Since Qt4, there is a bug which causes Qt to drop dead key modifiers
(like graves and acutes) if the user types enough fast on MS Windows.
This happens because of an extrange behavior of Windows, which drops
dead keys on ToUnicode() calls.
This patch tries to workaround that.
Task-number: QTBUG-8764
Task-number: QTBUG-10032
Change-Id: Ifdde25817743194fd5c0b7533c27f46a7a108ca4
Reviewed-by: Friedemann.Kleint@digia.com
Reviewed-by: oliver.wolff@digia.com
Reviewed-by: marc.mutz@kdab.com
Reviewed-by: bjoern.breitmeyer@kdab.com
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
Reviewed-by: Oliver Wolff <oliver.wolff@digia.com>
Useful for not having to fall back to QPainterPath drawing when using
the raster engine with a retina screen (which has a 2x scale).
Change-Id: I0a9f754d31b0ecd8e8daf7a01331d19716bab680
Reviewed-by: Gunnar Sletta <gunnar.sletta@digia.com>
This currently fails on case-insensitive file
systems since the check for existence then triggered
and indicated "file already exists".
Check on the file id (inode or file id) whether
the target file is really a different file for a
case-changing rename.
Task-number: QTBUG-3570
Change-Id: I1b2d40850692e02142ee23d2c753428de00aedc6
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@digia.com>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
This reverts commit f15a73f254.
The note has been there for the 5.0.0 release, so it should now
be removed.
Change-Id: I8744f74834cc87002aa590de9c258f544bcf0c9b
Reviewed-by: Sergio Ahumada <sergio.ahumada@digia.com>
Release the table item below the table test instead of releasing
the text item twice.
Change-Id: I74d283d50a39b9a4570b73a8297ed3dbb2de2271
Reviewed-by: Frederik Gladhorn <frederik.gladhorn@digia.com>
This is necessary for initializing things in a library, which require
a QCoreApplication instance (unlike Q_CONSTRUCTOR_FUNCTION, which runs
before that). Example use cases: KCrash (segv handler), and KCheckAccelerators
(debugging tool triggered by magic key combination).
Change-Id: I5f4c4699dd4d21aea72b007989ba57467e86ed10
Reviewed-by: Olivier Goffart <ogoffart@woboq.com>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
This is a partial revert of 7abf623. The desktop components have been
fixed to call QStyle in the main GUI thread.
Change-Id: Ifd8364269b7d2e350f34647c128ff2fbde70afd6
Reviewed-by: Jens Bache-Wiig <jens.bache-wiig@digia.com>
When re-applying window flags in setParent, force top level on
or off according to state.
Task-number: QTBUG-28872
Change-Id: If931fcb38394f472a6cdf260aa935c1d03779611
Reviewed-by: Joerg Bornemann <joerg.bornemann@digia.com>
Make sure that dpy->request == dpy->last_request_read after setting
up a new window. If we don't do this, last_request_read might never
be updated until the difference hits a limit that can lead to hangs
in the application (see e.g. QTCREATORBUG-8373).
Task-number: QTBUG-29106
Change-Id: I390493ca6f966dc105d3ea3a2c48abec01177bc2
Reviewed-by: Uli Schlachter <psychon@znc.in>
Reviewed-by: Samuel Rødal <samuel.rodal@digia.com>
GCC 4.3 had support for decltype but not the new function syntax (which
we call "auto function" for short). That meant the code did not compile.
qtconcurrentrun.h:105: error: expected initializer before ‘->’ token
Task-number: QTBUG-28984
Change-Id: I792276ec59c4f6e73f6137c517636e70c71ed849
Reviewed-by: Olivier Goffart <ogoffart@woboq.com>
Drop the read and write permissions for group and other users in the
system.
Change-Id: I8fc753f09126651af3fb82df3049050f0b14e876
Reviewed-by: Richard J. Moore <rich@kde.org>
eventfd(7) uses less resources than a pipe, as it only needs to store a
single 64-bit integer, as opposed to a full buffer.
It was introduced first on Linux version 2.6.22 and glibc 2.7. However,
both the configure-time test and the runtime usage require the use of
EFD_CLOEXEC for thread-safety, so this code will be enabled only for
Linux 2.6.27 and up as well as glibc 2.9 and up.
Change-Id: Ic7e10b28d7b1d4ca24be614ed84055c4429a68e4
Reviewed-by: Robin Burchell <robin+qt@viroteck.net>
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@digia.com>
Some of it still referred to classes that were cleaned up in Qt 5.0
Change-Id: Ief4ed4b4fce074884f755b0d18a526ac40ad1b0b
Reviewed-by: Olivier Goffart <ogoffart@woboq.com>
This is a very common thing to do, e.g. in order to send urls via DBus.
Change-Id: I277902460ee1ad6780446e862e86b3c2eb8c5315
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Had to move QTextStreamPrivate to a private header, to be able to use
its new internal Params struct from qdebug.cpp
Change-Id: If28e25f27bbd04b1825a5eb3e2ef83ecad72e7b2
Reviewed-by: Olivier Goffart <ogoffart@woboq.com>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
This is needed by QSaveFile.
Move QTemporaryFilePrivate to that header too, to avoid any confusion.
Change-Id: I8dce8a4fb853a9823c49ec565867432331e748cd
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
QFSFileEngine::rename() on Windows doesn't overwrite the existing destination.
Keep that unchanged (it's the desired behavior in QFile::rename), and
provide cross-platform rename-overwrite behavior in the new method.
This is needed by QSaveFile.
Change-Id: I5e753d289d8a53692530a48a1783d62e26169cdc
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@digia.com>
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
ShellExecute fails to open a share folder due to using '/' instead of '\'.
Windows API doesn't support forward slashes for extended-length path.
Extended-length path are path that start with a "\\?\" prefix. For example,
"\\?\c:\very_long_path\foo\bar.txt", or in the case of a UNC path that
would be "\\?\very_long_path\foo\bar.txt". [1]
[1] http://msdn.microsoft.com/en-us/library/aa365247%28VS.85%29.aspx#maxpath
Task-number: QTBUG-13359
Change-Id: Ibb113abeebd56f106f76520bc23dba797de548fa
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@digia.com>
QMessageAuthenticationCode is HMAC implementation based on
QCryptographicHash abilities. HMAC is often used in OAuth and similar
authentication protocols.
Change-Id: Ifc73947ad06c36a1b770315b7e89ba5c01c5e79e
Reviewed-by: Richard J. Moore <rich@kde.org>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Restrict disconnect() to the clicked() signal, leaving connections
to destroyed() (as used by QStyleSheetStyle) intact.
Task-number: QTBUG-20292
Change-Id: I7471b4d1262ec0684e4446b5c17513717c502749
Reviewed-by: Joerg Bornemann <joerg.bornemann@digia.com>
Avoid using realpath(X,0) on Mac OSX at all, since
even on versions of OSX where realpath(X,0) is supported,
we still get the legacy version due to our compiler flags.
If we were to change the -mmacosx-version-min to 10.5 or
higher then this patch would be safe but unnecessary.
Task-number: QTBUG-28282
Change-Id: Iee21003f3e9616482483a05ceee706b476091914
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Reviewed-by: Robin Burchell <robin+qt@viroteck.net>
The platform specific CFLAGS, most notably -march=armv7-a, were only
added to QMake's _RELEASE flags but not to the _DEBUG flags.
This then resulted in strange compilation errors such as this one:
http://qt-project.org/forums/viewthread/22141
Change-Id: Ib47996c6946b043294437e8543827da270df836d
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@digia.com>
in qt5, gtkstyle use gtk_widget_send_focus_change, this had a side
effect that will trigger real gtk im module, if qt load the same qt and
gtk im module for same input method, it will cause conflicts for qt im
module. The only gtk widget have im-module property is GtkEntry and
GtkTextView, since only GtkEntry is used here, we only need to override
im-module for GtkEntry.
gtk-im-context-none exists in gtk+ since 2.19.5, and for older version,
it will also harmlessly fallback to gtk-im-context-simple.
Change-Id: I4ffb93453e77c8d5db3349b084342bdca8e4a571
Reviewed-by: Liang Qi <liang.qi@digia.com>
Reviewed-by: J-P Nurmi <jpnurmi@digia.com>
Because this method creates a QSocketNotifier, it needs
to be split into a part that is run on initialization, namely
QQNXLocaleData::initialize(), and one that is run delayed
through event loop invocation, namely QQNXLocaleData::installSocketNotifier().
Task-number: QTBUG-28701
Change-Id: Ib60000902692bbca4820d3d0bc7719212668dfa9
Reviewed-by: Laszlo Papp <lpapp@kde.org>
Reviewed-by: Kevin Krammer <kevin.krammer@kdab.com>
Reviewed-by: Sean Harmer <sean.harmer@kdab.com>
Change-Id: I3ea2556769703a8cd4c2931cc2332ab0733fbea6
Reviewed-by: Sean Harmer <sean.harmer@kdab.com>
Reviewed-by: Kevin Krammer <kevin.krammer@kdab.com>
Reviewed-by: Thomas McGuire <thomas.mcguire@kdab.com>
The minimum GTK+ version was bumped from from 2.10 to 2.18 (sep 2009)
in commit 2cce297b58.
Change-Id: I77a48c8a3b0955b00e399f714949d08293abbebd
Reviewed-by: Jens Bache-Wiig <jens.bache-wiig@digia.com>
CoreText segfaults when creating paths for color-glyphs:
0 0x00007fff8fd41e69 in TFont::FindColourBitmapForGlyph ()
1 0x00007fff8fd417ac in TFont::CreatePathForGlyph ()
2 0x000000010567d1af in QCoreTextFontEngine::addGlyphsToPath (...)
So we shortcut the code-path, since we don't support Emoji yet anyways.
Task-number: QTBUG-28615
Change-Id: Ife16ae4959077d9eaaf6ea5cd1f27a4e2e01e7f5
Reviewed-by: Eskil Abrahamsen Blomfeldt <eskil.abrahamsen-blomfeldt@digia.com>
A division by zero occurred since an empty model was used.
Task-number: QTBUG-28611
Change-Id: I1f9e869bda9f76a1c97840dff6ba48ab813bce9a
Reviewed-by: Janne Anttila <janne.anttila@digia.com>
Reviewed-by: Frederik Gladhorn <frederik.gladhorn@digia.com>
QDBusReply allows one to extract a QVariant and the type reply from an
error reply and getting a default-constructed value. This is useful when
a valid reply can never contain the default-constructed value (0, false,
empty strings, empty arrays, etc.), so it simplifies error checking.
More importantly, qdbusxml2cpp was changed a while ago from generating
QDBusReply to generating QDBusPendingReply, so we need to have the same
behavior.
Task-number: QTBUG-29046
Change-Id: Ia873b9fd4311c0d4e94f0ef623ba405c20bc0e8c
Reviewed-by: Olivier Goffart <ogoffart@woboq.com>
Currently /etc/openssl/certs is symlinked to
/var/certmgr/web/user_trusted, but this will be changed in the future.
/etc/openssl/certs is the folder to be used to read the root certs.
Change-Id: Ic037e5075ec7ee50c132fe08dc69abbe585e32e4
Reviewed-by: Sean Harmer <sean.harmer@kdab.com>
This test runs fine almost all of the time on systems with 1
processor, which were the norm when the test was written and are still
the way that the Qt Continuous Integration system works as of
today. But it falls flatly on multi-processor systems.
The root of the problem is that QSystemSemaphore recreates the
semaphore if it disappears underneath it. However, the recreation
process is not thread-safe at all: if two threads race to recreate it,
weird things might happen. strace on Linux shows that a thread got
stuck trying to acquire the semaphore:
<... nanosleep resumed> NULL) = 0
stat("/tmp/qipc_systemsem_market5c9f73af73334ffe350c60ec076e5744db0ecda3", {st_mode=S_IFREG|0640, st_size=0, ...}) = 0
stat("/tmp/qipc_systemsem_market5c9f73af73334ffe350c60ec076e5744db0ecda3", {st_mode=S_IFREG|0640, st_size=0, ...}) = 0
semget(0x51001388, 1, IPC_CREAT|IPC_EXCL|0600) = -1 EEXIST (File exists)
semget(0x51001388, 1, IPC_CREAT|0600) = 114786308
semop(114786308, {{0, -1, SEM_UNDO}}, 1 <unfinished ...>
This problem does not happen if the creation and destruction of the
QSharedMemory (which uses QSystemSemaphore) does not race with other
threads or processes attaching and detaching. For the threads test
it's easy. For the processes, we use stdin and stdout as a
communication channel.
Change-Id: Ie11b135431d4abfc59234654848b67f622eb03c9
Reviewed-by: Richard J. Moore <rich@kde.org>
On Mac, GCC and Clang print "(framework directory)" at the end of the
directory listing for includes if that's a framework dir. The directory
with "(framework directory)" at the end does not exist, so it doesn't
affect anything by being there.
But we don't have to keep the list longer just for that.
Change-Id: I3d031d3d15c75801ec0d6112b2c913bd63e5def3
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@digia.com>
These headers are technically private API, since they include private
headers. They should be _p.h actually, but that change I'll leave for
5.1.
Change-Id: I2dec222854e147da0fc166de311012472954012e
Reviewed-by: Mark Brand <mabrand@mabrand.nl>
Task-number: QTBUG-29056
QOpenGLContextGroup object is designed to be destroyed by deleteLater(),
but this method will not always work due to the fact that in many cases
event loop will exit before the deferred deletion of the
QOpenGLContextGroup object is queued. Think about the following case:
int main(int argc, char *argv[])
{
QApplication a(argc, argv);
QGLWidget w;
w.show();
return a.exec();
}
In the above program, the event loop will exit before QGLWidget object's
destruction. This will cause the QOpenGLContextGroup object hold by
QGLWidget object never been deleted.
This patch will delete QOpenGLContextGroup object directly with delete
operator if the current thread is the same as the thread which the
QOpenGLContextGroup lives in.
Change-Id: If835d7482474f4a668763fc7c21b293a27f075fd
Reviewed-by: Samuel Rødal <samuel.rodal@digia.com>
Reviewed-by: Sean Harmer <sean.harmer@kdab.com>
This convenience overload allows one to write
QRegularExpression re1, re2, ...;
QRegularExpressionMatch match;
QString subject;
if (subject.contains(re1, &match)) {
// ...
} else if (subject.contains(re2, &match)) {
// ...
} // ..
One can then inspect the results of a successful match in each block
(as well as extracting the captured substrings, etc.).
Change-Id: I0fb8be8b577656e8db994198f8105c26c4fe67b0
Reviewed-by: Lars Knoll <lars.knoll@digia.com>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
This allows to put them in containers, and to enable subsequent
features for QString.
Change-Id: I3b3fe695ffe6930331ed9f670738376722e0fc36
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Explicitely pass the number of capturing groups for which the
offsets should be reserved in the capturedOffsets vector,
instead of relying on it adding 1 (for the implicit capturing
group #0).
In case 0 is passed, don't allocate any space for that vector.
This is being used in case of NoMatch match type or failing
match (invalid regexp, out of bounds offset, etc.).
Change-Id: I0ec7646d5bd53e7a7973177100b163a5e5030307
Reviewed-by: Jędrzej Nowacki <jedrzej.nowacki@digia.com>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
