| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
| |
Reviewed-by: MArtin Jones
|
|
|
|
|
|
|
|
|
| |
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)
|
|
|
|
|
|
|
|
| |
Broke signal handlers in QML
Task-number: QTBUG-12335
Reviewed-by: Aaron Kennedy
(cherry picked from commit ddb5e1eef379c7f32a594d91b00ff3514c46b62a)
|
|
|
|
|
|
|
|
|
|
| |
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)
|
|
|
|
|
|
|
|
|
| |
Ancestors of the item with focus should only report wantsFocus as true
when they are a FocusScope or a top-level item.
Reviewed-by: Aaron Kennedy
Reviewed-by: Yann Bodson
(cherry picked from commit 686fca1c78e6d4d2ba597dd75d982c76647c7707)
|
|
|
|
|
|
|
|
|
| |
This allows dragging a MouseArea that contains a clickable MouseArea,
for example.
Task-number: QTBUG-12323
Reviewed-by: Michael Brasser
(cherry picked from commit 5dd0dfcd7a079065f99c6149c15b58e69f302729)
|
|
|
|
|
|
|
|
|
|
|
| |
Otherwise a Behavior may mistakenly not be triggered. This situation
can arise when the property in question has been manipulated via the
property system, followed by a direct function call (which correctly
bypasses the Behavior), followed by a another change via the property
system.
Task-number: QTBUG-12295
(cherry picked from commit 64833c0a648211f3fe7547436f022edc0ceb51ac)
|
|
|
|
| |
(cherry picked from commit 5efd577b1aea64f422e08ca8d54e041fa4b20783)
|
|
|
|
|
|
|
|
|
| |
The boundingRect depended upon the image cache which may not become
valid until after boundingRect is called.
Task-number: QTBUG-12291
Reviewed-by: Michael Brasser
(cherry picked from commit 3aeafb4839f49f524f10eae65be27fd189d37060)
|
|
|
|
|
|
|
|
| |
We need to create our cached image in boundingRect() if it doesn't
already exist, to prevents painting errors.
Reviewed-by: Warwick Allison
(cherry picked from commit 6c3cf7e2c5868bc77ffa66059889eda3132531b3)
|
|
|
|
|
|
|
|
| |
having different contents
Task-number: QTBUG-12339
Reviewed-by: Martin Jones
(cherry picked from commit 0767c2ff719a35c16c72fac97e6ff612e8c71e21)
|
|
|
|
|
|
|
|
| |
... and percentage is useless.
Task-number: QTBUG-12282
Reviewed-by: Warwick Allison
(cherry picked from commit 78a01438e5a37fd1778924f73ca8bfa55960b0d0)
|
|
|
|
|
|
|
|
| |
height explicit
Task-number: QTBUG-12273
Reviewed-by: Michael Brasser
(cherry picked from commit 164b2b54922d87a44a60efe6dbbe2fa3b7716820)
|
|
|
|
|
|
|
| |
Also parent them back to the VIM to ensure correct destruction.
Task-number: QTBUG-12261
(cherry picked from commit 17cc89e918a969d85d2b46631ee8743288c7586d)
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The way we detect the end of a default argument does not take in account
template parametter.
It is unfortunatelly not trivial to do it properly without semantic information
So we will use heuristics and if the number of < matches the number of >
we consider it is a template. Or if we have a '=' we consider it is not a
template.
Task-number: QTBUG-12260
Reviewed-by: Roberto Raggi
(cherry picked from commit 4da1a3b63445c04d4ca4acae448e9b6b046938c3)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
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)
|
|
|
|
|
|
|
|
|
| |
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)
|
|
|
|
|
| |
Task-number: QT-3613
(cherry picked from commit 2c574ed1bf9943130fb9af5435b557a47e3c462b)
|
|
|
|
|
|
|
|
| |
QDeclarativeVisualDataModel::stringValue() did not handle
QList<QObject*> model types.
Task-number: QTBUG-12005
(cherry picked from commit b6251d31652232a6ac3a5dff54ad4229542089b1)
|
|
|
|
|
|
|
|
|
|
|
| |
The scale of the delegate was used to offset the item. This was a
bad way of making the item appear to be positioned correctly when
the default transform origin was the top-left. Now that transform origin
is center, it is obvious that it was a bad idea.
Task-number: QTBUG-12245
Reviewed-by: Michael Brasser
(cherry picked from commit 5572ec653fe735c4f413195c1ef34382aa8c6105)
|
|
|
|
|
| |
Replace the usages of Follows with Behaviors, update docs.
(cherry picked from commit b22a5e13003ff9f23b075284a1a1e0a6b0e46250)
|
|
|
|
| |
(cherry picked from commit 31cd8c15c0fdb2f08c2652609a57413a715d1b13)
|
|
|
|
|
| |
Task-number: QTBUG-10658
(cherry picked from commit 237bc693a527f9cb2b6bbe7c518018cbf56b0eb6)
|
|
|
|
|
|
|
|
|
|
| |
Although the wait is only specified as 50ms, it might still take longer than
that for Qt to deliver the actual mouse release event (especially on
embedded devices). The problem here is that in the meantime the auto-repeat
on the button might have been triggered.
Reviewed-by: Dominik Holland
(cherry picked from commit 1e9348f52f06feb355245a9fffd786562e76e15f)
|
|
|
|
|
|
|
|
|
|
| |
The problem was caused by the fragment id being inserted
in front of a cell spanning over several rows instead of
the next logical cell (or fragment_end) in the cells structure.
Task-number: QTBUG-11282
Reviewed-by: Simon Hausmann
(cherry picked from commit d3a6f124dde7732311ad9312ebf41997712fc6bb)
|
|
|
|
|
|
|
|
| |
This reverts commit b2a4c7f0142a48f60e7ec4fc5866917e3da8b7c3.
Unit test tst_qtexttable::QTBUG11282_insertBeforeMergedEnding
fails on mac and Linux, reverting for now.
(cherry picked from commit b2002d0814e18f7a50bdbf8d17c4bc1662fa70ee)
|
|
|
|
|
|
|
|
|
|
| |
The problem was caused by the fragment id being inserted
in front of a cell spanning over several rows instead of
the next logical cell (or fragment_end) in the cells structure.
Task-number: QTBUG-11282
Reviewed-by: Simon Hausmann
(cherry picked from commit b2a4c7f0142a48f60e7ec4fc5866917e3da8b7c3)
|
|
|
|
|
|
|
| |
Merge-request: 731
Task-number: QTBUG-4455
Reviewed-by: Peter Hartmann
(cherry picked from commit 5f6018564668d368f75e431c4cdac88d7421cff0)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
These classes create a new QWidget and assign it to a static pointer
which is never cleaned up. Using Q_GLOBAL_STATIC ensures that they are
deleted on library unload. This is really just a cosmetic change that
removes a leak warning - the real fix should be to find a way to not
use a new QWidget like this. It seems odd that QGraphicsLayouts, which
don't use QWidget in any other way, should depend on QWidget like this.
Task-number: QTBUG-10768
Merge-request: 741
Reviewed-by: Jan-Arve Sæther <jan-arve.saether@nokia.com>
(cherry picked from commit ea6a5146397b668bf535ee7249bd4262d6185234)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Crashed because tabFocusFirst could end up being a dangling pointer when
removing an item from the scene before deleting it.
When setting tabFocusFirst in fixFocusChainBeforeReparenting, we now
check that the item is in the scene. If it is not, tabFocusFirst is set
to 0.
Autotest included.
Task-number: QTBUG-12056
Reviewed-by: Alexis Ménard <alexis.menard@nokia.com>
(cherry picked from commit 7d09f690d5b4c56699092444665d1879deb86a6e)
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
When setting focus on an item, subfocus chain was previously updated
only for items with a common ancestor. We now make sure that this chain
is updated also for the previously focused item (possibly not sharing
any common ancestor with the newly focused item).
Autotest included.
Task-number: QTBUG-12112
Reviewed-by: Alexis Ménard <alexis.menard@nokia.com>
(cherry picked from commit b5b011a181bd187dcef5ee8f46a3ac2c5dc2e09c)
|
|
|
|
|
| |
Task-number: QTBUG-12167
(cherry picked from commit 9adc85fdfa0af2b6948408932188ee1b79247fa6)
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This may mean that the cursor is to the right of the width, so components should
cater for that with a margin if they are boxed / clipped.
TextInput used to try to account for right bearing (and left bearing incorrectly since
it added it on the right). This is removed. Potentially this means that for some fonts
the text repaints incorrectly on the left or right, but if that is the case Text and TextEdit
already had such a problem (undetected), and all will need fixing.
Task-number: QTBUG-11983
(cherry picked from commit 628159323c60c434a202b036ecbaf5e433c703e8)
|
|
|
|
|
| |
Task-number: QTBUG-11936
(cherry picked from commit cabdb16f5ea6458dec9a2ec3b70a01e498b27dbc)
|
|
|
|
|
|
| |
Task-number: QTBUG-12086
Reviewed-by: Michael Brasser
(cherry picked from commit 0fdf5122c72eb86d49cba2b69f80d3a9c5949da6)
|
|
|
|
|
|
|
| |
QDeclarativeImageProvider to provide pixmaps instead.
Task-number: QTBUG-11980
(cherry picked from commit 33b664eabe5d9621d71e5bf3bfcc46ea385a399c)
|
|
|
|
|
|
|
|
|
|
|
| |
synchronous loading of QImages. (QPixmaps can only be created in the
main thread so they will always be loaded synchronously).
This changes request() to requestImage() and adds requestPixmap()
for pixmap support.
Task-number: QTBUG-11989
(cherry picked from commit a258456bcb35ec4211751a702ea94a1881d82a07)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The QDeclarativePixmapCache was both slow, and very trickey to use correctly.
Many QML elements did not correctly cancel outstanding requests, which leads
to pixmaps leaking indefinately. Other elements, such as Text, were
subject to race conditions that meant they may never actually load all their
images.
QDeclarativePixmap is a single class than encapsulates the action of
fetching a pixmap, as well as the pixmap itself and the responsibility of
canceling outstanding requests. Rather than relying on Qt's pixmap cache
that doesn't cache all the information QML needs, QDeclarativePixmap
implements its own cache, that correctly degrades over time (unlike
QPixmapCache that can stop expiring items in some conditions).
Reviewed-by: Warwick Allison
(cherry picked from commit 09f07b98dfdaec2e48749768b967a48e588d3f7f)
|
|
|
|
|
| |
QTBUG-11933
(cherry picked from commit 76a1804b0fff9ffd092a551defe448d3e9d4346e)
|
|
|
|
|
|
| |
Task-number: QTBUG-11932
Reviewed-by: Martin Jones
(cherry picked from commit 5500ffeeaee3412272f0f4af844fbc1d4d78a3bb)
|
|
|
|
|
|
| |
Task-number: QTBUG-11790
Reviewed-by: Martin Jones
(cherry picked from commit 37792d20eeef870a0d695fc15bc943f20275be9a)
|
|
|
|
|
|
|
|
|
|
|
| |
in the Compiz window manager which leads to the QML viewer mainwindow not
being composited anymore (at least until the next resize).
Since we need to somehow switch between fixed size and freely resizable
views, we have to work around that bug using the layout constraint hints.
Task-number: QTBUG-11771
Reviewed-by: kkoehne
(cherry picked from commit 07321dfceed41c1851781841e03c4148da47e41e)
|
|
|
|
|
|
| |
Task-number: QTBUG-10297
Reviewed-by: Martin Jones
(cherry picked from commit 98bfc8b8db811eb902290dbe87660ce799a44c27)
|
|
|
|
|
| |
fixing QProcess::execute() revealed that this test has always been
broken on windows.
|
|
|
|
|
|
|
|
|
|
| |
this function enables starting subprocesses which need command lines
which cannot be constructed via the portable list-based api. such cases
would be programs which need quoting rules which diverge from the msvc
runtime.
Reviewed-by: joerg
Task-number: QTBUG-7620 (and various others which boil down to that)
|
|
|
|
|
| |
Task-number: QTBUG-230, QTBUG-5866
Reviewed-by: joerg
|
|
|
|
|
|
|
|
|
| |
8601-2004 standard.
Task-number: QTBUG-11623
Reviewed-by: Denis Dzyubenko
Reviewed-by: David Boddie
|
|\
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
4.7-integration
* '4.7' of scm.dev.nokia.troll.no:qt/oslo-staging-2: (51 commits)
Symbian on Linux: $QTDIR/bin is not necessarily in the path.
Updated Symbian def files with new EGL exports.
Added another missing EGL stub.
Fixed compilation on Symbian.
Added missing EGL stub function.
Got rid of unused variable compiler warning.
QGLWindowSurface support for partial updates via EGL_NOK_swap_region2
Adding func prototypes for EGL_NOK_swap_region2 extension.
Check for EGLSurface leak only when paint device is a QGLWidget.
Fixed bug in drawTiledPixmap when width of pixmap matches target rect.
Update def files for symbian
Improved performance of 16 bit memrotates using NEON instructions.
Use built-in iconv on Solaris if available
Export various symbols needed to make a custom GL graphicssystem.
Moc: fix compilation when templated types with multiple arguments are used.
Fixes the documentation of QGraphicsEffect::update().
support BGRA textures on SGX
Avoid calling time.elapsed() twice in abstract animation
Adds slowdownFactor to UnifiedTimer in abstract animation
Add the conversion in-place for QPixmap::fromImageReader() on raster.
...
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Each template argument need to be normalized separately. Else, the const
type of one argument may end up moved in the first argument.
(And then the compilation of the generated moc file would fail as the types
would not match)
Reviewed-by: Kent Hansen
Task-number: QTBUG-11647
|
| |\ |
|