| Commit message (Collapse) | Author | Age | Files | Lines |
|\
| |
| |
| |
| |
| |
| |
| | |
Conflicts:
src/gui/graphicsview/qgraphicsitem.cpp
src/gui/graphicsview/qgraphicsitem.h
src/gui/graphicsview/qgraphicsscene.cpp
tests/auto/qgraphicsitem/tst_qgraphicsitem.cpp
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
This feature is essential for Declarative UI, but does not add much
value for C++ developers. A FocusScope provides a stack of focused
widgets, and it ensures that the topmost item on the stack has
focus if any of the items in the stack gains focus. When the topmost
loses focus, focus is passed to the "parent" focus scope, and so on.
You can get almost the same behavior using panels (ItemIsPanel),
except panels impose other behavior, like stopping clickfocus
propagation, and stopping event propagation in general. In a QML
world you would typically use FocusScope for controlling focus
locally, and panels when you need to maintain separate focus stacks.
Reviewed-by: akennedy
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Make sure you can't set focus on an inactive scene, allow items
with subfocus to gain focus even if added to a scene that's not
active, and finally ensure that activating a panel in an active,
but unfocused scene, gives focus to the scene.
Also added a test that checks adding normal vs. panel items to an
active vs. inactive scene, and what happens if you initially say
the item should have focus.
Reviewed-by: TrustMe
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
This line was added to fix crashes when deleting items that had
a subFocusItem pointing to a child that was already deleted. This
bug was fixed by ebb1162f54a29baeccb71d1e283146892629518f. After
this, subFocusItem is always 0 at this point.
The original change was eb3d5a73148cd7206c6b3b6672ed47b44611f745.
Reviewed-by: TrustMe
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
The widget must be created before calling QInputContext::setFocusWidget.
Otherwise we run into an assertion. Yes, this only occurs in debug
configuration but its still annoying...
Cherry-pick of commit d6b8f81a2440e7a507ecbb1becd90ef284510787 from master.
Reviewed-by: thartman
Conflicts:
tests/auto/qwidget/tst_qwidget.cpp
|
| | |
|
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Bauhaus uses grabMouse to implement their own Drag and Drop mechanism.
Cocoa port never considered the grabber widget. All the mouse events
will now be routed to the mouse grabber widget.
Task-number: 261245
Reviewed-by: MortenS
|
| |
| |
| |
| | |
Reviewed-by: Trond
|
| |
| |
| |
| | |
Pointed out by Caio.
|
| |\ |
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
The bug appeared only when calling collidingItems right after setPos.
When calling setPos on a parent the sceneTransform is mark as dirty, so
when we paint the parent and its children if the scene transform of the
parent was dirty then we update all children sceneTransform. In our
case here, collidingItems call ensureTransform on one of the children
which go recursively to the top most dirty item and update the
sceneTransform. The problem is that all sibling children are not mark
their sceneTransform dirty so next paint will skip them (since the parent
is not dirty anymore).
Task-number:260711
Reviewed-by:bnilsen
|
| |/
| |
| |
| |
| | |
Reviewed-by: Samuel
Reviewed-by: Tom Cooksey
|
| | |
|
| |\ |
|
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Uses the primary key from the index in the query, not the resulting
location in the modified dataset.
Task-number: 222678
|
| | |
| | |
| | |
| | | |
Reviewed-by: Justin McPherson
|
| | |
| | |
| | |
| | |
| | |
| | | |
There should be no semi-colon after Q_DISABLE_COPY.
Reviewed-By: Trust Me
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Ammend commit 41dbc4406 and 21cf7b4c431742
We do not always get the mouseReleaseEvent (because the popups grabs the
mouse. So set mouseDown to false in the mouseMoveEvent if no buttons are
pressed.
We also need to call setCurrentAction(0) in the case we are hovering on
the empty area with no popup open. (in otder to remove the highlight)
The code is now much closer to the code before the two previous commit
Reviewed-by: Prasanth
|
| |\ \
| | |/ |
|
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
max_keycode wasn't retrieved.
Merge-request: 1308
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@nokia.com>
|
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
We have to munmap() what we have mmap()ed
Reviewed-by: Gunnar
Reviewed-by: Jeremy
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
This also creates a new header called QtTest/QtTestGui. This header is
not supported. You should #include <QtTest/QtTest>. But if you're
using a build system that is not qmake, QT_GUI_LIB isn't defined, so
it's possibly easier to #include this new header.
Reviewed-by: Trust Me
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
This allows #including qtestmouse.h and qtestkeyboard directly to get
mouse testing. All QtTest headers are still unsupported (you should
always include QtTest/QtTest) but we can still fix issues like this.
I have verified with the preprocessor that no GUI code is enabled by
including QtTest if QT_GUI_LIB isn't defined. This should not bring
the MinGW autotest failures back that the a250ca3a3c0 commit was
trying to fix.
Reviewed-by: Trust Me
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
When a button is pressed, a popup message is shown and the background
(everything, every items, behind the popup message) is faded to light
green.
Reviewed-by: Bjørn Erik Nilsen
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
An SVG element can now reference a gradient or solid-color defined
anywhere in the same SVG file.
Task-number: 245602
Reviewed-by: Trond
|
| | |
| | |
| | |
| | |
| | | |
Autotest: included
Reviewed-by: Bjørn Erik Nilsen
|
| | |
| | |
| | |
| | | |
Reviewed-by: Bjørn Erik Nilsen
|
| | |
| | |
| | |
| | | |
Reviewed-by: Bjørn Erik Nilsen
|
| | |
| | |
| | |
| | | |
Reviewed-by: TrustMe
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Change c72eaee91136bbe1a9fa99cdb0a7593bec60264b was wrong, we should
call ensureActive(), not syncState().
Task-number: 261113
Reviewed-by: Trond
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
In the QPF1 engine if the font does not exist, then the default
constructor of glyph_metrics_t is used which sets the x and y values to
the unlikely value of 100000. These glyph_metrics_t instances are not
suppose to be used in the text layout calculations.
Task-number: pending
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
When dumping the rectangle clip to the depth buffer due to needing to
intersect with a more complex clip, we didn't take into consideration
that writeClip will transform the path by the current matrix, whereas
the rectangle clip is in device coordinates. Thus, we need to map the
path by the inverse matrix.
Task-number: 260701
Reviewed-by: Trond
|
| |\ \ |
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
Previously it always returned the C locale's language, which is of not
much help for an application.
Reviewed-by: Morten Sorvig <msorvig@trolltech.com>
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
To allow fading between the original and the colorized version of the
pixmaps, a new strength factor is introduced, 0.0 means the filter
has no effect at all, 1.0 means full colorization.
Still missing is the non-raster implementation.
Autotest: included
Reviewed-by: Bjørn Erik Nilsen
|
| |/ /
| | |
| | |
| | | |
Reviewed-by: Bjørn Erik Nilsen
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
We need to call ensureActive() when save() is called, to make sure
systemStateChanged() gets called and updates the scissorTestEnabled
flag, so that we don't lose it on the next restore().
Task-number: 261113
Reviewed-by: Trond
|
| | |
| | |
| | |
| | | |
Reviewed-by: Trond
|
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
QWS uses alien widgets too, so we need the same logic as the other
platforms.
Reviewed-by: bnilsen
|
| | |
| | |
| | |
| | | |
Reviewed-by: bnilsen
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Say hello to QGraphicsAnchor, move the spacing (and removeAnchor)
functionality over to that class.
This also opens up for a cleaner API when we add support for size
policies or min/pref/max sizes for anchors.
Also remove
- addLeftAndRightAnchors()
- addTopAndBottomAnchors()
- addAllAnchors()
in favor of
- addAnchors(itemA, itemB, Qt::Orientations)
API change discussed with Caio and Andreas.
Reviewed-by: Alexis
|
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Regression introduced by commit 21cf7b4c431742
Task-number: 261199
Reviewed-by: Prasanth
|
| | |
| | |
| | |
| | | |
Reviewed-by: Trust Me
|
| | |
| | |
| | |
| | | |
Reviewed-by: Trust Me
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Change 93ba0035f4eadfaf7217d95f18a442d418a064b8 removed truncation of
a listview item's bounding rect to the viewport as documented in the
commit log. An unwanted result was that the style option passed to the
item delegates would sometimes contain a rectangle which was larger than
the delegated area. The direct result was that a QComboBox drawn with a
style with the SH_Combobox_Popup style hint in an RTL language would
align its item text to the right edge of a rectangle which was wider
than the popup menu, and thus only part of the text would be visible.
Task-number: 260974
Reviewed-by: Olivier
|
| | | |
|
| |\ \ |
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
OpenGL ES does not support GL_REPEAT wrap modes for NPOT textures. So
instead, we emulate GL_REPEAT by only taking the fractional part of the
texture coords in the fragment shader.
Task-number: 260982
Reviewed-by: Samuel
|
| | |\ \ |
|
| | | |\ \
| | | | |/ |
|