| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
|
|
|
|
|
| |
Property access to id objects is optimized at compile time, but we cannot do
that for QQmlPropertyMap instances (or generally fully dynamic types).
This issue was a regression against Qt 5.1
Task-number: QTBUG-35906
Change-Id: I759a1a899f6a3a1f6466282f455b289ad7451086
Reviewed-by: Albert Astals Cid <albert.astals@canonical.com>
Reviewed-by: Lars Knoll <lars.knoll@digia.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This is a regression against 5.2.0 (which didn't have this bug), due to
optimizations introduced in the stable branch after the release. The code path
for optimizing access to the members of C++ based singletons through the
regular meta-object properties would end up excluding access to enums when the
lookup happens at run-time. The run-time getter for the singleton itself would
return a wrapped QObject instead of a QQmlTypeWrapper, and only the latter
includes enums.
As QML based singletons (composite singletons) cannot declare enums, we can
continue to do fast lookups on these, but otherwise have to fall back to the
slower code path.
Task-number: QTBUG-35721
Change-Id: Icc66bdaf3572622cdb718f82b706e3204afa0167
Reviewed-by: Lars Knoll <lars.knoll@digia.com>
|
|
|
|
|
|
|
|
|
| |
Calculation of all the children of nodes in the dominator tree is now
calculated as a local variable right before computing the dominator
frontier. The effect is that they are not retained after their only use.
Change-Id: I83c962c691b78cb767708eb04cf30d3b7a760deb
Reviewed-by: Lars Knoll <lars.knoll@digia.com>
|
|
|
|
|
|
|
|
|
|
| |
Both type inference and the optimization pass do not do anything with
unconditional jumps. So, instead of adding them to the worklist and
later on removing them again, it’s faster to never add them in the first
place.
Change-Id: Ib81d43e9ea6df2b1a70e9dd1e9b9c29cb6d345d2
Reviewed-by: Lars Knoll <lars.knoll@digia.com>
|
|
|
|
|
|
|
| |
Task-number: QTBUG-34047
Change-Id: Idcce254f3594e1f7021705704dbe6a2330aa7e65
Reviewed-by: Lars Knoll <lars.knoll@digia.com>
|
|
|
|
|
|
|
|
| |
Replace hashes from basic-block to basic-block with vectors that
associate basic-block index to basic-block index.
Change-Id: I834ea3d825e4d2b02c075b1b0f080f5a65f41317
Reviewed-by: Fawzi Mohamed <fawzi.mohamed@digia.com>
|
|
|
|
|
| |
Change-Id: I4c4577edde96978a986606bf30fbb925f871bd42
Reviewed-by: Topi Reiniö <topi.reinio@digia.com>
|
|
|
|
|
|
| |
Task-number: QTBUG-35206
Change-Id: Ia6b4f743cd9bea937eb891ddd16776f975d42bcd
Reviewed-by: Jan Arve Sæther <jan-arve.saether@digia.com>
|
|
|
|
|
|
|
|
|
| |
Widget-based dialogs are allowed if the application is a
QApplication, but subclassing is OK too.
Task-number: QTBUG-35769
Change-Id: Iec329792d7c1d85adcb1c3542444f0e92792b20b
Reviewed-by: Liang Qi <liang.qi@digia.com>
|
|
|
|
|
|
|
|
|
|
| |
The example image scales an arrow, but uses an arrow to indicate that
the next image (an arrow) is a result of the first arrow being scaled.
Instead, use a circle.
Change-Id: I8de18d4c2f3cb309681f63c31b0c9f7255d1868c
Reviewed-by: Topi Reiniö <topi.reinio@digia.com>
Reviewed-by: Venugopal Shivashankar <venugopal.shivashankar@digia.com>
|
|
|
|
|
|
| |
Task-number: QTBUG-34100
Change-Id: Ic06ed2933b76704b1a298de826cbe6b8587fee76
Reviewed-by: Jan Arve Sæther <jan-arve.saether@digia.com>
|
|
|
|
|
|
|
|
|
|
| |
Couple of \snippet references in the ListModel docs
were pointing to files outside the exampledirs boundary.
Update those references to use the relative path.
Change-Id: Ibf9d0fda9e5280345b08cb66a4fe0db9142a42a6
Reviewed-by: Nico Vertriest <nico.vertriest@digia.com>
Reviewed-by: Topi Reiniö <topi.reinio@digia.com>
|
|
|
|
|
|
|
| |
"warning: No QML property group command found; using..."
Change-Id: Iafbdbc09cbd76bf81a5baf3a5a4fab843778b5cb
Reviewed-by: Topi Reiniö <topi.reinio@digia.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The existing HTTP url for logo.png was automatically
redirected to an HTTPS url which fails without SSL. So
updated the example to use different HTTP url.
Also updated another image URL, which is invalid now.
Task-number: QTBUG-35557
Change-Id: I793fc5ebda64b21f443da7db9ecda76ae951d5bf
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
Reviewed-by: Topi Reiniö <topi.reinio@digia.com>
|
|
|
|
|
| |
Change-Id: I1f51b7d043f49f2f2849eb370ac71366cbaa6da5
Reviewed-by: Alan Alpert <aalpert@blackberry.com>
|
|
|
|
|
| |
Change-Id: I4b58eee6243c92f67bdd9c8f9da36ef82a489c12
Reviewed-by: Alan Alpert <aalpert@blackberry.com>
|
|
|
|
|
|
|
| |
Also fixed some comments.
Change-Id: I4aedff84bdbf8de248025bc48805a859704bdd8a
Reviewed-by: Lars Knoll <lars.knoll@digia.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Before blocking to wait for the thread to finish ensure the thread
itself is not waiting for the main thread to complete some action
otherwise both threads will be waiting on the same wait condition that
neither is able to wake. Likewise ensure the shutdown action is
processed if the last scheduled event is currently being processed.
Task-number: QTBUG-35581
Change-Id: I483fc56d0d398493f9fa907f3ab203439bfb9221
Reviewed-by: Robin Burchell <robin+qt@viroteck.net>
Reviewed-by: Lars Knoll <lars.knoll@digia.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
When items prior to the first visible item are removed from the instantiated
items list the first item is moved forward by the size of the item removed so
it position remains correct relative to the visible items. But if the removed
item is the first instanitated item then it's size shouldn't contribute to the
offset as the next instatiated item is already offset by its size.
Task-number: QTBUG-33619
Change-Id: I05c33f505e2856afa08cd9cd89d8eae97c20679d
Reviewed-by: Alan Alpert <aalpert@blackberry.com>
|
|
|
|
|
|
|
|
| |
On high-resolution displays it's important to size the window so that
the content will fit. Also show the DPI conversion for pixel density.
Change-Id: Ie35852ecff0bb0ee5ab9b77c6c18e82cfc300448
Reviewed-by: Richard Moe Gustavsen <richard.gustavsen@digia.com>
|
|
|
|
|
| |
Change-Id: Iefda2d3611ed0f940255f1375ca4c356827fbc70
Reviewed-by: Topi Reiniö <topi.reinio@digia.com>
|
|
|
|
|
| |
Change-Id: I85b986c8b3dca6ccc872807a1d2e8b7b71688c0e
Reviewed-by: Topi Reiniö <topi.reinio@digia.com>
|
|
|
|
|
|
|
|
| |
Bionic pthreads reports too small stack size for main thread.
Change-Id: I3d33229e76101a847309c723d534844ffb2d2042
Reviewed-by: Andy Nichols <andy.nichols@digia.com>
Reviewed-by: Lars Knoll <lars.knoll@digia.com>
|
|
|
|
|
|
|
| |
Task-number: QTBUG-35603
Change-Id: Ia337845004290d0e01c391130bbf55ec31e52c9e
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
Reviewed-by: Alan Alpert <aalpert@blackberry.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
C++11 allows you to do = delete to the declaration. But some version
of Clang on Mac doesn't like it. I'm guessing it's a compiler bug in
an older version of Clang used in the Qt CI system. I can't reproduce
the issue, so no further investigation was done.
So just remove the "inline" keyword, which is what ICC 14.0 complains
about:
MathExtras.h(270): error #864: extern inline function "clampTo(double, T, T) [with T=long long]" was referenced but not defined
Change-Id: I3887ccb527b5995ffa7ece612f9c7d5a80e71079
Reviewed-by: Alan Alpert <aalpert@blackberry.com>
|
|
|
|
|
|
|
|
|
| |
It makes more sense to show the dialogs as items in the scene
rather than as undecorated windows.
Task-number: QTBUG-35561
Change-Id: I2fb29ed8eeb999b99829a9d5ffc088f401d5d602
Reviewed-by: Liang Qi <liang.qi@digia.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The Delegate.qml and view.qml files were not in the
scope of Qt QML exampledirs. These files were in the example
directory under quick, so had to make of copy of them in the
snippets directory and update the snippet path to get the snippets
in the html output.
Task-number: QTBUG-35445
Change-Id: Ic3fa19dbd36b91d76fbc497ab524f5c0ff91325f
Reviewed-by: J-P Nurmi <jpnurmi@digia.com>
Reviewed-by: Nico Vertriest <nico.vertriest@digia.com>
Reviewed-by: Topi Reiniö <topi.reinio@digia.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Having mouse events synthesised from both QtGui and internally in QtQuick is
not a great way togo about things, especially when QtGui doesn't have the same
degree of knowledge as QtQuick about the items in the scene.
Thus, we now accept all events inside QtQuick to block QtGui synthesis, which
should fix a significant amount of edge-case touch breakage/bad behavior.
Change-Id: I14e1c87761c8f43160049b5e6f9da15b4e5edbb7
Done-with: Martin Jones <martin.jones@jollamobile.com>
Reviewed-by: Shawn Rutledge <shawn.rutledge@digia.com>
|
|
|
|
|
|
|
|
| |
This makes time- and memory-complexity a lot better when compiling
big JavaScript functions.
Change-Id: I2a7cb9b5979844254747fa5cf7355cca0b113904
Reviewed-by: Fawzi Mohamed <fawzi.mohamed@digia.com>
|
|\
| |
| |
| |
| |
| |
| | |
Conflicts:
src/qml/qml/qqmlcompiler.cpp
Change-Id: I802731139d47c5b733dd805f7bf432d67d7331e1
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Ie76dbe5fd4ab935db3da34f3ff63d217a3ba5013 fails to set the format
for the QOffscreenSurface, resulting in BAD_MATCH failures when trying
to make it current.
Task-number: QTBUG-35410
Change-Id: I1d420556fad4df96a1893cb3513e398eeb6e71aa
Reviewed-by: Lars Knoll <lars.knoll@digia.com>
|
| |
| |
| |
| |
| |
| | |
Task-number: QTBUG-35412
Change-Id: I09feca09030a2eb07aa1abebf65481d68b880c79
Reviewed-by: Simon Hausmann <simon.hausmann@digia.com>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Properly set members that are actually array indices
and don't crash when trying to set those.
Task-number: QTBUG-35383
Change-Id: I04d4b65c27e97a2e9db19541ed46ee1bb202f780
Reviewed-by: Simon Hausmann <simon.hausmann@digia.com>
Reviewed-by: Milian Wolff <milian.wolff@kdab.com>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
toUtf8 would return a temporary, and constData would hold a pointer inside that
temporary. This isn't even remotely safe. Move the pointer use down to the
initializeEngine call so it is kept around long enough for us to do our stuff.
This is a backport of cf51cdb8fb002ae3602a4c886e7c67913d77373a.
Task-number: QTBUG-35355
Task-number: QTBUG-35343
Change-Id: I338ad7f4d4137445ed9a311a293ea82bf023aafd
Reviewed-by: Robin Burchell <robin+qt@viroteck.net>
Reviewed-by: Lars Knoll <lars.knoll@digia.com>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Add classname entry to qmldir, otherwise qmlimportscanner cannot
utilize it when doing static builds (on iOS).
Task-number: QTBUG-35369
Change-Id: If5440ffcdd805a832d4ff1d129e9f24d222b45d5
Reviewed-by: Morten Johan Sørvig <morten.sorvig@digia.com>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
This reverts commit 27052dcad9810869a9065da4c06e9f14379411d7.
While the additional flexibility would be nice, I've been reminded that
we already did commit to it back in July.
Change-Id: Iaf990dda98ee46eb028b4737bdeeafd050d9513f
Reviewed-by: Alan Alpert (Personal) <416365416c@gmail.com>
|
| |
| |
| |
| |
| |
| |
| |
| | |
Task-number: QTBUG-35210
Change-Id: I44bd9f19acba5b59711aa4ca3d2b12c246afcc59
Reviewed-by: Akseli Salovaara <akseli.salovaara@digia.com>
Reviewed-by: Alan Alpert <aalpert@blackberry.com>
Reviewed-by: Sebastian Wozny <swozny@blackberry.com>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
When shutting down, we left the gl context current on the window even
when it was hidden. On mac this was a problem as it would optimize
away our makeCurrent when the surface was made visible again, leading
to nothing being rendered. So we call doneCurrent regardless during
invalidateGL(). We also check and verify that makeCurrent in
invalidateGL() actually succeeds.
This lead to another problem which is that closing the app using [x]
will call QWindow::destroy() on all windows, removing their platform
windows and causing makeCurrent to fail. To still gracefully clean up
resources, we introduced the concept of an offscreen fallback surface
which is temporarirly used during the cleanup when needed.
The problem is still present on QWindow+QOpenGLContext level, and
I've opened QTBUG-35363 to fix this.
Task-number: QTBUG-35234
Change-Id: Ie76dbe5fd4ab935db3da34f3ff63d217a3ba5013
Reviewed-by: Laszlo Agocs <laszlo.agocs@digia.com>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
This reverts commit 707bbe5dea9d7398b205124a54422f2fafb6f151.
By itself this is completely broken and the image will
display at twice the size. We need to add code to handle
the @2x image but revert for now to get sane behavior.
Change-Id: Iccdb26e9d19930b5d0ef4b6c7f596c5766f77a8a
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@digia.com>
|
| |
| |
| |
| |
| | |
Change-Id: Idff1507fdce960ad7dabc2662bb273bc6103c1ca
Reviewed-by: Jerome Pasion <jerome.pasion@digia.com>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
When a Window element is used, the QML window item is deleted before
the ShaderEffectSource component and the deleteLater to delete the
texture gets handled too late. We now register for the
sceneGraphInvalidated signal and release resources synchronously
while shutting down.
Task-number: QTBUG-35294
Change-Id: Id83b669ddc16723551e5612264ccbad6d3a9bbcb
Reviewed-by: Laszlo Agocs <laszlo.agocs@digia.com>
Reviewed-by: J-P Nurmi <jpnurmi@digia.com>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
In debug mode fixes a crash. The next Q_ASSERT dereferrenced
QCoreApplication::instance().
In release mode fixes hundreds of messages about timers:
"QObject::startTimer: Timers can only be used with threads started
with QThread"
A case where this is reproducible is having a QQuickItem that's child
of qApp, or in some child hierarchy of qApp.
Change-Id: I3b506aefdb537b74dc0d6c755b6874d31be159a1
Reviewed-by: Gunnar Sletta <gunnar.sletta@digia.com>
|
| |
| |
| |
| |
| |
| |
| |
| |
| | |
If another GL context is bound to another surface on the GUI
thread, we can run into issues while cleaning up the SG nodes.
Task-number: QTBUG-34898
Change-Id: Ifa02b7cdbc7ab38b3a149a21452cc5071498a7d1
Reviewed-by: Laszlo Agocs <laszlo.agocs@digia.com>
|
| |
| |
| |
| |
| |
| |
| |
| |
| | |
A slight variance on this number is acceptable, it should be using
extremelyFuzzyCompare like the rest of the checks on that property.
Task-number: QTBUG-35470
Change-Id: I9114a3a5d3bbc3c63253320637f7af1e5f21a896
Reviewed-by: Robin Burchell <robin+qt@viroteck.net>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
String lists and other QList property types (wrapped as QQmlSequence) should
behave like arrays and have the Array prototype. Therefore it should be
possible to pass them also as parameter to concat and they get composed
correctly, i.e. the individual items get appended instead of the list being
appened as one item. In the spec for concat this "special" casing should be
applied if the "class internal property" is "Array", and concat appears to be
the only place where this check is done. Therefore this patch adds another
exception to match the expected behavior in QML and extends the "internal
class is Array" meaning to QML list types.
This is a regression from Qt <= 5.1.x
Task-number: QTBUG-33149
Change-Id: Iab9522ac3c4ae6b746e790a99d87501b1cc1b655
Reviewed-by: Michael Brasser <michael.brasser@live.com>
Reviewed-by: Lars Knoll <lars.knoll@digia.com>
|
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Out of 200 runs on 10.9 + stable branch, I have 0 failures.
Task-number: QTBUG-26696
Change-Id: I84de15b68dfff2cb5f141c5f2a783154db16b77e
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
Reviewed-by: Alan Alpert <aalpert@blackberry.com>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
The size is not a part of the matrix, so the matrix does
not need to be recomputed when the size of an element changes.
The exception from this is when using transform origin which
is already specialcased in the QQuickWindow::updateDirtyNode()
function.
Change-Id: I2c517880d39aa69232974f65da93333b7cba341d
Reviewed-by: Alan Alpert <aalpert@blackberry.com>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
The assumption that all list delegates are the same size (dependent on ListView
orientation) is not a correct one, even if it is correct most of the time.
Task-number: QTBUG-31626
Change-Id: Iba6f3bc5f38d60e3be7632ab17d0c66ab8e73965
Reviewed-by: Matthew Vogt <matthew.vogt@qinetic.com.au>
Reviewed-by: J-P Nurmi <jpnurmi@digia.com>
Reviewed-by: Alan Alpert <aalpert@blackberry.com>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Due to the high resolution the display default flicking speed
is rather slow. This change introduces increased values of
QML_FLICK_DEFAULTMAXVELOCITY and QML_FLICK_DEFAULTDECELERATION
which feel closer to the native experience.
Also cleans up the flick variables a little by moving them to
their own header, so that QQuickPathView can also benefit from
the improved behavior on BB10.
Change-Id: I80878d4cc2a53a8f530b1dc8674f729f70a83119
Reviewed-by: Alan Alpert <aalpert@blackberry.com>
Reviewed-by: Robin Burchell <robin+qt@viroteck.net>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
A lot of work has gone into this test already since
a79839dc5287de1448d1fd858db312a7bfa7b581, such as
6ba7d88df623df5bce85bfdae853fc49006e76f2.
Everything passes for me locally, so let's skip specific tests if they break (or
better: fix them) instead of having no test coverage at all.
Change-Id: I29a7aeb26b106955c442da26df486cb6ce4bc92a
Reviewed-by: J-P Nurmi <jpnurmi@digia.com>
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
|