This was a tweak to fix a bug in 2005 which has probably outlived its
usefulness, plus an accumulation of workarounds on top. (started as
48b5266e8ff9b472a16290dd923fe24dd0b6989b in the historical repo)
Task-number: QTBUG-34760
Change-Id: I2c01269e43636385ee5c89305c6b90f4a7f2c537
Reviewed-by: Gabriel de Dietrich <gabriel.dedietrich@digia.com>
One less obvious part of this patch: the fontCache pointer in engineData was not
safe. It isn't safe to rely on pointer addresses to verify we're cleaning up the
right thing, as a sequence of malloc()/free()/malloc() can return the same
pointer, and nothing was cleaning up the dangling pointer in engineData.
With this, it is possible to safely drop OpenGL contexts in QtQuick under all
conditions with no possibility of crashes.
Done-with: Aaron Kennedy <aaron.kennedy@jollamobile.com>
Change-Id: I7b91384251593730124323a74737d41333a05f59
Reviewed-by: Konstantin Ritt <ritt.ks@gmail.com>
Reviewed-by: Eskil Abrahamsen Blomfeldt <eskil.abrahamsen-blomfeldt@digia.com>
QBBSystemLocaleData emits qwarnings when it fails to open or read a pps object.
If the user code installs a message handler that will invoke QLocale API again
(i.e QDate, QDateTime, ...) which leads to a deadlock situation, since
the QBBSystemLocaleData global static object's ctor() is not yet done.
This patch logs the QBBSystemLocale's warnings to stderr and
skips the Qt message handler.
Change-Id: I3d51f85761253e09b14a44179dd14a887733b392
Reviewed-by: Rafael Roquetto <rafael.roquetto@kdab.com>
This mechanism is used by QtWebEngine to extract the platform GL
context. In the QNX case, the platform context is an EGL context, so
the resource you need to ask for is the "eglcontext". Compare to the
xcb native interface which has a similar implementation already.
Change-Id: I873eaadf96898abb24de347ac624c88cd54254cb
Reviewed-by: Jocelyn Turcotte <jocelyn.turcotte@digia.com>
Reviewed-by: Kevin Krammer <kevin.krammer@kdab.com>
Reviewed-by: Fabian Bumberger <fbumberger@rim.com>
When you move a dockwidget from one screen to another then it needs to
be kept at the same position on the new screen, i.e. with the mouse
pointer. This fix ensures that this is happening including when the
screens may not 100% align up with each other.
Task-number: QTBUG-33369
Change-Id: If414effdd0e0415629ca31a86f8bbe382dd29f80
Reviewed-by: Richard Moe Gustavsen <richard.gustavsen@digia.com>
Whenever a window gets blocked, then the qt_button_down variable should
be reset as it is not going to receive the release event for the mouse.
This fixes problems with the mouse cursor not being updated when moving
over widgets (such as QLineEdit) after a dialog is shown in response to
a press event.
Change-Id: Idfd072eaf36f51b816a0b2a0391cdc447d7a5d9d
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
Give processes a more complete environment (openSUSE
requires more variables to start an X11-process),
add test column indicating whether a test is
supposed to crash. If it is not, report crash reason.
Change-Id: I1e0ad59824963f69ee425f331d845741be2b3928
Reviewed-by: Tony Sarajärvi <tony.sarajarvi@digia.com>
Reviewed-by: Frederik Gladhorn <frederik.gladhorn@digia.com>
The code was using the "long" type when a 32 bit type was actually
needed. This can cause bugs in those systems where "long" is 64 bits
wide, such as Linux x86-64 (which is LP64).
Task-number: QTBUG-34861
Change-Id: Iab289b2af3847dd62d8b4ecea51896936ca4c7a2
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
Reviewed-by: Gunnar Sletta <gunnar.sletta@digia.com>
Reviewed-by: Jørgen Lind <jorgen.lind@digia.com>
... so it will be found when included. That file was added with
commit 3b6b615334,
"export QT_NO_<foo> equivalents to the build system"
Change-Id: I38208aafe5b274d5976cec5d5149a41e6a963798
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@digia.com>
When clicking on a widget currently focused, w/o having Qt::ClickFocus
set as focus policy, the focus should stay on the widget and not get
propagated to the widget's parent.
Task-number: QTBUG-34042
Change-Id: I53f1153829cc7228de02a90e38125b5cf4ee5008
Reviewed-by: Marc Mutz <marc.mutz@kdab.com>
This is an improvement over 6800728d where we only fixed it if
the window was in "normal" state (Qt::WindowNoState) before
minimizing.
With this patch, if the window was previously maximized or
full-screen it will also get the expose events when being
restored.
Change-Id: I4a30423145e0999c5d0a5ee2a989730b83f4e3f2
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
Reviewed-by: Nicolas Arnaud-Cormos <nicolas@kdab.com>
We claim that -1 is a special automatic value, but calling
setMinimumSectionSize with -1 is unfortunately ignored due a
regression in 524c3e05e8
Change-Id: I7d7e5dbbf78e561849d2f2352c9edb2df36aa181
Reviewed-by: Giuseppe D'Angelo <giuseppe.dangelo@kdab.com>
GCC refuses to use a merely static const uint array in a constexpr function.
Fix by making the array constexpr if supported by the compiler.
Change-Id: Idd59d3f74f8f4e98aad82bc892f4a6469932df9f
Reviewed-by: Olivier Goffart <ogoffart@woboq.com>
Reviewed-by: Lars Knoll <lars.knoll@digia.com>
This code was duplicated in every qcollator_platform.cpp and identical everywhere,
except in _icu, which uses a QByteArray m_key and the implementation used
QByteArray::operator<, which is semantically and probably code-wise identical to
what the other implementations did (after inlining).
Inlining this function removes a potential maintenance problem and increases
speed without violating encapsulation.
Change-Id: If3e9d38a7d4326b49f0611a9f4187c53960e8a03
Reviewed-by: Lars Knoll <lars.knoll@digia.com>
This enables specialisations of (std and q) swap using member-swap
and declares the types movable for efficient use in Qt containers,
and QList in particular.
This is a binary-incompatible change, so cannot wait for 5.2.1.
Change-Id: I7e90b6397ac4d00d0a7a5c42bae166c1b43e1994
Reviewed-by: Aleix Pol Gonzalez <aleixpol@kde.org>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
This necessitates adding d==0 checks in QCollator. By documenting that
moved-from instances can only be assigned to or destroyed, we can
limit the functions in which to check for d==0 to the assignment
operator and the destructor.
Doing otherwise would destroy all advantages of move semantics by
introducing a heap allocation to re-populate other.d.
Add a test for this (QCollator didn't have any before).
Change-Id: Ic6ff202072822bebfd5e48259c3d0fa345a63118
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Rather than breaking at column 79 precisely, break entire words,
to improve readability.
Change-Id: Ie30db00f0e6ed95cce87480c3b91804826c6076b
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@digia.com>
Vertical scrollbar may get out of sync. When this happens, the calculation of
firstVisibleItem will retrun "-1". This must be handled in ::sizeHintForColumn().
Added an auto-test for the crashes.
Task-number: QTBUG-34717
Change-Id: I867fd144ef3ce45e382337c5eafe345f573cd944
Reviewed-by: Thorbjørn Lund Martsum <tmartsum@gmail.com>
This is standard for (QObject*) ctors.
Change-Id: I4756ba50b1f3148d72e95e581d52a37ebd47a7ae
Reviewed-by: John Layt <jlayt@kde.org>
Reviewed-by: Alan Alpert <aalpert@blackberry.com>
Reviewed-by: Olivier Goffart <ogoffart@woboq.com>
This ensures that the possible window flag combinations are respected
where possible in Cocoa.
Task-number: QTBUG-34645
Task-number: QTBUG-31616
Change-Id: I6be8ca666b7cbc397575e97cd95ea298f52a7113
Reviewed-by: Gabriel de Dietrich <gabriel.dedietrich@digia.com>
Center windows on screen to avoid conflicts with Unity taskbars,
observe minimum geometry for Windows, move cursor out of the
way for Mac. Introduce QScopedPointer to ensure cleanup in
case of failures.
Change-Id: Ic169e015e795597457b8e85a28ff7fcb7b9b9430
Reviewed-by: Robin Burchell <robin+qt@viroteck.net>
Task-number: QTBUG-32308
Change-Id: Ibbab3852e5cc289faa63d0a66a3816ab8062ccb9
Reviewed-by: Stephen Kelly <stephen.kelly@kdab.com>
Reviewed-by: Brad King <brad.king@kitware.com>
Ensure that custom style hints returned by subclassed
QPlatformIntegration::styleHints() are used if not explicitly overridden
by subclassed QPlatformTheme::themHints().
Task-number: QTBUG-34589
Change-Id: I0a3114cb00b532d2a5cd40941eca192b32412bcf
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
Reviewed-by: Paul Olav Tvete <paul.tvete@digia.com>
The inplace_converter_map never had an entry for ARGB4444_Premultiplied
this leads to the possibility of accessing outside of the array, and
means the RGBA8888 formats are misplaced.
Change-Id: Ida0d94912b53a7730b8fb5f6ccc31e7879ea3d27
Reviewed-by: Gunnar Sletta <gunnar.sletta@digia.com>
qcocoaapplication.mm:118:61: warning: cast to 'id' from smaller integer type 'int' [-Wint-to-pointer-cast]
id a1 = ([args->arg1 isKindOfClass:[NSNumber class]]) ? (id)[args->arg1 intValue] : args->arg1;
^
qcocoaapplication.mm:119:61: warning: cast to 'id' from smaller integer type 'int' [-Wint-to-pointer-cast]
id a2 = ([args->arg2 isKindOfClass:[NSNumber class]]) ? (id)[args->arg2 intValue] : args->arg2;
Change-Id: Ibcf3d5d5698ff863f3c9bd65e0388ccca147f419
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@digia.com>
This is required for a Qt value type these days.
Change-Id: Ibd4e1581a4f4791a410caa10fede92c26b35dd9d
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
QCollator and QLocale are not equivalent types, so there should
be no implicit conversion between them.
Change-Id: I395f8dc3c35b4202c9276c3eea0686176f8e07cc
Reviewed-by: Aleix Pol Gonzalez <aleixpol@kde.org>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
It was providing all possible types, but now MessageDialog is a new
native dialog type, and only on Android at the moment.
Task-number: QTBUG-34784
Change-Id: I2fb288c8d5e176ca4dafbbc310de2f29bbcfc000
Reviewed-by: J-P Nurmi <jpnurmi@digia.com>
This was actually causing test failures in qtdeclarative
and blocking the CI there.
Change-Id: I4538342f16b6468ad60b283c19948863b20ad5d4
Reviewed-by: Gunnar Sletta <gunnar.sletta@digia.com>
Otherwise wheeling over disabled items won't let the wheel event
propagate to the parent.
An easy way to reproduce the problem is by having a flickable with
some disabled item in it.
Task-number: QTBUG-33477
Change-Id: Ib327cc34dd9f415617ffbc41064f689507446c05
Reviewed-by: Nicolas Arnaud-Cormos <nicolas@kdab.com>
Reviewed-by: Giuseppe D'Angelo <giuseppe.dangelo@kdab.com>
It's not really clear if styles *must* return a non-negative value for
QStyle::pixelMetric(PM_Layout{Vertical,Horizontal}Spacing), but both
QBoxLayout and QGridLayout seems to be robust enough to handle this.
They will simply make sure that the spacing is never negative.
We therefore make QFormLayout equally robust.
Task-number: QTBUG-34731
Change-Id: I62235bfcd8adf7757cf15bc9927b29650ae6459d
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
Only align if we actually *need to* (due to SSE/XOP usage)
The code will currently not compile as it is now with UseSSE or
UseXOP defined, (so the ALIGN macro could just
be totally removed). In order to keep the diff to upstream as small
as possible, we keep the ALIGN macro definition.
Task-number: QTBUG-34362
Change-Id: I6cf420205896e3a476cdcfbf3308a07e0ae0c84e
Reviewed-by: Richard J. Moore <rich@kde.org>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Touch positions reported by libscreen have to be adjusted relative to
the window position to be properly interpreted by Qt.
Task-number: QTBUG-34812
Change-Id: I68744dc9da95fb1d0d1704d12154fb24c148fe03
Reviewed-by: Sean Harmer <sean.harmer@kdab.com>
Previously matched Qt.platform.os, however that can only provide one
string. Multiple selectors can be present at once, so we can provide
both unix and linux instead of having to pick the most specialized one.
Task-number: QTBUG-34796
Change-Id: I219517d740fa7385e923a9e09cb7e241378fbaee
Reviewed-by: David Faure <david.faure@kdab.com>
This is used by QML comboboxes, menu buttons, etc.
Task-number: QTBUG-31420
Change-Id: I6d3f32fd80ce91c4be34eae636d1b95a4e251a49
Reviewed-by: BogDan Vatra <bogdan@kde.org>
!build_all is a sufficient condition for bundling headers in both debug
and release configurations, as it means that we will likely make only
one of the builds, and that needs to be self-contained.
Task-number: QTBUG-34575
Change-Id: I1a19e4b619eeff207cfbfd50d3b761053aeaa667
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Up to now, the feature classe Uniform Initialization was subsumed by the
Q_COMPILER_INITIALIZER_LISTS flag together with support for
std::initializer_list.
This caused at least two problems:
1. On QNX, the standard libray does not ship <initializer_list>, even
though the compiler (a GCC 4.6, IIRC) supports it. But since there
was only one Q_COMPILER flag for both, support for the compiler-only
part of the feature had to be disabled, too.
2. MSVC 2013 supports initializer lists, but has a bug that renders full
uniform initialization support, as required for QUuid, useless.
By splitting the feature into two, we can separate them better, and do
so in QUuid, which is the only class that currently takes advantage of
uniform initialization (to provide constexpr constructors).
Since Q_COMPILER_INITIALIZER_LISTS worked as a flag for uniform
initialization so far, with the two known exceptions above,
UNIFORM_INIT is defined whenever INITIALIZER_LIST is, except that
I don't revert UNIFORM_INIT on QNX as I do for INITIALIZER_LISTS
and that I expect the MSVC 2013 features to set INITIALIZER_LIST,
but not UNIFORM_INIT.
Task-number: QTBUG-34705
Change-Id: I81916e950a0f3aab3de7977e0326d2de3d31b14c
Reviewed-by: Yuchen Deng <loaden@gmail.com>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
