Refactored certain bits of qarraydataops.h: picked exception-related
building blocks and put them into one place, (somewhat) documented
the usage, added tests
Personally, the existing code seemed rather complicated to analyze
(and do mental experiments for corner cases), especially when staring
at the whole thing for a while or "returning back" from some other work
and I still have my doubts that everything works correctly. Testing the
building blocks that are used should:
a) increase trust into existing code (provided the usage is correct)
b) give more use cases of how to use the building blocks, which in turn
would allow to compare and contrast tests vs implementation
Task-number: QTBUG-84320
Change-Id: I313a1d1817577507fe07a5b9b7d2c90b0969b490
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Replaced copyAppend implementations with insert(this->end()) where
possible. This forced an update of the preconditions in insert
Unified moveAppend between generic and movable operations
Change-Id: I388c14436e32152ebb969bdd94753ed5452c1b7c
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Left the translation NOOP for later, pending advice on how to fix
QIODevice, which doesn't compile without them.
Task-number: QTBUG-85700
Change-Id: Icc423ecabb43714d98b5d9b0f9a96c5bb6ef1d78
Reviewed-by: Martin Storsjö <martin@martin.st>
Reviewed-by: Lars Knoll <lars.knoll@qt.io>
Updated insert() methods:
* Refined Q_ASSERT() checks
* Fixed implementation issues (some of which resulted in
actual crashes)
* Allowed to insert at the end. This is safe as far as I can
tell and actually would allow to simplify considerable chunks
of code (mainly, copyAppend versions to just return
insert at the end)
Updated tests accordingly
Change-Id: I0ba33ae5034ce8d5ff95b753894e95d71ba00257
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Extended existing tests with QArrayData's allocation options
Added extra tests on array operations covering append, insert,
emplace, erase and truncate. "Raw" QArrayDataPointer is used instead of
test-specific SimpleVector to check the behavior without some custom
logic in-between
The change targets future updates to array operations in the light of
prepend optimization: as the array operations would become more complex,
these tests should give a much better coverage (specifically due to
likely non-trivial implementation details and optimizations)
Task-number: QTBUG-84320
Change-Id: I6581e2cb48f81b82ee5052d1dcea3da2819df47a
Reviewed-by: Sona Kurazyan <sona.kurazyan@qt.io>
There's no way that Qt 6 will be binary-compatible with Qt 5.1.
Change-Id: Ic1abfbdc8dc497dcb24494a8f3e2d96bcf27ef83
Reviewed-by: Alexandru Croitor <alexandru.croitor@qt.io>
The std::is_nothrow_assignable_v template takes two template
parameters. Current clang nightly errors out on the previous code,
while earlier versions, and GCC, fail to diagnose the missing
template parameter.
Change-Id: I8ae7d5b9a9e89c5579e5526fb29d4da5cd1ab859
Reviewed-by: Fabian Kosmale <fabian.kosmale@qt.io>
It is a valid use case to have a vertex shader with no
vertex attributes. Using gl_VertexID or gl_InstanceID
is enough to generate vertex data out of thin air.
Change-Id: If7689914624a84723923fd2d7aef355bda592e24
Reviewed-by: Laszlo Agocs <laszlo.agocs@qt.io>
Our deployment target is 10.14, which enables layer-backing by default,
and our layer-backing support nowadays is stable enough that we don't
need to maintain any of the old code paths for compatibility.
The wantsBestResolutionOpenGLSurface property on NSView is only relevant
for surface-backed views, so we no longer need to deal with it.
Change-Id: I8aef4ac99371113d463ac35eee648a8a2fd1ea72
Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
Reviewed-by: Morten Johan Sørvig <morten.sorvig@qt.io>
This makes it easier to reliably maintain input-event related states
in widgets, in particluar the state of keyboard modifiers. Instead of
testing for all possible event types, code can just test the flag before
safely static_cast'ing to QInputEvent to access the modifiers.
Simplify the code in QAbstractItemView accordingly.
Task-number: QTBUG-73829
Change-Id: Idc7c08e2f3f1e8844f5c6693c195153038ee6490
Reviewed-by: Shawn Rutledge <shawn.rutledge@qt.io>
...now that a recent path enables MultisampleTextures for ES >= 3.1 and
GL 3.x and up.
Just need to make sure the .qsb contains something for OpenGL as well.
While we are at it, make updated .qsb files for all the commonly used
shaders since what we had before was version 4. Bump them to version 5.
Change-Id: If2040f4894e6360d1ebd5daf7e698508e5e6e42e
Reviewed-by: Andy Nichols <andy.nichols@qt.io>
...to pull in extra defines, includes, libdirs and framework paths.
Change-Id: I7c252f5edbcf15f7e2ad69ace4e15fbacf6cca8a
Reviewed-by: Alexandru Croitor <alexandru.croitor@qt.io>
Introduce new CMake variables and map
-D to QT_EXTRA_DEFINES,
-I to QT_EXTRA_INCLUDEPATHS,
-L to QT_EXTRA_LIBDIRS,
and -F to QT_EXTRA_FRAMEWORKPATHS.
Those variables only affect the Qt build, not user projects.
Fixes: QTBUG-85878
Change-Id: I229df2eed1505a2619068d0d32975962b052569a
Reviewed-by: Alexandru Croitor <alexandru.croitor@qt.io>
CMAKE_SYSTEM_PROCESSOR has the value i686 when targeting the x86
Android ABI, which is set by the Android CMake toolchain.
This is not the value that Qt expects, and potentially
breaks loading of Qt plugins. The same with aarch64 instead of
arm64-v8a.
In qmake the proper values for QT_ARCH variable are the equivalent of
the ANDROID_ABI variable set by the CMake toolchain file, so use
ANDROID_ABI.
Task-number: QTBUG-85399
Change-Id: Ia2eaceb75168cddb2d28414f1f4bf481ec76f29e
Reviewed-by: Joerg Bornemann <joerg.bornemann@qt.io>
QPixmap doesn't implement operator==, but QCursor had a non-explicit
constructor from a QPixmap, which the compiler would use if two pixmaps
get compared.
Making that QCursor constructor explicit prevents that implicit
conversion, at the cost of potential (but trivially resolved) source
incompatibility.
Also, make QCursor's comparison operators into hidden friends to avoid
any other implicit conversions, and allow only QCursor object to be
compared.
In addition, delete operator== and operator!= in QPixmap so that the
compiler generates a useful error message, rather than a screen full
of global operator== candidates.
Delete these operators in QIcon as well for consistency.
Change-Id: I9b3f770da5718360ecc41c35cf327ef4f60d169b
Fixes: QTBUG-85993
Reviewed-by: Lars Knoll <lars.knoll@qt.io>
QDesktopWidget is not a public class anymore and only exists for
compatibility reasons. So we might just as well move the only method
used by other classes out of QDesktopWidgetPrivate, and get rid of the
friend declarations and the access-to-private code in QApplication.
Change-Id: I74ff14233e59912e0930f98995f8af60dbadfdb6
Task-number: QTBUG-62094
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
Follows the QWindow semantics, and is a replacement for creating
a QWidget with a QDesktopScreenWidget as the parent.
We can now remove much of the special handling of QDesktopWidget and
the Qt::Desktop window type, and get rid of QDesktopScreenWidget.
Add a manual test that allows local testing. Our CI environments
only have a single screen, and no multi-head display server setup
which is the primary case where QWidget::setScreen is interesting.
For the more common case of a virtual desktop, QWidget::setScreen
has no real impact (just as QWindow::setScreen doesn't).
Change-Id: Id0099e069d316741bacd8c795c396ccad37be297
Fixes: QTBUG-85483
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
The event handler listens to the relevant input events already to
read the state of the modifiers. The implementation doesn't care about
touch, wheel, or tablet events either way.
The call was introduced in 28a21d98ef,
with no explanation why it would be necessary, and no test cases to
exercise that case.
Task-number: QTBUG-73829
Change-Id: Ibe408e86f697ac5acae5944b941f0d935ca2e408
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
One warning about unused variable (dataLeftover), which was at some
point replaced by the class variable 'decoderHasData'.
The second warning was a fault of logic: checking that the unsigned
value retValue was greater than or equal to zero. Which only came about
because they initially did different things but the branches got merged
and the logic became flawed.
Change-Id: Ia3a04516c1b7b5f962226998bf3f4d101dd38148
Reviewed-by: Fabian Kosmale <fabian.kosmale@qt.io>
After 9ee554ac1d, the various static
inline versions of randTLS are no longer used, which clang warns
about.
Change-Id: I6dc21c0aab35fb6a8ca9e8d43ee4960ee844ef47
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Since a QProxyStyle could be easily used for just a style hint or
similar we should account for the baseStyle() too to see if it is
effectively a QMacStyle too.
Pick-to: 5.15
Change-Id: I5d1df31f594cac2a7f5f5d1384cb7c278633b29d
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@qt.io>
We extend configurejson2cmake to read the "commandline"
information from configure.json. This data is then translated to CMake function
calls and written it into commandline.cmake files.
We extend QtProcessConfigureArgs.cmake to pick up those commandline.cmake
files to feed our command line handling code, which is a
re-implementation of the command line handling in qt_configure.prf.
The command line handler sets INPUT_xxx variables, similar to
configure/qmake's config.input.xxx variables. The INPUT_xxx values are
translated
- to -DFEATURE_xxx=ON/OFF arguments if the input represents a feature,
- to corresponding CMake variables if such a variable is known,
- or to -DINPUT_xxx=yyy CMake arguments.
Configure arguments that have an entry in
cmake/configure-cmake-mapping.md are actually implemented. Other
arguments are likely to need more work.
Task-number: QTBUG-85373
Change-Id: Ia96baa673fc1fb88e73ba05a1afb473aa074b37d
Reviewed-by: Alexandru Croitor <alexandru.croitor@qt.io>
For Qt 6.0, all versioned style option types have been merged into the base
class, and version number has been reset to 1.
Task-number: QTBUG-84221
Change-Id: If729720813c0cf284271a9084311ed6eb820ecca
Reviewed-by: Christian Ehrlicher <ch.ehrlicher@gmx.de>
Move the QIODevice::OpenMode enum into a base class, so that
we can remove the full QIODevice (and thus QObject) dependency
from qdatastream.h and qtextstream.h.
This is required so that we can include QDataStream in qmetatype.h
without getting circular dependencies.
As a nice side effect, QDataStream and QTextStream can now inherit
QIODeviceBase and provide the OpenMode enum directly in their
class scope.
Change-Id: Ifa68b7b1d8d95687ed032f6c9206f92e63bfacdf
Reviewed-by: Paul Wicking <paul.wicking@qt.io>
Reviewed-by: Maurice Kalinowski <maurice.kalinowski@qt.io>
This is a distinct types in C++, and should be supported out of the
box in QDataStream. This is also required so we do find a data stream
operator for the types when searching for it using template magic.
Change-Id: Iea57780621e2aab7ed253f1cc896bebada43b9f0
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
This will require some more work in QFlags and the metaobject
system for Q_ENUM.
Change-Id: I3687c7d2529dc18e7bc1abfc326e56ff7a365bc5
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Those are problematic as they are also interpreted as
forward declarations of methods that are defined inline in
qdatastream.h and might never get instantiated. This can
lead to problems if template code checks for the existence
of the method.
Change-Id: I4550a6bc70ebd7edc57fe0420b89b453195971d0
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Reviewed-by: Fabian Kosmale <fabian.kosmale@qt.io>
Check that we can successfully instantiate the data
stream operator for a container before we actually try.
This is required so we can automate registration of debug
stream operators with QMetaType.
Change-Id: Ib100a5242470d7fc8067058cc4d81af2fa9354b0
Reviewed-by: Fabian Kosmale <fabian.kosmale@qt.io>
Check that we can successfully instantiate the debug
stream operator for a container before we actually try.
This is required so we can automate registration of debug
stream operators with QMetaType.
Change-Id: I3943e7a443751d250c33b2ca1b9cf29207cfe6c4
Reviewed-by: Fabian Kosmale <fabian.kosmale@qt.io>
We should probably deprecate both constructors in addition.
Change-Id: Id62448b2abe14248c134feae899fa053ef373e58
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Reviewed-by: Mårten Nordheim <marten.nordheim@qt.io>
Add metaType()/setMetaType() methods to be used instead
of the type() methods taking a QVariant::Type.
Change-Id: Ieaba35b73f8061cd83288dd6b50d58322db3c7ed
Reviewed-by: Maurice Kalinowski <maurice.kalinowski@qt.io>
It was marked internal anyway. Use the constructor taking a
QMetaType instead.
Change-Id: I15b9cd0911aac063a0f0fe0352fa2c84b7f7c691
Reviewed-by: Lars Knoll <lars.knoll@qt.io>
It's unclear that the actual file searched for is different from the
error output. Make it clearer.
Task-number: QTBUG-85399
Change-Id: Ia5eb6d03c42b399604ce452b88408e0e5071d17c
Reviewed-by: Assam Boudjelthia <assam.boudjelthia@qt.io>
Add a default argument for the data pointer, so one can
construct a null variant of a certain metatype. This
is meant to help replace the QVariant(Type) constructor
that will get deprecated.
Change-Id: If75a1491ffcaa82d28eb5a7efb547da0ef1f2a87
Reviewed-by: Fabian Kosmale <fabian.kosmale@qt.io>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
configure script has started to fail on presence of unknown parameters.
Change-Id: I7a18730e0579a36aadd8e92881886e2cc397594a
Reviewed-by: Alexandru Croitor <alexandru.croitor@qt.io>
To fix CMake Qt For Android projects to configure, we need to move
some functions from a private CMake API file only, to a public one.
Add Qt6AndroidMacros.cmake which will be loaded by Qt6Core package.
We'll have to decide how we proceed with Qt5AndroidSupport.cmake,
because that file automatically runs code when included in Qt5, and we
usually don't want to do it.
We'll also have to decide how to handle the define_property() calls
that are still left in the private QtPlatformAndroid.cmake file.
With this fix, Qt example CMake projects that use
add_qt_gui_executable should now be buildable. An APK can be created
with 'ninja apk'.
Unfortunately Qt Creator 4.13 does not currently seem to support
opening and building CMake Qt For Android projects properly.
While the build succeeds after fiddling with the Kit settings, the APK
deploy step fails to run (at least on my machine).
So the simplest way to run the built APK is to open the android-build
dir with Android Studio and launch the example application from there.
Task-number: QTBUG-85399
Change-Id: I77f246331de7a6e9e6d4ba7d973730190138f136
Reviewed-by: Cristian Adam <cristian.adam@qt.io>
Fix detection of qt android platform plugin by globbing inside
the install qt6 prefix location. This is just a sanity check.
Fix platform plugin detection for CMake standalone tests configured
using qt-cmake-standalone-test, which used to look into the fake
standalone prefix location instead of the real Qt location.
Fix detection of stdlib path using CMAKE_SYSROOT.
Add a global apk target that allows easier building of all apk targets
defined in the project. Creation of this target can be opted out by
setting QT_NO_GLOBAL_APK_TARGET to TRUE.
Amends b1f8ca8032.
Task-number: QTBUG-85399
Change-Id: Ic9c1646b4f00e0084fe3f4397df471b8f925afd8
Reviewed-by: Cristian Adam <cristian.adam@qt.io>
