On a small display leftLay is left initialized to a null pointer,
therefore don't try to dereference it.
Task-number: QTBUG-43643
Change-Id: I9d22dac88a3a853ce154a6f64b35fc113abd9262
Reviewed-by: Marc Mutz <marc.mutz@kdab.com>
Yes, this is necessary. The noexcept operator looks for noexcept tagging,
not at the contents of the function to determine whether to return true.
The more conditionally-noexcept functions are used, the more important it
becomes that low-level classes are correctly marked noexcept. In that, it
is like constexpr.
Change-Id: Ia1aebf9b8d73fd8164c10dfca27a710934ba79a8
Reviewed-by: Olivier Goffart <ogoffart@woboq.com>
QCursor already dealt with d == nullptr, so this is trivial.
Change-Id: Ib2c6a3f0dc8b93035ffe3059cdce925463706e10
Reviewed-by: Olivier Goffart <ogoffart@woboq.com>
Under DBus it is possible to have methods and signals with the same name
or have methods which are reserved c++ keywords.
For example the logind session interface has a signal and method both
called Lock.
This patch allows generated methods to use a different method name
specified in the annotation that the original DBus name in the DBus
interface in a similar manner to how one can rename accessors.
[ChangeLog][QtDBus] Add annotation org.qtproject.QtDBus.MethodName to allow
autogenerating C++ methods with different names to the original DBus method
Change-Id: I08bbe77554fbdd348e93f82d45bab0d75d360c27
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Linux gracefully allows us to do that and treat the v6 socket as if it
were v4. Other OS (notably OS X) aren't so forgiving.
Change-Id: I13dd3274be2a4b13e8b1eef93cbc2dd17b648f96
Reviewed-by: Richard J. Moore <rich@kde.org>
This makes it possible for the flags that RegisterTouchWindow takes to be
specified after the window has been created.
Task-number: QTBUG-41433
Change-Id: I166143875ef54ab6a249cffb31d017845a694a01
Reviewed-by: Jørgen Lind <jorgen.lind@theqtcompany.com>
We already let enter/return key presses to reach text edits instead
of closing the editor. Do the same for tab/backtabs.
[ChangeLog][QtWidgets][Important behavior changes] QItemDelegate will
now not close a QTextEdit/QPlainTextEdit editor when the tab key
is pressed; instead, the key will reach the editor.
Task-number: QTBUG-3305
Change-Id: Ife9e6fdc5678535c596d1068770b0963134d8d5a
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@theqtcompany.com>
Reviewed-by: Thorbjørn Lindeijer <bjorn@lindeijer.nl>
Reviewed-by: Thorbjørn Lund Martsum <tmartsum@gmail.com>
Instead of leaking the QLockFile instance, which causes
leak-checkers to emit false positives, simply call exit(),
which doesn't run the destructors, yet doesn't lead to
leak-checker warnings.
Change-Id: Ia61010671e5218ae412e2bcf873e66255a2c5a99
Reviewed-by: David Faure <david.faure@kdab.com>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
toString() is both a function template and a set of overloaded
functions (for const char* and const void*). So don't explicitly
specify the function template arguments (they're deduced from the
arguments anyway). This enables overloading toString() as well as
specializing the template.
[ChangeLog][QtTest][Important Behavior Changes] toString() can now be
overloaded (instead of just specialized) for custom types, and is now
reliably found through argument-dependent lookup (ADL).
Change-Id: Ic4a622d236ad7f0e4142835353f1b98bf2dc6d4c
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
QtMetaTypePrivate::QAssociativeIterableImpl::{find,begin,end}() allocate a new _iterator,
so when they're used outside of the ref-counted world of QAssociativeIterable::const_iterator,
their lifetime needs to be manually managed.
Instead of going to that length, which failed in previous iterations of this patch, implement
value() in terms of (new) find() and let find() operate on const_iterator.
Because of forwards compatibility between patch releases, use (unexported) friend functions
for now with the intention to make them proper member functions come Qt 5.5.
Task-number: QTBUG-41469
Change-Id: I43b21eae0c2fc4c182369e669a8b3b457be68885
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Change-Id: Ia624fcefe77a456424bcaa00b106ef7f69abb6b9
Reviewed-by: Simon Hausmann <simon.hausmann@digia.com>
Reviewed-by: Jason McDonald <macadder1@gmail.com>
Add a new mkspec variable, QMAKE_LIBS_EXECINFO, for platforms where
backtrace(3), backtrace_symbols(3) and others are not in libc, but
rather in a separate library -- on the BSDs, this is libexecinfo.
Use it in corelib/global/global.pri so that libqt5core links against it
and has the proper dependency when necessary.
Change-Id: I62ac36c9b3ba7ab0719420cb795087d43ec138a4
Reviewed-by: Olivier Goffart <ogoffart@woboq.com>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
The "int" was a left over when this was documentation for QAtomicInt.
Change-Id: If7b7688982d27cbbd42f080eff7d08344b587f44
Reviewed-by: Richard J. Moore <rich@kde.org>
So we can get the output when running with QDBUS_DEBUG=1.
Change-Id: I6a6b8e0d82727c522914fb90a7ce594c86307d8f
Reviewed-by: Frederik Gladhorn <frederik.gladhorn@theqtcompany.com>
Instead of killing them outright (and note that terminate() doesn't work
on Windows), ask them nicely to exit on their own. This way, if we run
them in valgrind, valgrind gets a chance to print the leak check output
and summary.
Change-Id: Ib6cc8d4560ff0bf255f94980eb220e97592c00f0
Reviewed-by: Frederik Gladhorn <frederik.gladhorn@theqtcompany.com>
GCC said:
qtldurl.cpp:51:50: error: loop exit may only be reached after undefined behavior [-Werror=aggressive-loop-optimizations]
qtldurl.cpp:51:48: note: possible undefined statement is here
while (tldIndices[index] >= tldChunks[chunk] && chunk < tldChunkCount) {
^
That's because we check whether chunk is still valid (less than
tldChunkCount) after we've dereferenced tldChunks[chunk]. That is, we've
already read tldChunk[2].
Change-Id: I79b6a1ea9a2454813d6cce7596fc2bb6d972d097
Reviewed-by: David Faure <david.faure@kdab.com>
Platformsupport doesn't need network in order to compile and link.
Change-Id: Ifda2db9c39d89619ff21cbcd21dc5bf97fe58f55
Reviewed-by: Jørgen Lind <jorgen.lind@theqtcompany.com>
This update explains how to use the optional 'square bracket' argument
for the link command (\l), the new single execution mode for qdoc, and
how to generate and use the QA pages.
Change-Id: I7232e05677c6ea97636cdb4fb20bd9708c83f311
Task-number: QTBUG-35495
Reviewed-by: Topi Reiniö <topi.reinio@digia.com>
Commit 99357e32a0 introduces a change that
causes <cstddef> to be included instead of <stddef.h> under QNX. That causes
symbols such as size_t to be placed on the std namespace only - QNX does not
put those in the global namespace, since it is not really required by the
standard, and therefore the build fails.
Merry Xmas!
Change-Id: I70c6976203a9d7beadd0076e122e2ac633a4ba69
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
The ROUND function for PostgreSQL only accept NUMERIC field as argument
Change-Id: I0c3753bfe4167cd47158e21b407cca8771816104
Reviewed-by: Mark Brand <mabrand@mabrand.nl>
Added support on QDateTime::fromString to read correctly dates on ISO
format with Time zone designators at format [+-]HH
Change-Id: Ied5c3b7950aee3d0879af0e05398081395c18df5
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Reviewed-by: Mark Brand <mabrand@mabrand.nl>
SQL Server 2012 Native Client (version 11.0.2100.60) or later introduced
a change in the behavior of the SQLGetData method when converted string values
are involved. In older version a (sometimes wrong) size was returned. Now always
SQL_NO_TOTAL is returned which signals to read as much data as available.
SQL_NO_TOTAL was handled like SQL_NULL_DATA in the code before which indicates a
NULL value so the returned string was empty.
See link for more info: http://msdn.microsoft.com/en-us/library/jj219209.aspx
Change-Id: Ia0d2296caf593890b301ee1848d1bf3eb8d7b6fe
Reviewed-by: Mark Brand <mabrand@mabrand.nl>
This patch adds correct timezone support in PSQL plugin. Prior to this
patch, no timezone support was provided, so only the following case
worked :
* using local time in both client application and postgresql server
* datetime were using second precision
This patch tries to take care that postgresql has two different
datatypes for date time, respectively :
* timestamp with time zone
* timestamp without time zone
Both are internally stored as UTC values, but are not parsed the same.
* timestamp with time zone assumes that there is a time zone
information and will parse date time accordingly, and then, convert
into UTC before storing them
* timestamp without time zone assumes that there is no time zone
information and will silently ignore any, unless the datetime is
explicitly specified as having a time zone, in case it will convert
it into UTC before storing it
Both are retrieved as local time values, with the following difference
* timestamp with time zone includes the timezone information
(2014-02-12 10:20:12+0100 for example)
* timestamp without time zone does not include it
The patch does the following :
* parse the date retrieved by postgresql server using QDateTime
functions, which work correctly
* always convert the date to UTC before giving it to postgresql
* force time zone so that timezone information is taken into account
by postgresql
* also adds the milliseconds when storing QDateTime values
The following configurations are tested to work :
* client and server using same timezone, timestamp with or without tz
* client and server using different timezone, timestamp with tz
The following configuration will *not* work :
* client and server using different timezones, timestamp without tz
Because data will be converted to local time by the postgresql server,
so when returned it will be different from what had been serialized.
Prior to this patch, it gave the illusion to work because since TZ
information was lost, time was stored as local time from postgresql.
Lots of inconsistencies occurred, though, in case client tz changes...
I don't expect this to be an issue since having different TZ in server
and client and *not* handling this is a broken setup anyway.
Almost based on changes proposed by julien.blanc@nmc-company.fr
[ChangeLog][QtSql] Added timezone support for datetime fields in PSQL
Task-number: QTBUG-36211
Change-Id: I5650a5ef60cb3f14f0ab619825612831c7e90c12
Reviewed-by: Mark Brand <mabrand@mabrand.nl>
The toolchain is basically the same as on Linux, so we can get the
program interpreter and print the build information when Qt5Core.so
is run.
Change-Id: I02a910e691622e24e882015716c5f74dd5a20c4a
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Extend the support for pre-MIPS32 architectures (done mostly in
60b6b28c21) also to orderedMemoryFence.
Change-Id: I50b9091c16166b8434a07988053c1f901d528237
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
This covers the only real additions over QVector: push and pop. Really, there
isn't too much specific to benchmark here, but we're interested in one specific
case: that of pushing and popping a single item repeatedly.
With the current QVector behavior, this causes constant deallocation, which
makes it morbidly slow. This behavior will be reviewed in a subsequent commit.
Results (not that anyone really cares) for me:
PASS : tst_QStack::qstack_push()
RESULT : tst_QStack::qstack_push():
1.9 msecs per iteration (total: 61, iterations: 32)
PASS : tst_QStack::qstack_pop()
RESULT : tst_QStack::qstack_pop():
8.2 msecs per iteration (total: 66, iterations: 8)
PASS : tst_QStack::qstack_pushpopone()
RESULT : tst_QStack::qstack_pushpopone():
80 msecs per iteration (total: 80, iterations: 1)
Change-Id: I3530888abbfcfcef39318d6be6d5b07306a4704e
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Before:
Could not find output file: No such file or directory
After:
Could not find output file 'arch.exe' or 'arch' in C:/Qt/qt5/qtbase/config.tests/arch : No such file or directory
(it turned out that linking in that directory failed because of a wrong %PATH%)
Change-Id: I948d7f10f7e82f77a08ac9a8db76d97536c42dd0
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
The interface macro needs to be in the Qt namespace, otherwise the
template specialization in Q_DECLARE_INTERFACE fails.
Task-number: QTBUG-41959
Change-Id: I818da5ba8460219d73c0d248ee410ac92f7498f4
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@theqtcompany.com>
Reviewed-by: Tony Sarajärvi <tony.sarajarvi@digia.com>
Reviewed-by: Stephen Kelly <steveire@gmail.com>
- remove the X11R6 paths, since they were gone for years, and their
lack may cause build issues with CMake config files
- add empty QMAKE_*_XCB variables, as done in the common linux.conf
- add to QMAKE_PLATFORM, not just reset it (just done in other mkspecs)
- borrow QMAKE_LFLAGS_GCSECTIONS from the common linux.conf
Change-Id: I94e05032f8195bbda73dffe1da02eec7ac679045
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
When running the compositor on xcb the compositor backend will call
nativeResourceForWindow with EglDisplay as resource. When this is done
the window->handle() will be null and result in a crash. By returning
just the eglDisplay() when the window->handle() is null solves the
problem and also makes the compositor work.
Change-Id: I72f4341402facc8c44a41151c4f76d6447bd8070
Reviewed-by: Jørgen Lind <jorgen.lind@theqtcompany.com>
...and get rid of workarounds for older versions,
which also makes the code more readable.
Change-Id: I087110c5f60cd664dad241776e1a0df901989c75
Reviewed-by: Lars Knoll <lars.knoll@digia.com>
Replace C-style cast with static_cast for casts from void*.
Change-Id: I4d6c602037e70afe2746a179a31fea4c5043c3d5
Reviewed-by: Lars Knoll <lars.knoll@digia.com>
There is no point in modifying font tables, so mark them RO for clarity.
Change-Id: I8ddbda9ace9c5a0eeabc0c5e5f64558060b5ee69
Reviewed-by: Lars Knoll <lars.knoll@digia.com>
This fixes the tst_QSqlQuery::isNull test case when run against a
MySQL database driver.
Change-Id: I8248ba956472bae97a64247594055e6f02840557
Reviewed-by: Mark Brand <mabrand@mabrand.nl>
We want to use "localhost" if the server's address is "any", as some OS
can't send datagrams to "any" (e.g., OS X and FreeBSD).
Change-Id: I1004bc2282e7f930cdb7ed394aa9f4b5a1cfcf82
Reviewed-by: Richard J. Moore <rich@kde.org>
On my Mac Mini, port 5000 is in use, which means the broadcasting test
fails.
Change-Id: Ifb0883263e277f388342430349ea7315d42f324a
Reviewed-by: Richard J. Moore <rich@kde.org>
It was unconditional. Someone forgot to check for IPv6 support before
skipping IPv6 tests.
Change-Id: I7b11528ad02560f0db9defde3c64f76f48a6c1f8
Reviewed-by: Richard J. Moore <rich@kde.org>
With toString(), it was printing "0.0.0.0", which is the same as
QHostAddress::AnyIPv4, making it difficult to tell the two apart.
Change-Id: I4668ec3337c25ddfdc2fa3bbacc83b9d34316b1f
Reviewed-by: Richard J. Moore <rich@kde.org>
Fix indention, add empty line as separator, undef a local macro.
Change-Id: Ie0d8529322f87ca2bdedb9a7f2f4cf9f463a6e12
Reviewed-by: Olivier Goffart <ogoffart@woboq.com>
Split off default-value check from getSetCheck().
This is in preparation of extracting getSetCheck_data().
Change-Id: I869c34014a92e29a07dc08584a93a4cbba372518
Reviewed-by: Olivier Goffart <ogoffart@woboq.com>
There's no reason not to have it, and it allows better debug output.
[ChangeLog][QtWidgets][QSizePolicy] QSizePolicy::ControlTypes is now
introspectable through QSizePolicy's meta object.
Change-Id: Ia274c708f31e735e4605bd4769aa65a9b78d690f
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Was including qtest_gui.h instead, for QT_WIDGETS_LIB.
Not a problem in practice, since qtest_widgets.h doesn't
currently add anything on top of qtest_gui.h. But that will
change soon.
Change-Id: I942870cddc7b1ccfda262e2fe26834594336dd0b
Reviewed-by: Olivier Goffart <ogoffart@woboq.com>
