a1a48194a786bba64da11606077d2.
I am happy that this commit fixed three bugs at once. But
Actually, I am not sure if QTBUG-15513 should be fixed at
this point. Fact is that the patch as it is would have changed
the point->pixels calculation back to how it was in Qt 4.6.
This means that the fonts which are defined with pointSize
would now (in Qt 4.7.4) suddenly be bigger than they were
in Qt 4.7.3. Imho this is unacceptable, as it would break
all layouts which were developed for Qt 4.7 apps, when
point size (instead of pixle size) was used.
I will need to fix QTBUG-17844 without fixing QTBUG-13009
If QTBUG-13009 will be fixed for 4.8 will be discussed.
|
|
|
|
| |
Reviewed-by: Jani Hautakangas
|
|
|
|
|
|
|
|
| |
As QTBUG-19880 highlighted, the old S60 naming is not suitable for
these classes anymore.
Task-number: QTBUG-19913
Reviewed-by: Jani Hautakangas
|
|\ |
|
| |
| |
| |
| | |
Reviewed-by: TRUSTME
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
When drawing such pixmaps (used by both the openvg and opengl graphics
systems) onto another pixmap or to a QImage, the performance was
sub-optimal due to missing and accidentally disabled support
specific to QVolatileImage. This is now fixed and drawing pixmaps into
a QImage is also made optimal by using the QS60PaintEngine for
QImage too. This will cause a 5-7x (or even up to 12x on certain
hardware and platform) increase in offscreen pixmap drawing
performance.
Task-number: QTBUG-19880
Reviewed-by: Jani Hautakangas
|
|/
|
|
|
|
|
|
|
|
|
|
|
|
| |
Due to the fact that native side re-opens keyboard before sending
rotation event, the original size of graphicsview before auto-translate,
is stored incorrectly. We store the height of graphicsview in previous
orientation. Now, if the window was maximized, graphicsview was drawn
correctly, since closing vkb would change the window state back to
maximized and thus force a resizing of the view. With fullscreen gv
this didn't happen. Fullscreen graphicsview thus needs a forced
resize back to fullscreen.
Task-number: QTBUG-19856
Reviewed-by: Miikka Heikkinen
|
|
|
|
|
|
|
|
|
| |
Switching orientation double-committed any preedit string in progress,
so added inputcontext reset to KEikDynamicLayoutVariantSwitch handling,
ensuring the string will get only committed once.
Task-number: QTBUG-19864
Reviewed-by: Sami Merila
|
|
|
|
|
|
|
|
|
|
| |
New Symbian root theme defines KAknsIIDQsnFrSetOpt as empty.
Drawing the frame through native API produces white box.
Since the frame is no longer used on the native side,
skip drawing it from style.
Task-number: QTBUG-19782
Reviewed-by: Tomi Vihria
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Unlike QFont::pixelSize(), which may return -1 if the font size was
defined in points, QFontInfo::pixelSize() always needs to return a
valid value.
c4ef479906f073fa84999eb950f00e264ebd4e8e which was a fix for
QTBUG-13009 tried to fix a similar issue, but failed to do that
properly, which resulted in QTBUG-15513 and QTBUG-17844.
This commit is supposed to fix all three bugs.
Task-Number: QTBUG-13009
Task-Number: QTBUG-15513
Task-Number: QTBUG-17844
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
When a new window is shown and splitview is open, native side claims
that available screen area for the application is only the screen area
above keyboard. However, since opening a new window, will eventually
lead to keyboard getting closed, the new window will look strange
occupying only top part of the screen.
Fix it so that when a new window opens and splitview is open, window
will still get its extent set to fullscreen area (minus native panes,
if available).
Task-number: QT-5103
Reviewed-by: Miikka Heikkinen
|
|
|
|
|
|
|
|
|
|
| |
SVG icons automatically resize to the size of the pixmap, so the
size of the icon pixmap needs to be what the platform expects. Added
custom pixel metrics for CBA icon size that can be utilized for this
purpose.
Task-number: QT-5115
Reviewed-by: Sami Merila
|
|
|
|
|
|
|
|
|
|
| |
The client area dimensions reported by native side only encompass
the screen not covered by virtual keyboard and ignore the status
pane. Fixed by recalculating proper client area in
handleClientAreaChange() using layout metrics.
Task-number: QT-5105
Reviewed-by: Sami Merila
|
|
|
|
|
|
| |
Example application for OpenGL shading language and QML. Added also a link from qml examples main documents labs section.
Reviewed-by: Kim Gronholm
|
|
|
|
|
|
| |
Added a link from qml element index, removed some unnecessary doc markings, added missing example image and some other minor fixes.
Reviewed-by: Kim Gronholm
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The original architecture of the QtNetwork bearer support hosted the
engines in the application's main thread, but this causes some problems.
If the QNetworkConfigurationManager is constructed in a worker thread,
then it is populated asynchronously without any notification when it is
done (the app gets incomplete or missing results)
Fixing that by restoring the earlier behaviour of using blocking queued
connections to wait for the lists to be populated caused a regression,
as some applications deadlock because the main thread is waiting on the
worker thread at this time.
By introducing a dedicated worker thread for the bearer engines,
QNetworkConfigurationManager can be safely constructed in any thread
while using blocking queued connections internally.
Task-number: QTBUG-18795
Reviewed-by: mread
|
|
|
|
|
|
| |
first""
This reverts commit daba0c0d588c55e3f1591ab8ce0ef0946d1447fd.
|
|
|
|
|
|
|
|
|
|
|
| |
Removal of StyleChange event sending to dialogs in to previous
optimization (QTBUG-17840) caused dialogs to no longer adjust their
position properly at orientation switch. Added sending of a dummy
resize event for dialogs in case of KEikDynamicLayoutVariantSwitch
to trigger the adjustment.
Task-number: QT-5101
Reviewed-by: Sami Merila
|
|
|
|
|
|
|
|
| |
Enlarge SSL readbuffer as per performance testing results.
Checked from Shane Kearns that larger buffer is OK.
Task-id: http://bugreports.qt.nokia.com/browse/QTBUG-18943
RC id: ou1cimx1#779022
|
|
|
|
|
| |
Task-number: QTBUG-19180
Reviewed-by: TRUSTME
|
|
|
|
|
|
|
|
|
| |
If input hints change call reset() to avoid duplication of predicted
text into input widget. Otherwise input widget will commit the
predicted word AND whatever the user has so far typed.
Task-number: QTBUG-19689
Reviewed-by: Miikka Heikkinen
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Remove ugly TemporarilyCachedBindOption and use QGLTexture
objects as texture pool entries instead of QGLPixmapData.
Make texture pooling totally Symbian specific, remove
VG like texture pooling code and use common texture binding
path which is used on other platforms also on Symbian.
QGLPixmapData should be only used to bind SgImage based
textures (will be implemented by another commit).
Task-number: QTBUG-19180
Reviewed-by: Samuel Rødal
|
|
|
|
|
|
| |
SGX GPU requires precision definitions for all variables. Additionally some unnecessary semicolons were removed.
Reviewed-by: Kim Gronholm
|
|
|
|
|
|
| |
This fix is originally submitted by George Tavares. APIENTRY is a Win32 macro defined as stdcall__ *
Reviewed-by: Kim Gronholm
|
|
|
|
|
|
|
|
| |
When symbol menu is opened, input context needs to be reset or
preedit string duplication may occur after symbol menu is closed.
Task-number: QTBUG-19528
Reviewed-by: Sami Merila
|
|
|
|
|
|
|
|
| |
There is no check if graphics view pointer is null in the SLOT
function translateInputWidget().
Task-number: QTBUG-19734
Reviewed-by: Miikka Heikkinen
|
|
|
|
|
|
|
|
| |
Make non-native QMenus, i.e. context menus, work like in Avkon.
Reviewed-by: Miikka Heikkinen
Reviewed-by: Sami Merila
Task-number: QTBUG-19656
|
| |
|
|
|
|
| |
Reviewed-by: TRUSTME
|
|
|
|
|
|
|
| |
Make two autotest cases pass on Symbian/VGA devices.
Task-number: QT-5057
Reviewed-by: Miikka Heikkinen
|
|\ |
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Pixel metrics for focus frame (PM_FocusFrameVMargin and
PM_FocusFrameHMargin) had invalid values for VGA screens. Additionally
these same pixel metrics were adjusted for nHD.
Now, all but one QTreeView test cases pass for either screensize.
Task-number: QT-5056
Reviewed-by: Tomi Vihria
|
|/
|
|
|
|
|
|
|
|
|
| |
Reading incoming data in Symbian is slow when it is done from the native RSocket in 1kB blocks.
Typically other native apps use 16 kB or even 24 kB (browser).
This contributes (among other slow tasks such as writing to mass memory) to the TCP window filling up.
This case, even though it should be recoverable, has proven to be problematic in some scenarios including downloading from Ovi Store via 3G.
The fix just increases the amount of data read, which speeds things up and makes the problematic window fill-up less common.
Reviewed-by: Shane Kearns
Task-number: QTBUG-18943
|
|
|
|
|
|
|
|
|
|
|
| |
This reverts commit 5f241ec1426447380b1e938ac7888fb16cde94f8.
Reason for reverting: Some already published applications suffer
from the deadlock behaviour, causing regressions.
Conflicts:
tests/auto/qnetworkconfigurationmanager/tst_qnetworkconfigurationmanager.cpp
|
|
|
|
|
|
|
|
|
| |
Some devices have bigger than expected minimum width for scrollbars,
which broke the sorting test case as some objects that were expected to
be painted were fully obscured by scrollbars and therefore skipped.
Task-number: QT-5048
Reviewed-by: Janne Koskinen
|
|
|
|
|
|
|
|
| |
Minimum section size is bigger than expected in VGA devices, which was
not taken into accound in two test cases.
Task-number: QT-5049
Reviewed-by: Sami Merila
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
QSymbianGraphcisSystemEx::hasBCM2727() uses
bool QApplicationPrivate::useTranslucentEGLSurfaces to decide if
Symbian is running on BCM2727 chip which is not entirely correct.
bool QApplicationPrivate::useTranslucentEGLSurfaces should be
assigned according to QSymbianGraphcisSystemEx::hasBCM2727()
and QSymbianGraphcisSystemEx::hasBCM2727() should be
also static function.
Task-number: QTBUG-19578
Reviewed-by: Laszlo Agocs
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Effectively three fixes to reach same autotest results as on other
Symbian devices and to make the case not crash (itself, or Qt).
a) Fix null pointer usage in QWidget.
This is mostly theoretic case, since it requires that previous focus
widget has widget, yet it doesn't have internal winId. Still test case
manages to make this happen, so lets prevent the null pointer use.
b) Skip activatedCount_noQt3() test case,
since it would require shortcut support and leads to test crash.
Qt for Symbian should have shortcut support as a result of task
http://bugreports.qt.nokia.com/browse/QTBUG-5730
c) Ensure that menu has at least 360 width in tests that send keypresses
to the menu. Otherwise, menuitems might get set into menu extension,
which makes highlight tests fail (since item is not visible).
Task-number: QT-5053
Reviewed-by: Tomi Vihria
|
|
|
|
|
|
|
|
| |
Contains check in mmf.pro was spelled in upper case, but qmake's
internal contains function isn't case-insensitive and needs to be
spelled in lower case.
Reviewed-by: Miikka Heikkinen
|
|
|
|
|
|
|
|
| |
In E6 device, QToolButton autotest trigger() fails due to mouse
click not hitting the context menu at all.
Task-number: QT-5055
Reviewed-by: Miikka Heikkinen
|
|
|
|
|
|
|
|
|
|
| |
In Symbian QDesktopWidget::availableGeometry() is not guaranteed to
return correct values at the time the resize event related to the
geometry change is passed to the widgets. There is a similar issue
with QDesktopWidget::screenGeometry(). Documented this fact.
Task-number: QTBUG-14058
Reviewed-by: Sami Merila
|
|
|
|
|
| |
Task-number: QT-5065
Reviewed-by: Sami Merila
|
|
|
|
|
|
|
|
| |
QFileDialog and QSoftkeyManager broke in S60 5.3 platform because
incorrect version ifdeffing.
Task-number: QT-5065
Reviewed-by: Sami Merila
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
QCoeFepInputContext is very aggressive committing its preedit string.
When AVKON FEP opens any of its subwindows, it steals the focus from
editable widget, which causes preedit string to get committed.
This makes the input context almost unusable with T9 word prediction.
As it is rather difficult to prevent focus loss in these types of use
scenarios, replace the committed string with user selected word when
suggested word list is closed.
Task-number: QTBUG-15031
Reviewed-by: Miikka Heikkinen
|
|
|
|
|
|
| |
These are additional fixes for CI-issues caused by db20b6c03b6a93ab3e483cd85d5d0a923c3d3430
Reviewed-by: Kim Gronholm
|
|
|
|
|
|
| |
These are fixes for CI-issues caused by db20b6c03b6a93ab3e483cd85d5d0a923c3d3430
Reviewed-by: Kim Gronholm
|
|
|
|
|
|
|
|
| |
The patch applies everything from the original, except for the
filename case changes in LIBS which are handled differently
based on the auto-detected is_using_gnupoc CONFIG value.
Reviewed-by: Laszlo Agocs
|
|
|
|
|
|
| |
The akndoc.h was moved to epoc32/include/mw.
Reviewed-By: axis
|
|
|
|
|
|
|
|
| |
This issue is about backporting Scenegraph's ShaderEffectItem and ShaderEffectSource elements into Qt Quick 1.1 as a Qt labs plugin.
Purpose of these elements is to provide an interface for utilizing OpenGL shaders in QML applications.
Task-number: QTBUG-18346
Reviewed-by: Kim Gronholm
|
|
|
|
|
|
|
|
|
|
| |
Changing the library names to lowercase breaks 5.0-based
Linux builds. It will be investigated if it is possible to
change the SDKs instead. If not, the patch will have to be
extended to use the original names for 5.0 and the lowercased
ones otherwise.
This reverts commit 5933d4e4fb8b48ebed641e7f6b1d032df253df30.
|