Observe Qt::BypassWindowManagerHint in XCB and use it for the
tray. Remove the call to XSync(), which sometimes was causing
the window to appear in the wrong place for a short time.
Task-number: QTBUG-31762
Change-Id: I7b4fc5e42b8c8bb9d03b1575be99d6d1c727be37
Reviewed-by: Shawn Rutledge <shawn.rutledge@digia.com>
Change-Id: I19df59453156a95c5075a065268333a74cec2476
Reviewed-by: James Turner <james.turner@kdab.com>
Reviewed-by: Gunnar Sletta <gunnar.sletta@digia.com>
Reviewed-by: Giuseppe D'Angelo <giuseppe.dangelo@kdab.com>
The behavior introduced by qtdeclarative's 475d1ed4f6a2 mandates that an
expose event follows a geometry change event. The fix for this uncovered a bug
on the raster backing store which caused the rotation to break on QtWidgets.
The problem was rooted on the assumption that the
QQnxRasterBackingStore::flush() method was only called after paints to the
backing store surface, which was discovered to be a false assumption. Flushing
the backing store can occur in other circunstances, which are out of scope on
the context of the QNX plugin.
Task-number: QTBUG-31329
Task-number: QTBUG-31934
Change-Id: Icfc3e85324f5a8745f6b5f189f27f8b763f770c7
(cherry picked from commit 0343b92660)
Reviewed-by: Nicolas Arnaud-Cormos <nicolas@kdab.com>
Reviewed-by: Vladimir Minenko <vminenko@rim.com>
Setting an icon on a QWindow before creating the platform
window did not have any effect.
Task-number: QTBUG-31961
Change-Id: I8e58a3b239c8bc69af58f765e7826337463d2201
Reviewed-by: Shawn Rutledge <shawn.rutledge@digia.com>
Reviewed-by: Joerg Bornemann <joerg.bornemann@digia.com>
Set initial directory and title correctly in native file dialog.
Task-number: QTBUG-31098
Change-Id: I87c4383bdf0b26beda6a075a186d0a951c86f8a7
Reviewed-by: Joerg Bornemann <joerg.bornemann@digia.com>
qvariant_p.h(226): error: name followed by "::" must be a class or namespace name
template<class C> static Yes test(char (*)[(&C::isNull == 0) + 1]);
^
Change-Id: I42eef44d46ca0fb7aac8f82150e917c4fb912b15
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@digia.com>
The macro comes from MSVC's yvals.h header and it actually indicates
absence of C++11 support in MSVC. It is defined if
_HAS_CHAR16_T_LANGUAGE_SUPPORT isn't defined, which is a C++11 feature.
Task-number: QTBUG-31701
Change-Id: I155b8e764f786e95a08de660f70094b2bee38937
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@digia.com>
We forgot to set the ShowIsFullScreen property for the raster
engine case. OpenGL windows already did the right thing, since
the GL backend is based on eglFS. The iOS backend also has
this logic.
Task-number: QTBUG-31984
Change-Id: I1cf5df32d79f441768126f00632107a7971eb5ad
Reviewed-by: Eskil Abrahamsen Blomfeldt <eskil.abrahamsen-blomfeldt@digia.com>
The delayed root window posting introduced by
77a06e7e6c broke the OpenGL support. In that patch, the posting of
the root window is trigerred by QQnxRasterRasterBackingStore's call to
QQnxWindow::post(), that obviously does not happen when OpenGL is enabled,
therefore requiring the OpenGL context to explicitly post() the root window.
Task-number: QTBUG-31934
Change-Id: Ifd302c1dde612a03b79c778ec4586aa70f88260d
(cherry picked from commit 6a9333841d)
Reviewed-by: Nicolas Arnaud-Cormos <nicolas@kdab.com>
The control key results in modifiers having Qt::MetaModifier, and then
the correct character is found in charactersIgnoringModifiers. The
rest of the time, [nsevent characters] seems to be correct. If we
use charactersIgnoringModifiers too much of the time, then the keycode
will be wrong in some cases even though typing is still possible.
Task-number: QTBUG-29005
Task-number: QTBUG-31811
Task-number: QTBUG-31977
Change-Id: Ib23b89f03bc9a61fe6d177320fa603c05649e979
Reviewed-by: Eike Ziller <eike.ziller@digia.com>
Reviewed-by: Frederik Gladhorn <frederik.gladhorn@digia.com>
Make sure to stop any previously active (auto-default button)
animation when the auto-default button changes.
Task-number: QTBUG-31668
Change-Id: Iaa9bfa485b11e49cd7f765d0e54b4e6391ad2f6d
Reviewed-by: Jens Bache-Wiig <jens.bache-wiig@digia.com>
Remove virtual from QPageSetupDialogPrivate::init() (which was
called from the constructor) and keep it as a non-virtual
function for QUnixPageSetupDialogPrivate.
Fix breakage introduced by
54b899f448 .
Task-number: QTBUG-31790
Change-Id: I6d552bc250e4d72c1bb50eceadd2852b1b957aea
Reviewed-by: Lars Knoll <lars.knoll@digia.com>
On desktop platforms, widgets have traditionally received
focus on mouse press. On touch platforms (iOS, Android) this
is different, there you need to delay setting the focus
until a touch release (probably to check if the press starts
a flick or tap'n'hold etc).
This patch will add a new style hint SetFocusOnRelease that
can be set by the plugin to control this behavior in Qt.
Change-Id: I2e4d714894e327822c855eb48a3b28e354726e95
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@digia.com>
Reviewed-by: Paul Olav Tvete <paul.tvete@digia.com>
When a window is active in Qt, it means that is has keyboard
focus. And on iOS, we only want a window to be rendered with
keyboard focus when the input panel is open. Therefore we
choose to call QWindowSystemInterface::handleWindowActivated()
as a response to the input panel opening or closing, rather than
from QPlatformWindow::requestActivateWindow(). And becoming or
resigning first responder is that same as showing or hiding
the input panel.
Change-Id: I33b1bad769bec1fdd7c6ae4119b4b445da2f930f
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@digia.com>
When the user is tranferring input focus between line edits
(or similar controls), the edit that lost focus will close
the input panel, just to see that the input that gained
focus will open it again. This will cause the input panel
to "jump", and can also trigger other input panel related
code/signals unnecessary.
Change-Id: Iac0a103e8d2f0f7cdcc04b8ec5b10e07587d1ad6
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@digia.com>
This adds a property that specifies how long a tooltip is displayed.
This partly solves:
Task-number: QTBUG-1016
Change-Id: Ieea218bbcb869f6b48e72913d967e74fa792f2e2
Reviewed-by: J-P Nurmi <jpnurmi@digia.com>
The HTML pages need to display minimal navigation such
as a link to the module landing page or Qt 5's index.html.
-new variables available in qdocconf files:
navigation.landingpage
navigation.homepage
navigation.cppclassespage
navigation.qmltypespage
buildversion
The navigation variables are for the new navigation bar.
The buildversion variable is to insert documentation build
information into the pages.
Note about compatibility with Qt Creator docs:
The HTML template files for Qt Creator is retained to keep
Qt Creator compatible with 5.0.x and 5.1.x.
Change-Id: Ibb4d7cada19644204457d822b6b77b2aa7b70f8d
Reviewed-by: Martin Smith <martin.smith@digia.com>
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@digia.com>
-The navigation bar requires the titles of the landing page and
relevant pages.
Change-Id: I5d5986b2bf74205cd49957b63fd6ac4e32cbb36b
Reviewed-by: Martin Smith <martin.smith@digia.com>
Both Android and iOS needs the style to use RSIP_OnMouseClick (and
qwindowscestyle and qwindowsmobilstyle as well, it seems) so set it
as default (instead of expanding the #if test even more).
Change-Id: I93a78e11031b506c2cb80f30f1946fe25f7945a9
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@digia.com>
Reviewed-by: Paul Olav Tvete <paul.tvete@digia.com>
Both ways of getting the children are the same for
nsview and qcocoaaccessibleelement.
Factor out a function.
Change-Id: I4be091c6acde0de519a358f879c24eae04ec4c8d
Reviewed-by: Gabriel de Dietrich <gabriel.dedietrich@digia.com>
Flip !Q_OS_IOS conditions to Q_OS_MACX where it seems appropriate,
remove a redundant condition in qtextcodec_p.h.
Change-Id: I21c8c0c490f1eb4a9337a7f2f3e907c125489438
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@digia.com>
I.e. let the user know whether he is on 2G / 3G / 4G.
Since usually apps are rather interested whether they are on Wifi or
cellular, change of cellular type (i.e. transition from 3G to 4G etc.)
will not trigger emission of the
"QNetworkConfigurationManager::configurationChanged" signal. The
cellular type check will only be performed if the user queries it by
calling the respective member methods (bearerType() and
bearerTypeName()).
Reading the cellular type takes around 1-2 milliseconds.
In addition, add new fields "BearerEVDO" and "BearerLTE" to the public
API for QNetworkConfiguration.
Change-Id: I4c4ec52926f862b84487c91a88b1e20e590dd793
Reviewed-by: Kevin Krammer <kevin.krammer@kdab.com>
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
Reviewed-by: Thomas McGuire <thomas.mcguire@kdab.com>
This fixes a regression introduced in 3e2cd8ef6f. We shouldn't try
to call mkdir for just a drive letter.
Task-number: QTBUG-31862
Change-Id: I56035857aec03f9f1cceef62e82f372486b7737b
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@digia.com>
Reviewed-by: Shawn Rutledge <shawn.rutledge@digia.com>
Reviewed-by: Kai Koehne <kai.koehne@digia.com>
The current implementation does not allow colons in attribute values,
as it's already used as a separator in the form of 'name:value'.
This change allows colons in the value string - one use case is to
have URL attributes, as in 'imageUrl:qthelp://path.to.image'.
Change-Id: I05c5cb32ffc79a39fbe5e4102f7a4b5bdcc2be53
Reviewed-by: Martin Smith <martin.smith@digia.com>
Reviewed-by: Jerome Pasion <jerome.pasion@digia.com>
BSD licensing should be used for QtOpenGLExtensions instead of LGPL as
it is a static library that is linked into applications. Using LGPL
would impose additional requirements for application developers to
provide object files to end users to be able to relink the application
with a modified or updated QtOpenGLExtensions static library.
Task-number: QTBUG-29918
Change-Id: I0bb80f8ba5158be0a71fe4c5e1c37787ce21337d
Reviewed-by: Sean Harmer <sean.harmer@kdab.com>
Reviewed-by: Lars Knoll <lars.knoll@digia.com>
We were not comparing all fields.
Spotted-by: Richard J. Moore <rich@kde.org>
Change-Id: I5c1a773b00b64a26606c47ced292808696c9b4bb
Reviewed-by: Richard J. Moore <rich@kde.org>
Each pair of (normal request, preconnect request) requires only one
socket. E.g. if there is 1 preconnect request in-flight and 2 normal
requests, we need only 2 sockets in total, and not 3.
Therefore, we need to keep track of whether a request is
preconnecting or a normal one.
Task-number: QTBUG-31594
Change-Id: If92ccc35abadfa6090d64ee92bd466615909c94c
Reviewed-by: Richard J. Moore <rich@kde.org>
This is similar to 9a5ade8abe which
modified the qt5_use_modules macro.
Change-Id: Ib63ee693ed2e25de6ac9d969e8be27ddbf8948c3
Reviewed-by: Peter Kümmel <syntheticpp@gmx.net>
Reviewed-by: Stephen Kelly <stephen.kelly@kdab.com>
NSPICTPboardType is deprecated. Mac OS will translate
pasted images of that type to a different image format.
Change-Id: Id7a68509615816a852abf1c11a70f0f0c862f0f2
Reviewed-by: Jake Petroules <jake.petroules@petroules.com>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Useful to have the same behavior in QLineEdit and QTextEdit.
Change-Id: I02d6ac0a6072403c7aa3e8f76a728c46a9d66b14
Reviewed-by: David Faure (KDE) <faure@kde.org>
Setting the buttons in the constructor is a use-case that happens
more often than setting the orientation. Yet, there was a
(Qt::Orientation,QWidget*) constructor, but no
(StandardButtons,QWidget*) one.
This patch adds it.
Change-Id: If6a5c9f7450a388cd77bd93c8dd144b2fdc11847
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
Reviewed-by: David Faure (KDE) <faure@kde.org>
Reviewed-by: Kevin Ottens <kevin.ottens@kdab.com>
Reviewed-by: Konstantin Ritt <ritt.ks@gmail.com>
Reviewed-by: Thorbjørn Lund Martsum <tmartsum@gmail.com>
Reviewed-by: Giuseppe D'Angelo <giuseppe.dangelo@kdab.com>
We can not trust the iterator when we do erase. This patch
makes an index while we do a takeAt and then we restore
the iterator from that index.
This both ensures that we can merge an iterator check
and ensures consistent behavior.
Change-Id: Icbe67051cb055d24541475e826109212eba9dd30
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Even if it currently works it is highly non standard to
modify (incl. free/delete) the key for the iterator and then
use it afterwards.
The current code prevents a merge of a patch that uses the
key to check if the iterator is valid.
Change-Id: Ia89553ea4ba53e24488a6bfe06b8d7d0e351c1ea
Reviewed-by: J-P Nurmi <jpnurmi@digia.com>
Reviewed-by: Jens Bache-Wiig <jens.bache-wiig@digia.com>
The cmake variable for the mkspec dir must specify the source
location if used in the build dir, and must specify the install
location if used in the install dir.
Change-Id: I2fee8cd0c7198e9fc5cbb63972e20c75636672d1
Reviewed-by: Thorbjørn Lund Martsum <tmartsum@gmail.com>
Reviewed-by: Stephen Kelly <stephen.kelly@kdab.com>
Even though the documentation says that "This should never normally
happen", we do get a "navigation" change when an external keyboard
is connected or disconnected.
Thanks to BogDan for finding the solution.
Task-number: QTBUG-31806
Change-Id: Id92639800e617821976f842f93e09517529e14ef
Reviewed-by: BogDan Vatra <bogdan@kde.org>
Change-Id: Ica84e9ce715702b690906ef748970e16d9162c4b
Reviewed-by: Gabriel de Dietrich <gabriel.dedietrich@digia.com>
Reviewed-by: Aurélien Gâteau <agateau@kde.org>
it needs to come after the vista and xp style blocks.
Change-Id: I66b5c1eff41b19a76802a77b6926552d7343b29c
Reviewed-by: Joerg Bornemann <joerg.bornemann@digia.com>
Reviewed-by: Jens Bache-Wiig <jens.bache-wiig@digia.com>
they are not used. only the negative QT_NO_FOO_STYLE defines are used.
Change-Id: Ic39081240c78cf66bd50c300b4aad04abca6f38c
Reviewed-by: Joerg Bornemann <joerg.bornemann@digia.com>
Reviewed-by: Jens Bache-Wiig <jens.bache-wiig@digia.com>
we have commonstyle for that for a while now.
Change-Id: I1ab3c86af994fc9cc7aaf65d28a2428b5e02be99
Reviewed-by: Joerg Bornemann <joerg.bornemann@digia.com>
Reviewed-by: Jens Bache-Wiig <jens.bache-wiig@digia.com>
The behavior introduced by qtdeclarative's 475d1ed4f6a2 mandates that an
expose event follows a geometry change event. The fix for this uncovered a bug
on the raster backing store which caused the rotation to break on QtWidgets.
The problem was rooted on the assumption that the
QQnxRasterBackingStore::flush() method was only called after paints to the
backing store surface, which was discovered to be a false assumption. Flushing
the backing store can occur in other circunstances, which are out of scope on
the context of the QNX plugin.
Task-number: QTBUG-31329
Change-Id: Icfc3e85324f5a8745f6b5f189f27f8b763f770c7
Reviewed-by: Kevin Krammer <kevin.krammer@kdab.com>
Reviewed-by: Sean Harmer <sean.harmer@kdab.com>
Qt itself has no API for disabling screen power save, although this
might make an interesting addition at some point.
Instead we make the QNX QPA plugin determine if power save should be
disabled from the QQNX_DISABLE_POWER_SAVE environment variable. This
would typically be set in the application's bar descriptor file.
Change-Id: Ie195ecd0496efd533ad9a32b65883283f471c534
Reviewed-by: Kevin Krammer <kevin.krammer@kdab.com>
Reviewed-by: Rafael Roquetto <rafael.roquetto@kdab.com>
Unit test by Friedemann Kleint <Friedemann.Kleint@digia.com>
Task-number: QTBUG-31569
Change-Id: I526d33d4f88a41f6ac349098476bc45af6c841b0
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
__clear_cache is missing from Android; use cacheflush instead.
Backport in Qt's own PCRE copy of the upstream fix:
http://vcs.pcre.org/viewvc?view=revision&revision=1340
Task-number: QTBUG-31754
Change-Id: I4996265f7b19343ba24983d4f8b2f75e2b578504
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
This was a regression from 5.0. Since we still require a minimum
height to be present in order for the spin buttons to draw correctly,
we require the same height from a spinbox without a frame as one
with a frame.
task-number: QTBUG-31538
Change-Id: I999f59a29e826e8e621faebffce9cccd9bedb8ca
Reviewed-by: Gabriel de Dietrich <gabriel.dedietrich@digia.com>
Reviewed-by: Jens Bache-Wiig <jens.bache-wiig@digia.com>
We do not decide which socket a HTTP request is sent on until the
socket is actually ready for sending a request (i.e. it is connected
for HTTP requests and encryption is done for HTTPS requests).
When deciding how many sockets we need for the queued requests, we
consider an in-flight socket as free if it is still connecting.
However, we considered a socket that was connected but needed to
complete encryption as busy, and would instead open another socket.
Now, we consider an encrypting socket as in-flight as well.
Change-Id: I93d6743da6fc430d1424c6965e1475865fd97243
Reviewed-by: Richard J. Moore <rich@kde.org>
The issue is already fixed in 5.0 but let's be nice and ensure the issue
won't be reintroduced later.
Task-number: QTBUG-30931
Change-Id: Ia6944acaf6e7217f8d0f1fa75d0e9977db11d892
Reviewed-by: Lars Knoll <lars.knoll@digia.com>
Change the example to Q_OS_UNIX.
Change-Id: I8aed12237408a0e526839a96867aceae33b993b9
Reviewed-by: Shawn Rutledge <shawn.rutledge@digia.com>
Reviewed-by: Jerome Pasion <jerome.pasion@digia.com>
KEYCODE_MEDIA_FAST_FORWARD is mapped to Qt::Key_Forward,
which undoes Back. I believe the best match in Qt would be
Qt::Key_AudioForward which is also mapped by XCB from
XF86XK_AudioForward /* fast-forward audio track */
KEYCODE_MUTE is mapped to Qt::Key_Volume_Mute, but the mute button is
for muting the microphone, the volume mute is called
KEYCODE_VOLUME_MUTE.
Change-Id: Id0b78c9bde78faef1f5d31019693e6c466941d70
Reviewed-by: BogDan Vatra <bogdan@kde.org>
QDateTime.toString() is writing out milliseconds since change
15da0a5af2. Unfortunately this breaks
QDateTime::fromString() with Qt::TextDate which can't handle the new
format.
Fix by making QDateTime::fromString split up seconds and milliseconds
on a period, if any. Now
QDateTime dt = ...;
assert(QDateTime::fromString(dt.toString(), Qt::TextDate) == dt)
works again.
Change-Id: Ibfe9032e357ceaf894e33f3e33affe94f56dbf5c
Reviewed-by: Lars Knoll <lars.knoll@digia.com>
It is deprecated and clang is starting to warn about it.
Patch mostly generated by clang itself, with some careful grep
and sed for the platform-specific parts.
Change-Id: I8058e6db0f1b41b33a9e8f17a712739159982450
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
On stock QNX it is quite common that libscreen doesn't provide the physical
screen size. The warning about this has been replaced with a debug statement,
because it was annoying especially when running auto tests.
Change-Id: Iab07f0d4a6293c40678f2ec3e77352e68797c8c7
Reviewed-by: Kevin Krammer <kevin.krammer@kdab.com>
Reviewed-by: Sean Harmer <sean.harmer@kdab.com>
The mouse over event on upper level menu separator should
automatically close the sub menu listing. Manual test is not
needed for this commit since it is easy to test it against common
examples.
Task-number: QTBUG-31664
Change-Id: I323d7ba206352a5d533584543b9a2ebf842b4dfc
Reviewed-by: J-P Nurmi <jpnurmi@digia.com>
Currently, in save mode, if selectFile has not been called, the native
cocoa file dialog puts the folder name in the line edit. This patch
restores the old behavior where the line edit is presented empty to
the user.
Task-number: QTBUG-31619
Change-Id: I938eb6d968e0e2c343e70bc19f29663e112d0496
Reviewed-by: Richard Moe Gustavsen <richard.gustavsen@digia.com>
Also check if the selectedFile is a bundle to correctly
set currentDir.
Patch for Qt 4.8 has been submitted
Task-number: QTBUG-31562
Change-Id: I72b0e8484b3c3a610932c03cd7fdab4ddee70277
Reviewed-by: Richard Moe Gustavsen <richard.gustavsen@digia.com>
because if it is automatically restored it's out of the application's
control, so the user's interaction will be ignored. Change
I8ce3cd94f5ae81d7877a346743ca4e0e188baa02 did this for normal windows
by default, but the dialog helpers generate windows which aren't
affected by that.
Task-number: QTBUG-31750
Change-Id: I636bd87b664a489a2dc8693dad5370a715b1cf7b
Reviewed-by: Gabriel de Dietrich <gabriel.dedietrich@digia.com>
Silently failing doesn't seem the right way to handle this.
Updated docs for DontUseNativeDialog option: it's not just for
the Mac anymore. Ensure that the Qt dialog and Mac panel
will never be shown at the same time.
Change-Id: Ia9e80754df6c7622d9039c8dd050ac4de771a030
Task-number: QTBUG-29161
Reviewed-by: Richard Moe Gustavsen <richard.gustavsen@digia.com>
QNSColorPanelDelegate::finishOffWithCode sets mResultSet = true
and it needs to be set false again the next time the dialog is shown.
Task-number: QTBUG-31566
Change-Id: If1d4bb9d4e76273c6423f5bf2ae37790e8a9704e
Reviewed-by: Liang Qi <liang.qi@digia.com>
... because otherwise this would crash.
Apparently there are cases where the header name is empty.
Task-number: QTBUG-31667
Change-Id: I31b3e964502c05b7614c23876bb3752fa75ab22d
Reviewed-by: Richard J. Moore <rich@kde.org>
Reviewed-by: Shane Kearns <shane.kearns@accenture.com>
Change-Id: I840f963c3648d123b31f79aa2c8902c0ad74e982
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@digia.com>
Reviewed-by: Stephen Kelly <stephen.kelly@kdab.com>
During a call to QDir::mkpath(), the same path could be created
by another process, in which case the OS mkdir will fail with EEXIST.
But the docs for mkpath() state that it's not an error if it
already exists, whereas for mkdir() it is an error. So
QFileSystemEngine::createDirectory should accept the EEXIST error
silently if it occurs while creating the sequence of parent directories
and the final leaf directory, but should fail if EEXIST happens when
it was called from QDir::mkdir(), which is when the createParents
parameter is false. We assume the operating system mkdir() and
CreateDirectory() are atomic, so there should be no race condition
in QDir::mkdir(). It's not necessary for mkpath() to call stat()
at each level, only to check whether an existing entry is a directory
or a file. Also added to the autotest to verify that if the
path is an existing file, creating a dir with the same name will
fail in either mkdir or mkpath.
Task-number: QTBUG-30046
Change-Id: I926352f10654fdf3b322c8685bb85ad8b8844874
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@digia.com>
There is a mismatch how QML and C++ converts QJsonValue. This patch
unifies conversions by adding QJsonValue support in QVariant::convert().
Change-Id: I8a1db3d77c517945ef48064b4b66ba03aa4f2fd0
Reviewed-by: Frederik Gladhorn <frederik.gladhorn@digia.com>
QWidget::setWindowIcon() and similar call createTLExtra()
which creates a QWindow without setting the native attributes
on the parent, which can cause crashes when setParent_sys()
decides to delete the window.
Task-number: QTBUG-31672
Change-Id: I4c40ee12741be88b2281df90329ffb698d4009eb
Reviewed-by: Shawn Rutledge <shawn.rutledge@digia.com>
The time format should depend on the device settings (24 hour format)
Change-Id: I452d9b7158d39c4a657adfd9e64c99549eeda4ff
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@digia.com>
Reviewed-by: Rafael Roquetto <rafael.roquetto@kdab.com>
The delayed root window posting introduced by
77a06e7e6c broke the OpenGL support. In that patch, the posting of
the root window is trigerred by QQnxRasterRasterBackingStore's call to
QQnxWindow::post(), that obviously does not happen when OpenGL is enabled,
therefore requiring the OpenGL context to explicitly post() the root window.
Change-Id: Ifd302c1dde612a03b79c778ec4586aa70f88260d
Reviewed-by: Sean Harmer <sean.harmer@kdab.com>
We destroy the thread data for the main thread when the QCoreApplication
is destructed, and then delete the pthread key for the thread data in
the global static destructor function 'destroy_current_thread_data_key'.
The user may have its own Q_DESTRUCTOR_FUNCTION though, which may or may
not run after we've destroyed the key. If it runs after we've destroyed
the key, we'll end up trying to re-create the tread-data, as expected,
but set_thread_data() will fail to persist it, as pthread_setspecific
is called with an invalid key. The result is an infinite recursion:
...
6 in QThreadData::current () at qthread_unix.cpp:216
7 in QObject::QObject (this=0x48e1b30, dd=@0x48e1b40, parent=0x0) at qobject.cpp:703
8 in QThread::QThread (this=0x48e1b30, dd=@0x48e1b40, parent=0x0) at qthread.cpp:396
9 in QAdoptedThread::QAdoptedThread (this=0x48e1b30, data=0x48e1af0) at qthread.cpp:120
10 in QAdoptedThread::QAdoptedThread (this=0x48e1b30, data=0x48e1af0) at qthread.cpp:130
11 in QThreadData::current () at qthread_unix.cpp:219
12 in QObject::QObject (this=0x48e1a20, dd=@0x48e1a30, parent=0x0) at qobject.cpp:703
...
To solve this, we reset current_thread_data_once when destroying the key,
so that subsequent calls to pthread_once to potentially create the key
will call create_current_thread_data_key once more. This means we'll leak
the key for this particular use-case, since we don't end up calling
pthread_key_delete a second time, but this leak is small and happens
typically only for a short duration during application shutdown.
Change-Id: I580484a3239849e891172e24e7f77b75afd2c51b
Reviewed-by: Olivier Goffart <ogoffart@woboq.com>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
qguiapplication.cpp:792:133: warning: format not a string literal and
no format arguments [-Wformat-security].
Introduced by 0d7d53fd46 .
Change-Id: I70268a8b597be33226efb9bb4d7d9f672f0d7440
Reviewed-by: Kai Koehne <kai.koehne@digia.com>
The previous implementation of the nested runloop was derived from the
Mac Cocoa implementation(?), and did not correctly deal with UI
animations and other UIKit functions which are run in a different mode
to the default mode. This version corrects that (in most cases) and
switches the implementation to use CoreFoundation instead of NextStep
APIs.
Change-Id: I45802d22044465749a1e5b6207d745268f6ae8a1
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@digia.com>
Move iOS event dispatcher from platform plugin to platform support, so
that it can be used by multiple iOS platform plugins.
Change-Id: I9041b2de5e00e5fe8f30af2dfd922b4f5c594802
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@digia.com>
QDockWidgetPrivate::mouseMoveEvent calls q->move() which can result
in a call to QDockWidgetPrivate::endDrag(), which will delete the
state struct. Therefore mouseMoveEvent needs to check whether it's
null or not before accessing it.
Task-number: QTBUG-31672
Change-Id: I91a31620f16a80b31b65b2742e7937ae8960f0fd
Reviewed-by: Gabriel de Dietrich <gabriel.dedietrich@digia.com>
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
this is for shadow builds during build time, where the respective files
are expected in the source dir.
Change-Id: I18dcfbdef99e1562a51dacac333642cae8105ebd
Reviewed-by: Joerg Bornemann <joerg.bornemann@digia.com>
When behind a mouse-event-transparent window, synthesized leave events
are emitted, because the transparent window is considered as the window
under the mouse, not the Qt window.
This change skips transparent windows when searching for the window
under the mouse.
Task-number: QTBUG-31464
Change-Id: I85c8b46a1af37b4d1c5d1d77566ab045657aa9ae
Reviewed-by: Björn Breitmeyer <bjoern.breitmeyer@kdab.com>
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
Initialize undockedGeometry to roughly the current position.
Task-number: QTBUG-31044
Change-Id: I03cbe280d1215bb58ab721b60e29b45359cde76d
Reviewed-by: Andy Shaw <andy.shaw@digia.com>
We can combine the hybrid and non-hybrid use-cases into a single static
library if we are careful about which symbols are included in which
object files. By limiting the main() and qt_user_main() functions to
their own translation units, the linker will only pick them up if they
are missing at link time (the user's program do not provide them).
This technique is resilient to the -ObjC linker flag, which includes all
object files that implement an ObjectiveC class or category, but will
fail if the -all_load flag is passed to the linker, as we'll then have
duplicate symbols for either main() or qt_user_main(). The latter should
not happen unless the user provides the flag manually, and in the case
he or she does, there's ways to work around it by providing less global
flags such as -ObjC or -force_load.
Change-Id: Ie2f8e10a7265d007bf45cb1dd83f19cff0693551
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@digia.com>
Reviewed-by: Richard Moe Gustavsen <richard.gustavsen@digia.com>
Instead of force-loading the whole static library of the platform plugin
we tell the linker to look for the missing symbol qt_registerPlatformPlugin.
This symbol is provided by the same object file as the plugin's static
initializer, so the object file is included in the final binary and
the static initializer is run, resulting in the plugin registering with
Qt.
We could have marked the actual static initializer wrapper provided by
Q_IMPORT_PLUGIN(QIOSIntegrationPlugin) as undefined, but due to the C++
mangling this would look less intuitive on the linker command line than
the custom dummy function that we provide, which has C linkage.
Change-Id: I6805537e1f49260a41d48c555376964cb1fe75d8
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@digia.com>
Reviewed-by: Richard Moe Gustavsen <richard.gustavsen@digia.com>
Cocoa sets the NSWindow hidden when it's minimized, and we don't
propagate that further to QWindow or QWidget (QWidget's 'visible'
documentation is quite ambiguous about that state).
The obvious fix is to check for QWindow's visibility. However, this
might need to be corrected later if we truly want the visible state
to be in sync.
Task-number: QTBUG-31117
Change-Id: I457452eff75e6ea746a0ef5351104f8bbb65f98b
Reviewed-by: Richard Moe Gustavsen <richard.gustavsen@digia.com>
Refactors some code from previous changes addressing the same issue
for maximized and fullscreen states.
Change-Id: I8a27274a90e6610b424ea7af3269369a1d724c48
Reviewed-by: Richard Moe Gustavsen <richard.gustavsen@digia.com>
There's no NSWindow notification we can listen to, but we can override
behavior from its delegate.
Change-Id: I61cebf4119f83c770fe4e7f45ff0d4e8bf9d0df9
Reviewed-by: Richard Moe Gustavsen <richard.gustavsen@digia.com>
We update the internal/synched state by assigning it its QWindow's
state. Since that one is not a flag, it can only be in one state at
a time, meaning that we may lose the maximized state (but we've been
losing it for some time).
Task-number: QTBUG-30139
Change-Id: Idf30713c6ae912cafe3bbdd7be18214cf4de34f7
Reviewed-by: Richard Moe Gustavsen <richard.gustavsen@digia.com>
Change-Id: Ifdf487fbae8acbffb32b6db5f8dd93e9eb213d77
Reviewed-by: Gabriel de Dietrich <gabriel.dedietrich@digia.com>
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@digia.com>
Fix multiple gcc warnings about statements in accessibleDebug arguments
have no effect.
Change-Id: I34bf05d911707f0b50592bf76442a3805c560d56
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
Reviewed-by: Frederik Gladhorn <frederik.gladhorn@digia.com>
Since we don't use focus to show/hide place holder text, we don't need
to check widget focus but if text is empty.
Change-Id: I37f2fdb9e20a64ca38c61e60190f95635695c613
Reviewed-by: Aurélien Gâteau <agateau@kde.org>
Reviewed-by: David Faure (KDE) <faure@kde.org>
When qdoc searched for QtConncurrent::blockingFilter(),
it found the module node for QtConcurrent instead of the
namespace. This was because qdoc wasn't given specific
enough instructions on how to perform the search. Now
it searches for the namespace first, then the C++ class,
then the module.
Task-number: QTBUG-31535
Change-Id: I4f8aec503903508789738f2a77c76f47a3e80a93
Reviewed-by: Jerome Pasion <jerome.pasion@digia.com>
Forgetting to deploy the platform plugin, or deploying it to the wrong
folder, is a common mistake. The old error message didn't made it however
explicit what was happening. Make the log more verbose, and explicitly
state that the missing plugin is the reason for termination.
Change-Id: I810a0ef8da5f8e898e7e0c6f853972514a05c75d
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
Some functions rely on the size to be sure that the file can be read
from. Therefore when nativeSize() fails to access the file we should
return 0 instead of a random size to prevent these functions from
acting as if there is a valid file to read from.
Task-number: QTBUG-31571
Change-Id: I589fc199760ba40ef2e63047b2f226cb196f7e56
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Reviewed-by: Janne Anttila <janne.anttila@digia.com>
Remove the mapping from press-and-hold to right-mouse-button
press. This logic was implemented in the android plugin, and
there was no way to disable it for applications that needed
to handle long presses in a different way.
Add an environment variable to enable the mapping for those
who really need it, but do not advertise it. (We want to
add a proper API for this in 5.2)
Task-number: QTBUG-31644
Change-Id: Ia5ae75d6147485b73c78b7c3fa88d11af5d84c06
Reviewed-by: BogDan Vatra <bogdan@kde.org>
This was broken by a previous fix which aimed to fix gaps in
polylines with tiny line segments. The result was that we
skipped updating the origin point when stroke() didn't produce
pixels which accidentally included the case of the line
being completely outside the deviceRect. I fixed this
by returning the value of clipLine in drawLine to the caller
so we could still update the origin for this case.
Task-number: QTBUG-31579
Change-Id: Iac29436f042da7658bbeaf9370351dc6f2c95065
Reviewed-by: Lars Knoll <lars.knoll@digia.com>
This is actually a list in CMake, not a value to be prepended
to paths. Specify the QT_SYSROOT instead to root the location
of include directories. CMake will soon get a CMAKE_SYSROOT
variable which will replace this.
Change-Id: I239f69f127f3676a3835aa4f29638f44ef209819
Reviewed-by: Volker Krause <volker.krause@kdab.com>
Reviewed-by: Stephen Kelly <stephen.kelly@kdab.com>
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@digia.com>
IFileDialog::SetDefaultExtension() needs to be called for the dialog to
automatically update the suffix when the user chooses a new file type.
Calling it IFileDialogEvents::OnTypeChanged as introduced by
42760e43ea causes it to use the previous
suffix when switching.
Call in setNameFilters() using the first available suffix instead.
Task-number: QTBUG-31381
Task-number: QTBUG-30748
Change-Id: Idd532e4190c752aa80abce963dd0389be7d22a34
Reviewed-by: Andy Shaw <andy.shaw@digia.com>
Reviewed-by: Oliver Wolff <oliver.wolff@digia.com>
Reviewed-by: Joerg Bornemann <joerg.bornemann@digia.com>
The code that sets 'text' variable was executed only for KeyPress
events. This patch moves 'text' initialization higher up where it
will be run for both - KeyPress and KeyRelease events.
Task-number: QTBUG-31285
Change-Id: I140de9ef5d29beedad276d348bed964ecdf46f9d
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@digia.com>
Reviewed-by: Andy Shaw <andy.shaw@digia.com>
QDoc uses .qdoc files as snippets in other .qdoc files. These snippet
.qdoc files are also parsed, which is not the intent. This change
resolves the issue by renaming the files to have a different
extension.
Task-number: QTBUG-31574
Change-Id: I399c8b714478b9c4b8fbe2db53f601e6a1386cea
Reviewed-by: Jerome Pasion <jerome.pasion@digia.com>
The class documentation missing the \wrapper command, making QDoc
complain a lot.
Change-Id: Iad8fa3d2facadfb3393642dc08ff4dc416bdccc7
Reviewed-by: Martin Smith <martin.smith@digia.com>
Reviewed-by: Jerome Pasion <jerome.pasion@digia.com>
make the include dir in the source tree the "main" include path, as
that's where the majority of the headers is. then selectively add the
shadowed dirs.
Change-Id: I03ad13cfcf77175c141b94d41b1221740d851faf
Reviewed-by: Joerg Bornemann <joerg.bornemann@digia.com>
otherwise even non-angle builds with pre-generated headers will get them
into their include path, which is Not Good (tm).
Change-Id: Ie98354297baf3564ef82b3d4a32e5763e625b8f2
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
Reviewed-by: Joerg Bornemann <joerg.bornemann@digia.com>
warning C4267: 'argument' : conversion from 'size_t' to ' int', possible
loss of data.
Change-Id: I79af7497420d468b5bc7c48c9ae21b86117519a9
Reviewed-by: Mitch Curtis <mitch.curtis@digia.com>
Mitch Curtis
Friedemann Kleint
Frederik Gladhorn
Sean Harmer
David Gil
Gabriel de Dietrich
Rafael Roquetto
Thiago Macieira
Paul Olav Tvete
Shawn Rutledge
J-P Nurmi
Richard Moe Gustavsen
Thorbjørn Martsum
Jerome Pasion
Oswald Buddenhagen
Richard J. Moore
Gunnar Sletta
Jake Petroules
Peter Hartmann
Kai Koehne
Topi Reinio
Jonathan Liu
Stephen Kelly
Morten Johan Sørvig
Benjamin Port
Marc Mutz
Jørgen Lind
Debao Zhang
David Faure
Tasuku Suzuki
Liang Qi
Giuseppe D'Angelo
Jens Bache-Wiig
Konstantin Ritt
Allan Sandfeld Jensen
Bernd Weimer
Leonard Lee
Samuel Gaist
Dominik Holland
Jędrzej Nowacki
El Mehdi Fekari
Tor Arne Vestbø
Ian Dean
Sebastian Eifert
Nico Vertriest
Laszlo Papp
Martin Smith
Ivan Komissarov
Andy Shaw
Gatis Paeglis
Robert Griebl
Albert Astals Cid
Hannu Lyytinen