| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
|
| |
Rather than requiring an allocation of the "shared data" region,
QObject *'s should be stored directly in the data structure. This very
marginally, but measurably, improves QVariant performance.
Reviewed-by: Thiago Macieira
|
|
|
|
|
|
|
|
| |
There's a restriction to output only 255 characters per line.
The loop we had before was wrong and did some read operations somewhere
behinde the end of the message string.
Reviewed-by: thartman
|
|
|
|
|
|
|
|
|
| |
QProcess is supposed to be reentrant but was not on Unix. The
constructor of QProcessManager could be exectued several time when
QProcess is created. The construction is now protected by a mutex.
Task-number: 254246
Reviewed-by: Olivier Goffart
|
|
|
|
| |
mingw doesn't support format with %llu or %ll. So I used a QString
|
|
|
|
|
| |
moved static CRITICAL_SECTIONs to cpp files. In header they would be
declared more than once depending on what file includes those headers.
|
|
|
|
| |
getQApplicationPrivateInternal was not declared
|
|
|
|
| |
Reviewed-by: jesper
|
|
|
|
| |
Cocoa: Add support for native gestures
|
| |
|
|
|
|
|
|
|
| |
setCursorPosition() should be able to set the cursor at the end of the
string. This is a copy-paste error made when refactoring QLineEdit.
Reviewed-by: Gunnar
|
|
|
|
| |
We should not unlock too early, or node might get deleted.
|
|
|
|
|
|
|
|
|
|
|
| |
When selecting the text and setting the cursor position, we need to
emit the cursorPositionChanged signal. This seems to be a copy-paste
error from refactoring QLineEdit, as the function call is there in the
original version of setSelection(). The effect was two test breakages
in tst_QDateTimeEdit, because the section and cursor position of the
QDateTimeEdit would be out of sync.
Reviewed-by: Gunnar
|
| |
|
|
|
|
| |
Reviewed-by: olivier
|
|
|
|
|
|
|
|
|
|
|
|
| |
Changing a style on a widget should invalidate the widget's layout,
since the layouts spacings and margins may depend on the style.
One optimization could be to check to see if the spacing and margins
have changed due to the style change, but the old style is not available
in changeEvent, so we'll keep it simple for now.
RevBy: paul
Task: 256986
|
|
|
|
| |
Reviewed-by: trustme
|
|
|
|
|
|
|
|
|
| |
The mac style ignored the QAbstractSpinBox::NoButtons flag for the
SC_SpinBoxEditField sub-control, effectivly always making room for
the spinner buttons. Now it's in line with both the Cleanlooks
style and the Windows (common) style.
Reviewed-by: Jens Bache-Wiig <jbache@trolltech.com>
|
|
|
|
|
| |
We needed to call ensurePolished before updating the
action rects.
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Use the known factorization trick to speed-up quaternion multiplication.
Now we need only 9 floating-point multiplications, instead of 16 (but
at the cost of extra additions and subtractions).
Callgrind shows that the function now takes 299 instructions instead of
318 instructions, which is not a big win. However I assume the speed-up
has a better effect for mobile CPU, where multiplications are more
expensive.
Reviewed-by: Rhys Weatherley
|
|
|
|
| |
Reviewed-by: trustme
|
|
|
|
|
|
|
| |
Instead of ignoring _NET_WORKAREA, merge it with each screen geometry
to give a meaningful availableGeometry() for all screens.
Reviewed-by: denis
|
|
|
|
|
|
|
|
|
| |
Setting a minimum size to a value that is larger then the biggest
allowed widget size, we should costrain the widget to that max allowed
size as we did before. Change 6a2621b6832dbdd349f77cf1f3242b4a6ba3c740
broke it.
Reviewed-by: Bradley T. Hughes
|
|
|
|
|
|
| |
We should close the gestureinfo handle only we managed to open successfully.
Reviewed-by: trustme
|
|
|
|
|
|
|
|
|
| |
Two-finger panning will work on all QAbstractScrollArea based widgets, however
single-finger panning will be enabled only on some special widgets that always
wants it - for example we don't want it on QGraphicsView, and we might want it
on itemviews *sometimes*.
Reviewed-by: trustme
|
|
|
|
|
|
|
|
|
|
|
|
| |
Instead of having a protected setter for the state and forcing the application
developer to emit signals manually (which leads to misunderstanding - i.e. if
the started() signal should be emitted only once, or of the triggered() signal
should be emitted before the finished() signal, etc). So I've added an
protected updateState(state) function that sets the internal state and emits
appropriate signals depending on the old and new states.
Reviewed-by: Volker Hilsheimer
Reviewed-by: Richard Moe Gustavsen
|
|
|
|
|
|
|
|
| |
Moved the code that subscribes to native gestures on Windows to a private
function in QWidget which will check which gestures the widget is subscribed to
and enable native gestures as requested.
Reviewed-by: trustme
|
|
|
|
|
| |
Unfortunately mingw doesn't know about the IFileDialog interface.
So it might have issues on Vista with a file dialog with search enabled.
|
|
|
|
|
|
| |
QAbstractItemViewPrivate::selectionModel may be null.
Reviewed-by: olivier
|
| |
|
|\
| |
| |
| |
| |
| |
| |
| | |
Conflicts:
src/gui/dialogs/qfiledialog.cpp
src/gui/dialogs/qfiledialog_win.cpp
tests/auto/qicoimageformat/tst_qicoimageformat.cpp
tests/auto/qscriptqobject/qscriptqobject.pro
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
The HideNameFilterDetails was not taken in account for windows
file dialog. This patch fix that. For Windows Vista and later we use
now the new COMMON_ITEM_DIALOG API (in master) where it add always
the filter extension like *.txt so no point to take that flag in account.
Task-number: None, during testing
Reviewed-by: prasanth
|
| |
| |
| |
| |
| |
| |
| |
| | |
Note however you still need to -nomake examples -nomake demos
when configuring Qt.
Reviewed-by: Thiago
Task: 259179
|
| |
| |
| |
| |
| |
| |
| |
| |
| | |
If there is severals matching items but all of them are disabled, it
would have produced an infinite loop.
Block the loop once we reach an already matched item.
Task-number: 258949
Reviewed-by: Thierry
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
We never told Cocoa that it needed to redraw the window view
when a window was shown. This is implicit if the window is
shown for the first time, but needs to be done explicit
if you hide and show it again.
Task-number: 254672
Reviewed-by: bnilsen
|
| |
| |
| |
| |
| |
| |
| | |
Somehow forgot to hook up the blinking signal (in the new QLineControl)
to QLineEdit.
Reviewed-by: Trust Me
|
| | |
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
I added the documentation to QWeakPointer::data, but I forgot to
declare the class in the qdoc stub.
Also, apparently I cannot link to a specific overload of a given
function.
Reviewed-by: Volker Hilsheimer
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
With the previous commit, you could create a QWeakPointer from any
QObject-derived object. It's possible because QObject now has a
pointer to the QWeakPointer's d-pointer.
However, if you did:
QSharedPointer<QObject> obj(new QObject);
QWeakPointer<QObject> weak1(obj);
QWeakPointer<QObject> weak2(obj.data());
Then weak1 would shared d-pointers with QSharedPointer, but weak2
wouldn't. Also, weak1.toStrongRef() would work, but
weak2.toStrongRef() wouldn't.
This change makes QObject know where the d-pointer created by
QSharedPointer is, so weak2 would get the same d-pointer.
As a nice side-effect, you can check if a given QObject is shared by
trying to promote its QWeakPointer to QSharedPointer.
Reviewed-by: Bradley T. Hughes
|
| |
| |
| |
| | |
with d=value=0
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
The problem with QPointer is that it's a simple QObject*. So the only
way for QPointer to do what it's supposed to do is if the object it's
pointing to clears all QPointers when getting deleted. That means the
QObject must know each and every QPointer pointing to it. To make
matters worse, QPointers can be deleted while the object they're
pointing to also gets deleted. So deleting QObjects must do locking.
The solution to the QPointer problem is that both QObject and the
"QPointer" reference something outside the QObject. This way,
QObject doesn't have to lock anything to destroy itself: it's simply
setting a volatile integer to zero when it gets deleted. Since the
integer is outside the QObject, the integer is also refcounted. It's also
O(1), so there's no problem having as many "QPointer".
The two-atomic-ints structure is exactly what QSharedPointer and
QWeakPointer use internally. We just abuse this structure for QObject
needs, setting the strong reference count to -1 to indicate that it's
a QObject that cannot be managed by a QSharedPointer. But QWeakPointer
can still work and replace QPointer neatly.
Reviewed-by: Bradley T. Hughes
Reviewed-by: Jarek Kobus
|
| |
| |
| |
| |
| |
| |
| | |
Previously it was only switching between checked and unchecked.
Task-number: 259008
Reviewed-by: ogoffart
|
| | |
|
| |
| |
| |
| |
| |
| |
| | |
GCC has template friends, so I wasn't seeing the error that d is
protected. Thanks to Robert for pointing out.
Reviewed-by: Robert Loehning <robert.loehning@nokia.com>
|
| |
| |
| |
| |
| |
| |
| | |
If you only set the position, of the button without setting a geometry,
it did not work.
Task-number: 259226
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
The commit 31edb4d5a63b9b3c28cce1c4ca6eb99f62a09759 introduced
regressions in filters which breaks the mac native dialog, the non
native dialog with HideNameFilterDetails activated.
Task-number:259105
Reviewed-by:olivier
Reviewed-by:prasanth
|
| |
| |
| |
| |
| |
| | |
A non painted (ie. garbage) area could appear if you had a fixed size
dock widget and you moved the separator. The previous place where the
separator was painted was not updated
|
| |
| |
| |
| |
| |
| | |
This was disabled by patch bf305bc2e488ad4f08c49767246f31a81218991e
Reviewed-by: Thomas Hartmann
|
| |
| |
| |
| |
| | |
We were calling pixelMetric and sizeFromContents with a NULL styleoption
That crashes on mac and might also with other styles.
|
| | |
|
|\ \
| |/
| |
| |
| |
| | |
Conflicts:
src/gui/graphicsview/qgraphicsitem.cpp
|