João Abecasis [Wed, 22 Feb 2012 00:25:28 +0000 (01:25 +0100)]
Workaround compiler issue
I can't figure this one out, but it seems to be a clang compiler bug
that is triggered in association with -DQT_NO_DEBUG. Changing the
test from QVERIFY to QCOMPARE keeps the intent and the check, but makes
the failure go away.
It can't hurt...
Change-Id: Ib34e5e850e5b731d729e417430dec55e372805ac
Reviewed-by: Chris Adams <christopher.adams@nokia.com>
João Abecasis [Tue, 21 Feb 2012 15:32:35 +0000 (16:32 +0100)]
Fix warnings on shadowing data members with arguments
Example emitted by GCC 4.2.1:
warning: declaration of ‘iter’ shadows a member of 'this'
Change-Id: I288da01c511a1404bf41881a6c96a5f3cd00d0a7
Reviewed-by: Jędrzej Nowacki <jedrzej.nowacki@nokia.com>
João Abecasis [Tue, 21 Feb 2012 13:58:57 +0000 (14:58 +0100)]
Merge remote-tracking branch 'gerrit/master' into containers
Conflicts:
src/corelib/tools/qbytearray.h
Change-Id: I03b1f3e05c9b7a45130887c522fcd9b7aa387129
Friedemann Kleint [Tue, 21 Feb 2012 07:38:37 +0000 (08:38 +0100)]
Fix qlogging test.
Build app sub-process first.
Change-Id: I87a11f7fd5d8a82584e496722f79e236191b0fb3
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@nokia.com>
Jason McDonald [Tue, 21 Feb 2012 02:57:27 +0000 (12:57 +1000)]
Improve tst_QPrinterInfo::testForDefaultPrinter().
Be explicit about skipping the test when no default printer is present.
Change-Id: If69b275eb0f490411471ec42798d8aefcc57fd83
Reviewed-by: Rohan McGovern <rohan.mcgovern@nokia.com>
Jason McDonald [Tue, 21 Feb 2012 02:52:15 +0000 (12:52 +1000)]
Allow printer names to contain hyphens in QPrinterInfo test.
Change-Id: I473627413d2f1cd5637704f2948eb07a264be49e
Reviewed-by: Rohan McGovern <rohan.mcgovern@nokia.com>
Jason McDonald [Tue, 21 Feb 2012 02:00:14 +0000 (12:00 +1000)]
Fix warning about unregistered metatype in QWindow autotest.
Change-Id: If9d8d7e1cd52815ef7231294e4890dfafcd28ec8
Reviewed-by: Rohan McGovern <rohan.mcgovern@nokia.com>
Toby Tomkins [Tue, 21 Feb 2012 01:03:00 +0000 (11:03 +1000)]
Flag two tests (Win+A, Simon+G) in parseString as expect fail on OSX.
Remove previously defined insignificant test flag as the number of
tests failing has been reduced.
Task-number: QTBUG-24406
Task-number: QTBUG-23058
Change-Id: I01b41f30469cf7a440e21195e105cb30a8db76e2
Reviewed-by: Rohan McGovern <rohan.mcgovern@nokia.com>
Jason McDonald [Mon, 20 Feb 2012 06:01:43 +0000 (16:01 +1000)]
Make the QApplication autotest work for shadow builds.
Use QFINDTESTDATA to locate the helper applications.
Change-Id: I604d10e37c9367f2e95225864edf5bf705f1d961
Reviewed-by: Rohan McGovern <rohan.mcgovern@nokia.com>
Rohan McGovern [Mon, 20 Feb 2012 05:12:50 +0000 (15:12 +1000)]
tst_qapplication: mark expected failure on mac
The tab key behavior in Qt5 no longer respects the "Text boxes and lists
only" tab navigation option in OSX, which is the default. This is a
regression since Qt4.
Task-number: QTBUG-24372
Change-Id: I54c1663f8fb259dd847083432102a0bfad7dd69c
Reviewed-by: Toby Tomkins <toby.tomkins@nokia.com>
Kurt Korbatits [Mon, 20 Feb 2012 01:36:24 +0000 (11:36 +1000)]
Changed qnetworksession unittest to work from installation dir
- Made test depend on subprogram
- added install of subprogram
Change-Id: Ib263e9e75ed3c900b52fb1c9b6d319e71d19bdbb
Reviewed-by: Rohan McGovern <rohan.mcgovern@nokia.com>
João Abecasis [Fri, 17 Feb 2012 13:11:28 +0000 (14:11 +0100)]
Change meaning of offset in QByteArrayData
It used to be an index into the first element in 'd' that came after
'offset'. It is now the byte offset from the beginning of the
QByteArrayData structure.
By no longer using an actual array to access characters, we also steer
clear of GCC bug #43247:
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=43247
This aligns this data structure with QArrayData. The intention is to
have QVector, QString and QByteArray share the same memory layout and
possibly code.
Change-Id: I8546e5f51cd2161ba09bd4ada174b7f5e6f09db7
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Rohan McGovern [Tue, 14 Feb 2012 02:55:49 +0000 (12:55 +1000)]
Stabilize tst_qdbusabstractinterface and make it XFAIL
This test was written incorrectly in a way which happened to allow it to
pass most of the time (but not all the time).
Reset the state of test objects between each test function, and mark the
broken functions with QEXPECT_FAIL and a link to a task.
Replace the unusual WaitForPinger construct with the usual
QTRY_VERIFY/QTRY_COMPARE method of verifying asynchronous operations.
Task-number: QTBUG-24262
Change-Id: I82d09002307c0b500bf60cd5b583674321b37609
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Richard Moore [Sun, 19 Feb 2012 15:43:57 +0000 (15:43 +0000)]
Remove many unneeded virtuals.
Cleanup a ### Qt 5 TODO.
Change-Id: I384e868ecc3ca4d80e4e71bb54d402f4ec1337a8
Reviewed-by: David Faure <faure@kde.org>
Richard Moore [Sun, 19 Feb 2012 15:31:44 +0000 (15:31 +0000)]
convert length functions in QDom from uint to int.
Fixes a ### Qt 5
Change-Id: I7a385f2b704d38ae626094f27b06a918d4a9bc48
Reviewed-by: David Faure <faure@kde.org>
Richard Moore [Sun, 19 Feb 2012 13:41:31 +0000 (13:41 +0000)]
Merge overloads.
Fix a ### Qt 5 by merging the overloads.
Change-Id: If33e7592191c81b32caa6d68c73dbf2282437886
Reviewed-by: David Faure <faure@kde.org>
Andreas Holzammer [Mon, 20 Feb 2012 17:38:52 +0000 (18:38 +0100)]
Include qt_windows.h for Windows CE.
The include is needed for OutputDebugString,
as it is defined in winbase.h. To include
use the centralized header.
Change-Id: I486da20d7b054cee352be085e65a0ede1394653d
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@nokia.com>
Kent Hansen [Mon, 20 Feb 2012 13:26:00 +0000 (14:26 +0100)]
Silence warnings/debug output from qobject tests
They create noise in the test results.
Change-Id: I40e7239ba7cd41bec577fe8220c86476553a6502
Reviewed-by: Olivier Goffart <ogoffart@woboq.com>
Reviewed-by: João Abecasis <joao.abecasis@nokia.com>
Reviewed-by: Bradley T. Hughes <bradley.hughes@nokia.com>
Xizhi Zhu [Sun, 19 Feb 2012 09:15:18 +0000 (10:15 +0100)]
Merges several overloaded functions in QCoreApplication.
The source compatibility is kept.
Change-Id: If66053b271d65062b3c0ce6ec66c8394a37b4e3e
Reviewed-by: Lars Knoll <lars.knoll@nokia.com>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Oswald Buddenhagen [Fri, 3 Feb 2012 21:15:07 +0000 (22:15 +0100)]
don't attempt to build host tools which are gone from qtbase
Change-Id: If9a3fb9792fc5d90496ff834b2c3adc177e1311e
Reviewed-by: Bradley T. Hughes <bradley.hughes@nokia.com>
Reviewed-by: Marius Storm-Olsen <marius.storm-olsen@nokia.com>
Shane Kearns [Fri, 17 Feb 2012 20:33:57 +0000 (20:33 +0000)]
Windows - fix connecting to a socket using IPv4 mapped IPv6
Connecting to an IPv4 mapped IPv6 address (e.g. ::FFFF:127.0.0.1)
requires the IPV6_V6ONLY socket option to be cleared.
This was causing tst_qtcpserver::ipv6ServerMapped autotest to fail.
The same change is not required on MacOS X - the test passes there.
Task-number: QTBUG-24351
Change-Id: I6c08b19f0daa12765da2d44792ffb17299322695
Reviewed-by: Markus Goetz <markus@woboq.com>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
João Abecasis [Fri, 17 Feb 2012 10:23:41 +0000 (11:23 +0100)]
Have QVectorData::grow, take size of header w/ padding
This includes padding necessary to align the data array, but excludes the first
element as was done before. Size of header is the interesting piece of
information, anyway.
This simplifies calculations in a couple of places, harmonizes code with the
QRawVector fork and paves the way for further changes in QVector, namely the
memory layout.
When Q_ALIGNOF is not available, default to pointer-size alignment. This
should be honoured by malloc and won't trigger use of more expensive
aligned allocation.
Change-Id: I504022ac7595f69089cafd96e47a91b874d5771e
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Friedemann Kleint [Mon, 20 Feb 2012 14:36:57 +0000 (15:36 +0100)]
Remove QPlatformIntegration/QPlatformTheme from QGuiAppPrivate.
Forward-declare instead.
Change-Id: I3851994e8bc05b389e94e948478339ba33d521c1
Reviewed-by: Samuel Rødal <samuel.rodal@nokia.com>
Friedemann Kleint [Mon, 20 Feb 2012 14:44:21 +0000 (15:44 +0100)]
Fix warning about converting false to pointer value.
Change-Id: I70c324e6fcfd2bba3ab44837c5ce2c007de8896f
Reviewed-by: Bradley T. Hughes <bradley.hughes@nokia.com>
Reviewed-by: Stephen Kelly <stephen.kelly@kdab.com>
Jørgen Lind [Tue, 14 Feb 2012 15:38:32 +0000 (16:38 +0100)]
Make sure that we pick up xcb from pkg-config
and run the configure checks with the values
Change-Id: Ie8e0072c686c6a7dce1d02e25a9c1abce4679d34
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@nokia.com>
Morten Johan Sorvig [Mon, 20 Feb 2012 11:05:17 +0000 (12:05 +0100)]
Remove Native Client special casing from configure.
Change-Id: I20f044b267b6c18e5a3050de22710c45e7573cc3
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@nokia.com>
Jan-Arve Saether [Mon, 20 Feb 2012 13:42:41 +0000 (14:42 +0100)]
Remove QAccessibleInterface::relationTo().
QAccessibleInterface::relations() replaces both relatedTo() and
some of the RelationFlags that navigate() handled before.
Change-Id: I4b9d2c28ba3d753efe7cc9bfa0ad68bed3ee02fd
Reviewed-by: Frederik Gladhorn <frederik.gladhorn@nokia.com>
Jan-Arve Saether [Mon, 20 Feb 2012 10:19:10 +0000 (11:19 +0100)]
Updated for accessibility
Change-Id: I13b5c2f293e9c1c238c7b7c6aae9d26f5c150f24
Reviewed-by: Frederik Gladhorn <frederik.gladhorn@nokia.com>
Kent Hansen [Sat, 18 Feb 2012 20:33:28 +0000 (21:33 +0100)]
Port QDBusAdaptorConnector meta-object to revision 6
We want to drop support for old revisions in Qt 5.
This commit brings the QDBusAdaptorConnector meta-object in sync with
current moc output.
The QtDBus implementation was assuming that the relaySlot() slot
would be created at index methodOffset() in the meta-object, but
since revision 4 that's no longer the case (signals always come
first). Made the code more robust by actually querying the
meta-object what the index is.
Change-Id: Ie0791680cc2e9e5fb1472c4462c391f92ea22ea6
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Kent Hansen [Fri, 17 Feb 2012 22:00:41 +0000 (23:00 +0100)]
Remove unused QtDBus meta-object extensions
The inputSignature and outputSignature fields were not being used
anymore.
All tests still pass.
Change-Id: Icbc8fdcd5179a2b1b4843d58b90af925f6bef133
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Marc Mutz [Thu, 16 Feb 2012 06:42:11 +0000 (07:42 +0100)]
QPoint/QPointF: inline manhattanLength(); QLineF: inline isNull()
There's no reason for them not to be, and it's a prerequisite
for making these functions constexpr.
Change-Id: I03c9965147b51014c7af60a4c2d7f25a3f6e21a7
Reviewed-by: David Faure <faure@kde.org>
Reviewed-by: Gunnar Sletta <gunnar.sletta@nokia.com>
Bradley T. Hughes [Sat, 4 Feb 2012 13:32:16 +0000 (14:32 +0100)]
Move QTypeInfo out of qglobal.h and into a separate header
qtypeinfo.h is still included from qglobal.h, and defines QTypeInfo
as well as all the specializations for built-in and default Qt types.
Change-Id: I24116174de288a730cbb7aa2f80d346d2b2f9408
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Reviewed-by: João Abecasis <joao.abecasis@nokia.com>
Samuel Rødal [Fri, 17 Feb 2012 15:58:09 +0000 (16:58 +0100)]
Added note in QPlatformWindow::setGeometry() docs about position policy.
Change-Id: I0ac76b7aac1aa717592c34f414b1dd8fbee92be2
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@nokia.com>
Marc Mutz [Sun, 11 Dec 2011 15:55:01 +0000 (16:55 +0100)]
uic: use QStringLiteral() instead of QString::fromUtf8() where applicable
Many (most?) strings written aren't in fact UTF-8, and we can check at
compile-time which are and which aren't, so don't hard-code fromUtf8()
but use the much more efficient QStringLiteral() where applicable.
This is low-hanging fruit. This patch only optimises US-ASCII string
literals, not those that are latin-1 or even UTF-8, because that would
require more extensive changes to the original fixString() function.
Likewise, there are also other calls to QString::fromUtf8() being
generated (e.g. in the pixmap code) that could benefit from being
turned into QStringLiterals, but their code paths are more involved
than those this patch fixes.
This patch at least suffices in turning all the setObjectName()
arguments into QStringLiterals, which was the main goal.
The autotest baseline has been updated with the new expected results.
Change-Id: Ic1ef67f500f9ff92d36164d515f4e004ef2a10bc
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@nokia.com>
Harald Fernengel [Mon, 20 Feb 2012 08:25:24 +0000 (09:25 +0100)]
Fix LSB mkspec
Use the standard g++ specs, just override the compiler name
Change-Id: I4adacd7204d3880155bdfc3405e20f322ad593ea
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@nokia.com>
Joerg Bornemann [Mon, 20 Feb 2012 11:31:37 +0000 (12:31 +0100)]
fix QProcess for Windows XP
CreateNamedPipe supports the flag PIPE_REJECT_REMOTE_CLIENTS since
Windows Vista. On earlier Windows versions the system call would fail
with ERROR_INVALID_PARAMETER.
This does not open a security hole on Windows XP as there can be only
one pipe instance.
Change-Id: I5a1c7fdf756678009857317c7b563c884afeef2c
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@nokia.com>
Matthew Vogt [Sun, 19 Feb 2012 23:40:12 +0000 (09:40 +1000)]
Ensure the directory exists to receive a deprecated header.
When creating a deprecated header, create the directory it will be
located in, if it does not yet exist.
Change-Id: Id0ac7327e7dfe5b3e34b431c7bc3844b72251a3d
Reviewed-by: Rohan McGovern <rohan.mcgovern@nokia.com>
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@nokia.com>
Jan-Arve Saether [Sat, 18 Feb 2012 22:29:24 +0000 (23:29 +0100)]
Remove QAccessibleInterface::navigate()
This is replaced by parent(), child() and relations()
Change-Id: Iabff6ec56176a1ca8465d6480860f6e0174fd134
Reviewed-by: Frederik Gladhorn <frederik.gladhorn@nokia.com>
Harald Fernengel [Mon, 20 Feb 2012 08:37:07 +0000 (09:37 +0100)]
Use standard unix mutexes on LSB
LSB doesn't allow syscalls, so fall back to the normal _unix
implementation
Change-Id: I8aba6147da8b46e3f85b0454cf9aca219811c9fe
Reviewed-by: Olivier Goffart <ogoffart@woboq.com>
Kent Hansen [Sun, 19 Feb 2012 18:34:50 +0000 (19:34 +0100)]
Move QMetaMethod::parameterTypes() implementation to helper function
Because of an upcoming change to the meta-object data format, the
ability to extract parameter types from a signature will be needed by
meta-object builders (such as QMetaObjectBuilder) soon.
Change-Id: I1f21b2be41761a5db2f1a05976fad29eb3aebb03
Reviewed-by: Olivier Goffart <ogoffart@woboq.com>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Kent Hansen [Sat, 18 Feb 2012 20:15:45 +0000 (21:15 +0100)]
Port QSignalEventGenerator meta-object to revision 6
We want to drop support for old revisions in Qt 5.
This commit brings the QSignalEventGenerator meta-object in sync with
current moc output.
Change-Id: Id5f4954b08cf49169850bd77543b2a7ab8274c18
Reviewed-by: Olivier Goffart <ogoffart@woboq.com>
Kent Hansen [Fri, 17 Feb 2012 21:26:27 +0000 (22:26 +0100)]
Use QMetaMethod::isValid() to check method validity
The signature() method will soon be replaced by a function that
returns a QByteArray rather than a pointer, and calling it will be
more expensive than calling isValid(). Plus, isValid() looks nicer.
Change-Id: I6844988c84dcadf876fc86bc71c8b310a21f88de
Reviewed-by: Olivier Goffart <ogoffart@woboq.com>
Reviewed-by: Sergio Ahumada <sergio.ahumada@nokia.com>
Kent Hansen [Fri, 17 Feb 2012 21:12:05 +0000 (22:12 +0100)]
Add QMetaMethod::isValid() function
This function provides a proper way of determining whether a function
returned by QMetaObject::method() is valid. (Checking whether
signature() returns a 0 pointer, which e.g. testlib does, is not an
ideal API -- especially given that signature() will soon be removed
and replaced by a function that returns a QByteArray.)
Change-Id: I644f476b09904925f2042945f5d0ad744482b682
Reviewed-by: Olivier Goffart <ogoffart@woboq.com>
Reviewed-by: Lars Knoll <lars.knoll@nokia.com>
Kent Hansen [Sun, 19 Feb 2012 18:02:16 +0000 (19:02 +0100)]
Fix typo in QByteArray(QConstByteArrayData) constructor
There is no "str" member in QConstByteArrayData, it should be "ba".
Change-Id: Ife76460e9332733480c1ceded21e78388656092d
Reviewed-by: João Abecasis <joao.abecasis@nokia.com>
Kent Hansen [Fri, 17 Feb 2012 20:48:52 +0000 (21:48 +0100)]
Don't call metaObject() several times
It's sufficient to call it once per invokeMethod.
Change-Id: I1db826027eca87c799f216d65e27c801a23e64e2
Reviewed-by: Bradley T. Hughes <bradley.hughes@nokia.com>
Reviewed-by: Olivier Goffart <ogoffart@woboq.com>
Kent Hansen [Sun, 19 Feb 2012 17:52:19 +0000 (18:52 +0100)]
Small moc cleanup: Pass lists by const reference
generateFunctions() shouldn't (and doesn't) modify the list of
functions, so the list should be passed by const reference.
Change-Id: If5ff810e5623e734816c3089eb8b3fd4f54f2345
Reviewed-by: Bradley T. Hughes <bradley.hughes@nokia.com>
Reviewed-by: Olivier Goffart <ogoffart@woboq.com>
Kent Hansen [Sat, 18 Feb 2012 17:51:46 +0000 (18:51 +0100)]
moc: Get rid of implicit conversions to const char *
moc mostly operates on QByteArrays. When an actual const char * is
needed, it should be explicit.
Change-Id: I0b3e262830128306688f4512a4b59ce8966c2579
Reviewed-by: Bradley T. Hughes <bradley.hughes@nokia.com>
Reviewed-by: Olivier Goffart <ogoffart@woboq.com>
Kent Hansen [Mon, 20 Feb 2012 05:48:17 +0000 (06:48 +0100)]
Minuscule moc cleanup: Don't declare i in function scope
This is not C.
Furthermore, the function-level i was already shadowed in several
places by "for (int i = 0; ..." blocks later in the function.
Change-Id: Ic1777f78d9838bc2921d68337be05de454bbeabe
Reviewed-by: Olivier Goffart <ogoffart@woboq.com>
Morten Johan Sorvig [Fri, 17 Feb 2012 13:21:37 +0000 (14:21 +0100)]
Don't force CFG_ENDIAN to "auto" on Mac.
We no longer support universal ppc/x86 builds.
Change-Id: I8c4a1d087d02da1ad80d91a7a04147b37e81d74f
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@nokia.com>
Morten Johan Sorvig [Fri, 17 Feb 2012 13:34:25 +0000 (14:34 +0100)]
Don't use -rpath-link when building on Mac.
The logic here was wrong: test if the
target linker supports -rpath-link, and then set
it for both host and target via mkspecs/qmodule.pri.
Change-Id: Ie4da7ed2e06e784f9edb65a27290913ab838a8c2
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@nokia.com>
Jonathan Liu [Sun, 19 Feb 2012 11:33:35 +0000 (22:33 +1100)]
Fix compilation with MinGW-w64
This fixes a "cast from 'void*' to 'int' loses precision" error and adds
a missing argument to qWarning.
Change-Id: I38544c4e317a3f6d1ecf26ebb3e35c66b1878d24
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@nokia.com>
Bradley T. Hughes [Fri, 17 Feb 2012 13:17:54 +0000 (14:17 +0100)]
Rename Q_PROCESSOR_POWERPC to Q_PROCESSOR_POWER
IBM's POWER and the PowerPC architecture have been merged into a single
ISA, the Power ISA (see http://www.power.org). Use this unified name in
Qt.
Change-Id: Ia41492b0031d890843e43c5f7ecd1e60c65bb75b
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Holger Hans Peter Freyther [Sun, 19 Feb 2012 18:20:49 +0000 (19:20 +0100)]
qwidget_p.h: Remove DontSetCompositionMode from the internal flags
This flag was used by QWS to set or not set the composition mode
when filling the background. E.g. this was used by 'transparent'
windows to set the alpha bits. QPA didn't have this enabled in 4.8,
the code paths are gone in Qt5 and it is the responsibility of the
QPlatformBackingStore to initialize the alpha bits properly.
Change-Id: I4011d842b718d6a40b19922b5f75c597e63079c8
Reviewed-by: Samuel Rødal <samuel.rodal@nokia.com>
Holger Hans Peter Freyther [Wed, 1 Feb 2012 12:32:53 +0000 (13:32 +0100)]
qpa: Verify that the platform returns a WId that is not WId(0)
This is motivated by visiting a customer that re-implemented the
::winId method and returned WId(0) that resulted in a crash. Currently
there is only a comment inside the implementation of the ::winId
default implementation. Add a note to the API documentation, add
a Q_ASSERT to check if our assumption holds true.
Change-Id: I8607a4efc4f561f7849c976cd2454f6fbcb20eaa
Reviewed-by: Samuel Rødal <samuel.rodal@nokia.com>
Richard Moore [Sun, 19 Feb 2012 16:31:39 +0000 (16:31 +0000)]
Remove dead method.
Fix Qt 5 TODO
Change-Id: I9a12cc07ebc7c039bbeb53922b0539f177b226f8
Reviewed-by: Samuel Rødal <samuel.rodal@nokia.com>
Kurt Korbatits [Mon, 20 Feb 2012 04:05:47 +0000 (14:05 +1000)]
Changed qabstractnetworkcache unittest to use TESTDATA
- Changed qabstractnetworkcache to install testdata
Change-Id: I8f2ae6103214755ee7898dbc0ee50c0e4d7d45ab
Reviewed-by: Rohan McGovern <rohan.mcgovern@nokia.com>
Reviewed-by: Jason McDonald <jason.mcdonald@nokia.com>
Toby Tomkins [Mon, 20 Feb 2012 03:29:13 +0000 (13:29 +1000)]
Modify helper functions to remove embedded QSKIP call and return status.
The helper functions contained QSKIP macros which do not also skip their
parent function. These QSKIP function were removed and replaced with a
return success value.
Change-Id: I533f57842fc95beaeb1fcde5235678e9807db056
Reviewed-by: Rohan McGovern <rohan.mcgovern@nokia.com>
Toby Tomkins [Mon, 20 Feb 2012 05:46:46 +0000 (15:46 +1000)]
Set tst_qtoolbutton triggered function skipped.
This autotest seems to be caught in an event loop preventing further
correct signalling. Mark test function skipped with QSKIP.
Task-number: QTBUG-24374
Change-Id: Ic943a33b71fa87d0873278cb7b7b134c22602be3
Reviewed-by: Rohan McGovern <rohan.mcgovern@nokia.com>
Jason McDonald [Mon, 20 Feb 2012 01:39:31 +0000 (11:39 +1000)]
Make QLocale autotest pass for shadow builds.
For shadow builds where "make install" had not been run, the
tst_QLocale::emptyCtor test function would fail to find its helper
application because QFINDTESTDATA was not searching for it in the build
directory due to the test not instantiating QCoreApplication.
This commit fixes the test by instantiating a QCoreApplication using the
QTEST_MAIN macro, which also allows some special case code for Windows
CE to be removed.
Change-Id: Ic81d4699da2538c24b36b3d6bd52c4a02ad417f4
Reviewed-by: Rohan McGovern <rohan.mcgovern@nokia.com>
Jason McDonald [Fri, 17 Feb 2012 03:39:30 +0000 (13:39 +1000)]
testlib: Report one test result per benchmark test.
Prior to this commit, a benchmark test could report 0..n passes and 0..m
fails or skips, where n is the number of accumulation iterations used to
collect benchmark data and m is the number of times the test function
was invoked. Depending on the type of benchmark measurer being used,
this could result in a very large volume of test output and inconsistent
pass, fail and skip counts between test runs.
This commit changes the behaviour so that each benchmark test reports
one pass, fail or skip, regardless of the number of iterations used to
collect benchmark data.
This commit also prevents benchmark data being reported in the test
output if the benchmark test failed or skipped, as any benchmark data is
of dubious value in such cases.
The latter change in behaviour requires a minor modification to the
badxml selftest, which now tests quoting of literal strings in xml test
output for both passing and failing benchmarks.
Finally, this commit also adds a new selftest specifically for verifying
correct behaviour for benchmarks that fail or skip.
Task-number: QTBUG-24313
Change-Id: I3426dc659a7511b62fd183a031c7235bc753f497
Reviewed-by: Rohan McGovern <rohan.mcgovern@nokia.com>
Kurt Korbatits [Mon, 20 Feb 2012 00:18:01 +0000 (10:18 +1000)]
Fixed qobject unittest to work from installation directory
- Made test depend on subprogram to make sure it was there when
test ran.
- install signalbug subprogram
Change-Id: Ie0a19e52d131adcd17c97b263389aecffb81520e
Reviewed-by: Rohan McGovern <rohan.mcgovern@nokia.com>
Reviewed-by: Jason McDonald <jason.mcdonald@nokia.com>
Jason McDonald [Mon, 20 Feb 2012 00:15:57 +0000 (10:15 +1000)]
testlib: Add selftests for XPASS and XFAIL involving QCOMPARE.
Change-Id: I4f897b411aaa51ff453208eda50e12665489074b
Reviewed-by: Rohan McGovern <rohan.mcgovern@nokia.com>
Jason McDonald [Fri, 17 Feb 2012 05:11:20 +0000 (15:11 +1000)]
testlib: make XPASS output slightly less confusing.
Prior to this commit, when an unexpected pass occurred the test output
showed the message "XPASS : tst_foo::function() 'expr' returned FALSE",
where the problem was actually that the expression evaluated to true
when it was expected to be false.
This commit changes the output to make it clear that the expression
evaluated to true unexpectedly.
Task-number: QTBUG-22118
Change-Id: Id22c178073d3b75789675ca37a8ef019029b1f91
Reviewed-by: Rohan McGovern <rohan.mcgovern@nokia.com>
Andrew den Exter [Mon, 20 Feb 2012 00:50:06 +0000 (10:50 +1000)]
Remove unnecessary image height adjustments from layout.
cb8445f0323b0eefbb04f1d8adad81a00b53abd8 removes an implied +1 from
document line height so we need to remove the compensating -1 from
image ascent/descent for image heights to be correct..
Change-Id: I310c06bcf1ce2eefef42c2828faa0eb435ad71cf
Reviewed-by: Yann Bodson <yann.bodson@nokia.com>
Bradley T. Hughes [Thu, 16 Feb 2012 08:47:32 +0000 (09:47 +0100)]
Speed up tst_QWaitCondition
Reduce the thread and iteration counts to make this test execute faster.
This change reduces the runtime to 14 seconds (from 5 minutes, 38
seconds).
Change-Id: Id5ea056cfd33022da5a06809f0598a5cdb02b27b
Reviewed-by: Rohan McGovern <rohan.mcgovern@nokia.com>
David Faure [Sat, 18 Feb 2012 15:00:09 +0000 (16:00 +0100)]
Import QMimeType / QMimeDatabase into QtCore.
History of the development before the import:
ssh://codereview.qt-project.org/playground/mimetypes.git
Mimetype definitions come from shared-mime-info where available (UNIX systems),
loaded using a mmap'ed binary cache generated by update-mime-database.
As a fallback if no cache is found, we parse the raw XML files otherwise.
This makes the MIME type support fast and with very low memory usage on UNIX,
and it makes it easy to use on Windows (no dependency on shared-mime-info,
Qt even includes a freedesktop.xml file to use if none are found on the system).
Change-Id: I27b05008216ff936dc463bd80d3893422bfb940e
Reviewed-by: Richard J. Moore <rich@kde.org>
Oswald Buddenhagen [Mon, 30 Jan 2012 21:11:43 +0000 (22:11 +0100)]
make configure ignore $CC and $CXX
this was inconsistent with how qmake itself works, and was actually
wreaking havoc.
Change-Id: I5aa83cc88ffe7141cc0c31b03b76c48274f1ebdb
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Reviewed-by: Bradley T. Hughes <bradley.hughes@nokia.com>
Gunnar Sletta [Fri, 17 Feb 2012 10:30:56 +0000 (11:30 +0100)]
Introduced QWindow::isExposed().
The visible property along with show/hideEvent tracks the
windows visibility from the application perspective and is
really a request. The exposeEvent() along with the isExposed()
accessor is used to notify the application of the actual
state of the window in the windowing system.
Change-Id: I7f5b7ed74a168e34aaa21ce0ae9042ddfb0bf6d8
Reviewed-by: Samuel Rødal <samuel.rodal@nokia.com>
Thorbjørn Lund Martsum [Mon, 6 Feb 2012 08:24:50 +0000 (09:24 +0100)]
Adding hasHeightForWidth as a virtual Widget funcion
Just implements what the note states (and removes the private function)
Change-Id: I9a6fd5134460712accf09ba01691df8b9b1f5d0d
Reviewed-by: Jan-Arve Sæther <jan-arve.saether@nokia.com>
Andrew Christian [Fri, 17 Feb 2012 20:38:23 +0000 (15:38 -0500)]
Added error reporting to QJsonParser
Change-Id: Ib2390c0faf1ed7ada3fc185abce83740ad112929
Reviewed-by: Lars Knoll <lars.knoll@nokia.com>
Marc Mutz [Wed, 15 Feb 2012 17:55:25 +0000 (18:55 +0100)]
QEasingCurve: return QVector for cubic spline representation
QEasingCurve internally holds the spline data in a
QVector<QPointF>. For the return from cubicBezierSpline(),
the vector is transformed into a QList<QPointF>.
This involves copying, and into an inefficient (for QPointF
payloads) container at that, so deprecate cubicBezierSpline()
in favour of a new toCubicSpline() returning the QVector
directly.
Change-Id: Ie4827fe7c6e289ad97a0b09772e47298779c76ca
Reviewed-by: Lars Knoll <lars.knoll@nokia.com>
Richard Moore [Sat, 18 Feb 2012 10:57:42 +0000 (10:57 +0000)]
Trivial doc fix
Change-Id: I9b63e0b63f225b245eec68ea4211cb0f2ccf9bb5
Reviewed-by: Sergio Ahumada <sergio.ahumada@nokia.com>
João Abecasis [Thu, 16 Feb 2012 22:11:26 +0000 (23:11 +0100)]
Add header to .pri
Commit
7d16ea40 introduced QArrayDataPointer and respective header file, but
missed making it known to qmake.
Change-Id: I48c1831730cc4d27c6600c090a719861bb71a301
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Bradley T. Hughes [Fri, 17 Feb 2012 08:41:35 +0000 (09:41 +0100)]
Document Q_PROCESSOR_* macros
All known processors and their variants/revisions are documented. I
added Q_PROCESSOR_MIPS_V based on the MIPS64 online documentation, which
documents MIPS64 as a superset of MIPS IV and MIPS V.
Change-Id: Ie2796d4f03499283aa2c96d60f5e37bd74a36ab0
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Reviewed-by: Casper van Donderen <casper.vandonderen@nokia.com>
Friedemann Kleint [Wed, 15 Feb 2012 11:34:04 +0000 (12:34 +0100)]
Fix qclipboard autotest on Windows.
Emit changed signal only if the clipboard is not owned,
in which case QClipboard does it.
Task-number: QTBUG-24184
Change-Id: I27420583a718a5f8cd93b9d361b1e422a75df300
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@nokia.com>
João Abecasis [Thu, 16 Feb 2012 22:53:58 +0000 (23:53 +0100)]
QVector: always grow exponentially
For non-movable types (QTypeInfo<T>::isStatic), QVector would grow the
array linearly, and defer to qAllocMore otherwise. That property,
however, gives no indication as to how the vector will grow.
By forcing additional allocations for growing containers of such types,
this penalized exactly those objects which are more expensive to move.
We now let qAllocMore reign in growth decisions.
Change-Id: I843a89dcdc21d09868c6b62a846a7e1e4548e399
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Reviewed-by: Lars Knoll <lars.knoll@nokia.com>
João Abecasis [Thu, 16 Feb 2012 18:59:07 +0000 (19:59 +0100)]
qAllocMore: Always grow exponentially
qAllocMore is used by growing containers to allocate additional memory
for future growth. The previous algorithm would grow linearly in
increments of 8 up to 64 and then progress exponentially in powers of
two.
The new (constant time) algorithm does away with the linear segment and
always progresses exponentially. It also has the nice benefit of cleanly
avoiding undefined behaviour that the old implementation tried hard to
circumvent.
Besides always progressing exponentially, the next-power-of-two
algorithm was tweaked to always include space for growth. Previously
queries at boundary values (powers of two) would return the same value.
The test was updated to verify sanity of results. As the algorithm is
well behaved, testing of bogus data was dropped. Whatever happens in
those cases is irrelevant, anyway: the bug lives elsewhere.
Change-Id: I4def473cce4b438734887084e3c3bd8da0ff466b
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Reviewed-by: Lars Knoll <lars.knoll@nokia.com>
Shane Kearns [Fri, 17 Feb 2012 15:14:01 +0000 (15:14 +0000)]
Fix error handling in QHostInfo windows backend
If the DNS server returns a non authoritative host not found response,
then windows returns WSATRY_AGAIN error code.
This is now reported as HostNotFound and not UnknownError
Change-Id: I212985acd4e85ff4b2bdb6c57ec403405a7695fb
Reviewed-by: Miikka Heikkinen <miikka.heikkinen@digia.com>
Reviewed-by: Richard J. Moore <rich@kde.org>
Friedemann Kleint [Fri, 17 Feb 2012 15:08:17 +0000 (16:08 +0100)]
Windows: Fix QWindow-test.
- Save & Restore style and geometry when switching to
full screen and back since it is not a real state on
Windows.
- Obey the positioning policy in setGeometry.
Task-number: QTBUG-24185
Change-Id: I18dea4fd372e0b2e46273a7a27e0c6f4f4bde771
Reviewed-by: Samuel Rødal <samuel.rodal@nokia.com>
Rick Stockton [Tue, 14 Feb 2012 16:52:09 +0000 (08:52 -0800)]
qnamespace.qdoc: Add doco for new values of Qt::MouseButton
QtTBUG-22642 added about 25 new values for the Qt::MouseButton Enum.
This change adds documentation for the new values. Note that the
special value 'Qt::AllButtons' is documented at the top (next to
the other special value, 'Qt::NoButton'.) The new internal value
'Qt::MaxMouseButton' is listed as a \omitvalue.
Task Number: QTBUG-22642
Change-Id: Iaec623754156fff8a2c73e357fef82e1fe36354b
Reviewed-by: Alan Alpert <alan.alpert@nokia.com>
Reviewed-by: Samuel Rødal <samuel.rodal@nokia.com>
Stephen Kelly [Wed, 15 Feb 2012 21:12:41 +0000 (22:12 +0100)]
Only define QT_NAMESPACE if it is used.
Fixes non-namespaced builds.
Change-Id: I0ec3c29f58042313dfa864f15c2b2d47914cdedb
Reviewed-by: Clinton Stimpson <clinton@elemtech.com>
Reviewed-by: Stephen Kelly <stephen.kelly@kdab.com>
David Faure [Fri, 17 Feb 2012 08:05:39 +0000 (09:05 +0100)]
Ensure that Qt public headers compile with strict flags
Those from
http://wiki.qt-project.org/Coding_Conventions#Conventions_for_public_header_files
(unfortunatey -Wold-style-cast cannot be used due to the glibc macro bswap_16)
and many Qt defines that disable casts.
Change-Id: I97ac707a101df9819e8c031fa75a31b30e20247f
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Rafael Roquetto [Fri, 17 Feb 2012 11:17:30 +0000 (12:17 +0100)]
QProcess documentation typo fix and improvement.
The word 'momentaneously' does not exist.
Change-Id: I3d2201d5b7b3a01af5989bb1c3fcd55110d5482c
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Reviewed-by: Till Adam <till@kdab.com>
Reviewed-by: Robin Burchell <robin+qt@viroteck.net>
Simon A. Eugster [Tue, 14 Feb 2012 18:04:11 +0000 (19:04 +0100)]
Mention QT_NO_STL in the QString documentation
Defining QT_NO_STL disables STL functions (toStdString() etc.) too.
Change-Id: Id5c8e12d933af6af63ee7f80fa2d5d7577cd689a
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
David Faure [Fri, 17 Feb 2012 13:16:55 +0000 (14:16 +0100)]
No need to stat with QFile::exists before QFile::open, for reading.
Change-Id: I14ca7f8a377bca39004defc96d69d62c151181f5
Reviewed-by: João Abecasis <joao.abecasis@nokia.com>
Reviewed-by: Jeremy Katz <jeremy.katz@nokia.com>
Shane Kearns [Thu, 16 Feb 2012 17:02:11 +0000 (17:02 +0000)]
Remove unnecessary locking from QNetworkProxy constructor
QGlobalNetworkProxy (a singleton) had two phase construction, with
the second phase being called from QNetworkProxy's constructor.
This isn't necessary, and has been reported as causing deadlocks.
Although constructing socket engine handlers has side effects
(they add themselves to a list on construction and remove themselves
on destruction), this appears to be safe. The socket engine handlers
are only used while holding the list mutex, and any socket engines
created don't have any reference to the factory that created them.
With the new version, it is possible that two instances of
QHttpSocketEngineHandler and QSocks5SocketEngineHandler exist
temporarily if a Q_GLOBAL_STATIC initialisation race occurs.
This appears safe, because the loser of the race deletes its
handlers, which remove themselves from the global list as above.
Task-number: QTBUG-13088
Change-Id: I8cf520da717d8ab7d862ab89c6de13aea6d60ac3
Reviewed-by: Richard J. Moore <rich@kde.org>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Friedemann Kleint [Fri, 17 Feb 2012 10:59:48 +0000 (11:59 +0100)]
Fix warnings about missing file.
Change-Id: I8644d2b80bab9c18f666d742da389e84bf3f124c
Reviewed-by: Xizhi Zhu <xizhi.zhu@nokia.com>
Shane Kearns [Wed, 15 Feb 2012 17:24:14 +0000 (17:24 +0000)]
Fix handling of urls containing username/password in QNetworkAccessManager
QNetworkAccessManager was ignoring the supplied credentials, although
webkit seems to support these urls at a higher level.
Following the behaviour of browsers:
We use supplied credentials if authentication is required.
We add supplied credentials to the authentication cache.
We emit authenticationRequired signal if the credentials were wrong.
We do not use previously cached credentials for that url
Synchronous http requests fail, if the credentials were wrong.
Task-number: QTBUG-18107
Change-Id: If46e8eab1511ba8a0f4bbe0d4efaabc4df0b8ab4
Reviewed-by: Richard J. Moore <rich@kde.org>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Shane Kearns [Wed, 15 Feb 2012 17:58:11 +0000 (17:58 +0000)]
Fix incorrect use of QObject::disconnect in synchronous http request
In certain circumstances, this could cause the request to time out
(and repeatedly send bad authentication credentials to the server)
instead of failing with AuthenticationRequiredError.
Change-Id: Iff66b32f1d7268f21fd77b6620aae4b5d49d857f
Reviewed-by: Richard J. Moore <rich@kde.org>
Shane Kearns [Wed, 15 Feb 2012 15:34:09 +0000 (15:34 +0000)]
Update documentation to show NTLMv2 is supported
This feature was implemented in 4.8, but documentation was not
updated at the time.
Task-number: QTBUG-18181
Change-Id: I657d7ab7aaf43b73b7bf8fd1cb76086522cf5c2b
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Marc Mutz [Thu, 16 Feb 2012 15:38:52 +0000 (16:38 +0100)]
QSize/QSizeF: add const versions of scale()/transpose()
Non-mutating functions can (potentially) be constexpr.
On top of that, they often make for more readable code.
Change-Id: I3547327cf5a7162737353a864a1025d0d02ccc2f
Reviewed-by: Gunnar Sletta <gunnar.sletta@nokia.com>
Reviewed-by: David Faure <faure@kde.org>
Miikka Heikkinen [Thu, 16 Feb 2012 11:36:27 +0000 (13:36 +0200)]
Make 'nmake check' succeed for widgets tests
Marked a bunch of tests insignificant, and skipped one crashing test
case in QApplication test, as that couldn't be made to pass simply by
marking the test insignificant.
Once the underlying issues are fixed, the tests need to be re-enabled.
Task-number: QTBUG-24203
Change-Id: I9aea4fa207d307793445efdcaead72219fbf6c4f
Reviewed-by: Sergio Ahumada <sergio.ahumada@nokia.com>
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@nokia.com>
Friedemann Kleint [Fri, 17 Feb 2012 12:59:20 +0000 (13:59 +0100)]
Fix another typo in the evdev keyboard plugin.
Amend
3296b7327bd84f420df7e5561d192c0a8d0bb409
Change-Id: Ide58a6b8cbd980eda6ff91948c8f4fb45a30e527
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@nokia.com>
Friedemann Kleint [Tue, 14 Feb 2012 15:46:31 +0000 (16:46 +0100)]
Widgets: Remove obsolete QGuiPlatformPlugin.
It has been superseeded by QPlatformTheme.
Task-number: QTBUG-24204
Change-Id: I8c4131c7bfd8201e747984d772ed8042610192b8
Reviewed-by: Morten Johan Sørvig <morten.sorvig@nokia.com>
Miikka Heikkinen [Fri, 17 Feb 2012 10:28:50 +0000 (12:28 +0200)]
Disable 'check' target generation for network autotests for Windows.
A number of network autotests are unstable in Windows, so don't
generate check target for them as is done for mac.
Once the tests are acceptably stable, this needs to be reverted.
Change-Id: I18262e28ce40eba541aecf3cfb651bff34698ead
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@nokia.com>
Miikka Heikkinen [Fri, 17 Feb 2012 08:48:22 +0000 (10:48 +0200)]
Fix "check" target when load(testcase) is used in .pro file.
The "check" target always depended on debug-check on projects that
explicitly loaded the testcase.prf, even if Qt was configured with
-release parameter. This obviously caused build failure. Changed the
testcase.prf to check which configuration is preferred using
CONFIG(debug, debug|release) like is done with similar cases in other
.prf files.
Task-number: QTBUG-24332
Change-Id: Ib5140b106e99efe51932bdd5a48914786de23230
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@nokia.com>
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@nokia.com>
Friedemann Kleint [Fri, 17 Feb 2012 10:01:26 +0000 (11:01 +0100)]
Qt 5 plugin system: Fix handling of namespaced plugin classes.
- Add 'using namespace' to moc code as was the case in 4.8.
Change-Id: I26cba9ad74bf05eecc5205714c32c3176695e3b4
Reviewed-by: Lars Knoll <lars.knoll@nokia.com>
Bradley T. Hughes [Wed, 15 Feb 2012 12:49:55 +0000 (13:49 +0100)]
Clean up default arch setting on Mac OS X
Remove config.tests/mac/defaultarch.test, but do the same type of
default arch detection, only using the qmake binary instead of compiling
an empty file. Qt 5 will only support 10.6 and up, which means we
only support i386 and x86_64 now.
Change-Id: I24949ac803b965c523b1ee45cf769e266dcde134
Reviewed-by: Morten Johan Sørvig <morten.sorvig@nokia.com>
Shane Kearns [Wed, 15 Feb 2012 20:12:15 +0000 (20:12 +0000)]
Fortune client example - list all IP addresses
The first IP address on the machine might not be a usable one.
Changed the QLineEdit to a QComboBox, and populated it with
all IP addresses of the machine, along with the machine names
if they are configured.
Task-number: QTBUG-13121
Change-Id: I7c443f5ce6efb0d0b525c5abad1671d3dcbba33c
Reviewed-by: Casper van Donderen <casper.vandonderen@nokia.com>
Samuel Rødal [Fri, 17 Feb 2012 09:19:35 +0000 (10:19 +0100)]
Fixed API in QWindow: visible() -> isVisible().
Change-Id: I14706abb8441c153f738563cb1a46205fdb2dae6
QWindow::visible() did not follow the API guidelines.
Reviewed-by: Gunnar Sletta <gunnar.sletta@nokia.com>