If we do not process the event, q->underMouse may return false in
QGraphicsViewPrivate::centerView (as the QEvent::Enter has not been
received yet.)
It was working before if by luck, the window appears under the cursor.
Change-Id: I9b4497683eae20915680297013a9c21fd6275f4b
Reviewed-on: http://codereview.qt.nokia.com/2385
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: Sergio Ahumada <sergio.ahumada@nokia.com>
Since we now have QBasicMutex as a POD, we can simplify the mutexpool.
This remove the call the the Q_GLOBAL_STATIC and some others tests that
are taking CPU cycles when activating a signal.
The QMutexPool class itself can't be simplified because its mutex are
recursive mutexes, and the size is dynamic. also it is harder to get all
the mutexes initialized to 0.
Change-Id: Ie781655635907d2ad620eb189099cba14638414f
Reviewed-by: Bradley T. Hughes <bradley.hughes@nokia.com>
Reviewed-on: http://codereview.qt.nokia.com/2171
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: Olivier Goffart <olivier.goffart@nokia.com>
And added a POD QBasicMutex. (QBasicMutex* can safely be
static_cast'ed to QMutex*)
The d pointer is not anymore always a QMutexPrivate.
If d == 0x0: the mutex is unlocked
If d == 0x1: the mutex is locked, uncontended
On linux:
if d == 0x3: the mutex is locked contended, waiting on a futex
If d is a pointer, it is a recursive mutex.
On non-linux platforms:
When a thread tries to lock a mutex for which d == 0x1, it will try to
assing it a QMutexPrivated (allocated from a freelist) in order to wait
for it.
Change-Id: Ie1431cd9402a576fdd9a693cfd747166eebf5622
Reviewed-by: Bradley T. Hughes <bradley.hughes@nokia.com>
Reviewed-on: http://codereview.qt.nokia.com/2116
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: Olivier Goffart <olivier.goffart@nokia.com>
This one-line change makes the binary search in QTextEngine::findItem
behave consistently with the linear search that it replaced in commit
acf678e57ed088f3e56a551cac6c7c3322005750. The new behavior seems to
cause crashes in kword (and perhaps other applications) by triggering a
logClusters assert, although I have been unable to create a unit test
that reproduces this.
Task-number: QTBUG-17209
Done-by: Dr. Robert Marmorstein <robert@narnia.homeunix.com>
Change-Id: I68b79f024e9836e1cc8b0f3514889120541eb2ea
Reviewed-on: http://codereview.qt.nokia.com/2343
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: Jiang Jiang <jiang.jiang@nokia.com>
QtDBus requires a QVariant argument to be passed to property reads and
writes. For performance reasons QtDeclarative does not do this. By
flagging the meta object as requiring this, QtDeclarative can do so only
required.
Task-number: QTBUG-15052
Change-Id: I032c946f079523f5f10217ed56642fb315265d9f
Reviewed-on: http://codereview.qt.nokia.com/2365
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: Martin Jones <martin.jones@nokia.com>
When a table that is related to in a QSqlRelationalTableModel gets
updated in some way (e.g. a new row, or the data is changed) then the
related model could not be updated without recreating the
QSqlRelationalTableModel.
Now, to get around this, select() can be called on the related model to
get it to be updated.
Task-number: QTBUG-7885
Reviewed-by: Charles Yin
Reviewed-by: Michael Goddard
Change-Id: Ic589e840234f3a809bcb112a807a87afe0bc25ca
(cherry picked from commit 2c60a4f67f9fb02f3b711fe749b2f293a07b4e02)
Reviewed-on: http://codereview.qt.nokia.com/2224
Reviewed-by: Charles Yin <charles.yin@nokia.com>
Mark the selection as dirty if an input method event contains a
selection and emit selectionChanged() if it's not emitted by
finishChange().
Task-number: QTBUG-19731
Change-Id: Ief6f06f40071f64dae4db0ba365676c059a39c7e
Reviewed-on: http://codereview.qt.nokia.com/2081
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: Andrew den Exter <andrew.den-exter@nokia.com>
Previously, the mac arch was only put into qconfig.pri for carbon
and cocoa builds. Make sure it is also available for qpa builds.
This is needed for projects which need to select architecture-specific
sources in their .pro file.
Change-Id: I5f72e3b699b11dafc4dae052620913f2b9b81d0a
Reviewed-on: http://codereview.qt.nokia.com/2313
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: Marius Storm-Olsen <marius.storm-olsen@nokia.com>
Test started to fail, complains about outdated SSL certificate.
Task-number: QTBUG-20582
Change-Id: I2622375d9bc7b446dbbc18f72403896c60a8dee1
Reviewed-on: http://codereview.qt.nokia.com/2307
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: Jonas Rabbe <jonas.rabbe@nokia.com>
And try to fail a bit more gracefully if it isn't.
Change-Id: I62e01c0536aa0a032940d6a9a5ccf5edcfeef221
Reviewed-on: http://codereview.qt.nokia.com/2109
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: Rohan McGovern <rohan.mcgovern@nokia.com>
Reviewed-by: Michael Goddard <michael.goddard@nokia.com>
Try a lot harder to remove the old $TARGET output, since make
clean isn't sufficient. Also fix a bug in program invocation
that was hidden because of the stale files.
Change-Id: I0a365409d81efb74c5836eaf9f129fd9b2cca77e
Reviewed-on: http://codereview.qt.nokia.com/2052
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: Rohan McGovern <rohan.mcgovern@nokia.com>
The timer id allocator doesn't need a paylod (hence T=void), but we want
to tune the allocation 'strategy.' We allocate a maximum of 6 blocks
(like before), but the first block is twice as large as before and is
not static writable anymore. The initial value is 1 (0 is not a valid
timer id), but otherwise the constants are the same as the defaults.
Change-Id: Ied49ff4d7a6a8d69bc8c7bfa5475e4111446659f
Reviewed-on: http://codereview.qt.nokia.com/2161
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: Olivier Goffart <olivier.goffart@nokia.com>
This is a generic implementation of the lock-free free list found in
qabstracteventdispatcher.cpp. Use next() to get the next free entry in
the list, and release(id) when done with the id.
This version is templated and allows having a payload which can be
accessed using the id returned by next(). The payload is allocated and
deallocated automatically by the free list, but *NOT* when calling
next()/release(). Initialization should be done by code needing it after
next() returns. Likewise, cleanup should happen before calling
release(). It is possible to have use 'void' as the payload type, in
which case the free list only contains indexes to the next free entry.
Autotest included.
Change-Id: Ifd12a961d47f3d76593c45061f72e55c9b80a43b
Reviewed-on: http://codereview.qt.nokia.com/2160
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: Olivier Goffart <olivier.goffart@nokia.com>
This is very much useful to be able to write things like
myTabWidget->tabBar()->setSelectionBehaviorOnRemove( QTabBar::SelectPreviousTab );
without subclassing QTabWidget.
Change-Id: Ic7c42709ea1086631d37f90f184b058c4b6e9601
Merge-request: 3
Reviewed-by: Olivier Goffart <olivier.goffart@nokia.com>
Reviewed-on: http://codereview.qt.nokia.com/2172
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
`load(qt_module_config)' clobbers DESTDIR.
Rearrange the order so that our setting of DESTDIR works as intended.
Change-Id: Id6f02e9fb55069fae9b2a75c9d0f51578b84f4d1
Reviewed-on: http://codereview.qt.nokia.com/2130
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: Michael Goddard <michael.goddard@nokia.com>
There has been a few reports of user confusion from the fact that the
constData and data functions return results in column-major format.
There is nothing in the doc anywhere that states this, and nothing
states that the class is especially for OpenGL which would give a clue
at least.
Change-Id: I3a9afde0fbeb8b9d2bcba6a387620b60a56774b9
Reviewed-on: http://codereview.qt.nokia.com/2066
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: Julian de Bhal <julian.debhal@nokia.com>
Check if the input method removes the selection and force emit
selectionChanged if it sets a new selection.
Task-number: QTBUG-19727
Reviewed-by: Martin Jones
Change-Id: Ic8ea1044d0917aac4e52368f431ac9e5c7db7c56
Reviewed-on: http://codereview.qt.nokia.com/2076
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: Andrew den Exter <andrew.den-exter@nokia.com>
A reversed selection will have the same resolved start and end
positions but a different cursor position so testing the end points
alone doesn't guarantee the selection is the same.
Task-number: QTBUG-19456
Reviewed-by: Martin Jones
Change-Id: I516e5a501ec878d673f21e54d688fd2d21b624ef
Reviewed-on: http://codereview.qt.nokia.com/2080
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: Andrew den Exter <andrew.den-exter@nokia.com>
The actual file defined in `sync.profile' and generated
by 'syncqt' is `qttestversion.h' instead of `qttestlibversion.h'
Change-Id: I90ed2d5b7cd9f98855a878c97afceced01e7e8b0
Reviewed-on: http://codereview.qt.nokia.com/2050
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: Jason McDonald <jason.mcdonald@nokia.com>
The headers are just skipped, information stored in them is ignored.
Merge-request: 824
Change-Id: I48f37757114ed83cd5c92cb3d5a43eeaca4b91b3
Reviewed-on: http://codereview.qt.nokia.com/2108
Reviewed-by: Kim M. Kalland <kim.kalland@nokia.com>
According to the C++ standard, there is no guarantee that
you can cast between function pointers and void pointers
without data loss (section 5.2.10-6).
Change-Id: I27f4d835e4c8ca8ecca0d76cfea9ce34491956bd
Reviewed-on: http://codereview.qt.nokia.com/1995
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: João Abecasis <joao.abecasis@nokia.com>
Putting the logic inside supportsTransformation() is a bit confusing and
that name is misleading.
Also move the same check in GL2 paint engine to the same place.
Change-Id: I182500a0ff375122e6be966b7ce2495c84d113d0
Reviewed-on: http://codereview.qt.nokia.com/2096
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: Eskil Abrahamsen Blomfeldt <eskil.abrahamsen-blomfeldt@nokia.com>
If QT_GUI_PASSWORD_ECHO_DELAY is defined in qplatformdefs.h with an
integer value in milliseconds, QLineEdit and TextInput will display
the last character entered unmasked for that delay period and then
mask the character as normal. If QT_GUI_PASSWORD_ECHO_DELAY is
not defined then the behaviour is unchanged.
Task-number: QTBUG-17003
Reviewed-by: Martin Jones
(cherry picked from commit f9e7aee2019d321edd655bfde7de43f20a106971)
Conflicts:
src/declarative/graphicsitems/qdeclarativetextinput.cpp
tests/auto/declarative/qdeclarativetextinput/tst_qdeclarativetextinput.cpp
Change-Id: I3683223189b7176e4ef5081ee315c95a0efb9cfe
Reviewed-on: http://codereview.qt.nokia.com/2060
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: Andrew den Exter <andrew.den-exter@nokia.com>
An extra script is added (qtmodule-configtests) which is currently
invoked from syncqt (with some derived parameters passed to it).
The module can optionally have an entry in the module's sync.profile
file in the form of a perl map of "test name" => parameters. Tests
can print an advisory message if they fail (e.g. "Install this
SDK/dev package"), or abort the syncqt process (e.g. mandatory
prereq missing). Also, if the test has a "requires(foo)" line
that results in it being skipped, this is also supported.
Change-Id: Ic3c820a488a0992c944994d4d7dc283da36742d6
Reviewed-on: http://codereview.qt.nokia.com/928
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: Sarah Jane Smith <sarah.j.smith@nokia.com>
Reviewed-by: Marius Storm-Olsen <marius.storm-olsen@nokia.com>
Before, it was the length + 1, to include the ending \0 (for historical
reasons)
Having it the actual length is more intuitive and less error prone
Also added QT_ASCII_CAST_WARN to QConcatenable<QByteArray>::appendTo
to show the warnig that convertion from ascii to qstring occurs.
Change-Id: Ie7c8552b6b4e7ccb393cb09f5f0ca9b00336c714
Reviewed-by: thiago
Reviewed-on: http://codereview.qt.nokia.com/1988
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: Olivier Goffart <olivier.goffart@nokia.com>
This requires a fix for QByteArrayLiteral to work too.
Change-Id: I3c2a50ad431d5b0c014a341e675fa54e7b206e70
Merge-request: 27
Reviewed-by: Olivier Goffart <olivier.goffart@nokia.com>
Reviewed-on: http://codereview.qt.nokia.com/1967
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
The commit 412ef92162 introduced a fix, but the
fix was incomplete. Fix it for good.
Change-Id: I3e7fbdb294f8e960fbbf2e830790750240ed813a
Merge-request: 30
Reviewed-by: Olivier Goffart <olivier.goffart@nokia.com>
Reviewed-on: http://codereview.qt.nokia.com/1991
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Makes it easier to track which library output is coming from,
when building Qt with -j > 1.
Change-Id: I9acda04e84014dc441e409a0b24b2f78762dcc1c
Reviewed-on: http://codereview.qt.nokia.com/1765
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: Joerg Bornemann <joerg.bornemann@nokia.com>
By default, syncqt will now compress the console output of what it's
doing. Increasing the verbosity level will make syncqt output the same
as before
Change-Id: I542072504f022f87997b4036eda5747a5da88839
Reviewed-on: http://codereview.qt.nokia.com/1764
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@nokia.com>
Also add a fixme, since the version headers are currently
being created in the source directory.
Change-Id: If6e9eeba854a1f35561b69518eb8739dc28a58be
Reviewed-on: http://codereview.qt.nokia.com/1763
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: Joerg Bornemann <joerg.bornemann@nokia.com>
Their paths may vary depending on configuration and
build/install state going forward.
Change-Id: I60c8bcd1dc6c2fb729bf84d211875a3d910d5e4a
Reviewed-on: http://codereview.qt.nokia.com/1761
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@nokia.com>
In debug_and_release builds, we have three passes, the glue
pass, and a pass for each build type. We only need to run
syncqt when we are in a glue pass, which is when
'build_pass' is not set. In non-debug_and_release builds,
build_pass is never set, so this condition works for all
build types, with multiple configurations or not.
Change-Id: I191da0df3ad422bb5fb8994391c2b2d2e23efc32
Reviewed-on: http://codereview.qt.nokia.com/1717
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: Joerg Bornemann <joerg.bornemann@nokia.com>
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@nokia.com>
If you ran qmake with 'qmake -r', default_pre.prf would only be run once
while default_post.prf would run for every sub-project.
This makes it more symmetrical and correct.
Change-Id: I1d096c38dffb16f1d256c511ed9e2912cfaefe66
Reviewed-on: http://codereview.qt.nokia.com/1716
Reviewed-by: Joerg Bornemann <joerg.bornemann@nokia.com>
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@nokia.com>
"default" was used a very long time ago, and it's time to let it go.
Change-Id: I230573ef778789f6e1a5a7df3543e660392da39b
Reviewed-on: http://codereview.qt.nokia.com/1746
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@nokia.com>
