| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
| |
Just a few small fixes to get things compiling with the QScopedPointer
changes.
|
|\
| |
| |
| |
| |
| | |
Conflicts:
demos/demos.pro
src/gui/graphicsview/qgraphicsitem_p.h
|
| | |
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
This change ensures that mouse presses received by the scene when there
are no current mouse grabbers trigger hover event delivery. This is
useful when the scene only receives presses, and no mouse moves (e.g.,
disabling mouse tracking on the viewport, or on systems where the
mouse press is the first received event).
Reviewed-by: Michael Brasser
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Change f68fed3 introduced a few regressions in the QGraphicsWidget
autotests. It turned out those autotests relied on behavior that this
fix "fixed". The exact bugs were 1) that setting focus on a window
or a child of a window that isn't active will automatically give that
item focus, despite that its window is inactive (in contrast it should
just set up subfocus and give the item focus when the window is
activated), and 2) that adding a window to a scene that is active did
not immediately activate that window.
So one fix in the test and one in QGraphicsScene. The autotests were
modified so that the respective tests operate on an active scene (by
assigning the scene to an active view). The change in QGraphicsScene
ensures that the first window that gets added to an active scene that
does not have any active windows already, automatically gets activated.
Reviewed-by: Michael Brasser
|
| |
| |
| |
| |
| |
| |
| | |
Add using to prefer QObject::children() over the obsolete
QGraphicsItem::children() function.
Reviewed-by: Henrik Hartz
|
| | |
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
The idea of having separate rotationX/Y/Z, shearX/Y, etc.
methods in QGraphicsItem turned out to be not giving us
the flexibility we need and wanted.
The new code now implements a different scheme, where we
keep simple rotate (around z-axis), scale and
transformOriginPoint methods, but remove the other ones.
Instead we now have an additional list of QGraphicsTransform
object. QGraphicsTransform is an abstract class that inherits
QObject. Several specializations are provided and can be
used to transform (and through property bindings animate)
the item.
Reviewed-By: Andreas
|
| | |
|
| |
| |
| |
| |
| | |
Clarified that the format used in QImage::fromData() is the image format,
not the pixel format.
|
| |
| |
| |
| | |
to respective APIs.
|
| | |
|
| |
| |
| |
| | |
This closes task 235801.
|
| |
| |
| |
| | |
Reviewed-by: Trustme
|
| |
| |
| |
| |
| |
| |
| |
| |
| | |
When you have an icon set we do not show the text label,
but the previous code would still use the text for the size
hint calculation.
Task-number: 218836
Reviewed-by: ogoffart
|
|\ \
| |/ |
|
| | |
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Checkboxes would animate when first shown if they were checked.
This is unintentional and looks a bit odd in wizard for instance.
To fix this we simply check if the old state was set. Note that
this is safe because we will at least require the enabled state
flag to be set.
Task-number:253075
Reviewed-by: ogoffart
|
| |
| |
| |
| |
| | |
As pointed out on IRC, setTransform is used most frequently in code and in
an ideal world would be the only such function.
|
| |
| |
| |
| | |
Reviewed-by: TrustMe
|
| |\ |
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
The arrow was reported to be on the wrong side of the control.
Technically the arrow part seems to cover the whole rect on
Vista and Gtk+ but due to compatibility it is probably safer to keep
the old rects for now.
Task-number: 252857
Reviewed-by: ogoffart
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
QTransform and respective APIs should be used. Still some changes required
- Some references to QMatrix left in documentation
- Qt code uses QMatrix APIs (ie translationX)
Reviewed-by: Samuel
|
| | | |
|
| | | |
|
| | |
| | |
| | |
| | |
| | |
| | | |
Task: 258573
Reviewed-By: TrustMe
|
| |/
| |
| |
| |
| |
| |
| |
| |
| | |
Handle the case when insertText is called with no corresponding
keyDown.
This fix is for the Cocoa port.
Task-number: 147379
|
| |
| |
| |
| |
| |
| |
| |
| | |
Call [NSToolbar setShowsBaselineSeparator] on the (unified) toolbar
if the window contains tabs in document mode.
Task-number: 252660
Reviewed-by: Richard Moe Gustavsen
|
|\ \
| |/
| |
| |
| |
| |
| |
| |
| | |
Conflicts:
configure.exe
src/corelib/io/io.pri
src/corelib/io/qfilesystemwatcher.cpp
tests/auto/qfileinfo/tst_qfileinfo.cpp
tools/configure/configureapp.cpp
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
If you set this flag on an item, and descendant item that gains input
focus will become this item's focus proxy. This simplifies how focus
proxy items are assigned from QML; instead of binding the possible
focusProxy property to a named child widget, this assignment happens
automatically as you set the focus property of a descendant to true.
As part of this change, QGraphicsWidget::focusWidget behavior has been
improved and moved into QGraphicsItem. For example, if you set focus on
an item that it's part of a scene, it can gain focus once the parent
has been assigned (which is how object trees are built in QML).
Autotests are included.
Reviewed-by: Michael Brasser
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| |
| |
| |
| |
| | |
This reverts commit 1368c210ef9976f68eb9fb1c3e4dc14f4fa4edd2, which
accidentially reverted previous commits.
|
| | |
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
The problem was that on Windows, we would activate on WM_LButtonUp,
but a double click after activating will also generate a second
WM_LButtonUp. Hence we get three activations. The fix was basically
to filter out the second WM_LButtonUP, something we also do in
qapplication_win.cpp.
Task-number: 205499
Reviewed-by: denis
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
When opening a menu by shortcut on Windows, we would loose
the keyboard mnemonic when navigating around. This is incorrect
compared to native applications and somewhat inconvenient.
The fix is basically to enable the keyboard mode when shortcuts
are triggered, not only when the alt-key is pressed.
Task-number: 254496
Reviewed-by: denis
|
| |
| |
| |
| |
| |
| |
| | |
ends up on the right X11 screen.
Merge-request: 987
Reviewed-by: Olivier Goffart <ogoffart@trolltech.com>
|
| |
| |
| |
| |
| |
| | |
This makes the QPixmapCache autotest link and pass on Windows.
Reviewed-by: Trustme
|
| | |
|
| |
| |
| |
| |
| |
| |
| | |
Panther was the last reason for having this around. We don't touch this
code anywhere else in Qt. As a result it's orphaned and can be safely
removed. It truly is the end of an era, but it's definitely worth
celebrating. Quartz4Life!
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
There was an attempt to do this earlier, but it was a bit more complex
than it needed to be. We now do the update on show in Cocoa. Carbon
actually does it all for us, we just need to flip the bit. We may do the
updates to often, but it's better than not enough.
Task-Id: 195445
Reviewed-by: Denis
|
| |\ |
|
| | |
| | |
| | |
| | |
| | |
| | | |
The check in QPainter::checkEmulation was just plain wrong.
Reviewed-By: Eskil
|
| |\ \
| | |/
| |/|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Conflicts:
src/3rdparty/webkit/VERSION
src/3rdparty/webkit/WebCore/ChangeLog
src/3rdparty/webkit/WebCore/bridge/qt/qt_instance.cpp
src/3rdparty/webkit/WebCore/bridge/qt/qt_instance.h
src/3rdparty/webkit/WebCore/page/DragController.cpp
src/3rdparty/webkit/WebKit/qt/Api/qwebframe.cpp
src/3rdparty/webkit/WebKit/qt/ChangeLog
src/3rdparty/webkit/WebKit/qt/tests/qwebpage/tst_qwebpage.cpp
src/gui/painting/qpaintengineex_p.h
tools/linguist/lupdate/main.cpp
|
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Polygonal vector paths may have types==null, in which case this
would have crashed.
Reviewed-by: Eskil
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
The reason being that there was an assumption that any non-curved path
was a continous polyline. For paths with multiple subpaths in it
we need to split this up into multiple strokePolygonCosmetic calls.
Task-number: 257621
Reviewed-by: Kim Motoyoshi Kalland
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
We normally pad the clip rect with the size of the pen and miterlimit
to avoid this, but this didn't handle the case where there was a long
diagonal dash. We also need to multiply the padding with the longest
dash.
Reviewed-By: Tom Cooksey
|