Destroying the instances is a problem because the instances might
have been created in a plugin. In that case, the destructor of
the instance will be implemented in the plugin. As the instance
destruction code is run very late, the plugin might have already
been unloaded by the time an attempt is made to destroy the
instance.
The workaround is to create static instances on the stack.
Task-number: QTBUG-31142
Change-Id: Ic2632c3548a734b742da46d90249916c35705d46
Reviewed-by: Stephen Kelly <stephen.kelly@kdab.com>
Reviewed-by: Jędrzej Nowacki <jedrzej.nowacki@digia.com>
When Xcursor LoadCursor fails to find a cursor
Task-number: QTBUG-27537
Change-Id: I46086c78c3d5bea3c39187dcc55b4774e58f5749
Reviewed-by: Samuel Rødal <samuel.rodal@digia.com>
This allows you to easily retrieve properties in the XSETTINGS
specification. It is also possible to add listeners to get notified when
a specific property changes.
XSETTINGS is lazy initialized, so it will not be instansiated before
someone uses it. For now the intended use is a fallback for finding
cursor theme
Change-Id: Id47f0613f5876424cd47d721b40da17d3f63429e
Reviewed-by: Samuel Rødal <samuel.rodal@digia.com>
Objective-C++ does not support namespaces. In order to make this work,
we decorate each symbol with suffix using the macro QT_MANGLE_NAMESPACE.
However, with such a technic each public symbol needs to be explicitly exported through
this macro. The use of @compatibility_alias allows to define alias to use it
automatically and transparently.
Task-number: QTBUG-23946
Change-Id: Id521b8160bab126fda40a9d960277b1c04cc8b66
Reviewed-by: Gabriel de Dietrich <gabriel.dedietrich@digia.com>
We were setting the frame origin and content size as two separate steps.
The frame origin was asynchronous, while the content size was synchronous.
This resulted in the window visually jumping/moving, instead of applying
the new geometry in one single step.
We now set both in one go using setFrame, by manually calculating the
frameRectForContentRect. This makes the whole process asynchronous.
Task-number: QTBUG-31039
Change-Id: Iaba46106b9f200c707507f330fa801c99e2e7d89
Reviewed-by: Gunnar Sletta <gunnar.sletta@digia.com>
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@digia.com>
After Apple added the resize functionality to window borders,
they stopped rendering the size grip in most apps. Hence we do not
need to either.
Since the status bar height depends on the status bar being
there, we considered the safest option to keep it but make
it hidden.
Change-Id: Ida06e8f0de0f50d746a7d25bc5a472e22a370645
Reviewed-by: Gabriel de Dietrich <gabriel.dedietrich@digia.com>
We permanently enable 'wantsBestResolutionOpenGLSurface' when creating
an Open GL window/surface. We were enabling it once we would detect a
high-dpi Cocoa backend, whitout ever turning it off. Leaving it 'on' on
non-high-dpi displays should not change anything.
Task-number: QTBUG-31000
Change-Id: I3d0c77f7b36f3c22dacd04a50fcc5a7d88af32c0
Reviewed-by: Samuel Rødal <samuel.rodal@digia.com>
Reviewed-by: Gunnar Sletta <gunnar.sletta@digia.com>
If the menu bar is subject to height for width (HFW) we should of
course respect that, but in addition we should ensure that the HFW is
within the minimum and maximum height. This also is consistent with how
QGridLayout calculates the effective minimum row height.
This fixes a regression because change 4780f94e391b5e881497c5228661dead
turned QTabWidget into a proper height-for-width citizen, and when
setting a QTabWidget as a menuwidget, the buggy codepath for HFW was
suddenly hit in menuBarHeightForWidth().
Task-number: QTBUG-31057
Change-Id: I3c1bb8063c92d6eda7e9433e44f08967d8e1c43e
Reviewed-by: Paul Olav Tvete <paul.tvete@digia.com>
Instead of waiting for -drawRect to call -setView on the
NSOpenGLContext, we go ahead and attempt to set the context as soon as
possible. If it is indeed required that we call -drawRect first then
will try to call -setView again during -drawRect with the new
NSOpenGLContext.
Change-Id: I33d9f2ba241b49e8cfa6c9156dd5bf5e4cc6b164
Reviewed-by: Liang Qi <liang.qi@digia.com>
Reviewed-by: Gunnar Sletta <gunnar.sletta@digia.com>
The HTTPS links fail in Qt Assistant on Windows as the
qt installation package includes qt libraries that are
built without SSL support for legal reasons.
Task-number: QTBUG-31073
Change-Id: I86909abadb1e8164749d924cc53ee05aa57f8f31
Reviewed-by: Jerome Pasion <jerome.pasion@digia.com>
Reviewed-by: Karsten Heimrich <karsten.heimrich@digia.com>
... and introduce -hostlibdir configure option for symmetry.
the libraries built for the host have no business in the target prefix.
in principle this code would even support dynamically linked host
libraries, but that's currently unused.
Task-number: QTBUG-30591
Change-Id: I8e600fa4911a020fb0e87fbf7ef2f35647c7c4d5
Reviewed-by: Joerg Bornemann <joerg.bornemann@digia.com>
Reviewed-by: Ivan Romanov <drizt@land.ru>
Currently Qt looks at 1000 rows when scaling a column.
That can be slow in some situations and too inaccurate in others.
With this patch we leave it up to the user to decide how precise
e.g resizeToContents should be.
Change-Id: I6ef60f9a3bb40fc331ce1a1544fdc77488d20ca3
Reviewed-by: Stephen Kelly <stephen.kelly@kdab.com>
This makes sizeHintForColumn and sizeHintForRow considering indexes that
are outside the visual area - which is what most people want. It is
a bit weird that scrolling gives '...'
That will make the behavior a bit more like QTreeView, that considers
indexes outside the visual area.
Furthermore it is important since the user cannot just resize a column
when QHeaderView::ResizeToContens is used.
Task-number: QTBUG-4206
Change-Id: Icb209a19ce9e62cd30ca7159a4ad2aa9aadc1b6a
Reviewed-by: Stephen Kelly <stephen.kelly@kdab.com>
Fix the inline editor of the QMenuBar in Qt Designer not working
due to the key events being sent to the QMenu which is open
at that time.
Task-number: QTBUG-31059
Change-Id: Ic96bc119d0d2566d8f8d6ee62858445a70a447b7
Reviewed-by: Samuel Rødal <samuel.rodal@digia.com>
When transferring html formated data through the clipboard
some of the html markup was lost when converting from Windows
clipboard native mimedata to Qt mimedata type.
On X11 we are sending everything - inluding <!DOCTYPE ...><html> .. </html>.
This patch enables the same behavior on Windows.
Task-number: QTBUG-30984
(cherry picked from commit 9e28245da1580e25e77a665b7738c59eb298c422)
Change-Id: Ic0e339ad53955f1d31e8dff92ccf38b6eeec369a
Reviewed-by: Andy Shaw <andy.shaw@digia.com>
The function pointer "maybeCancelWaitForMoreEvents" was never
initialized to any value. This caused qt_mac_socket_callback
to call that "function" when it had a non-zero value, which caused
a crash under some circumstances.
Change-Id: I1d31be7b5b7c103fdc0204ffa6e9a5ecbb28d062
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@digia.com>
QFontComboBox had convenient filtering options but somehow not
QFontDialog, so provide the same type of flags and a similar behavior.
Change-Id: Ia8efabc60ae795673c772ff8ed63fd49244a5bb9
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
Reviewed-by: David Faure (KDE) <faure@kde.org>
Reviewed-by: Giuseppe D'Angelo <giuseppe.dangelo@kdab.com>
Maybe spans would be better in QTreeView than QHeaderView - but
from what I have seen of them (complex and slow) I doubt that
it is what we want in QTreeView.
Beside that the code removes a dead line of code, that is bug.
This is a comment fix that does not change the code.
Change-Id: I4c67a95c4bd2904612d6146077b7cbd59de7bdd6
Reviewed-by: Stephen Kelly <stephen.kelly@kdab.com>
In the Cocoa platform plugin the hovered signal() was not being emitted
for the corresponding QMenu. This now ensures it is emitted for the
action itself and its QMenu.
Task-number: QTBUG-30841
Change-Id: I3cfe634a71de7b55550e2561eb1c6a939536216d
Reviewed-by: Gabriel de Dietrich <gabriel.dedietrich@digia.com>
When users created a modal window in a dropEvent(QDropEvent* event),
the cursor was not restored to its original shape until after the modal
window had been dismissed.
Change-Id: Id4723e3cb66b36a3ca18ff097a3993563816d9e1
Reviewed-by: Lars Knoll <lars.knoll@digia.com>
This is not needed. The DnD initiated scrolling is done by
installing an eventFilter which forwards all MouseMove events
to the QGuiApplicationPrivate::processDrag which creates and forwards
all the necessary QDrag* events. QAbstractScrollArea has its own logic
in the timerEvent() to make the view scroll when the pointer is standing
still on the edge and the drag process is still ongoing.
With the current implementation, widgets (during the DnD) were receiving
DragMove events even when the pointer was standing still outside auto scroll
areas.
Task-number: QTBUG-28171
Change-Id: I355d88f3eab0ad39f916f84d66f5d0af7c0ff93e
Reviewed-by: Lars Knoll <lars.knoll@digia.com>
Reviewed-by: Stephen Kelly <stephen.kelly@kdab.com>
We removed all Q_WS_WIN code when going to Qt 5. One of the things
removed was the condition that we do painter path text rendering
for transformed, non-ttf fonts, since the GDI engine does not support
transforming those. This has now been reintroduced and adapted to the
QPA way of doing things, by checking for it in the font engine subclass.
Then there was the problem that QStaticText only supports cases
where the font engine can transform the glyphs. Thus we need to fall
back to regular text drawing in drawStaticText() for unsupported cases,
and we need to skip the optimized path in the raster engine (which
goes to drawStaticTextItem)
Task-number: QTBUG-30932
Change-Id: I17ba7355ee127811b0e77bb3a9b9db092e99893b
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
The new enum value was introduced in f3612f39ff
Change-Id: I9ab5d8fba459408246e9a6f40c5c086af5abbf82
Reviewed-by: Mitch Curtis <mitch.curtis@digia.com>
Reviewed-by: Shawn Rutledge <shawn.rutledge@digia.com>
The comment to the code said it was to avoid double click.
However it actually breaks wanted double clicks.
The reason for it must be that the replayed event in earlier
code versions could be changed into a double click (together
with the first event).
However (now) we only test qt_replay_popup_mouse_event in
void QWidgetWindow::handleMouseEvent(QMouseEvent *event);
Regardless what kind of event we receive as input it will send
QEvent::MouseButtonPress when it sends replay mouse event.
I.e. it will then call QCoreApplication::sendSpontaneousEvent(r,e)
=> QCoreApplication::notifyInternal(receiver, event)
=> QCoreApplication::notify(receiver, event)
=> QApplicationPrivate::notify_helper(receiver, event) (+filters)
=> (probably) QWidgeWindow::event(receiver, event)
=> further handling in widget classes.
That especially means that it will *not* get into the function
QGuiApplicationPrivate::processMouseEvent where doubleclicks are
created. Therefore no doubleclick can be made from the extra event.
That makes the statement have no good effect - just side effects.
Change-Id: I190baff3c060548775201695e324059560bb7106
Reviewed-by: Samuel Rødal <samuel.rodal@digia.com>
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
PostScript support was removed back in 2011:
34f712e79e
- The QPSPrint* classes are gone
- The QPrinter::OutputFormat enum only has NativeFormat and PdfFormat
constants
- QPaintEngine::PostScript is unused now
Change-Id: I27d595b92341f41e14bd1e4fd860eac6826d7890
Reviewed-by: Lars Knoll <lars.knoll@digia.com>
Set SCREEN_PROPERTY_SENSITIVITY to SCREEN_SENSITIVITY_NO_FOCUS of the
screen window when the Qt::WindowDoesNotAcceptFocus window flag is set.
Change-Id: Iac0764a84186c382dcd3f4bdd31dd24ac15df768
Reviewed-by: Sean Harmer <sean.harmer@kdab.com>
There is no point in not being able to get the message.
Sometimes it can be the case that we just want to append a dot
(or something else) on the existing message.
Change-Id: I171495f1edc568ae1654f43b3cf734a4eb67d374
Reviewed-by: Stephen Kelly <stephen.kelly@kdab.com>
Reviewed-by: Giuseppe D'Angelo <giuseppe.dangelo@kdab.com>
PostScript support was removed back in 2011, but not documented then:
34f712e79e
Change-Id: Ie818bf5f733422db910efadd57efa0ddabcc5175
Reviewed-by: Jerome Pasion <jerome.pasion@digia.com>
Reviewed-by: Lars Knoll <lars.knoll@digia.com>
When transferring html formated data through the clipboard
some of the html markup was lost when converting from Windows
clipboard native mimedata to Qt mimedata type.
On X11 we are sending everything - inluding <!DOCTYPE ...><html> .. </html>.
This patch enables the same behavior on Windows.
Task-number: QTBUG-30984
Change-Id: Ic0e339ad53955f1d31e8dff92ccf38b6eeec369a
Reviewed-by: Jan Arve Sæther <jan-arve.saether@digia.com>
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
No need to use old style Carbon Event Manager function when there is
Cocoa [NSEvent keyCode] equivalent for retrieving virtual key code.
Change-Id: I5cceb76fc662e42a46a17dc69131091a64ca6060
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@digia.com>
Several modules, including DBus, MySQL, and OpenSSL have
configure options of the form <MODULE>_PATH, which is used
on Windows (where pkg-config is not present) to specify the
locations of third-party libraries. These switches had been
implemented by adding extra variables which were referenced
in .pro files, to add the appropriate compiler and linker
switches. This is undesirable because it means there are
two independent paths for adding the switches to the build,
which can get out of sync with each other, and indeed this
had happened for some of the DBus tools.
To remedy the situation, all three of the switches were
reworked so that they added values directly to the principal
variables that are used in the project files. This reduces
maintenance, by ensuring that the pkg-config and non-pkg-config
paths appear the same to the rest of the build system.
Change-Id: Iae342f1d14b79fbcfef9fe38aadc803ad3141799
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@digia.com>
You are not supposed to call NSOpenGLContext -setView: for a view that
has not yet called drawRect. We we attempted to do this, we would get
the invalid drawable error, leading to QGLWidgets just drawing garbage.
Task-number: QTBUG-28175
Change-Id: I47aef07b4676f2db8591f98fc1661f6f447bdef9
Reviewed-by: Gabriel de Dietrich <gabriel.dedietrich@digia.com>
Added the same information for QFontMetricsF::elidedText.
Task-number: QTBUG-30865
Change-Id: I57791de21f5721d7a505c3d13f450e5ec6cd2730
Reviewed-by: Jerome Pasion <jerome.pasion@digia.com>
Reviewed-by: Eskil Abrahamsen Blomfeldt <eskil.abrahamsen-blomfeldt@digia.com>
QDoc adds document nodes as files into the .qhp, including nodes with
image subtype. This will generate a lot of unnecessary warnings for
missing '<image>.png.html' files from qhelpgenerator when running
'make docs'.
This change makes qdoc skip the above step for image nodes to remove
the warnings. The images are still always added to .qch in a separate
step.
Change-Id: Ib07fc8a3ce65f5d4dd700daf00f5f6a36f74a696
Reviewed-by: Martin Smith <martin.smith@digia.com>
Reviewed-by: Jerome Pasion <jerome.pasion@digia.com>
Added an upper bound check for socket notifier file descriptors. Too
high FDs have been a real source of failure.
Added compiler hints to allow error-free code path run faster (errors
are highly unlikely) and adjusted warning messages (some were misleading
and too long).
Change-Id: I1c9c41f5d006ca9d3a28214c3a464555b8a1c71f
Reviewed-by: Thomas McGuire <thomas.mcguire@kdab.com>
Session tickets can be cached on the client side for hours (e.g.
graph.facebook.com: ~ 24 hours, api.twitter.com: 4 hours), because the
server does not need to maintain state.
We need public API for it so an application can cache the session (e.g.
to disk) and resume a session already with the 1st handshake, saving
one network round trip.
Task-number: QTBUG-20668
Change-Id: I10255932dcd528ee1231538cb72b52b97f9f4a3c
Reviewed-by: Richard J. Moore <rich@kde.org>
Numpad is unnecessarily long.
Change-Id: I19a6ce129e26a4f6f8344f514317214c48abde6e
Reviewed-by: Andy Shaw <andy.shaw@digia.com>
Reviewed-by: David Faure (KDE) <faure@kde.org>
The implementation that GCC came with from 4.4 to 4.6 used external
locks for most platforms, other than x86. It's unsuitable to be called
"atomic" at all. More importantly, it's not behaviour-compatible with
Qt's own QBasicAtomic classes.
Change-Id: I92a0beab58504e6bd7cd236d358fef905c69821c
Reviewed-by: Olivier Goffart <ogoffart@woboq.com>
You can't write:
connect(proc, &QProcess::finished, [](){});
because of the overload.
Change-Id: I651cc56ee15481392590dc44942d8e814fad75f6
Reviewed-by: Olivier Goffart <ogoffart@woboq.com>
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@digia.com>
... instead of scoping the defines in qconfig.h, which relied on the
Q_PROCESSOR_xxx defines and meant that we had to include qconfig.h
after qprocessordetection.h, which added a whole bunch of other
dependency issues.
We now let configure write QT_COMPILER_SUPPORTS_xxx to qconfig.h as
before, without any scoping, and then undefine the ones that don't
apply for the given processor. This means we need to include
qprocessordetection.h before qcompilerdetection.h in qglobal.h,
but the former does not depend on the latter, so this should be
fine.
Change-Id: If00c00d405463e9626fa0f7f5e6b17f68778904f
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@digia.com>
FILEICONPROVIDER isn't listed in qfeatures.txt and Qt doesn't
build without it.
If you need QT_NO_FILEICONPROVIDER, besides reverting this commit,
you'll need to add a good amount of ifdefs to QFileDialog,
QFileSystemModel and QFileInfoGatherer classes.
Change-Id: Ic288c90268f6a900e160ba8eb4457580d1fd7a5a
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
Reviewed-by: Giuseppe D'Angelo <giuseppe.dangelo@kdab.com>
Because there is only one QBackingStore per top-level window, child windows do
not need to have their own buffer, since Qt will use the TLW backing store as
their drawing surface. Since QNX's libscreen does not support windows without
buffer, a 1x1 buffer is used instead.
Change-Id: I4b81fdd0f3e6059c46b1d4302d2a754d72dc8a68
Reviewed-by: Kevin Krammer <kevin.krammer@kdab.com>
Reviewed-by: Thomas McGuire <thomas.mcguire@kdab.com>
Folders can have a custom icon, set by the user. Some system
folders also have one, for example c:\windows\fonts.
This option allows you to disable this behavior, you'll get the
folder directory icon.
As a side-effect, you'll get a very big performance improvement
on removable/network media: 2 seconds vs 60 seconds on a SDCard
with 10000 folders.
Change-Id: Id55ea628186e0a6523585ec7a4ff622d6f5da505
Reviewed-by: Giuseppe D'Angelo <giuseppe.dangelo@kdab.com>
If a transient scrollbar was hidden during the fade out animation,
it got stuck visible after being shown again. A regression introduced
by f52177829a.
Task-number: QTBUG-31060
Change-Id: I1cda16c6afa4370cce4702d3b4fdaee9f44f9ab9
Reviewed-by: Gabriel de Dietrich <gabriel.dedietrich@digia.com>
All occurrences of `#if defined(Q_OS_MAC) && !defined(Q_OS_IOS)` have
been replaced with `#if defined(Q_OS_MACX)`.
Change-Id: I5055d9bd1845136beb8ed1c79a8f0f2c0897751a
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@digia.com>
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@digia.com>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
The equivalent was done earlier for qmake, and this will enable
developers writing OS X-specific code to use #ifdef Q_OS_MACX as
opposed to overly verbose #if defined(Q_OS_MAC) && !defined(Q_OS_IOS).
The sole usage of Q_OS_MACX within qtbase has been changed to the
now appropriate value and documentation has been updated.
Change-Id: I2c59eea02e94b691b705170b2f96a97940fdc756
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@digia.com>
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@digia.com>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Compact an object in regular intervals when inserting
data into it, to avoid the object becoming huge.
Compact an object/array before inserting into another
array or object.
Check that the document doesn't get so big it's overflowing
the internal data structures.
Task-number: QTBUG-29288
Change-Id: Id39d80dac1e7d5a11f40819f41b4b336bce16947
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Fix connection to pointer to member signal that belongs to the base
class, but whose type is a pointer to a member of the derived class.
The current code only use the QMetaObject of the type coming from the
function type to look up the signal id. But if the signal was casted
to a pointer to member function of a derived type, then we also need to
look in the base classes
Change-Id: Ib98fc38f63942946acb34d9f83c100991d58e4e5
Reviewed-by: Giuseppe D'Angelo <giuseppe.dangelo@kdab.com>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
QTreeView actually had a better model than QTableView, but after
previous patches QTableView now has a better one.
This patch makes sizeHintForColumn similar to what QTableView has.
Change-Id: I2f2d35e7aa66fc8990f54e2f4a12d97f490840e5
Reviewed-by: Stephen Kelly <stephen.kelly@kdab.com>
Some of sizeHintForColumn is now moved into widthHintForIndex.
(and in QTreeView some of sizeHintForRow into heightHintForIndex)
This makes the code a bit more readable and it prepares some
extensions that will use these functions more. There should
be no semantic changes in this patch.
In releasemode this does not seem to have a performance cost.
(QTableView actually seemed to be a bit faster)
Change-Id: I940432ee01715ce94cd6aab5f3b2aa00dcd19ace
Reviewed-by: Stephen Kelly <stephen.kelly@kdab.com>
In some situations we can get into resizeSections in a hidden
QHeaderView. If the headerView is hidden then we look at all the
rows, and that can be extemely expensive for a large model.
This patch limits the sizeHint with only looking at a maximum
1000 rows. Though this is more inaccurate it is also faster -
and it is not much different from what QTreeView does.
Change-Id: Ief4b54c5a3c5a0db02e8b595c9b9b3162633ee67
Reviewed-by: Stephen Kelly <stephen.kelly@kdab.com>
This reverts commit 79e729e111, which
partially fixed QTBUG-30122 and will need to be reopened.
On the Creator side, we think using the Qt::ToolTip flag is a bit unsafe
since tooltips are supposed to have very limited interactivity. We would
like to propose to use a similar pattern as in QComboBox instead.
Task-number: QTBUG-30702
Change-Id: I245b7d8b16ac870564ea44f24f481ebfec8ca9a1
Reviewed-by: Shawn Rutledge <shawn.rutledge@digia.com>
moc currently silently ignores them, but I have a version which display
a warning.
Change-Id: I9a239cb7e99d40a57a013fb66357c4a6426d6e8b
Reviewed-by: Giuseppe D'Angelo <giuseppe.dangelo@kdab.com>
Reviewed-by: Stephen Kelly <stephen.kelly@kdab.com>
During qHash refactorings, this line was changed as qt_hash didn't have
an overload taking a QStringRef. This causes a performance regression
w.r.t. the same code in Qt 4.
Task-number: QTBUG-30821
Change-Id: I17b27a54a73cb9061c20f1bd7f79d0c405050edd
Reviewed-by: hjk <hjk121@nokiamail.com>
qdoc's faux C++ parser did not recognize class declarations
of the form:
class Name1::Name2 {
...
};
...where class Name2 is nested in class Name1. Now it does,
but this fix doesn't handle deeper nestings. doc needs a
proper C++ parser.
Task-number: QTBUG-28664
Change-Id: I5adf88cc1b2ce03f5565250734416bf9592914b5
Reviewed-by: Jerome Pasion <jerome.pasion@digia.com>
The brief text for a documented thing is now output as an
attribute of that thing in the module's index file, and it
is reconstituted in the thing's tree node, when qdoc reads
the module's index file later. Only the verbatim text of
the brief is saved in the index file, i.e. no links or
other markup.
The effect is that brief texts can be used in other modules.
Task-number: QTBUG-31021
Change-Id: I932a0c85259b6d1901138f0c0959ddb9815b7db5
Reviewed-by: Jerome Pasion <jerome.pasion@digia.com>
There is no longer any difference in install directories between demos
and examples (QT_INSTALL_DEMOS is obsolete and points to the same dir
as QT_INSTALL_EXAMPLES for compatibility). This change removes any
special handling of demos' paths used in manifest xml, enabling the
demos to be visible in Creator welcome mode as well.
Change-Id: I5afdf578a42d001fab05979d3ea1102fd52b51da
Reviewed-by: Martin Smith <martin.smith@digia.com>
Reviewed-by: Thomas Hartmann <Thomas.Hartmann@digia.com>
In Qt 5, I managed to break the guarantee that a deserialised local
datetime is the same time of day (potentially different UTC time),
regardless of which timezone it was serialised in. This happened after
I fixed QTBUG-4057 with If650e7960dca7b6ab44b8233410a6369c41df73a,
which serialised datetimes as UTC.
This patch reverts QDateTime serialisation to pre-Qt 5 behaviour to
restore the guarantee and consequently re-opens QTBUG-4057.
Change-Id: Iea877f7ed886f530b928067789b53534e89fe8cb
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Necessary to make QWindow::fromWinId() work on Mac.
Change-Id: I0dc3a1b0d63cff61d484693aee711dc54fea6b2f
Reviewed-by: Samuel Rødal <samuel.rodal@digia.com>
The warning causes a test failure since it occurs with XCB.
Introduced by enabling frame strut events for all widgets
by f2fcf75381 .
Task-number: QTBUG-30530
Task-number: QTBUG-29012
Change-Id: I025b55f5198a270072b33cae033221709db256aa
Reviewed-by: Samuel Rødal <samuel.rodal@digia.com>
Don't send Qt::Key_unknown down the shortcut handling pipe, because
unknown keys are not part of a shortcut sequence and results in an assert.
Task-number: QTBUG-30871
Change-Id: Idd451fb793bd07124a96d75ed7b936f7ff584f28
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@digia.com>
If tabFocusFirst is not cleared or set to another valid item,
there will be crash later if the removed item is deleted after
removal.
Task-number: QTBUG-30923
Change-Id: Iba9a6ce9334c52f8e552b0accda95ebb5fcfcdb1
Reviewed-by: Jan Arve Sæther <jan-arve.saether@digia.com>
To allow copying one QPair of two types to a QPair of two different
types.
Change-Id: Ie7aa84cf5a361be6b7d242517578172c35bcacfd
Reviewed-by: Olivier Goffart <ogoffart@woboq.com>
A new set of classes is introduced for iterating over the contents
of a container within a QVariant without knowing the exact type of
the container, but with the guarantee that the element type within
the container is a metatype.
The implementation of the iterable interface uses
the stl-compatible container API so that we can also iterate over stl
containers, or any other container which also conforms to stl norms.
This enables the functionality in the bug report.
Task-number: QTBUG-23566
Change-Id: I92a2f3458516de201b8f0e470982c4d030e8ac8b
Reviewed-by: Stephen Kelly <stephen.kelly@kdab.com>
This patchs allows the user to convert defined QMetaType types like
MyType to be converted by using e.g. QVariant::toString(), mapping to
MyType::toString(). Also all the other QVariant::toXYZ() methods are
supported so far.
The patch adds static methods QMetaType::registerConverter supporting:
- implicit convertion
- conversion using member method of source type
- conversion using unary functor
Change-Id: I4f1db83d9c78bcc9df5c42f82f95cce0480cdcc3
Reviewed-by: Jędrzej Nowacki <jedrzej.nowacki@digia.com>
Reviewed-by: Christoph Schleifenbaum <christoph.schleifenbaum@kdab.com>
Reviewed-by: Stephen Kelly <stephen.kelly@kdab.com>
The custom printf formatter Qt ships supports e.g. %lld, which is part
of gnu_printf, but not ms_printf. This fixes a lot of MinGW warnings.
Change-Id: Iff600f20ac23ecb88c4b569d2e668f5d4af6ef27
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Reviewed-by: Kai Koehne <kai.koehne@digia.com>
qWarning now depends on QT_MESSAGE_PATTERN, depending on that variable.
It will show things like the moc process id or the Parser::error
function name. We don't want that.
Change-Id: I5b35401200f0f7de2442aa77d700a82402081489
Reviewed-by: Simon Hausmann <simon.hausmann@digia.com>
The current implementation kept a list of TouchPoints that
was reused when sending active touces to QPA. This list was
never cleaned up, so if you pressed three fingers, and released
one, we would still continue to sendt three touches to QPA.
Especially, since this list was not cleaned up when receiving
a touch cancel, mouse events sometimes stopped working when trigging
a system gesture (like a four finger swipe). This can be seen by
using the fingerpaint example.
Since we cannot rely on TouchPoints having IDs that corresponds to
their index in the touch point list, it ends up being
simpler (and results in less code) to rewrite the implementation to use
a hash table of UITouch to TouchPoints instead.
Change-Id: I5b32f57a8d72a0b8759a64ac7cdfa6700109d2b3
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@digia.com>
QWidgetWindow will always redirect mouse events to the active popup
(if any). The same logic is not implemented for touch events, which
means that touch events are always delivered to the widget under the
finger.
It is therefore possible to interact with widgets that are modally
shaddowed by the popup. It is also not possible to close popups
without touching them directly.
This patch will ignore touch events when a popup is active, and
as such, force a synthesised mouse event to be sent instead.
Implementing proper touch support also for popups is out of scope for
widgets.
Change-Id: I023c09c3e1fd4e5495df990c11419c69ecafb8f9
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@digia.com>
Reviewed-by: Frederik Gladhorn <frederik.gladhorn@digia.com>
Reviewed-by: Samuel Rødal <samuel.rodal@digia.com>
The current documentation is not terribly clear on this topic, and there's a
couple of posts on various forums where people want to do this. In fact, the old
wording suggested (at least to me) that it is OK to explicitly override a
disabled state, which is apparently not true.
Change-Id: I10c54e0089e9ba5d16958aea62df27feafdf7b3d
Reviewed-by: Samuel Rødal <samuel.rodal@digia.com>
The underlying rename() of the operating system simply does
nothing when renaming 'foo' to 'Foo' in a case insensitive file
system. Work around by moving in 2 steps.
Change-Id: Ibc73724bfca402a5ce7fcf2a83e8fea32ff71093
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@digia.com>
Reviewed-by: David Faure (KDE) <faure@kde.org>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
It makes more sense to keep this workaround header together with the other
libxkbcommon files for a better access point since it's used by several *.pro
files.
Change-Id: I63d4eb58f6e7f3852834e41c4b6e058a2c962233
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@digia.com>
Reviewed-by: Samuel Rødal <samuel.rodal@digia.com>
To have a properly working key input in the xcb plugin in the case when
xcb-xkb library is not available we can update the xkb_state struct with the
keyboard state information available in the X11 core events.
The current modifier state is reported to clients in a number of core
protocol events and can be determined using the QueryPointer request.
This is how it is done in Weston, Wayland's reference implementation.
Note: In case the X server doesn't have a xkb support on it (which is very
unlikely), then xkbcommon will only pick up the user's primary layout.
The X server with the xkb support stuffs unused bits (13 and 14) of 'state' in the
core events with the effective keyboard group, which we can use to determine layout
changes.
Change-Id: I9f1ef635109870e7412ef1157ca592f3c8f9271c
Reviewed-by: Samuel Rødal <samuel.rodal@digia.com>
This is required for a modifier+key shortcut support.
Also fixes old Qt bug when keymap group changes were not handled
properly for shortcut functionality.
Task-number: QTBUG-26902
Task-number: QTBUG-4845
Change-Id: I04d2c2ad7049df7420999816154605848fa670e1
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
Reviewed-by: Samuel Rødal <samuel.rodal@digia.com>
libxkbcommon is a keymap compiler [1] which utilizes xkb
keyboard configuration database [2].
xkbcommon simplifies keyboard state handling by providing xkb_state
object which holds the current state of all modifiers, groups, LEDs,
etc, relating to the provided keymap.
Detecting modifier mapping should become simpler once it gets supported
in libxkbcommon. Also with xinput2 we could get rid of the XkbStateNotify
events, because xinput2 key press/release already include all the mod/group
info.
[1] http://xkbcommon.org/
[2] http://www.freedesktop.org/wiki/Software/XKeyboardConfig
This patch:
- Removes a dependency to -lxcb-keysyms
- Removes a dependency to XLib call - XLookupString
- Enables required functionality to fix QTBUG-27681
Task-number: QTBUG-27680
Change-Id: I10e10f873821ee02f6df72238e215a541150f38f
Reviewed-by: Samuel Rødal <samuel.rodal@digia.com>
1) -qt-xcb
a) Use xkb from the 3rd party libs. As it is done for the other xcb
dependencies when qt configure with -qt-xcb.
2) -system-xcb (default)
a) If xkb found then use xkb from the system. (Currenly xkb is not
enabled by default when configuring libxcb library).
b) If xkb can't be found on the system then keyboard state will be
updated from X11 core events.
Change-Id: I7c3dbce6daa2cec52067cd5af80f19040233a0db
Reviewed-by: Samuel Rødal <samuel.rodal@digia.com>
This library is required by the XCB platform plugin. As we depend on
very recent version of this library and it might not be available
in base repositories of distributions, users can use -qt-xkbcommom
switch to build Qt with the bundled version.
Change-Id: I0ed2a5cc2f1df98b0e7cc926cabfa69818674e08
Reviewed-by: Samuel Rødal <samuel.rodal@digia.com>
Similar to what change a298216bb4 does for update(QRect) we clip
the update region against the widget's rect and return if it's empty.
Otherwise we risk ending up with update rects that are larger than
INT_MAX due to multiple update rects being merged.
Task-number: QTBUG-30876
Change-Id: Idf695b1fdca50449a1e5ddf37500653de290590c
Reviewed-by: Gunnar Sletta <gunnar.sletta@digia.com>
Set up for downloading from the qt-project.org servers instead
of KDE servers.
Change-Id: I6f9e69f91ea956165300d75109e1683d0b1aa5f0
Reviewed-by: Paul Olav Tvete <paul.tvete@digia.com>
Reviewed-by: BogDan Vatra <bogdan@kde.org>
We have to use the 4.4.3 compiler toolchain for producing binaries
for ARMv5 on Android, since the 4.7 toolchain has a regression for
this architecture. The regression has been fixed, but the fix has
not been released yet, so until it has, we need to use the 4.4.3.
However, the 4.4.3 toolchain has a different bug, which breaks
compilation in qtimageformats with a message about redefinition of
uint. This works around that bug for Android builds. When the
patched compiler has been released, we can revert this.
Task-number: QTBUG-30921
Change-Id: I620c186c6e932413a4de1dd331fbf4b9401f2e72
Reviewed-by: Paul Olav Tvete <paul.tvete@digia.com>
Change-Id: Ida382a80dba882bbeb920756adc0c16321efe37e
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@digia.com>
Reviewed-by: Stephen Kelly <stephen.kelly@kdab.com>
Change-Id: I98145fb3149dd8f7aba60d305da98c148aa9c705
Reviewed-by: Volker Krause <volker.krause@kdab.com>
Reviewed-by: Stephen Kelly <stephen.kelly@kdab.com>
Breakage was caused by two typos
Change-Id: Ic9ca3a3db8dbf191b3bee8b2ef6e3513f7c9adc4
Reviewed-by: Kevin Krammer <kevin.krammer@kdab.com>
Reviewed-by: Nicolas Arnaud-Cormos <nicolas@kdab.com>
Reviewed-by: Fabian Bumberger <fbumberger@rim.com>
Reviewed-by: Giuseppe D'Angelo <giuseppe.dangelo@kdab.com>
Provide a new displayIntegerBase property which control the base used
by the spin box to display the value in its internal line edit.
Change-Id: Ibadc37107db8770d757b64350946bf19142e8f6c
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
Reviewed-by: Lars Knoll <lars.knoll@digia.com>
Reviewed-by: Giuseppe D'Angelo <giuseppe.dangelo@kdab.com>
If a process dies before all output is read into the internal buffer
of QProcess, we might lose data. Therefore we must drain the output
pipes like we already do in the synchronous wait functions.
Task-number: QTBUG-30843
Change-Id: I8bbc5265275c9ebd33218ba600267ae87d93ed61
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@digia.com>
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
If \wrapper appears in a \class comment or a
\qmltype comment, qdoc will not print warnings
when it finds public members of the class or the
QML type that are not documented.
The \wrapper command is added to several
opengl classes. This reduces the number of
qdoc warnings by several thousands.
Task-number: QTBUG-30755
Change-Id: Iba1eebc1590ccf54100e40fe91423240c1b3d09d
Reviewed-by: Topi Reiniö <topi.reinio@digia.com>
Reviewed-by: Venugopal Shivashankar <venugopal.shivashankar@digia.com>
Reviewed-by: Martin Smith <martin.smith@digia.com>
Reviewed-by: Jerome Pasion <jerome.pasion@digia.com>
This is a spin-off of 00b11cc.
Change-Id: I042f066bdb1d248a04473d4b0a31002330a73907
Reviewed-by: Stephen Kelly <stephen.kelly@kdab.com>
Reviewed-by: Thorbjørn Lund Martsum <tmartsum@gmail.com>
QApplicationPrivate::translateTouchToMouse always sets
buttons() to Qt::LeftButton for synthesised events. This is wrong
for a mouse release, since 'button' should in that case be
Qt::LeftButton, and 'buttons' should be Qt::NoButton (since no buttons
are actually being pressed).
This caused problems for QGraphicsView, which refuses to
release any mouse grab set on a QGraphicsItem if at least one
button is being pressed (which was always true).
This resulted in broken drag behavior on touch platforms.
Change-Id: Iefe63cd753f9f8bb04278fd04a4d728e3deda25e
Reviewed-by: Samuel Rødal <samuel.rodal@digia.com>
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@digia.com>
Reviewed-by: Frederik Gladhorn <frederik.gladhorn@digia.com>
Reviewed-by: Shawn Rutledge <shawn.rutledge@digia.com>
When QGraphicsView has sceneInteractionAllowed == false (e.g
when dragMode == QGraphicsView::ScrollHandDrag), all touch
events are accepted. This is wrong, and will stop mouse synthesising
from happening on touch platforms. This in turn will make
ScrollHandDrag not work (since no mouse events will come through).
This patch will call QEvent::ignore() if the touch event isn't send
to the scene, which will cause a mouse event to be synthesised.
Note that according to http://doc.qt.digia.com/qq/qq11-events.html
the correct approach would probably be to just return 'false', rather
than calling QEvent::ignore(). But this logic is not followed
consistently elsewhere (e.g in QApplication::notify), so I choose
to follow what the code actually expects for this bugfix.
Change-Id: Ida777647134c41661bab156d7b164ebd882a6bb1
Reviewed-by: Shawn Rutledge <shawn.rutledge@digia.com>
One reason is when we are using XInput 2.2, because touch devices
will then generate touch events only. For the other X11 scenarios,
X11 does its own mouse emulation.
QPlatformIntegration::styleHint() wasn't overridden yet. The remaining
hints are TODO for now.
Change-Id: I2e444a00a18b33ed840ebfa8d8218655c2c39aad
Reviewed-by: Samuel Rødal <samuel.rodal@digia.com>
It's possible to override event() to receive the QCloseEvent and
do event->ignore() to prevent the window from closing.
Task-number: QTBUG-31019
Change-Id: I9abed47fca02a002b78727f98d678a824854adfc
Reviewed-by: Samuel Rødal <samuel.rodal@digia.com>
Reviewed-by: Stephen Kelly <stephen.kelly@kdab.com>
-"qt-sql" removed in qtdoc
-no longer needed because "Data Storage" and Qt SQL pages already serve
as the main introduction to SQL support in Qt
Change-Id: I7384ae04ab4c09ecc9d76668e4e7f836095066c1
Reviewed-by: Topi Reiniö <topi.reinio@digia.com>
Reviewed-by: Martin Smith <martin.smith@digia.com>
Reviewed-by: Leena Miettinen <riitta-leena.miettinen@digia.com>
defaultFramebufferObject() may be called from anywhere, at any point,
not just makeCurrent(). One example is the glyph-cache, which uses it
to re-bind the default FBO after generating the texture cache.
If the default FBO had already been created, but the render-buffer was
out of sync with the window size, we would end up in the resize code
without the correct context current, and without the render-buffer's
owning FBO bound. This caused "Failed to make complete framebuffer
object 8cd7" warnings at runtime.
We now make the context current and bind the FBO, even though it might
already be bound and the context current from makeCurrent(), or when
initially creating the FBO. For the future we should move the whole
resize logic out of defaultFramebufferObject() and call it from
makeCurrent(), or possibly [EAGLView layoutSubviews]. That's a
higher impact change though, which we reserve for the 'dev' branch.
Change-Id: I50ea949c12a02ad1af6ec9fdc3215d5da85b324f
Reviewed-by: Richard Moe Gustavsen <richard.gustavsen@digia.com>
OpenGL ES2 doesn't support NPOT textures in combination with GL_REPEAT,
so for OpenGL ES2 we use a custom program that emulates repeat by taking
the fractional part of the texture coordinates.
This is not enough though, as merely setting GL_TEXTURE_WRAP_x to GL_REPEAT
with a NPOT texture is an error in some implementations, so we have to
guard the call to updateTextureFilter() in updateBrushTexture() with
a check for OpenGL ES2 and use GL_CLAMP_TO_EDGE instead.
This fixes missing/black backgrounds in the diagramscene example on iOS.
Change-Id: I5020090b5f17faeb06dcab9dc0292459e021af30
Reviewed-by: Richard Moe Gustavsen <richard.gustavsen@digia.com>
Reviewed-by: Samuel Rødal <samuel.rodal@digia.com>
Even though the intentions of this warning were good, the warning
was a bit harsh. In addition, in certain circumstances
(like the autotest demonstrates) we could end up calling object()
on an interface where the object was in the destructor. This
could happen because: *after* we got the destroyed() signal, the
widget would still notify the accessibility framework of a FocusOut
event.
Since the code even called object() from isValid(), we could not even
(as a defensive measure to circumvent this issue) check the isValid()
of an interface without getting this warning (duh).
So - for isValid(), the warning is not needed at
all, since the caller will of course check the result of isValid() and
act accordingly.
As for the result of object(), it should always be a pointer, but it
might point to a partially destroyed object. To detect this, you simply
check isValid() first:
if (iface->isValid())
doStuff(iface->object());
Change-Id: I206307fe618806133d8c6bc338c412d0009d7181
Reviewed-by: Frederik Gladhorn <frederik.gladhorn@digia.com>
Take into account the possibility that the target platform plugin does not
support platform services.
Change-Id: I48e7fac2e1230a9a7d450414044d23ed26b334be
Reviewed-by: Konstantin Ritt <ritt.ks@gmail.com>
At the moment if d->m_data == systemData() it calls
systemLocale()->query but forgets about the standalone part
so you get the wrong data
This patch introduces the new enums so that backends can implement
properly the standaloneMonthName feature properly. At the moment the Windows
and Mac ones still return the monthName, the Unix and Blackberry ones return
the data we store in months_data
Change-Id: Idc5a50b04ab1f914f16c7385be1dca2e027feae3
Reviewed-by: Lars Knoll <lars.knoll@digia.com>
Reviewed-by: Rafael Roquetto <rafael.roquetto@kdab.com>
Reviewed-by: Mehdi Fekari <mfekari@blackberry.com>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
For bundling Qt, we need two things:
1. We need to build a regular .jar file out of the Java files,
so that they can be built into the app package. Dexing the
classes first (i.e. compiling the JVM bytecode to Dalvik
bytecode) is required for loading the .jar file at run-time,
but cannot be used for building it into the app, so we need
two different paths.
2. We need to specify which extra files have to be bundled for
each module (this is primarily for plugins and imports). This
is because there is no static dependency on these files, so
it cannot be detected during deployment.
Task-number: QTBUG-30751
Change-Id: I733603ee5d1c64bd7c5b9357eb5d993b9d0298f7
Reviewed-by: Paul Olav Tvete <paul.tvete@digia.com>
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@digia.com>
Add the enablers so that Qt Creator (or another
deployment tool) can add a specification in the app's
meta data of which libraries are bundled and the
Java code required to extract plugins and imports into
the required directory structure inside the app's
data directory.
This is intended to be an alternative to using Ministro
for deployment, and the mechanism of extracting
libraries on first startup is a work-around for the
requirement in Qt of having this directory structure.
For Qt 5.2, the approach should be changed to load
plugins directly from the app's lib directory and
the other files in imports will be bundled as qrcs
in the native plugins.
Task-number: QTBUG-30751
Change-Id: Ibdb3a672548b4802f9bf3ecd05fc194426ac30e7
Reviewed-by: Paul Olav Tvete <paul.tvete@digia.com>
KeccakF-1600-opt32.c:497:5: error: comparison between signed and unsigned integer expressions [-Werror=sign-compare]
harfbuzz-thai.c:264:49: error: comparison is always false due to limited range of data type [-Werror=type-limits]
These warnings are caused by "char" defaulting to unsigned on ARM. In
particular, the second warning was introduced by commit
785e95ef0a, which is not upstream...
qbenchmarkvalgrind.cpp:224:5: error: variable ‘_qzz_res’ set but not used [-Werror=unused-but-set-variable]
This one was fixed for x86-64 in 7b54571ec2 but not
for the other platforms.
KeccakF-1600-opt32.c:250:5: error: comparison between signed and unsigned integer expressions [-Werror=sign-compare]
This one is wasn't caught before because it applies only to big-endian
code.
Change-Id: Ice33b639e55d95140cbf912bb81b6f508ed3744a
Reviewed-by: Richard J. Moore <rich@kde.org>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
51ee309a79 introduced this check, but it
was supposed to be >= (it's available in 4.8.1, not after 4.8.1)
Change-Id: Id993b128de5c3500684833aea8ef556b31aac5f2
Reviewed-by: Olivier Goffart <ogoffart@woboq.com>
The check in [QIOSOrientationListener orientationChanged] ensured we
never reported the two unsupported orientations through QPA, but we
were reporting back the orientation through QIOSScreen::orientation()
as well, and that didn't have a guard for -1. This resulted in crashes
in client code that assumed the range of QScreen::orientation() was
defined by the enum, such as the paintedwindow example.
The listener now ignores the two unsupported orientations, which leaves
us at the previous orientation. For the conversion function, we still
have to support all UIDeviceOrientations, so we fall back to portrait
for the two unsupported orientations. In the future we should consider
caching the previous value explicitly, or fall back to the interface
orientation.
Change-Id: Ic19d0ce86b4ddea250ea927d5e8664396b2b68fd
Reviewed-by: Richard Moe Gustavsen <richard.gustavsen@digia.com>
Internally iOS double-buffers its rendering using copy instead of flipping,
so we reported that our context was single-buffered so that clients could
take advantage of the unchanged buffer. This failed when clients (such as
Qt itself) then assumed that calling swapBufferes() was not needed.
We now properly report that we're double-buffered, and we'll have to find
another way to report the way double-buffering works if that's still an
optimization we'd like to provide to clients.
Change-Id: Id2e4faa68ed3b837ad01d6f22b2927fc9c9769c2
Reviewed-by: Richard Moe Gustavsen <richard.gustavsen@digia.com>
Reviewed-by: Samuel Rødal <samuel.rodal@digia.com>
Since 39a052c664, QAccessible::State is no
logner an enum that moc understand.
moc currently silently ignores it the Q_ENUMS
Change-Id: Iecc30ad57055fc9ccaa33e9e9c400d96997d0902
Reviewed-by: Jan Arve Sæther <jan-arve.saether@digia.com>
Reviewed-by: Frederik Gladhorn <frederik.gladhorn@digia.com>
QSharedDataPointer does not actually need a class derived from
QSharedData. All it needs is a member called "ref".
Change-Id: I2f7fe4cc143478ef7ef64681eada16e2d4c2e63a
Reviewed-by: Lars Knoll <lars.knoll@digia.com>
Reviewed-by: Jędrzej Nowacki <jedrzej.nowacki@digia.com>
It's not used anywhere, so we don't need to cache the locale data
index. We already have the pointer to the QLocaleData anyway.
This saves us a few roundtrips calculating the index from the data
pointer only to get the data pointer again.
Change-Id: I6905d20a382ddcb9fb04cc886a17499b467f905a
Reviewed-by: Robin Burchell <robin+qt@viroteck.net>
Reviewed-by: Lars Knoll <lars.knoll@digia.com>
Reviewed-by: Jędrzej Nowacki <jedrzej.nowacki@digia.com>
Direct benefit is that the code between the two QLocale constructors
taking language, country and (maybe) scripts is merged.
This will also allow us to cache the QLocale::c() result.
Change-Id: Ia46c5a37764dc287bfcd3a52a022ac413c53a582
Reviewed-by: Robin Burchell <robin+qt@viroteck.net>
Reviewed-by: Jędrzej Nowacki <jedrzej.nowacki@digia.com>
Reviewed-by: Lars Knoll <lars.knoll@digia.com>
We have to call DetachCurrentThread() for each time we call
AttachCurrentThread(). Fortunately we have this convenience
class that we prepared earlier.
Task-number: QTBUG-30847
Change-Id: I5ffb94b336d3787a3bae197bab22b91770d58848
Reviewed-by: Eskil Abrahamsen Blomfeldt <eskil.abrahamsen-blomfeldt@digia.com>
The back button would be non-responsive for 5 seconds after
hiding the software keyboard. This is a minimal change that
does not look into why we need to have a 5 second delay in
the first place.
Task-number: QTBUG-30752
Change-Id: Ied514b77650cea7accc37a03efef2ce861090f65
Reviewed-by: Eskil Abrahamsen Blomfeldt <eskil.abrahamsen-blomfeldt@digia.com>
Most of the logic was still following the 10.6 UI guidelines, and had
not yet been upgraded to the 10.7 new button look. We tried to keep
10.6 compatibility were possible, and improve 'small' and 'mini' Aqua
sizes support.
Change-Id: I64139f24cccd095e9349b27a987395210b55c586
Reviewed-by: Jens Bache-Wiig <jens.bache-wiig@digia.com>
The Config class is further modified to make
use of the current directory information it
stores with each configuration variable.
Task-number: QTBUG-30725
Change-Id: I34c845e6c05d7868266324f1d54e56f94d709f95
Reviewed-by: Jerome Pasion <jerome.pasion@digia.com>
The Config class is modified to build a single
multimap containing a record for each variable
found in each .qdocconf file. Each reacord
contains not only the name and value of the
variable, but also its location in the qdocconf
file it was read from and the path to that file.
This single multimap replaces 3 maps in the
Config class.
Task-number: QTBUG-30725
Change-Id: I049a69790f943b24c014a24b55b2b39725a1b56f
Reviewed-by: Jerome Pasion <jerome.pasion@digia.com>
When the control is read only. This is documented but not implemented.
Add functions to zoomIn and zoomOut.
Task-number: QTBUG-30845
Change-Id: I692b5f8cc5791498d34d35ea3dafa18b6e5d3e65
Reviewed-by: Frederik Gladhorn <frederik.gladhorn@digia.com>
One can zoom in/out text inside a QTextEdit when the widget is
read-only or when using the zoomIn/zoomOut functions.
Zooming inside a HTML document only works if the font-style is not set
to a fixed size inside the document though. Adding this information
to the documentation.
Change-Id: I66a62da53827e1ce3241ba16b91e86926b97c297
Reviewed-by: Jerome Pasion <jerome.pasion@digia.com>
Reviewed-by: Frederik Gladhorn <frederik.gladhorn@digia.com>
The majority format is "<Qt Module> C++ Classes" (see
http://qt-project.org/doc/qt-5.0/qtdoc/modules-cpp.html)
Also, fix a broken link (Qt Network C++ Classes)
"<Qt Module> C++ API" is perhaps the more correct format, but that's
part of a much bigger cleanup: QTBUG-30556
Change-Id: I753365e2bec8d85d9a5f686b4aa35c9eeeaf0871
Reviewed-by: Jerome Pasion <jerome.pasion@digia.com>
With QUrl variants of the static methods, it is possible to get to use
VFS facilities of the platform if available.
Since we can't predict if the application will use the VFS available in
the platform or its own mechanisms, an extra parameter is provided to
restrict the protocols allowed to the user. This extra parameter
defaults to no restriction, which is the most convenient if the platform
file dialog and the application use a matching VFS. It's likely to be
the most common use.
Change-Id: I4c9effde9d194d226cd8b7a140eb9036187ba87b
Reviewed-by: Olivier Goffart <ogoffart@woboq.com>
The first call to resize() was a left-over from before we had retina-support.
Change-Id: I637e8d40f443f81fe7cfc367650bb28b917da2bc
Reviewed-by: Richard Moe Gustavsen <richard.gustavsen@digia.com>
After 475d1ed4f6a21686828fbd018542b469a8b2dbcd in qtdeclarative,
orientation changes on Android were broken, because the resize
event no longer implicitly causes an expose event. So we need
to post both when doing the resize.
Task-number: QTBUG-30909
Change-Id: I87c8c38e14d96a03b3409ef6439c3ac6ef432005
Reviewed-by: Samuel Rødal <samuel.rodal@digia.com>
The mips/uclibc features.h of the toolchain used by a former key
account of PSO is defining both __USE_XOPEN2K and __USE_BSD this
will lead to POSIX_MADV_* and MADV_* being defined while only the
symbols for madvise are present. Change the order to make it link.
Change-Id: If324b978d72ad2b37b8cd624562e81503c9465d4
Reviewed-by: Rafael Roquetto <rafael.roquetto@kdab.com>
Reviewed-by: Andreas Holzammer <andreas.holzammer@kdab.com>
Reviewed-by: Giuseppe D'Angelo <giuseppe.dangelo@kdab.com>
As described in the QTBUG-30872, there may be a race condition involving
3 threads fighting for a mutex. I am surprised it was not caught
before despite all the Q_ASSERT and the stress test in tst_qmutex.
We do not need to call store(0) because the unlocking thread will
eventually remove the BigNumber flag. And perhaps it even did it
already, and another thread has incremented waiters (hence the Q_ASSERT
is wrong)
Here is a paste of part of the description from the bug report:
---
many threads, one of them is ready to release mutex, while at least two other trying to acquire it
d->waiters is 0
Thread 1 release mutex in unlockInternal:
if (d->waiters.fetchAndAddRelease(-QMutexPrivate::BigNumber) == 0)
d->waiters is now -QMutexPrivate::BigNumber
Thread 2 try to acquire mutex in lockInternal:
old_waiters = d->waiters.load();
if (old_waiters == -QMutexPrivate::BigNumber) {
if (d_ptr.testAndSetAcquire(d, dummyLocked())) {
It acquire 'about to release mutex' by changing d to dummyLocked
Thread 1 continue release procedure:
d->derefWaiters(0);
d->waiters is now back to 0
Thread 3 try to acquire mutex in lockInternal:
while (!d->waiters.testAndSetRelaxed(old_waiters, old_waiters + 1));
d->waiters is now 1
Thread 2 continue its dummy lock:
d->waiters.store(0);
d->waiters is force to 0
Thread 3 continue wait procedure
but it realize that mutex was already unlocked so decrease back waiters
if (d != d_ptr.loadAcquire()) {
if (old_waiters != QMutexPrivate::BigNumber) {
d->waiters.deref();
d->waiters became negative value of -1
Neither thread need internal data so it is released back to pool
The waiters counter in released internal structure is still -1
---
Change-Id: I1b22555db764482775db6e64a8c9ffa9e1ab0cf6
Task-number: QTBUG-30872
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
QKeySequence failed to find a match in the shortcut table when QKeyEvent
contained Qt::GroupSwitchModifier modifier. It's not a part of the shortcut,
it simply shifts character group in a keyboard mapping table.
Task-number: QTBUG-26302
Change-Id: Id91cd4999777f7085068e9dba5cb22b40653e23d
Reviewed-by: David Faure (KDE) <faure@kde.org>
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
We can simply clip the update rect against the widget's rect and return
if it's empty. Otherwise we risk ending up with update rects that are
larger than INT_MAX due to multiple update rects being merged.
Task-number: QTBUG-30876
Change-Id: I23bd0149fbe8d1a007a60b228e6bddb45dc4fc32
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Reviewed-by: Gunnar Sletta <gunnar.sletta@digia.com>
Introducing a new method which allows us to know before hand if an URL
scheme will be supported by QNetworkAccessManager. It is especially useful in
combination with QFileDialog URL based methods to pass this list as the
allowed schemes the user can select in the dialog.
Change-Id: If625b045e87959bfd78fea2c9213b69caf506886
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Add a new button with which any color present on the screen can be picked
up. This feature is very useful in any image processing application.
Notice:
--I have setMouseTracking to true because I consider that it makes much
easier to grab the desired color.
--I have set the cursor manually because for some unknown reason the
cursor wouldn't change with grabMouse(Qt::CrossCursor).
Task-number: QTBUG-14332
Change-Id: I39e3543d3ed55276f43d569a2f03087bbf89b27a
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
Reviewed-by: Kevin Ottens <kevin.ottens@kdab.com>
Reviewed-by: Lars Knoll <lars.knoll@digia.com>
This was added just so that moc could pick up the enums and so that
we could use the enums in Q_PROPERTY declarations, which was needed for
accessibility in QML. It turns out that Q_GADGET is enough for us.
This is a strictly a binary compatible change.
However, QAccessible was marked internal in 5.0, so we are free to
change it. In addition, this class is static and cannot be instantiated.
Change-Id: I27e2e97c5f4b45c38678264c6b593a4383db8d3e
Reviewed-by: Frederik Gladhorn <frederik.gladhorn@digia.com>
Setting mouseGrabEnabled means that the window should continue
to receive mouse events even when the mouse is not over the
application. This is not an issue on iOS, but the warning is
still annoying.
Change-Id: I0dd7c3828bcb1a51a4eae534aca1da5bfa258f03
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@digia.com>
The current implementation would never hit the Qt::Tool case, since
a tool is also a Qt::Popup. This patch fixes that by making the
logic more explicit.
Change-Id: I0e6898081a18289e1007c8a168b374740915b3ff
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@digia.com>
This makes it possible to listen for events on xcb_window_t which are
not platformwindows inside the xcb plugin
Change-Id: Ic9ec17ed757a7f9a5302ef2759c119a72bac573c
Reviewed-by: Samuel Rødal <samuel.rodal@digia.com>
Basically you don't want to grab the X server while your debugging.
Also added an environment variable which lets you force to not grab
the X server
Change-Id: Iba03f11c8f486ce71c55fac7716bffcb7cc8cb98
Reviewed-by: Samuel Rødal <samuel.rodal@digia.com>
QNSWindowDelegate was not handling windowShouldClose, which is how you
can tell Cocoa that your window should not close if the close button is
pressed. This change moves the close handling from windowWillClose to
windowShouldClose, and adds an optional "accepted" pointer to
QWindowSystemInterface::handleCloseEvent so that QNSWindowDelegate can
return a true/false value for whether the window should actually close
Task-number: QTBUG-28965
Change-Id: I67c6296ad42cbeeb71413e05411467d4e558adb4
Reviewed-by: Morten Johan Sørvig <morten.sorvig@digia.com>
Use CamelCase for module name(s) used in in .qdocconf - this is
required as qdoc will generate visible output (tags in example
manifest files) based on these names.
Change-Id: Ie246e740203ee0b996fea5dee612bf7f61638991
Reviewed-by: Jerome Pasion <jerome.pasion@digia.com>
Use a slightly better regular expression for splitting module names
into tags used for example manifest files.
This will correctly split words with consecutive capital letters
(e.g. QtDBus)
Change-Id: I1320e08a1fbd44f718b82a1fcfea19eabca035fc
Reviewed-by: Martin Smith <martin.smith@digia.com>
Instead, add QCocoaWindow::setEmbeddedInForeignView which can be called
via QPlatformNativeInterface::NativeResourceForIntegrationFunction
Task-number: QTBUG-30805
Change-Id: I05861e80ca664ddb430216388cf0fec573a4d32b
Reviewed-by: Morten Johan Sørvig <morten.sorvig@digia.com>
Apply 0293aff5c44202e5c62e229b74d8bd0bf9206185
from Qt 4.
Without this, calls to deleteLater() may create delete
later events with a loopLevel of 1. Those events will
not be processed until QApplication::exec() returns.
Add a QScopedLoopLevelCounter that increases the loopLevel
for the duration of the activated() call.
Task-number: QTBUG-30660
Change-Id: I7ab3bb3a53243691b8f7f64e025150e5cc7da2c8
Reviewed-by: Gabriel de Dietrich <gabriel.dedietrich@digia.com>
This way I can have in my QtQuick something like
Text {
text: "<font color='#ff0000'>H</font> <font color='#99ff0000'>H</font>"
}
and it works properly
QtQuick already supports #AARRGGBB for color: properties so I've
decided to go the notation
Once this is merged we can remove the extra code
in QQuickColorProvider::QColorFromString
I've also added some tests for the hex -> QColor conversion that where
non existent
Change-Id: I1dd4a2ec113293aec26968329b2e4930df6fdcb7
Reviewed-by: Gunnar Sletta <gunnar.sletta@digia.com>
Since we use native Cocoa menus, we cannot rely on the
normal shortcut handling. Shortcuts can be overridden by
the currently focused object in Qt.
In order to make that possible we need to send a
QShortcutOverride event before accepting any key event.
For menus the key event goes from the NSApp directly to
the menu, so the shortcutOverride would not work.
This is mostly an adaptation of the Qt 4 code.
Task-number: QTBUG-30695
Change-Id: Icb4979309d2d6f9606eb9c8abc4130dc79926593
Reviewed-by: Gabriel de Dietrich <gabriel.dedietrich@digia.com>
On Windows 8 it would end up changing the look of the QLineEdit when the
mouse hovered over it even though it was not enabled. None of the
Windows platforms show the lineedit changing when hovered over if it is
disabled so we can skip the whole thing.
Task-number: QTBUG-29224
Change-Id: Ib9495bf395477f114e91b744e1b1209c9e11f336
Reviewed-by: J-P Nurmi <jpnurmi@digia.com>
The hotspot was not taken from the QCursor so if one was set then it
was reset to 0x0.
Change-Id: Ie81f1c2ac15a16f10436738367e612c44dc42d38
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
Reviewed-by: Richard Moe Gustavsen <richard.gustavsen@digia.com>
Reviewed-by: Morten Johan Sørvig <morten.sorvig@digia.com>
Don't try anything after the original syncing, particularly after
changing the menu item's text. Also, don't try anything if the menu
item cannot be linked up to a menubar (see QTBUG-30756). This latter
point requires extra syncing after adding a menu in the menubar.
Finally, to be able to find the menubar, we need to clean the code for
moc's eyes.
Task-number: QTBUG-30756
Task-number: QTBUG-30812
Change-Id: I88fad663f1c35d03a0cb167d1723d16f590918c0
Reviewed-by: Morten Johan Sørvig <morten.sorvig@digia.com>
QCocoaMenu is child of either a QCocoaMenuBar, a QCocoaMenuItem as a
submenu, or nothing as a standalone menu. QCocoaMenuItem is child of
its containing QCocoaMenu.
The parent is set during insertion and cleared during removal.
QMenu needs to be updated to avoid double deletion and leaking its
own platform menu.
Change-Id: Iadf60d8062d7466fa616f84f3761fe322fc9aa2e
Reviewed-by: Morten Johan Sørvig <morten.sorvig@digia.com>
Instead, set the currently selected filter's suffix as
default suffix of the dialog unless another default suffix
has been set.
This emulates the behavior of Qt 4 behavior which would set
the selected name filter's suffix as default suffix in
QFileDialog::getSaveFileName().
Task-number: QTBUG-30748
Change-Id: I111cd6190ddab8775a0fa72b94b3c728dd411c5e
Reviewed-by: Oliver Wolff <oliver.wolff@digia.com>
Reviewed-by: Andy Shaw <andy.shaw@digia.com>
Previously, accessibleTree->child(0) would return an interface for the
header even if it was hidden.
Also, the assertion was wrong since the index would be 0 if both row
and column were 0. The assertion was actually found while using the
project explorer of Qt Creator (2.7)
Change-Id: I9f3cc2c13b6887569d10c4e062a64552f898231a
Reviewed-by: Frederik Gladhorn <frederik.gladhorn@digia.com>
Make sure the value of QWindowsWindow::isExposed is in sync
with regions we expose.
This provoked a couple of existing issues in the qwidget test.
setWindowGeometry tested that windows with invalid sizes got
exposed on screen. They didn't, but because the plugin sent
bogus events, these used to pass. Same with windowMoveResize.
The expect fails are also rather bogus. Showing invalid-size
widgets could be considered undefined behavior. The Window
manager could resize it, choose to not show it at all, etc,
but they now pass on windows.
resizeEvent has been broken since 5.0.0, but the test didn't
spin the event loop so the second event didn't get delivered
before the test completed.
Task-number: QTBUG-30744
Change-Id: I3a9efcd095f366126a87739f4248185b6c81d407
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
UNKNOWN now means generic unix theme, rather than Gtk: there
are other KDE/Qt-based desktops, and there's also the case of
no desktop at all (e.g. bare Xvfb, as used by automated builds).
To make this correct for GTK-based desktops, this commits improves
the detection of the current desktop environment.
Change-Id: Ib696624de39d5024527880df7adc26c65b838d15
Reviewed-by: J-P Nurmi <jpnurmi@digia.com>
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
Reviewed-by: Jens Bache-Wiig <jens.bache-wiig@digia.com>
Reviewed-by: Shawn Rutledge <shawn.rutledge@digia.com>
Change-Id: I8485031edc623f99b4b858d4f777be43f4bc3264
Reviewed-by: Samuel Rødal <samuel.rodal@digia.com>
Reviewed-by: Gunnar Sletta <gunnar.sletta@digia.com>
This change restores a proper function of the "(?)" button in the window
decorations which is used as a clue for the user to check what a particular
widget is supposed to do. The change is only implemented for QtWidgets because
the underlying QWhatsThis is inherently widget-specific -- which is why it sends
an event to QGuiApplication, but only processes it in the QtWidget-specific
QApplication.
Thanks to Alberto Mardegan and Gunnar Sletta for their feedback on this patch.
Change-Id: Ibb912e3960f1e9aec54c5ed77ade1c6744d6ca23
Reviewed-by: Alberto Mardegan <mardy@users.sourceforge.net>
Reviewed-by: Gunnar Sletta <gunnar.sletta@digia.com>
Reviewed-by: Samuel Rødal <samuel.rodal@digia.com>
- don't check if m_directWriteFontFace is not null, it MUST be initialized
just in the constructor;
- fix support for surrogate pairs in stringToCMap();
- reorder some affected code for better readability.
Change-Id: Ibdcf7b50f06910e95cefe074f33932b8fbcdcfe1
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
Reviewed-by: Konstantin Ritt <ritt.ks@gmail.com>
We create our QIOSViewController in didFinishLaunchingWithOptions,
and schedule a timer to run the user's main. If the device is
placed in landscape orientation at startup, we will receive a
willRotateToInterfaceOrientation message before the timer is
triggered to run the user's main, which means we do not yet
have a QApplication.
To fix this crash we exit early, but we might have to store the
new orientation for later, and make sure the initial QScreen is
then created with the correct orientation.
Change-Id: I0cc02f0d36b992d190736e98858dc7d002d595b7
Reviewed-by: Morten Johan Sørvig <morten.sorvig@digia.com>
Find effective screen by searching the virtual sibling that
contains the center as does QDesktopWidget::screenNumber().
Task-number: QTBUG-30724
Change-Id: I8441ab4f3e5ee8169613a82f150d1a4f1777b662
Reviewed-by: Samuel Rødal <samuel.rodal@digia.com>
Not trying to load the lib saves 30 - 50ms upon an apps' 1st host
lookup.
Task-number: QTBUG-30809
Change-Id: Id893cec09ff57494776625700c93f7efe96fcc6b
Reviewed-by: Thomas McGuire <thomas.mcguire@kdab.com>
Reviewed-by: Rafael Roquetto <rafael.roquetto@kdab.com>
Since change 3bb9024952 the documentation is invalid.
Task-number: QTBUG-29680
Change-Id: I7d5fcb6bc490aa5cba83439d33f798459640c42d
Reviewed-by: Jørgen Lind <jorgen.lind@digia.com>
Creating a second QFactoryLoader for the same plugins seems to trigger
an unload of the plugins loaded by the first factory loader. The
QIconEngine created by the SVG icon plugin thus gets an invalid virtual
table pointer, which causes a crash when attempting to call any virtual
function in the QIconEngine (pixmap(), the virtual destructor, etc).
Reusing a single QFactoryLoader instead fixes the crash.
Task-number: QTBUG-30496
Change-Id: I80c5fa8b52ab9b0db68499f8c37fad14a1ac4f3c
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
Failure to initialize the variable can cause spurious non-zero
values.
http://msdn.microsoft.com/en-us/library/windows/desktop/ms715438(v=vs.85).aspx
"..value can either be a SQLULEN value or a null-terminated character
string. If the value is a SQLULEN value, some drivers may only write the
lower 32-bit or 16-bit of a buffer and leave the higher-order
bit unchanged. Therefore, applications should use a buffer of SQLULEN
and initialize the value to 0 before calling this function. Also, the
BufferLength and StringLengthPtr arguments are not used."
Follow-up to 1509316a37
Change-Id: I2e92eb845a2590bea0849c52bde8902adff1b419
Reviewed-by: Andy Shaw <andy.shaw@digia.com>
... but rather throw an error, so the HTTP layer can recover from a SSL
shutdown gracefully. In case the other side sent us a shutdown, we should
not send one as well, as it results in an error.
Change-Id: Ie7a56cf3008b6ead912aade18dbec67846e2a87e
Reviewed-by: Richard J. Moore <rich@kde.org>
ignoreProxyFor() always returned true if the no_proxy was not set,
which resulted in the first token being an empty QByteArray, causing
the endsWith() check to always evaluate to true.
Add a unit test that is enabled for those platforms that use the generic
system proxy.
Change-Id: I6081ad5e0b8e2c3fee1568835907c32bde5b7772
Reviewed-by: Peter Hartmann <phartmann@blackberry.com>
This functionality was in Qt4's qapplication_x11.cpp and was missing
from the XCB QPA plugin. Ported the code from xlib to xcb.
This code was actually tested (with plasma), unlike the Qt-4.8 code which
skipped every other character...
for (uint i = 0; i < 20 && i + sent <= length; i++)
xevent.xclient.data.b[i] = message[i + sent++];
Provide a QPA native-function for accessing the startup id, for cases where
an application doesn't show a window, but starts another app instead, or asks
a running app to show the window on its behalf.
Change-Id: If392179efddd70a51c45a8fab4fb9d753913094a
Reviewed-by: David Faure (KDE) <faure@kde.org>
Currently, running a Qt5 widget app in Xvfb (with its default setup, 8 bit color depth)
outputs a stream of error messages, starting with:
QWidget::paintEngine: QWidget::paintEngine: Should no longer be called
Better output clearly the reason why we end up in that method: painting
on a null image, due to the unsupported image format.
Maybe this should even be a qFatal...
Change-Id: Iae8f1b057518c146bf8e034999c1b5e67ce6ef5e
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
Reviewed-by: Samuel Rødal <samuel.rodal@digia.com>
The pointer returned by unicode() was deleted right afterwards.
Change-Id: I7cef72386d02c7be42e71134f616926506d37ea6
Reviewed-by: Gatis Paeglis <gatis.paeglis@digia.com>
Reviewed-by: Konstantin Ritt <ritt.ks@gmail.com>
It was only used for toUpper/toLower but always computed in the
constructor, including QString::toLatin1 conversion and allocations.
This needlessly slows down all other uses, including supposedly "cheap"
operations QString::toDouble, or accesses inside QResourceFileEngine.
The benchmarks indicates that doing it always when needed is bearable.
There's still a lot of improvement potential on these code paths.
Change-Id: I88b637ee11f9f7ea614f8da4ec5df0bf40664fce
Reviewed-by: Konstantin Ritt <ritt.ks@gmail.com>
This also fixes a few "sloppy" mentionings of "new" into "new []".
Task-number: QTBUG-30777
Change-Id: I3f4add07777b59cd09cac97b672c73273b3b97eb
Reviewed-by: Jerome Pasion <jerome.pasion@digia.com>
Since QDateTime is pimpled, default allocation is expensive and
regularly shows up in profiles of code using QFileInfo.
For Qt 6, QDateTime's data members should be put into the class
proper, and this change here reverted.
Change-Id: I94a50e467b12772e1076181eb2ac6031984d8802
Reviewed-by: Thorbjørn Lindeijer <bjorn@lindeijer.nl>
Reviewed-by: Stephen Kelly <stephen.kelly@kdab.com>
If the QGLWidget's creation is triggered by a sibling or similar being
created it would end up without the surface type OpenGLSurface. We need
to make sure the QWindow has the correct surface type to prevent
QOpenGLContext::makeCurrent() from failing.
Task-number: QTBUG-30811
Change-Id: Ifc85f9120d89bd0bed4a3911233799830cb3a0f3
Reviewed-by: Jørgen Lind <jorgen.lind@digia.com>
Previously if l had a parent, addChildLayout would warn and skip the
reparenting, but it would still add the sub layout to the layout.
This caused some inconsistencies in the hierarchy which in worst case
could cause crashes.
Task-number: QTBUG-30758
Change-Id: I618ec3341636b97bd71e421201b22c746dcf43e1
Reviewed-by: Paul Olav Tvete <paul.tvete@digia.com>
Task-number: QTBUG-30125
Change-Id: I4e56fd3021b4ef5f344d4d36ae594dd88e2aa1bd
Reviewed-by: Morten Johan Sørvig <morten.sorvig@digia.com>
Reviewed-by: Gabriel de Dietrich <gabriel.dedietrich@digia.com>
Change-Id: Iae268f30b86a67fcf978983ca9b12f850948dd24
Reviewed-by: Volker Krause <volker.krause@kdab.com>
Reviewed-by: Stephen Kelly <stephen.kelly@kdab.com>
Change-Id: I138a29e1099e691770b9e3e094a108fc8c52f8cb
Reviewed-by: Andreas Holzammer <andreas.holzammer@kdab.com>
Reviewed-by: Giuseppe D'Angelo <giuseppe.dangelo@kdab.com>
Task-number: QTBUG-30346
Change-Id: I3d6dbe1e88bb5e2748eadabb2663f30be16f8d18
Reviewed-by: Stephen Kelly <stephen.kelly@kdab.com>
Reviewed-by: Mark Brand <mabrand@mabrand.nl>
If the application object is an ancestor of QMenu, dereferencing qApp
in its destructor will cause a crash.
Task-number: QTBUG-30756
Change-Id: I31a33db0fd783bb210a420618911ea8b412e9a0f
Reviewed-by: Frederik Gladhorn <frederik.gladhorn@digia.com>
The QAction may stay around while the parent/owner gets deleted.
This is for example the case for some dynamically created menus.
This is required for fixing QTBUG-30792
Change-Id: I7b6122edec6def69aed77502403134e1568e21c9
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
Reviewed-by: Jan Arve Sæther <jan-arve.saether@digia.com>
Then it could be used in QML world.
Task-number: QTBUG-28667
Change-Id: Ib65ad3b271a07ba50d3ca50d61f486f829f69119
Reviewed-by: Gunnar Sletta <gunnar.sletta@digia.com>
This is already fixed upstream in
https://codereview.appspot.com/8615046/patch/1/2
Change-Id: I4b9e865f6b5622c484418a8381334381bc256887
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
Added option grab=1/0 which allows user to choose using environment variable
if the application will exclusively grab the input device or let
the OS read it in paralel(default).
Task-number: QTBUG-30004
Change-Id: If3caa8419584be46f320931ddb152a41893d8693
Reviewed-by: Samuel Rødal <samuel.rodal@digia.com>
Reviewed-by: Andy Nichols <andy.nichols@digia.com>
Focus policy can be wider in this case given the limited scope of the
editor widget. This helps workaround platform specific focus policies,
like on Mac, were focus can be restricted depending on the widget type.
Task-number: QTBUG-30715
Change-Id: I69acf8ebff4ba16d473964c91680a1cb0235e3cf
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
Reviewed-by: Jake Petroules <jake.petroules@petroules.com>
Reviewed-by: Stephen Kelly <stephen.kelly@kdab.com>
This is the patch from the bug report with a few alterations to get it
to compile, and also with the GroupSwitchModifier code removed, as this patch
just focuses on Qt::KeypadModifier. The problem was determined to be in
QKeySequencePrivate::encodeString, which doesn't handle the
Qt::KeypadModifier flag.
Task-number: QTBUG-4022
Change-Id: Ic981eb8b5cd88c7b36892d3019b8175db4b7b6f2
Reviewed-by: David Faure (KDE) <faure@kde.org>
m_cocoaPlatformWindow is cleared when Qt is "done"
with the QNSWindow or QNSPanel to prevent messages
from Cocoa calling into possibly stale pointers.
Change-Id: Id42b794fedbd3e597a0c4b50584cca976ba74c0f
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
UnknownRole draws attention from the screen reader
and makes it focus on the unknown item. GroupRole
is ignored and passes focus to child items.
Change-Id: If7b4cd9ec02b1890929a709b84d897f452c39587
Reviewed-by: Frederik Gladhorn <frederik.gladhorn@digia.com>
Setting this flag causes scroll event lag, so we want
to keep it off for all widgets that do not need touch
events. QPanGestureRecognizer is installed on all
QAbstractScrollAreas. Prevent it from setting the
flag.
Change-Id: Idd4fcc545ff26377607b56f75db75c2865a5fc82
Reviewed-by: Gabriel de Dietrich <gabriel.dedietrich@digia.com>
It is necessary to properly initialize the icon loader in all case.
Otherwise some calls might give wrong results if a platform theme plugin
is involved.
For instance, we might miss the actual theme name reported by the
platform theme, eg it's what happen with QStyle::standardPixmap if no
one created a QIcon before its first call.
Also clean up the accesses to the global static and have only
QIconLoader::instance() use it. All other call go through the static
method. This way only instance() needs to call ensureInitialized(),
definitely safer and looks cleaner to me.
Change-Id: Id3ea6816edd5a65586004f69464960abc3602cf8
Reviewed-by: Olivier Goffart <ogoffart@woboq.com>
Simply say that the behaviour is undefined if you don't do what you must
do. I don't want to introduce a check: it can't be done reliably anyway.
Task-number: QTBUG-30806
Change-Id: Iba1bbbdfe62ffcb133f9c52215efdcc0ee7bd9bd
Reviewed-by: Olivier Goffart <ogoffart@woboq.com>
Some screen readers will pass in child id's that are negative
as response to notifications.
We should always check for negative id's on incoming calls.
Task-number: QTBUG-30792
Change-Id: Idaba3d1931d35ed068cfd9f20e70aa26da427616
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
Reviewed-by: Jan Arve Sæther <jan-arve.saether@digia.com>
There may be multiple libraries specified in the mkspec, such as
EGL and Mali, as used in devices/linux-sh4-stmicro-ST7108-g++, so
create an imported target for each one. Also populate the
Qt5Gui_EGL_LIBS variable with all created imported targets. Similar
variables are created for the used OPENGL implementation.
In the case of using the packaged ANGLE library, we already know the
exact locations of the binaries.
This makes it possible for third parties to use the same GL
implementation as used by the Qt build itself. As these are used only
privately by QtGui, they are also added to the DEPENDENT_LIBRARIES
of that target so that they are found for rpath-link usage.
On some platforms (eg Raspberry Pi), multiple include directories must
be set to include egl.h, as the headers it includes for vcos are a
bit scattered.
Task-number: QTBUG-29132
Change-Id: I1126da3d37cd51c88d3670347c8b6405b285efb5
Reviewed-by: Volker Krause <volker.krause@kdab.com>
Reviewed-by: Stephen Kelly <stephen.kelly@kdab.com>
Keep the content aligned to the bottom when the view has been scrolled
to the bottom and the content is relayouted (for example due to
sorting).
Task-number: QTBUG-30653
Change-Id: I9513e295e276d25ff2068036cd80dbf91314fe84
Reviewed-by: Thorbjørn Lund Martsum <tmartsum@gmail.com>
Reviewed-by: Stephen Kelly <stephen.kelly@kdab.com>
Fixes regression against Qt 4. This tries to recreate the logic
from 5c46d9a4c85abbcc0b5db2bbbafded3efd784cd9 in Qt 4, where we
on Gnome would override the default hint style specified in
FontConfig with Xft.hintstyle settings. This is the configuration
used for changing the hint style in the Gnome Tweak Tool.
Task-number: QTBUG-29582
Change-Id: I6b9fe2c8ff55ff080d034e5a53fc8cbb49f7651f
Reviewed-by: Konstantin Ritt <ritt.ks@gmail.com>
Reviewed-by: Paul Olav Tvete <paul.tvete@digia.com>
This patch adds SizeAdjustPolicy to QAbstractScrollArea.
If it is set to AdjustToContents it will make use of the new
protected viewportSizeHint() (BC since it was reserved in Qt5).
This function returns a suggested size based on contents.
Change-Id: I5d8aa517e88b8b21c5712e62b4d574c3aad99d3b
Reviewed-by: Stephen Kelly <stephen.kelly@kdab.com>
This reverts commit 88272d6c04,
which accidentally re-introduced a crash.
Change-Id: Ie60bcf55f940b24f982eed4f20c6af3320b040a7
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@digia.com>
When registering m_touchDevice with QWindowSystemInterface, it
will also transfer ownership to it. It will eventually be
deleted in clenupDevicesList on qtouchdevice.cpp.
This patch will also stop the app from crashing when telling it
to quit.
Change-Id: I97070efdf16b4db7d076935fbec62e60f094a7df
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@digia.com>
Make sure ImageSSE2.cpp is compiled with the right flags (-msse2)
Change-Id: I410bbd1fd78112d01ebd486908cd22d827e2e472
Reviewed-by: Jonathan Liu <net147@gmail.com>
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
There is already an enum to disable SSL session tickets, which has been
used to disable session sharing for now. However, SSL session sharing
is not the same as SSL session tickets: Session sharing is built into
the SSL protocol, while session tickets is a TLS extension (RFC 5077).
Change-Id: If76b99c94b346cfb00e47366e66098f6334fd9bc
Reviewed-by: Richard J. Moore <rich@kde.org>
Now that the relevant function is available we can do the conversion
inside QMacPasteboardMimeTiff when obtaining an image from the clipboard
Change-Id: Ie41b0fffedf4dd3f8ef49431482bd075a69722b1
Reviewed-by: Morten Johan Sørvig <morten.sorvig@digia.com>
The signal needs to be emitted directly as the event is not passed to
the QLineEdit if the QSpinBox gets the Key_Return. Since this signal
may be relied upon then we ensure it is emitted directly.
Change-Id: I17cdec62c9f995bacfd7d3cc66d6324f26c84c67
Reviewed-by: Stephen Kelly <stephen.kelly@kdab.com>
For most bootstrapped tools, the QT_EVAL code will not
have any effect, because most of the tools don't instantiate
a QCoreApplication. However, qdoc is bootstrapped for cross
compilation, and will instantiate QCoreApplication which
calls the QT_EVAL code. Since the QT_EVAL code requires QObject,
and QObject requires moc, it does not make sense to compile the
eval code into the bootstrap library. Instead, we simply disable
it to make sure the build succeeds.
Change-Id: I472803572b070df041014d337c23d3f3dc0749e4
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@digia.com>
Invoke slot "beep" on QPlatformNativeInterface. Implement for
Windows and X11.
Task-number: QTBUG-30416
Change-Id: I2be651165b899e5147818a012001d354827bb090
Reviewed-by: Gunnar Sletta <gunnar.sletta@digia.com>
Add QWindow::alert() and QPlatformWindow::setAlertState().
Add logic to clear alertion state when the window becomes
active. The platform plugins then only need to implement a setter
and a cheap getter and need not handle activation.
Prototypically implement X11 and Windows.
Task-number: QTBUG-30416
Change-Id: Ia70c4722d812462a21f4034b7d52735c9f2bc49c
Reviewed-by: Jerome Pasion <jerome.pasion@digia.com>
Reviewed-by: Gunnar Sletta <gunnar.sletta@digia.com>
That would set the full screen window in a weird state, where is was
full screen (but unable to exit that mode) and showing the title bar
at the same time.
Change-Id: I3ac913876f3b326504dd5af18c34181d002509d0
Reviewed-by: Shawn Rutledge <shawn.rutledge@digia.com>
Task-number: QTBUG-21150
Change-Id: I14c214e9f96892f0da4369e7253e363b7313c252
Reviewed-by: Kai Koehne <kai.koehne@digia.com>
Reviewed-by: Stephen Kelly <stephen.kelly@kdab.com>
Xcode builds projects for armv7s if a compatible device is set as
target device. If trying to include <QtCore/QtCore> into such
projects, the build fails complaining about "the usage of sizeof missing
QMutexData definition in qgenericatomic.h"
The reason is that qprocessdetection.h fails to pick up that we're
building for ARM, and includes qatomic_gcc.h instead of qatomic_armv7.h.
So we need to check for __ARM_ARCH_7S__ as well. In addition,
iPhoneOS6.1.sdk/usr/include/arm/arch.h will define _ARM_ARCH_7 if
any of the more specic ARM defines are defined, so I add this check
as well to more easy support new version of ARM7.
Change-Id: Ic51a4d7ac99f7f6ba1065f870b3ef82d1250b56c
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@digia.com>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
If building angle ourselves, that's just the basic Qt include dir,
and if using an external gl, look for it in the places specified
in the mkspec.
As the qopengl.h header includes the gl header, this is a 'public
include dependency' of QtGui, so it is added to the relevant
variable and the INTERFACE_INCLUDE_DIRECTORIES of Qt5::Gui.
Change-Id: I8c2c1782e0a2600032771175444b087da28433fc
Reviewed-by: Volker Krause <volker.krause@kdab.com>
Reviewed-by: Stephen Kelly <stephen.kelly@kdab.com>
As it stood, we always relied on the root view controller
being a QIOSViewController if isQtApplication() returned
true. For mixed application, this might not always be true
as native code can choose to replace the root view controller
at times, or even rip it out, and place it as a child of
another (e.g UISplitViewController).
This change will give an extra protection against that.
Change-Id: I0cb85796a8b82f9037c32f9e85e04e1dc7aad8e2
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@digia.com>
This change adds words from the module name (QHP 'project' name defined
in .qdocconf files) as tags for the module's examples. This makes
searching for examples easier in Qt Creator: For example, typing
'multimedia' will list all examples in Qt Multimedia and Qt Multimedia
Widgets modules.
Other minor changes:
- Exclude 'qt' as a tag (not needed)
- Exclude one-character strings as tags
Task-number: QTBUG-28720
Change-Id: I53751b7a87ff39ee7b648f865c9090c52444de76
Reviewed-by: Martin Smith <martin.smith@digia.com>
Unless we enable those buffers for the FBO that backs the
backingstore, setting a clip region on an associated QPainter
will not work.
One apparent bug from this was the menubar. Without this patch
it appeared to never be drawn. The reason was that we ended up
drawing the menubar background over the whole menubar instead
of inside the clip region.
Change-Id: I25660cec6ce9e43fe4cd693127dca6afeb8dcf65
Reviewed-by: Samuel Rødal <samuel.rodal@digia.com>
In 1b08e0307d, I removed the null check
by accident. It's possible for the Darwin API to return a null
property list.
Task-number: QTBUG-30760
Change-Id: Iaf0125767fe4b47c19810b70483a5219e94e4305
Reviewed-by: Morten Johan Sørvig <morten.sorvig@digia.com>
Followup to 7b9d4531 - the second part refactors
to call setDevicePixelRatio but the call itself
was left out.
Change-Id: I3e36452603fe1d7d53fa1a74d87169efea1c2e78
Reviewed-by: Gabriel de Dietrich <gabriel.dedietrich@digia.com>
This bug surfaced after we switched Mac over to use
clang's libstdc++, which has __builtin_popcount but
does not mark it as constexpr.
Change-Id: I4260af48b00c6db3322e52fb113075d305b1e1ec
Reviewed-by: Stephen Kelly <stephen.kelly@kdab.com>
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@digia.com>
We are passing a QSslConfigurationPrivate that is allocated on the stack
(in QSslSocketBackendPrivate::initSslContext()) to
QSslConfiguration::QSslConfiguration(QSslConfigurationPrivate *dd).
When the SSL context is destroyed, this object is not there any more.
So now we create a deep copy of the configuration like we do in
QSslSocket::sslConfiguration().
Task-number: QTBUG-30648
Change-Id: Iaefaa9c00fd6bfb707eba5ac59e9508bf951f8a5
Reviewed-by: Richard J. Moore <rich@kde.org>
Shown for example by Qt Designer.
Change-Id: Ia866a93a781a027aa3703f44314954888d75d436
Reviewed-by: Richard Moe Gustavsen <richard.gustavsen@digia.com>
Change-Id: I7d308a869fdc729bd46ace336b71c7e30556d65d
Reviewed-by: Sean Harmer <sean.harmer@kdab.com>
Reviewed-by: Samuel Rødal <samuel.rodal@digia.com>
Apply the logic from commit d7ae34fdfd
(List the Release library before the Debug library in cmake files.,
2013-02-21) to the Qt5::WinMain library too.
Task-number: QTBUG-29186
Change-Id: Ie465fef1cc0fc842d86c5bc69ab84ec65ec652d9
Reviewed-by: Andy Shaw <andy.shaw@digia.com>
The warning is useful for pointing out the the return
value will be null. However, code that correctly
checks the return value (such as qt_macWindowIsTextured),
still causes the warning to be printed.
Change-Id: I3828992b3d5e7b08451cf0e051b937fa9d9536d3
Reviewed-by: Jan Arve Sæther <jan-arve.saether@digia.com>
Also, make sure platform menu item is deleted on ActionRemoved event.
Change-Id: Ic07a81cb77833bdffd1464abf1c81ebdee4d16e9
Reviewed-by: Morten Johan Sørvig <morten.sorvig@digia.com>
The release() function also requires the QPoint to be set which was
missing from the code snippet.
Change-Id: I4a5cd2bed0915eb1fbaac3a34fb229eac9c284df
Reviewed-by: Topi Reiniö <topi.reinio@digia.com>
Reviewed-by: Geir Vattekar <geir.vattekar@digia.com>
If a native separator was put in the new name when renaming a file name
via the file dialog then it would correctly fail. But if a non-native
one was used on Windows then it would cause the file to be moved
instead if the directory existed.
Change-Id: If01760b8c54a69b600c9a44c7509017be70d33e3
Reviewed-by: Stephen Kelly <stephen.kelly@kdab.com>
While the raster platform plugin supports multiple top
level windows, this is not supported on the GL plugin,
so if you use GL or QtQuick2 in your app and use several
top levels, the app would crash with an error message.
A problem is that the top-level SurfaceView is a special
overlay View and does not support being stacked in a
layout. So instead, we let all windows share the same
GL surface and draw on top of each other. This works
fine for simple use cases.
We implement a new platform capability to make sure no
top level windows (even combobox popups and dialogs)
get non-fullscreen geometries. That has never
worked properly with the eglfs plugin.
Task-number: QTBUG-30473
Change-Id: Ia1438019638fc739cc93ffe79b46b81631254df2
Reviewed-by: Paul Olav Tvete <paul.tvete@digia.com>
This is simply removing some slow and rather ugly hacks that tried
to reconstruct broken data from the windows xp theme. After testing
without these hacks on both XP and Windows 7, I believe we can
conclude that these workarounds are no longer required and even
breaks our look and feel in several places.
Task-number: QTBUG-29888
Change-Id: Ifaffd660e8d9ed6dfd43657745c3fa1606d33a7c
Reviewed-by: J-P Nurmi <jpnurmi@digia.com>
Reviewed-by: Jens Bache-Wiig <jens.bache-wiig@digia.com>
Reviewed-by: Gunnar Sletta <gunnar.sletta@digia.com>
Make QOpenGLPaintDevice::metric(PdmDevicePixelRatio)
return d->devicePixelRatio instead of 1.
Change-Id: I4cf9dd552a700b958212edc8efb990a45e77fd66
Reviewed-by: Richard Moe Gustavsen <richard.gustavsen@digia.com>
An RTL menu containing a menu item which opens a submenu was showing
the wrong arrow if the application's direction was not also RTL.
So now the test for QTBUG-30595 can be simplified: no need to set
the application direction, and therefore less chance of failure.
Change-Id: Id140656206c6fefea3649289477dc54c77e2dd5e
Reviewed-by: Jens Bache-Wiig <jens.bache-wiig@digia.com>
The problem occurs at least with nvidia cards when vsync is enabled
in the control panel.
Task-number: QTBUG-29686
Change-Id: I6fd2a3560a5baeeac7c3fe0440db85904c45026d
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
Fix Qt Creator hang when copying a stack trace from an
application showing a runtime assert.
Change-Id: I874bd48643ebce1a3551644dc850cb7cf5869522
Reviewed-by: Kai Koehne <kai.koehne@digia.com>
Reviewed-by: Joerg Bornemann <joerg.bornemann@digia.com>
Using QFileSystemWatcher to update cache when directory is changed. It is
needed to correctly filter files created after the dialog opening.
Task-number: QTBUG-30604
Change-Id: I5479f9a54095de59ca2ba36b2bd4906de0907ac8
Reviewed-by: Morten Johan Sørvig <morten.sorvig@digia.com>
qdoc has been modified to emit a compact list of the
classes that have one or more obsolete members. The
command is:
\generatelist obsoletecppmembers
This generates an index of all such classes,
where each class name is a link to the class's
subpage of obsolete members. A class's subpage
of obsolete members is also accessible from the
class's reference page, but now it is also
accessible from this index.
Also, The command shown has been added to the
page obsoleteclasses.html in the generated
output. This page already contains the index
of obsolete classes.
Currently, no such output is generated for
QML types and QML types with obsolete members.
But qdoc does accept commands for those:
\generatelist obsoleteqmltypes
and
\generatelist obsoleteqmlmembers
...but qdoc doesn't know what to do with
those commands yet.
Task-number: QTBUG-30270
Change-Id: If19a3b977f64c948e4bd6f14a9e0a287419baa8a
Reviewed-by: Jerome Pasion <jerome.pasion@digia.com>
So that it can be used as the standard palette for QtGui
applications instead of the absolutely useless black palette.
Change-Id: Ie001439fcd8840a66275009c9f42cbf8bab4864a
Reviewed-by: Shawn Rutledge <shawn.rutledge@digia.com>
QUrl prefixes like file:// should be used. QFileDialog will use
QFileInfo later on, and that doesn't support QUrl prefixes.
Change-Id: I8f6d2fb9a4f502cff030a1cdd43b047b8ed5a120
Reviewed-by: Kevin Krammer <kevin.krammer@kdab.com>
Reviewed-by: Rafael Roquetto <rafael.roquetto@kdab.com>
Reviewed-by: Thomas McGuire <thomas.mcguire@kdab.com>
This time, popUpContextMenu:withEvent:forView: wouldn't respect the
menu's minimum width setting.
Change-Id: I7731851f2cf45d596d45a24dab5abe7b9239f07f
Reviewed-by: Morten Johan Sørvig <morten.sorvig@digia.com>
If the icon is null, don't force creation.
Task-number: QTBUG-30064
Change-Id: If639714f667fedfcc67a3393a7d75111a7dbff3f
Reviewed-by: Morten Johan Sørvig <morten.sorvig@digia.com>
With a global shortcut set it would be possible
to let the button re-open the menu again and again,
each time spinning an event loop.
Task-number: QTBUG-30399
Change-Id: If7eddc115c77fef3df3e751fd72e7414cedaf272
Reviewed-by: Gabriel de Dietrich <gabriel.dedietrich@digia.com>
Sounds non-sense but it means that QIODevice will not try to copy from
QBuffer's buffer onto its own.
Eventually we should figure out to make QBuffer use the
QIODevice::buffer member that is already there and avoid all of this
mistake. Something for the future.
Change-Id: Ib700c9cadb46cec20a8ea5a69a488ded7104ac76
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@digia.com>
Reviewed-by: Robin Burchell <robin+qt@viroteck.net>
Change-Id: I38d511ac0a53b65abfe47baaa6333629df5b578d
Reviewed-by: Kevin Krammer <kevin.krammer@kdab.com>
Reviewed-by: Sean Harmer <sean.harmer@kdab.com>
We don't need to keep an internal QBuffer position, we can just use the
one from QIODevice::pos(). It will keep track of goings ahead and
backwards for us, plus it will make the default bytesAvailable() work
out-of-the-box too.
This error was reported on IRC.
Change-Id: I8559e8ee56edaa01ca8732c1f1012082ebe3a3f2
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@digia.com>
Reviewed-by: Robin Burchell <robin+qt@viroteck.net>
The updated sqlite3 lacks a forward declaration of localtime.
Depending on the CE version that forward declaration
was sometimes available.
Change-Id: Iec7de1167702ccec46b62f63b65a4710231c8534
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@digia.com>
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
The ignored children do not show up in the
hierarchy, this should improve performance
significantly.
Previously the code would ignore the grouping
property which seems to break QGroupBox.
Change-Id: I4535af9c95bce76ded65f6d40fe07f17f3acffad
Reviewed-by: Gabriel de Dietrich <gabriel.dedietrich@digia.com>
Reviewed-by: Morten Johan Sørvig <morten.sorvig@digia.com>
This patch makes it possible to draw a checkable groupbox without
passing a widget pointer.
Task-number: QTBUG-29867
Change-Id: I9b74bcffa0401c88f9dcbcd9816081b7f03a5173
Reviewed-by: J-P Nurmi <jpnurmi@digia.com>
..introduced by 244cc5da55
Symbol fonts usually don't have code ranges support except Latin.
Don't load FT face for font that doesn't look like a symbol font.
Change-Id: Iec46aa84e27227e0bda5d50f96a2b2f75f58e950
Reviewed-by: Eskil Abrahamsen Blomfeldt <eskil.abrahamsen-blomfeldt@digia.com>
Although template<typename T> inline uint qHash(const T &t, uint seed) from
qhash.h is never instantiated because we have the two-argument version of
qHash() for both QOpenGLVersionProfile and QOpenGLVersionStatus, we need
the default argument, as the template in qhash.h uses noexcept, which
is evaluated regardless of instantiation, and uses qHash(t) without a
seed.
This behavior seems to not be the case with Apple clang 4.2, but has
been observed with Apple clang 4.1, Clang 3.2, and GCC 4.8.
Change-Id: If70e93f64eb9675a7c3ef7897ced2c6aebbec2d6
Reviewed-by: Eike Ziller <eike.ziller@digia.com>
Reviewed-by: Olivier Goffart <ogoffart@woboq.com>
Windows Embedded Compact 7 does not have MoveFileEx, simulate it with
the available file I/O functions [1]. For more details please check the
comments in source code.
[1] http://msdn.microsoft.com/en-us/library/ee490588.aspx
Change-Id: Ib0fb0ba84e2d070dee2f633af1b81bec132f4c17
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@digia.com>
Reviewed-by: Andreas Holzammer <andreas.holzammer@kdab.com>
Reviewed-by: Johannes Oikarinen <johannes.oikarinen@digia.com>
Those members were left uninitialised by inline constructors and/or
the destructor of those classes is/was also inline. Those members
cannot be used to store pointers that need managing during the Qt 5.x
lifetime.
They can be used to store simple values, as if they were integers.
Detected by Apple Clang 4.2.
Change-Id: I20e2def7c4006668e2d6a7e332c89e2dc8c2a184
Reviewed-by: Konstantin Ritt <ritt.ks@gmail.com>
Reviewed-by: Lars Knoll <lars.knoll@digia.com>
We want to override some of them in a subclass.
Change-Id: Ic7a22d7bc4589199b3f764157125094a4ff6ec28
Reviewed-by: Paul Olav Tvete <paul.tvete@digia.com>
We need to access these in the Android plugin.
Change-Id: I8c7f279bbe0b7087260cceb1f965c833c0097984
Reviewed-by: Paul Olav Tvete <paul.tvete@digia.com>
We don't support API levels < 9, so we can simplify this code
a little.
Change-Id: I9823d4c5a44d265e03dec0ceedeeb3c34a1e1eab
Reviewed-by: Paul Olav Tvete <paul.tvete@digia.com>
Just some code that was commented out and only adds to the
confusion.
Change-Id: Icfdf81de9731eeb2c473a2f6e2723742601a2037
Reviewed-by: Paul Olav Tvete <paul.tvete@digia.com>
Prevents mem leaks at startup on 10.6.
Change-Id: I5f8b72c54ab396ee0ec2e47320163a6a342b78f1
Reviewed-by: Gabriel de Dietrich <gabriel.dedietrich@digia.com>
With these Qt Creator startup is now warning-free
on 10.6.
Change-Id: Ie29d360611c0e516258920163e4e94d174217299
Reviewed-by: Gabriel de Dietrich <gabriel.dedietrich@digia.com>
qfeatures.h re-generated from qfeatures.txt because
QT_NO_IMAGEFORMATPLUGIN was missing
Change-Id: I1c9291529ec07f83f99c9cd08340cbfebda609b6
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@digia.com>
For some reason, we postpone clearing the stack of modal sessions until we call
processPostedEvents() again. However, it also means that when we clear the second
modal session, that session keeps running although we just closed its window. The
reason why it isn't stopped is because it wasn't the topmost modal session in the
stack.
This patch fixes the issue by stopping a modal session if any session above in the
stack has been stopped. This makes it less problematic if we don't call
processPostedEvents() in between ending modal sessions.
Task-number: QTBUG-30504
Change-Id: I9f898250ae629947d066647f9d5a0b9f75cf0070
Reviewed-by: Richard Moe Gustavsen <richard.gustavsen@digia.com>
I don't think this is an user error. If it is then
the warning should be in a cross-platform code path.
Change-Id: I4c615d9b2372660bb03b17ccb25204b2a8d17b1a
Reviewed-by: Gabriel de Dietrich <gabriel.dedietrich@digia.com>
Previously, we used -popUpMenuPositioningItem:atLocation:inView: but
this turns out not to work with modal windows. This can actually be
reproduced in Cocoa.
Change-Id: I9b750d1f97637f0cc30329d3da3acf5200f62206
Reviewed-by: Morten Johan Sørvig <morten.sorvig@digia.com>
The ref-counting and (native) touch enabling is
implemented in QCocoaWindow.
Change-Id: I4f3f5e16db59afec7dd7dd2f8360dd60b90bfe6d
Reviewed-by: Gabriel de Dietrich <gabriel.dedietrich@digia.com>
QGLWidgetGLPaintDevice::size now returns the size
in device pixels.
This fixes the QPainter "overpainting" and GraphicsView
with a QGLWidget viewport use cases.
Task-number: QTBUG-30217
Change-Id: I01998d39d7b4d755cf8b7b3fab5f75cd0d899ccf
Reviewed-by: Samuel Rødal <samuel.rodal@digia.com>
Reviewed-by: Gunnar Sletta <gunnar.sletta@digia.com>
Previously QPainter computed the devicePixelRatio
based on the physical and logical dpi, and expected
that the ratio between them would be either 1x or
2x.
This was problematic for paint devices like printers
where the physical dpi can be much higher than the
logical dpi, and also for QScreen where the physical
dpi would have to be defined as a multiple of the
logical dpi.
Add QPaintDevice::PdmDevicePixelRatio and QPaintDevice::
devicePixelRatio() getter and implement it for the
QPaintDevice subclasses. Use it when calculating the
highdpi scale transform in qpainter.cpp and when scaling
the clip rect in qwidget.cpp.
Remove physical dpi scaling for QImage, QPixmap and
QOpenGLPaintDevice, reverting to the old behavior.
Change-Id: I6c97510613196d4536ff39d08e9750b8782283d4
Reviewed-by: Samuel Rødal <samuel.rodal@digia.com>
Reviewed-by: Gunnar Sletta <gunnar.sletta@digia.com>
fontFamily has been acquired by IDWriteFont::GetFontFamily() and thus
must be released with IDWriteFontFamily::Release().
Task-number: QTBUG-26861
Change-Id: I314153f97b8b59c9ba728220f93b493b13337039
Reviewed-by: jian liang <jianliang79@gmail.com>
Reviewed-by: Lars Knoll <lars.knoll@digia.com>
We know the size of constructed list, so it may be worth to reserve
memory space for it.
Change-Id: Idad061bc1dbf5acecaec48d2e00ca3504b9db8b8
Reviewed-by: Stephen Kelly <stephen.kelly@kdab.com>
Previously only 32bit signed integers were ensured to be supported by
JsonValue type via API but it is expected that a larger integer range
should be supported by a JSON implementation.
This commit brings the Qt implementation into parity with NodeJS
JSON.stringify() in respect of the JavaScript Number type.
Change-Id: If91153cb3b13ecc14c50da97055b35ce42f341e7
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Previously Qt JSON writer would only emit 6 digits of precision as this
is the default with the formatter.
However with testing against NodeJS JSON.stringify() this behavior is
inconsistent with the defacto standard JSON implementation and conveys
a loss of precision.
Change-Id: Ie1845a6e0ee0b4c05f63ec0062f372e891855f0b
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
My interpretation of RFC4627, Section 2.4 "Numbers" of:
Numeric values that cannot be represented as sequences of digits
(such as Infinity and NaN) are not permitted.
I have also verified this matter with NodeJS JSON.stringify() that
emitting a null is consistent behavior with a JSON implementation
written in JavaScript.
Previously Qt would emit:
{
plusInfinity: inf,
minusInfinity: -inf,
notANumber: nan
}
Which maybe turned into a string values of "inf", "-inf", "nan" by
the receiving parser. Now it returns the JSON null value just like
NodeJS JSON.stringify().
Change-Id: I9f9c17f12b2606280806c47a9d90465c4ba5f786
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
This caused glitches up to unreadable text with i.e.
pixelSize 256 and stretch factor 4x /* ((256*4)<<16)<<6 */.
Change-Id: Ib6a038a043d820a94bd2019c50390a815a2a8277
Reviewed-by: Konstantin Ritt <ritt.ks@gmail.com>
QFontCache::clear() now frees everything it keeps and resets the cache costs.
Change-Id: I23ac2cab5c7c7d70db03f048b6fde151a18f92e1
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
Reviewed-by: jian liang <jianliang79@gmail.com>
Reviewed-by: Lars Knoll <lars.knoll@digia.com>
The comments in this file suggested that the bug was only present in clang/3.1.
However, qRegisterMetaType was failing even on clang 3.2
("tags/RELEASE_32/final"). The clang's bugzilla says [1] that the problem was
fixed two days before the release of 3.2, but apparently it didn't make it to
the release branch.
[1] http://llvm.org/bugs/show_bug.cgi?id=13470
Change-Id: I37db8f6f6b22ab939110e79240d92313c1786d6a
Reviewed-by: Olivier Goffart <ogoffart@woboq.com>
Use the correct metrics for the requested PageSize instead of always
defaulting to A4.
Task-number: QTBUG-30494
Change-Id: Ia3978afe3f7cc9b1ded1065416e5c3def44e7a05
Reviewed-by: Eskil Abrahamsen Blomfeldt <eskil.abrahamsen-blomfeldt@digia.com>
Reviewed-by: Gunnar Sletta <gunnar.sletta@digia.com>
uint32_t is in C++11, which we can not rely on being present. It is also
in C99, but some compilers do not support that in C++ mode.
Change-Id: I29bada724f5646a0a0562b3ab133cb49cf928d6d
Reviewed-by: Gatis Paeglis <gatis.paeglis@digia.com>
Reviewed-by: Tobias Hunger <tobias.hunger@digia.com>
There is no 'name' variable, but apparently 'd->serverName' is correct
variable here. Some other methods take 'name' as an argument and have
similar code block, so I think this is a copy/paste error introduced
somewhere in the past.
Change-Id: I2e82b42688b9928e0dcc5054df04e87d30a5e38e
Reviewed-by: Andreas Holzammer <andreas.holzammer@kdab.com>
Reviewed-by: Johannes Oikarinen <johannes.oikarinen@digia.com>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
QFontCache now references QFontEngineData-s it maintains,
so that QFont instances not freed prior to calling ~QFontCache()
would destroy QFontEngineData on their own.
Task-number: QTBUG-25434
Change-Id: Ia7679d64de436841f09ac7be62ceb570e50cce5b
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
Reviewed-by: Eskil Abrahamsen Blomfeldt <eskil.abrahamsen-blomfeldt@digia.com>
Reviewed-by: jian liang <jianliang79@gmail.com>
This class of bug seems to occur in 'flaky' CI tests, so give
more information about what the problem is. Found here:
https://codereview.qt-project.org/#change,44210
Change-Id: I73908e309bdecf1fd3bc323dd4e8febe47c1239a
Reviewed-by: Thorbjørn Lund Martsum <tmartsum@gmail.com>
Reviewed-by: Stephen Kelly <stephen.kelly@kdab.com>
The expected versions are only set if the cmake tests are set up by
qmake by running
qmake && make check
If instead someone uses cmake directly:
mkdir build && cd build && cmake .. && ctest
the expected versions are not known, and the test fails without
this patch. So, don't test the versions if they are not known.
Change-Id: I2e8a4f651a69a8817c819d881be75ca07bc1bfd4
Reviewed-by: Stephen Kelly <stephen.kelly@kdab.com>
Apple Clang 4.2 caught private members that aren't used in a class
with no friends. Since this code is in a tool, there's no binary
compatibility requirement.
databaseinfo.h:71:13: error: private field 'driver' is not used [-Werror,-Wunused-private-field]
quoter.h:86:10: error: private field 'validRegExp' is not used [-Werror,-Wunused-private-field]
qmlvisitor.h:123:11: error: private field 'tree' is not used [-Werror,-Wunused-private-field]
Change-Id: Iba9995f28ddef983c9ffc1dc551f042539252704
Reviewed-by: Alan Alpert <aalpert@blackberry.com>
Reviewed-by: Olivier Goffart <ogoffart@woboq.com>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
In QGuiApplication only Qt::AA_SynthesizeMouseForUnhandledTouchEvents
is taken into account when synthesizing mouse from touch events, in
QApplication only the PlatformIntegration syle hint
QPlatformIntegration::SynthesizeMouseFromTouchEvents.
With this patch both attributes are checked. Furthermore the check was
moved out of translateTouchToMouse in QApplication in order not to
influence the result which is returned to the user, when mouse events
are not be synthesized.
Change-Id: I87ac7299f0a9fbf0a083eff9c547f0dbfab75dfb
Reviewed-by: Fabian Bumberger <fbumberger@rim.com>
Reviewed-by: Shawn Rutledge <shawn.rutledge@digia.com>
Change-Id: I07525ac45a610f56f2bc72731379073f4aa67f7b
Reviewed-by: Kevin Krammer <kevin.krammer@kdab.com>
Reviewed-by: Thomas McGuire <thomas.mcguire@kdab.com>
Reviewed-by: Rafael Roquetto <rafael.roquetto@kdab.com>
When a string comes from the native APIs then it may need to be
normalized.
Task-number: QTBUG-14787
Change-Id: I5164e80efcd1d99a50263c72bcf97d7fb31b38f7
Reviewed-by: Morten Johan Sørvig <morten.sorvig@digia.com>
We should only warn about missing a11y plugins if iface is 0, since
we obviously have the plugin otherwise.
This got triggered by that a visible widget got destructed, which in
turn called hide(), which ultimately ended up calling
QAccessible::updateAccessibility(QAccessible::ObjectHide).
When QAccessibleEvent::accessibleInterface() then called
queryAccessibleInterface(), QAI::isValid() would return false because
the widget was in its destructor.
Change-Id: If655e00e7bb3737a39e24e26dbd1e23edd515bf5
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
Reviewed-by: Frederik Gladhorn <frederik.gladhorn@digia.com>
Similar to 70420ec3 (Look for cmake package dependents only in the parent
directory., 2013-04-02), make sure this cmake function only searches
for sibling packages as dependencies.
Change-Id: Icab23d333fa6a750ee262b592fae39f0ba334fee
Reviewed-by: Stephen Kelly <stephen.kelly@kdab.com>
Similar to 70420ec3 (Look for cmake package dependents only in the parent
directory., 2013-04-02), make sure cmake only searches for sibling
packages as components.
Change-Id: I9e0c13dfcdba28a9fd6892828b0f509dd63e6604
Reviewed-by: Brad King <brad.king@kitware.com>
Reviewed-by: Alexander Neundorf <neundorf@kde.org>
Reviewed-by: Stephen Kelly <stephen.kelly@kdab.com>
QFontEngine::FaceId::filename is intended to be local8Bit-encoded
but QFreetypeFace::getFace() always treats it like UTF-8-encoded.
Also replace explicit (to|from)Local8Bit and toLatin FaceId::filename
conversions with QFile::(en|de)codeName, where appropriate.
Change-Id: Ic7beabf0a160f2f02f1667bcb8e6067adaba1c16
Reviewed-by: Eskil Abrahamsen Blomfeldt <eskil.abrahamsen-blomfeldt@digia.com>
When populating the font database, FcPatternGetString(FC_FAMILY) gets
a localized font family name; but, in fallbacksForFamily, it gets
a non-localized font family name, so it unable to find the proper font to fallback.
Simply register all family name variants as aliases to localized name
and make sure they are checked when getting fallback families.
Task-number: QTBUG-28806
Task-number: QTBUG-30415
Change-Id: I71c03ae9b51a28736c2576f3442f1bbdb3497c09
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
Reviewed-by: jian liang <jianliang79@gmail.com>
Reviewed-by: Eskil Abrahamsen Blomfeldt <eskil.abrahamsen-blomfeldt@digia.com>
It should also have an effect for Visual Studio project files, not
just makefile generators.
Change-Id: I395071f09b29a6e8967a3d44e41d30480ae783f7
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@digia.com>
Reviewed-by: Stephen Kelly <stephen.kelly@kdab.com>
Perhaps this made sense when all our open gl examples were black,
but with components and I would assume the majority of use cases
we should use the default window background color to fill exposed
window background during a resize.
Change-Id: Ia439a7c3919243efa6026e2e07bee62c25557df0
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
Reviewed-by: Gunnar Sletta <gunnar.sletta@digia.com>
... before trying to update the configurations. Before, we would check
whether the pointer to the QNetworkConfigurationManagerPrivate instance
was 0 and in that case construct it. This would mean that this code path
was taken "at app shutdown", i.e. when the qAddPostRoutine had already
been called but the other statics were still accessed.
Note: This is not thread safe, but neither is the rest of the code;
making it thread-safe would require additional changes.
Task-number: QTBUG-30585
Change-Id: I8f6cf616e3f3ba1e84b8246589fb7210d2dae57a
Reviewed-by: Frederik Gladhorn <frederik.gladhorn@digia.com>
Reviewed-by: Stephen Kelly <stephen.kelly@kdab.com>
These macros are useless from Qt 5.1 on.
Change-Id: I4574e87ef36e7bc67d2f1caa38c685a837708764
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@digia.com>
Before it would skip out a row depending on the size of the tableview.
Now it will show the next set of cells for the direction it is going in.
Change-Id: I02926d0b2187ffafffb63f7fff43c233b4d7d6af
Reviewed-by: Mitch Curtis <mitch.curtis@digia.com>
Reviewed-by: Stephen Kelly <stephen.kelly@kdab.com>
The name field string was not set in native dialog when one was present.
The previous fix is not correct. In the doc of setNameFieldStringValue,
it said "The value must not be nil".
Task-number: QTBUG-28342
Change-Id: I847fb7472b8fecc0303433d8bf72b6a5bd7da72b
Reviewed-by: Andy Shaw <andy.shaw@digia.com>
Reviewed-by: Morten Johan Sørvig <morten.sorvig@digia.com>
Initial window position is at 0,0 by default, then it's corrected
in createNSWindow (by calling initialGeometry). After window
creation, QNSWindowDelegate will receive the windowDidMove
notification and call QCocoaWindow::windowDidMove() which then
calls QNSView::updateGeometry(). It will not call
QWindowSystemInterface::handleGeometryChange() because the window
is still being constructed. So we can fix this by directly
setting QWindow::geometry in the QCocoaWindow constructor.
Task-number: QTBUG-30382
Change-Id: I3db7a6e83693e5787ae82dcab1767de938d2e5d1
Reviewed-by: Gabriel de Dietrich <gabriel.dedietrich@digia.com>
Changing "C++ API" -> "C++ Classes" to match all the other modules.
Change-Id: I434cd9c32b1db18395242e73564fe309c8ebb99e
Reviewed-by: Jerome Pasion <jerome.pasion@digia.com>
This completes the change started with
0bdc86d9ef by providing the implementation
for the Oracle and DB2 sqldrivers
Change-Id: Ia14415c6d7dea51f1369a81236f79aff892b3af7
Reviewed-by: Mark Brand <mabrand@mabrand.nl>
QDoc needs correct capitalization to create a link to the class ref
Change-Id: If95ab1245fcad6c0f624b88eb4522ba78be7d6a6
Reviewed-by: Laszlo Papp <lpapp@kde.org>
Reviewed-by: Martin Smith <martin.smith@digia.com>
Reviewed-by: Jerome Pasion <jerome.pasion@digia.com>
We seem to have two tables. That's a bad idea, but they are there.
I won't add Qt 5.1 until we're certain that the datastream isn't
changing. It's preferable to lack information than to show the wrong
value.
Change-Id: Ibc7eb95bbbf3b35388f910154372fdd896735430
Reviewed-by: Martin Smith <martin.smith@digia.com>
The documentation used to have "Main Classes" list for C++ classes,
and qdoc kept track which classes were supposed to be listed as "main"
classes. This is no longer used, so this change removes the code that
marked a C++ class as a main class and tested whether a class was a
main class. This dead code was seen while preparing the way for
changing qdoc to output a documentation page for obsolete C++ classes
and functions and obsolete QML types and functions.
Task-number: QTBUG-30511
Change-Id: Iccc31b46b7c3c144038372cad4771d974f207937
Reviewed-by: Topi Reiniö <topi.reinio@digia.com>
Reviewed-by: Laszlo Papp <lpapp@kde.org>
Reviewed-by: Jerome Pasion <jerome.pasion@digia.com>
Also, get rid of qt_mac_toCGImage and QPainter warnings.
Change-Id: I9e2a0e93d74c318d99651de9222994615e082ac9
Reviewed-by: Morten Johan Sørvig <morten.sorvig@digia.com>
These are the same for normal builds, but differ when cross-compiling Qt.
Change-Id: I75eccc6f4b67b440a08c4aba41aabb7df686c9f9
Reviewed-by: Stephen Kelly <stephen.kelly@kdab.com>
By calling itemData() of the source model directly, the result cannot
contain data provided by the proxy model itself. The base class
implementation however will call data() on the proxy instead.
Change-Id: Ib0ef5f5621457adbfa4bd896a756dfcb98d0ae54
Reviewed-by: Stephen Kelly <stephen.kelly@kdab.com>
The evaluation message box caused the QtWebProcess
to quit due to the fact that the process has no other
windows and EvalMessageBox was both set to be closed
on quit and ended up being the last window in the process.
Change-Id: Iad6461d014258fdc5fade7dafe48da33903377bb
Reviewed-by: Simon Hausmann <simon.hausmann@digia.com>
Reviewed-by: Andy Shaw <andy.shaw@digia.com>
Reviewed-by: Jocelyn Turcotte <jocelyn.turcotte@digia.com>
With data=0, ANGLE can crash when setting the buffer data. As this
should be a legal operation, don't perform a memcpy when data is null.
Change-Id: I3fa1260482549b1da50d7a68001a65decb98f258
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
This allows the Direct3D 11 version of ANGLE to gracefully allow
surfaces with dimensions of 0. This is important because Qt may resize
the surface to 0x0 because of window minimization or other user
action (window resize). As EGL specifies that empty (0x0) surfaces are
valid, this makes sure an assert doesn't occur in the case that a valid
surface is resized to an empty one.
Change-Id: Ia60c4c694090d03c1da7f43c56e90b925c8eab6d
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@digia.com>
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
Upgrades ANGLE to dx11proto (dx11-MRT-support tag), which splits out support
for DirectX9 & DirectX11. The DX9 codepath is used by default;
CONFIG+=angle_d3d11 must be passed to the ANGLE project to build for DX11.
Existing patches to ANGLE have been updated (or removed if no longer
needed), and a patch has been added to make DX9/DX11 codepaths mutually
exclusive.
Change-Id: Ibe13befadb94f04883eca449d0ee1f0da955ff92
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@digia.com>
Reviewed-by: Gunnar Sletta <gunnar.sletta@digia.com>
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
Reviewed-by: Axel Waggershauser <awagger@gmail.com>
This allows re-using the function in QtQuick.
In addition TextBoundaryType is moved to QAccessible
and QAccessible2 as namespace ceases to exist.
Change-Id: I184bc2c181a22ca51ac4db4e5a080dc26d4acfe0
Reviewed-by: Jan Arve Sæther <jan-arve.saether@digia.com>
Change 595deca5 for QTBUG-22316 resulted in having titlebars only on
modal dialogs.
Change-Id: Ic23c957e0e95cf567d90831e105cc402465df982
Reviewed-by: Gabriel de Dietrich <gabriel.dedietrich@digia.com>
Note that GtkSettings is already used for retrieving the icon
theme name.
Change-Id: I935d903c703f7d832b0350d3cfd3a638f22d1391
Reviewed-by: Jens Bache-Wiig <jens.bache-wiig@digia.com>
This uses explicit handshakes between the application and the
gdbserver start and the host side by using the gdbserver socket
and two files ("ping" file in the application dir, "pong" file
in /data/local/tmp/qt)
The sequence is as follows:
host: adb forward debugsocket :5039
host: adb shell rm pong file
host: adb shell am start
host: loop until ping file appears
app start up: launch gdbserver --multi +debug-socket
app start up: loop until debug socket appear
gdbserver: normal start up including opening debug-socket,
not yet attached to any process
app start up: touch ping file
app start up: loop until pong file appears
host: start gdb
host: gdb: set up binary, breakpoints, path etc
host: gdb: target extended-remote :5039
gdbserver: accepts connection from gdb
host: gdb: attach <application-pid>
gdbserver: attaches to the application
and stops it
app start up: stopped now (it is still waiting for
the pong anyway)
host: gdb: continue
gdbserver: resumes application
app start up: resumed (still waiting for the pong)
host: write pong file
app start up: java code continues now, the process
is already fully under control
of gdbserver. Breakpoints are set etc,
we are before main.
app start up: native code launches
Change-Id: Iaa28b8664dbebc39022d1be7ff5533c52ce39715
Reviewed-by: Paul Olav Tvete <paul.tvete@digia.com>
Reviewed-by: Daniel Teske <daniel.teske@digia.com>
Reviewed-by: Eskil Abrahamsen Blomfeldt <eskil.abrahamsen-blomfeldt@digia.com>
The Android style is not fully implemented yet, so prefer the
Fusion style by default.
Task-number: QTBUG-29565
Change-Id: Ida413e6b34a41b618b8e2206688922195ffcc519
Reviewed-by: BogDan Vatra <bogdan@kde.org>
Reviewed-by: Eskil Abrahamsen Blomfeldt <eskil.abrahamsen-blomfeldt@digia.com>
The mathematically correct way would be to set logicalDPi to
160*scaledDensity, but then a 12 pt font would be gigantic. On
iOS, we use a factor of 72 to be compatible with the native APIs,
but that means that a 12 pt font is very small.
A factor of 100 means that desktop apps look reasonable by default.
Task-number: QTBUG-29674
Change-Id: I607f110150fb95685a6980b92f6f92f2b489f959
Reviewed-by: Eskil Abrahamsen Blomfeldt <eskil.abrahamsen-blomfeldt@digia.com>
Calling clear_mapping causes the persistent indexes to be queried, and
mapped using map_to_source, so that they can be restored later. That
is not the appropriate response to the source model being deleted
because there won't be anything to restore.
Simply clear the stored mapping information instead so that the source model
actually exists when mapToSource is called by the framework.
Change-Id: I99692ee7aa9c6714aec45c68fe4a2d62be189d60
Reviewed-by: Olivier Goffart <ogoffart@woboq.com>
The -redirect-documentation-to-dev-null option allows to get all qdoc
errors, without creating full documentation. It significantly reduces
amount IO operations.
Change-Id: I80fb6f9e304eb106f1c1d1e6320a1b38d788f6fc
Reviewed-by: Martin Smith <martin.smith@digia.com>
Reviewed-by: Frederik Gladhorn <frederik.gladhorn@digia.com>
When undocking a dock using the undock button on a main window
with native children, the dock widget goes to 0,0 (content
position) causing the window frame to be off-screen since
the frame is still 0,0. This change causes the frame to be
recalculated such that the frame position is 0,0.
Task-number: QTBUG-28872
Change-Id: I32896107cd7b982811f45de43dbad82e7407ea7a
Reviewed-by: Joerg Bornemann <joerg.bornemann@digia.com>
Reviewed-by: Oliver Wolff <oliver.wolff@digia.com>
Change-Id: I1d8061302fbb8494b5ae31e20a644745fe969f10
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@digia.com>
Reviewed-by: Brad King <brad.king@kitware.com>
Reviewed-by: Stephen Kelly <stephen.kelly@kdab.com>
These are easy, since they can be inline. Most types that would
benefit from move constructors can't have inline move constructors
because these types use smart pointers whose destructor is invoked
in the type's move constructor. Implementing move constructors
out-of-line would break binary compatibility between C++98 and C++11
builds of Qt and its users, so that is not attempted here.
Change-Id: I7f14437c2069cce54c498c7858f4e9060ff05e7b
Reviewed-by: Olivier Goffart <ogoffart@woboq.com>
Reviewed-by: Stephen Kelly <stephen.kelly@kdab.com>
Windows file dialogs are always modal (inheriting IModalWindow).
As the QtQuick dialog plugin has a widget-based fallback now,
there is no need to use native dialogs for the non-modal case.
Task-number: QTBUG-30513
Change-Id: I8b62af86fde34e8c0cf5e917d2e62260e16c82fb
Reviewed-by: David Faure (KDE) <faure@kde.org>
Reviewed-by: Shawn Rutledge <shawn.rutledge@digia.com>
Without this, it is not possible to do vsync drawing on NVidia
GeForce 210 on Windows 7.
Change-Id: I6064826eaa55266344393f50cdf5d17113350255
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
Reviewed-by: Samuel Rødal <samuel.rodal@digia.com>
The only way to make this possible is to disable the
atomic-rename-from-temp-file behavior. This is not done by default,
but only if the application allows this to happen.
https://bugs.kde.org/show_bug.cgi?id=312415
Change-Id: I71ce54ae1f7f50ab5e8379f04c0ede74ebe3136d
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Return correct writing system based on QLocale::uiLanguages() or
system locale.
Task-number: QTBUG-1573
Change-Id: Ia574b3772fd4d9b01bef47338498e875bf9d6aac
Reviewed-by: Stephen Kelly <stephen.kelly@kdab.com>
Reviewed-by: Konstantin Ritt <ritt.ks@gmail.com>
See https://www.khronos.org/bugzilla/show_bug.cgi?id=449 for upstream
bug in gl.spec which incorrectly marks glTexImage3D as deprecated.
Change-Id: Ib307a5315dd37b8b18389df54b1c93e6a43c61dd
Reviewed-by: James Turner <james.turner@kdab.com>
Reviewed-by: Samuel Rødal <samuel.rodal@digia.com>
This was spotted while tracking down a similar bug related to spans.
This now also eliminates the Q_EXPECT_FAILs in
heightForWidthWithSpanning(), since it now finally works.
The problem was only for the maximum size, since the size of an ignored
row/column was min: 0, pref: 0, max: FLT_MAX (the default constructed
values for a QGridLayoutBox).
Change-Id: Ibb33c26ede40ed02edd26f596ba6133d59c9962f
Reviewed-by: Paul Olav Tvete <paul.tvete@digia.com>
406c8ef6e6 introduced deriving the private SQL driver classes from
QSqlDriverPrivate. However, the drivers continued to keep their own
pointer to the private class, even though QObject provides the same
pointer. Worse yet, the private class is allocated too late and not
even passed to QSqlDriver. The result is that QSqlDriver allocates
a separate instance of QSqlDriverPrivate. This is likely to cause
all kinds of chaos.
The private class needs to be allocated in time pass it to QSqlDriver
which passes it on to QObject.
This commit covers the the base class and drivers:
ibase
mysql
odbc
psql
sqlite
tds
Fixes for the remaining drivers will follow.
Change-Id: Id8e7ec4205b0ca6cd00bd022c9cd24f137089245
Reviewed-by: Andy Shaw <andy.shaw@digia.com>
The former represents the physical device orientation, the latter the
UI orientation. We need to explicitly cast between them, as they are
different enums, but with compatible values for the subset we use.
Change-Id: I2926068802f35680cb6de5ced6dcf286014fdb2e
Reviewed-by: Morten Johan Sørvig <morten.sorvig@digia.com>
Reviewed-by: Richard Moe Gustavsen <richard.gustavsen@digia.com>
qcocoanativeinterface.mm:196:5: warning:
instance method '-qt_setDockMenu:' not found
(return type defaults to 'id') [-Wobjc-method-access]
[NSApp QT_MANGLE_NAMESPACE(qt_setDockMenu): menu];
Change-Id: I80838d0098e7f172d9b8124f04e91dba64764318
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
Reviewed-by: Morten Johan Sørvig <morten.sorvig@digia.com>
Reviewed-by: Gabriel de Dietrich <gabriel.dedietrich@digia.com>
Add commandline option to set the default for
fullscreen option.
Change-Id: Ieaf2be3858082fb0e572c8fbaabfd38a698f14f6
Reviewed-by: Frank Osterfeld <frank.osterfeld@kdab.com>
Reviewed-by: Thomas McGuire <thomas.mcguire@kdab.com>
Reviewed-by: Kevin Krammer <kevin.krammer@kdab.com>
We used to disable NEON completely, as the iOS toolchain does not
handle the GAS syntax of the pixman draw-helpers. But we can limit
the disabling to just the draw-helpers, which means we get NEON
optimization of eg. QImage and QString.
Change-Id: If350b06ce521cca8b24468be5a168ff21e9e7124
Reviewed-by: Richard Moe Gustavsen <richard.gustavsen@digia.com>
Having static QFont instance leads to a resources leaking, since QFontCache
is unable to clean-up font engines when the application exits.
Relates to QTBUG-25434
Change-Id: I71d91094de27c07ab2434c415e4c28b6acab3646
Reviewed-by: David Faure (KDE) <faure@kde.org>
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
If a row/column is only used only because of the spanning of an item,
the cell should be treated as it didn't exist. We keep track of this
with the "ignore" bit array.
The old code would always start from the row/column at position 1.
In the attached testcase this made the effectiveRowSpan become larger
than actually needed.
Task-number: QTBUG-30255
Change-Id: Ief0e7018ee8e5ee36272ce075a43312ffeac7b91
Reviewed-by: Paul Olav Tvete <paul.tvete@digia.com>
This change amends my earlier change to make
the function reference relevant to the example
being discussed.
Task-number: QTBUG-28204
Change-Id: I841a561ba117f83a96b8b4a3661cea2a7a3e47da
Reviewed-by: Sze Howe Koh <szehowe.koh@gmail.com>
Reviewed-by: Martin Smith <martin.smith@digia.com>
When a menu item is merged then we do not want to release it as it will
cause a problem with the Application menu. Therefore we just hide it
instead.
Change-Id: Ia9f7acfe063001111769841c9a360cac369f7ed7
Reviewed-by: James Turner <james.turner@kdab.com>
Reviewed-by: Morten Johan Sørvig <morten.sorvig@digia.com>
This functionality is used in multiple places in Qt itself, so it makes
sense to have a global function for this. This also allows to map this
onto specialized assembler instructions, should an architecture provide
them, later on.
Also added comprehensive tests, using a 4-bit lookup-table implementation
as a reference.
Change-Id: I8c4ea72cce54506ebb9fbe61141dbb5f1b7a660f
Reviewed-by: Stephen Kelly <stephen.kelly@kdab.com>
The orientations involved here can be easy to mix up,
so don't make it more confusing.
Change-Id: I73c4765ebbc89feced12898ac6001eb68db0e21c
Reviewed-by: Shawn Rutledge <shawn.rutledge@digia.com>
This file is used by QtCreator to decide if the java files
need to be updated.
Change-Id: Ib4875c32a071dd768f89950582c81ccdd420feca
Reviewed-by: Eskil Abrahamsen Blomfeldt <eskil.abrahamsen-blomfeldt@digia.com>
Reviewed-by: Paul Olav Tvete <paul.tvete@digia.com>
This patch will follow up on e27ca37 (add more state to
QT::ApplicationState), and add a new event class
QApplicationStateChangeEvent with type Qt::ApplicationStateChange.
Change-Id: Idee724f181f1fbb2321ddad7e0df00c88b3488b1
Reviewed-by: Eskil Abrahamsen Blomfeldt <eskil.abrahamsen-blomfeldt@digia.com>
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
Reviewed-by: Samuel Rødal <samuel.rodal@digia.com>
Some new ctor initializers were added that use
curly braces. qdoc's dumb C++ parser can't handle
curly braces in the ctor area; it thinks they are
the body of the function. In this update, the
Q_QDOC macro is used to isolate these problems
so qdoc doesn't have to deal with them.
Task-number: QTBUG-30254
Change-Id: Ie318e4ac58861f4ebd8a7b9f004d2677f850436e
Reviewed-by: Olivier Goffart <ogoffart@woboq.com>
Task-number: QTBUG-30134
Restore the lines in qcoreapplication.cpp removed by commit
950b35cf97 to ensure QThreadData and
QAdoptedThread object of main thread is destroyed at application exit.
We don't set QCoreApplicationPriavte::theMainThread to 0 as before since
it will be set to zero in QThreadData::~QThreadData()
Change-Id: I8ee56aff5a933ce1d812b07fb00a29ed0839ab6e
Reviewed-by: Olivier Goffart <ogoffart@woboq.com>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
There is potential trouble and no point in registering a QSocketNotifier
on an invalid file descriptor. This is prevented now in addition to the
existing warning.
Change-Id: I6fc3c639b167ea98d09a2738986360cc4b3fc210
Reviewed-by: Kevin Krammer <kevin.krammer@kdab.com>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Fix the documented function signature to match the declared one.
Task-number: QTBUG-30398
Change-Id: I0d4894e2597f70e29a3edd4ceda91a4dddb9941e
Reviewed-by: Shawn Rutledge <shawn.rutledge@digia.com>
evdevtouch moved to platform support, so the location in the plugin is now wrong
(as was the file name)
Change-Id: I976f5d7ee68f5da5ed1f61773b3d5da683e12c6f
Reviewed-by: Samuel Rødal <samuel.rodal@digia.com>
Check if the sequence is really a mnemonic and emulate grapheme cluster if so,
rather than blindly reset the next character's attributes.
This covers cases like "& ", "&<U+034F>", etc.
Change-Id: Ibb063a2d258aff6455b9bb41bbe1a58a5036d0d6
Reviewed-by: Lars Knoll <lars.knoll@digia.com>
Reviewed-by: Eskil Abrahamsen Blomfeldt <eskil.abrahamsen-blomfeldt@digia.com>
Stephen Kelly
Jørgen Lind
Romain Perier
Tor Arne Vestbø
Liang Qi
Gatis Paeglis
Jens Bache-Wiig
Gabriel de Dietrich
Jan Arve Saether
J-P Nurmi
Bjoern Breitmeyer
Andy Nichols
Venu
Oswald Buddenhagen
Thorbjørn Martsum
Friedemann Kleint
Nico Vertriest
Mitch Curtis
Richard Moe Gustavsen
Kevin Ottens
Andy Shaw
Eskil Abrahamsen Blomfeldt
Thiago Macieira
Sze Howe Koh
Andreas Holzammer
Matt Fischer
Tor Arne Vestbø
Topi Reinio
Bernd Weimer
Peter Hartmann
Sérgio Martins
Rafael Roquetto
Albert Astals Cid
Jake Petroules
Lars Knoll
Olivier Goffart
David Gil
Giuseppe D'Angelo
Kai Koehne
Martin Smith
Miikka Heikkinen
Christoph Schleifenbaum
Jonathan Liu
Jan Kundrát
Samuel Rødal
Ivan Komissarov
Sergio Ahumada
Joerg Bornemann
J-P Nurmi
Shawn Rutledge
Jerome Pasion
Paul Olav Tvete
Frederik Gladhorn
Caroline Chao
Holger Hans Peter Freyther
Morten Johan Sørvig
Josh Faust
Gunnar Sletta
David Faure
hluk
Konstantin Ritt
Mark Brand
Thomas McGuire
hjk
Libor Tomsik
Christoph Schleifenbaum
Tasuku Suzuki
Cyril Oblikov
Sergio Martins
Andreas Holzammer
Janne Anttila
Morten Johan Sørvig
Jędrzej Nowacki
Darryl L. Miles
John Layt
Tobias Hunger
Fabian Bumberger
Volker Krause
Keith Gardner
Michael Brüning
Andrew Knight
Ahmed Saidi
Dmitry Shachnev
BogDan Vatra
Marc Mutz
David Faure
Sean Harmer
jian liang
Robin Burchell