| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Reviewed-by: Trust Me
(cherry picked from commit cb086eb5340c4c41efaf45373aa05c37e8aa974a)
Conflicts:
doc/src/index.qdoc
src/dbus/qdbusconnection.cpp
src/declarative/util/qdeclarativeanimation.cpp
src/gui/s60framework/qs60mainapplication.cpp
src/gui/s60framework/qs60mainappui.cpp
src/network/access/qnetworkrequest.cpp
|
|
|
|
|
|
|
|
| |
The reason is that we tried to fix the stacking order of child windows
while we fixed a crash. But the patch turned out to break an auto-test
Reviewed-by: prasanth
(cherry picked from commit 749ad26d255327c9fa3fe857a30983cc6dfef0d9)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This reverts commit ebbab30af417dfbf3df47dec15c0e2f8d6a30fa6,
which broke fill / outline consistency, and when trying to fix that
by rounding the fill the same way that broke rendering in Creator.
Unfortunately the X11 paint engine is too sensitive to changes, there
have already been tons of patches to make it as consistent as possible.
It's simply not possible to get the same rounding for both fill and
outlines and at the same time have consistency between fill and outline
(no holes or fill outside the outline), while using the integer based
Xlib rendering API. Hopefully in 4.8 we'll switch to raster and the X11
paint engine will be a legacy.
Reviewed-by: Trond
(cherry picked from commit 041a68007413a20a9a9c97d0f2f04f9e03428f67)
|
|
|
|
|
|
|
| |
This reverts commit ff405f5623d7ed18c881c097368e3e9afd2e9443.
Reviewed-by: Trond
(cherry picked from commit 3014b427aa8700890f9bd8ea2a0300fc152bf3d2)
|
|
|
|
|
|
|
|
|
|
|
| |
Change ebbab30af417dfbf3df47dec15c0e2f8d6a30fa6 broke outline / fill
consistency in the X11 engine. Since the positioning of lines is more
important we'll round the fill coordinates the same way.
The bug was visible in the filltest.qps arthur test.
Reviewed-by: Gunnar Sletta
(cherry picked from commit ff405f5623d7ed18c881c097368e3e9afd2e9443)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The reason is the cocoa method that adds a window as a child of another,
it ends up showing both the parent and the child window regardless.
And this could in some cases also cause a crash.
So we therefore need to be careful when calling that function, and be
sure that the parent is actually visible. In addition, addChildWindow
reset the stacking level of the child window, and made e.g. normal
child windows pop in front of tool child windows. This could easily
be seen in e.g. Designer.
Task-number: QTBUG-12866
Reviewed-by: prasanth
(cherry picked from commit cc6dc0aeefde881a95f5fea2b26f2f3d7bdc6e15)
|
|
|
|
|
|
|
|
|
|
| |
The problem was in the toggleDrawer function. We were not checking the
status of the last transition before setting the new transition, so
all the transition were treated as non-intentional.
Task-number: QTBUG-11373
Reviewed-by: Prasanth
(cherry picked from commit 725d8e061ededee9a5ddf0914aabd0f6aa2ee741)
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Modal dialogs were sometimes hidden behind other normal windows
on screen. The reason was that, upon going modal for a window, we
actually resat the stacking level on the window in a faulty
attempt to respect any stays-on-top window flags. This patch
makes sure we avoid doing that, and at the same time, ensures
we don't reintroduce the original bug as well.
Task-number: QTBUG-12841
Reviewed-by: cduclos
(cherry picked from commit 6d0d767e63df3c997d6d7a8cb6cd4bf8c0dfc835)
|
|
|
|
|
|
|
|
|
|
|
|
| |
Every time [NSApp setMainMenu:] is called, Cocoa will add the 'Special
Characters' item to the 'Edit' menu. Before adding a new entry it will
make sure that menu items list doesn't contain an item with the selector
'orderFrontCharacterPalette' & a 'nil' target. We need to return the
index for the first entry (we have QCocoaMenuLoader as target).
Task-number: QTBUG-12842
Reviewed-by: Denis
(cherry picked from commit b52da4ab2179079a6aff95891786d2f6598048b5)
|
|
|
|
|
|
|
|
| |
Also emit the signal at the very end, so people can rely on the resize
event to adjust some stuff in their item.
Reviewed-by:yoann
(cherry picked from commit 3ee89bc0830f69d44f272eff5a0c886bff33c92e)
|
|
|
|
|
|
|
|
|
|
|
|
| |
Neither of the Mac font engines have implemented minimum right bearing,
which will cause them to return 0 for the minimum value. This will
cause the right bearing to never be calculated prior to breaking
and thus never be a part of the breaking width. Since actually
implementing the functions is too time consuming for right now, we
will disable the optimization for the time being.
Reviewed-by: Olivier
(cherry picked from commit bb696453f6f87cad58982af3455335478fedb8cc)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
When a line break was detected, we would retain the value of the
right bearing for the character after the break instead of resetting it
to the right bearing of the previous value. This could in some cases
cause the bounding rect of the text to be wrong, and could cause
unnecessary horizontal scrollbars to pop up. It was especially visible
when using WrapAnywhere. Visible e.g. in the compile output in Creator.
Done-by: mae
Reviewed-by: Eskil
Reviewed-by: Lars
(cherry picked from commit c5fa9eb1cb02d979502e2c9918d752c6708fb406)
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Internally, QTextEngine use int to calculate most of the layout data
required. If a string longer than 2^24 is passed into either QTextEngine
or classes using it (QStackTextEngine, QPainter, QFontMetrics, etc.),
overflow will happen because the memory size required to allocate for
layout will become too large for int to handle. This patch will prevent
these cases and add error handling mechanism for relevant code.
Task-number: QT-3658
Reviewed-by: Eskil Abrahamsen Blomfeldt
(cherry picked from commit 05bb249c2ad3ee15eb205a806f8546c105683096)
|
|
|
|
|
|
|
|
| |
Negative RightBearing was wrongly taken into
account when calculating the line wrap
Reviewed-by: Eskil Abrahamsen Blomfeldt
(cherry picked from commit 975b1913e44128a3e9b9055f9bf2ff40d86adf2a)
|
|
|
|
| |
(cherry picked from commit 30d6f7ed29a2a5723387768cfe82a807a2724b8b)
|
|
|
|
| |
(cherry picked from commit fb6aa3b0769c33685ad38a13fe5ad0ca48679a93)
|
|
|
|
|
|
|
|
|
| |
validating and adding some french translations.
Plus all the fixes from Gabriel's extensive review.
Doc: Fix a typo
Reviewed-by: gabi
(cherry picked from commit 209c017333fea9827fd20c7f8d23721bd8a6ee43)
|
|
|
|
|
|
|
|
|
| |
If the item is removed from the scene in response to the FocusOut event
of the previous item don't continue attempting to set focus.
Task-number: QTBUG-12620
Reviewed-by: Michael Brasser
(cherry picked from commit 7f2c9d716cba375ee3e576513898a39b3f2c39a5)
|
|
|
|
|
|
|
|
|
|
|
| |
The problem is that the qt_arcs... function will return point_count = 0
when the input rectangle is empty which it is in this case. Because we
previously passed 13 there, were passing random memory to downwards in
the pipeline, which caused problems.
Task: http://bugreports.qt.nokia.com/browse/QTBUG-12516
Reviewed-by: mariusso
(cherry picked from commit 08d09a78d3faf8b211e78cb63285c35c5544b5fc)
|
|
|
|
|
|
|
|
|
|
|
| |
Qt has a run-time assert for implementing drawPolygon() in any subclass
of QPaintEngine, which will be hit if set font decorations on the font
used by QStaticText. Since we don't care about line decorations or any
other shapes drawn in the QTextLayout, we simply leave the
implementation empty.
Reviewed-by: Gunnar
(cherry picked from commit 43f37074391b338b250883dfeb4a16844ca759fd)
|
|
|
|
|
|
|
|
|
|
|
| |
The temporary cursor position was not being updated after committing the
input (changing the cleartext into a *), as a result of which, the next
keypress was mistaken for a multitap input which should replace the last
character.
Task-number: QTBUG-11673
Reviewed-by: axis
(cherry picked from commit 7acf2417bd2ae17c2e3c289c7caed84219e5fecd)
|
|
|
|
|
|
|
|
|
|
| |
In some cases the NSKeyDown event won't have fully constructed string
in the [NSEvent characters]. So we need to add a guard before accessing
the elements of this NSString.
Task-number: QTBUG-12474
Reviewed-by: Richard Moe Gustavsen
(cherry picked from commit 307601da8d3b64de7b7d83a6de0b5dac290b0eb6)
|
|
|
|
|
|
|
|
|
|
| |
Our stringToCMap implementation for CoreText that bypasses the shaping needs
to have basic support for font cascading / glyph substitution. We now use
CTFontCreateForString to determine the substitute font for missing glyphs.
Task-number: QTBUG-12411
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@nokia.com>
(cherry picked from commit 00fb79714162ea3e434cf66a87a70a4fc26b2b86)
|
|
|
|
|
|
|
|
|
|
|
|
| |
If a window has several child windows, it was not be possible to
click on a child window to make it stack in front of the other
children. The reason is that cocoa held a fixed stacking order of
the child windows (sub-windows). This patch will release, and then
reestablish, the stacking order of the children upon acitvation.
Task-number: QTBUG-11780
Reviewed-by: prasanth
(cherry picked from commit 44e11692549f1d6a4fed01066a06085dacfaa444)
|
|
|
|
|
|
|
|
|
|
|
|
| |
If this is not done, later deletion of the backing store may cause
a crash. If the backing store is an EGL surface, its destruction
includes a call to eglDestroySurface(), which triggers an exception
if the window handle passed as a parameter is no longer valid.
Task-number: QTBUG-10643
Task-number: QTBUG-11376
Reviewed-by: Jason Barron
(cherry picked from commit cc3c416682791e1a2fd61d9a473ba3b4715c502d)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
On the X11 paint engine, when rendering lines with float coordinates,
the lines were one pixel off if the decimal was > 0.5.
This fixes the WebKit bug https://bugs.webkit.org/show_bug.cgi?id=42248
Autotest by Yoann Lopes.
Reviewed-by: Simon Hausmann
Reviewed-by: Yoann Lopes
Reviewed-by: Andreas Kling
(cherry picked from commit ebbab30af417dfbf3df47dec15c0e2f8d6a30fa6)
|
|
|
|
|
|
|
|
|
| |
We were making the wrong assumption that the selectedFilter string
will be valid always, when calling functions like getSaveFileName().
Task-number: QTBUG-12461
Reviewed-by: Denis
(cherry picked from commit fb47a99e1d0020e8dcc4cc4f9c188ccd2fd5ee3e)
|
|
|
|
|
|
|
|
|
|
| |
Sharing a NOTIFY signal can cause binding loop warnings with no apparent
cause.
Task-number: QTBUG-12333
Reviewed-by: Aaron Kennedy
Reviewed-by: Michael Brasser
(cherry picked from commit 5d5feaa86f3933a1e89b53267953b86e2b0459ed)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
After looking at QTBUG-11338 and Qt's behavior, it became clear that
we needed more state tracking in the key event handling code, to be
able to handle all the different ways that we can get key events.
This inspired having three states for each used scan code, instead of
the old code, which only took the current scan code into account.
This should make Symbian behave identically to Linux.
Task: QTBUG-11338
AutoTest: N/A, platform specific code. There was lots of manual
testing on 5800, N97 and N95.
RevBy: Jason Barron
(cherry picked from commit 041570ca54824ee358593f112b38e26632c3ebbf)
|
|
|
|
| |
(cherry picked from commit 5a3246b198ced3634bd974919e78310d536be27e)
|
|
|
|
|
| |
RevBy: Trust me
(cherry picked from commit 998ba966215482048f7fa93908212b1ee4a0c357)
|
|
|
|
|
|
| |
QTBUG-11938 and QTBUG-10801
(cherry picked from commit 5bdb85e3b21bcf624eb1830aa32d027e0c6870fe)
|
|
|
|
|
|
|
|
|
| |
The native border would be drawn over the one drawn by the QStyleSheetStyle
Regression since 68a498284e206e2bb000bc35331c966f3790f607
Reviewed-by: Gabriel
Task-number: QTBUG-10785
(cherry picked from commit 0318f0d5a43a78831f3bf4cb043d0b2b41cd9987)
|
|
|
|
|
|
|
|
|
|
|
| |
It used to fix an issue during logout but it
seems to be fine now.
Related issue: QTBUG-6296
Task-number: QTBUG-10993
Reviewed-by: Morten Sorvig
(cherry picked from commit 4942890f85e5fa74bf8c37cf7f53f6f3eaf40596)
|
|
|
|
|
|
|
|
|
| |
This fixes a crash when a static QWidget is destroyed after the QApplication
object.
Task-number: QTBUG-12025
Reviewed-by: Zeno Albisser
(cherry picked from commit 475a3c275ba173cb130a838b6053cb45d405887e)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
QDesktopWidget on Symbian was not properly emitting the resized() and
the workAreaResized() signals. Qt was properly translating the system
event and sending the QResizeEvent to the widget, but the signal
emission is determined solely based on changes in the QVectors of
QRects outlining the screen dimensions. On Symbian, these vectors were
never being set and as a result the signals were never omitted.
The fix is to set the rects correctly in the init() function which is
called by the resizeEvent() handler. Also modify the accessor functions
to return the rectangles stored in the vector instead of querying for
them each time.
Task-number: QTBUG-11433
Reviewed-by: axis
Autotest: PASSES
(cherry picked from commit 61908af46700c7c1f2e59ce98daeadc2529fef1b)
|
|
|
|
|
|
|
|
|
| |
This commit also updates QPalette's documentation regarding the current
ColorGroup and operator==.
Task-number: QTBUG-697
Reviewed-by: ogoffart
(cherry picked from commit 24cdab32de2abd8669f281dd54c8da1124514915)
|
|
|
|
|
|
|
|
|
|
| |
The wantsFocus property has been renamed to activeFocus, to better
reflect its value. Reading and writing the focus property is also now
consistent -- this property represents focus within a scope. Other small
changes were made to keep things consistent with the new naming.
Reviewed-by: Aaron Kennedy
(cherry picked from commit 21806ff0921641b4e4d9d39721ab4ebeae74dddc)
|
|
|
|
|
| |
Task-number: QTBUG-12132
(cherry picked from commit d9f5b8742f84124c3c0e2bdbd7b860dce2bc2225)
|
|
|
|
|
|
|
|
|
|
| |
While generating the QDragMoveEvent, we were always reusing the last
drop action set by the user. This is not correct, we should copy the action
only if a valid action set before. This is the behavior on Windows.
Task-number: QTBUG-9486
Reviewed-by: Denis
(cherry picked from commit 88410105978d5fb4344207cd5d296057ced0f27d)
|
|
|
|
|
|
|
|
|
| |
QTextBlock::layoutDirection() doesn't exist, and the QStaticText
constructor no longer takes a size argument.
Task-number: QTBUG-12072
Reviewed-by: Fabien Freling
(cherry picked from commit 04bc2b1f37750a2afbb88521001758249a8df383)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This only happens when the ItemHasNoContents and ItemClipsChildrenToShape
flags are set. Problem is that items with no content are threated as 'dummy'
items, which means they are never drawn or 'processed' otherwise, so the
cached bounding rect is not reliable/usable. This means that in case of
changing the geometry of such items, its children always have to take
care of invalidating the occupied areas and the update can not be
clipped to the item's bounding rect.
Regression after commit: c1c7dbf2
Auto test included.
Task-number: QTBUG-11504
Reviewed-by: yoann
(cherry picked from commit 4df9c96e2c213c39924e22e02621b0c61e83f8fe)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
QSymbianBitmapDataAccess is used to provide access to the bitmap heap
in a manner that locks correctly on all platform versions. The
heapWasLocked variable was meant to protect against the case where
the heap is locked recursively. However, it failed to take into
account the case where the same QSymbianBitmapDataAccess object was
used to lock recursively. In this case the variable would be changed
to true on the second lock, which means that the lock would never be
released again.
This was fixed by making the access reference counted instead. Since
the bitmap heap lock is global, the refcount was made global as well.
Task: QTBUG-11129
RevBy: Jason Barron
AutoTest: Works again. It was hanging before this fix.
(cherry picked from commit d4cc1dcea5b4116767cfee0ec45bfba72dc011ff)
|
|
|
|
|
|
|
|
|
| |
When the sections were moved calling logicalIndex on what was already
a logical index messed up the hidden sections.
Task-number: QTBUG-12268
Reviewed-by: Gabriel
(cherry picked from commit 8517f787b798d9e300438404aab359de2acc0978)
|
|
|
|
|
|
|
|
|
|
| |
When calling showNormal on a window that was maximized, we must remove
the [X] button from the task bar. But only if it was added by
qt_wince_maximize.
Task-number: QTBUG-8408
Reviewed-by: Martin Petersson
(cherry picked from commit ad4aff6e2d188d88a2c6b4b692932adb08491d22)
|
|
|
|
|
|
|
|
|
|
|
|
| |
The [X] or cancel button in the task bar shall just "minimize" the
widget on Windows mobile. A press on this button results in a
WM_COMMAND, IDCANCEL message. Before this patch we just sent a
QCloseEvent to the widget, which had basically no effect.
Now, we're calling showMinimzed(), which is the desired behaviour.
Task-number: QTBUG-8408
Reviewed-by: Martin Petersson
(cherry picked from commit 2c1aafa47b8915ea8aae8da229d65086e42543b3)
|
|
|
|
|
|
|
|
|
|
| |
We're now showing the cancel button explicitly in the taskbar on
maximize, if the widget does not have the widget flags
CancelButtonHint and OKButtonHint.
Task-number: QTBUG-8408
Reviewed-by: Martin Petersson
(cherry picked from commit 5f76c2b168ded91835d5d161b738a5dc03556cf6)
|
|
|
|
|
|
|
|
|
|
|
|
| |
compilation with QT_NO_DEBUG_STREAM
Qt in debug need to stay binary compatible with Qt in release.
See also commit cbbd7e084c7e46fd906db26b13032b8368c59093 that
introduced the problem
Task-number: QTBUG-11510
(cherry picked from commit 633349982422fec92df4ed06da5d2becf788c494)
|
|
|
|
|
| |
This reverts commit 5488e9213e1163ff7cadffd32cf8c8b94e88dbf1. A fix for
the symbian build issue causing the revert is now available.
|
|
|
|
|
| |
This reverts commit 734d4aff280f7f272a6abe8ed02fe95e2a1b4e6d, which was
not accompanied by the necessary .def file update for symbian.
|