This is so that the results of LayoutTests in WebKit are
more in line with what things were like in Qt 4.8.
Change-Id: I25962e03bd8e0316cb303c0d94c25ac4e73ea9a8
Reviewed-on: http://codereview.qt-project.org/6162
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: Jiang Jiang <jiang.jiang@nokia.com>
The current code was meant to be a thread-safe initialisation that
also ran a couple of extra steps. But it wasn't. While it's ok to call
qAddPostRoutine(), the call to updateConfigurations() was
thread-unsafe. It is possible that another thread got the pointer to
the Private before updateConfigurations() finished.
So instead protect the initialisation with a mutex.
It's possible that the value of the pointer becomes visible to
other processors before the other contained values, so use
atomics here.
To call qAddPostRoutine safely from the main thread, use the trick
of deleteLater() (which is thread-safe) in another thread connecting
to a slot.
Change-Id: If9bab88138755df95a791f34b0be8684207979d7
Reviewed-on: http://codereview.qt-project.org/5028
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: Bradley T. Hughes <bradley.hughes@nokia.com>
Reviewed-by: Lars Knoll <lars.knoll@nokia.com>
There was a big processEvents() method which was actually getting
called for a single type of events.
Change-Id: I66ac70c7ac6e2f9136b24541d42a4185f8797583
Reviewed-on: http://codereview.qt-project.org/6132
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: Lars Knoll <lars.knoll@nokia.com>
Reviewed-by: Joona Petrell <joona.t.petrell@nokia.com>
according to the Unicode specs, code point U+0085 should be treated
like a white space character (an exceptional Cc one)
Change-Id: Ib17ae0c4d3cdafe667cafa38b645138ef24c238c
Merge-request: 32
Reviewed-by: Jiang Jiang <jiang.jiang@nokia.com>
Reviewed-on: http://codereview.qt-project.org/6158
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Requesting the write permission marked the read permission as known
without having retrieved it. A subsequent request for read permission
would return false.
Task-Number: QTBUG-20714
Reviewed-By: mread
(cherry picked from commit bf69c7e0af67ec877da4fee244386e538fa1c01d)
Change-Id: I74fd118fcdd3b222098eda2718823a36931424a1
Reviewed-on: http://codereview.qt-project.org/6159
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: João Abecasis <joao.abecasis@nokia.com>
There were two fixes in 4.8 which each fixed a part of the problem.
Comparing canonical paths is more correct, but is only possible where
both directories exist. If neither directory exists, then compare
absolute paths instead.
Changed a regression test, because /tmp is a symbolic link on MacOS.
I.E. "/tmp/.." is canonically "/private" and not "/" as expected.
Task-Number: QTBUG-20495
Reviewed-By: joao
(cherry-picked from ad35d25e78c8252a72108a4ba931934047c4707e)
Change-Id: Ia4986e8337f0e512e1a3398a5a4dd36e62680b9c
Reviewed-on: http://codereview.qt-project.org/5813
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: Shane Kearns <shane.kearns@accenture.com>
Widget failed to update cursor position if selection remained the same, but
changed direction, and did not emit selectionChanged if change resulted
from input method event.
Change-Id: I6a4ae48d41efa5e126980749f89982f92e3ec499
Reviewed-on: http://codereview.qt-project.org/6131
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: Joona Petrell <joona.t.petrell@nokia.com>
Reviewed-by: Lars Knoll <lars.knoll@nokia.com>
This is in order to have a regression test for QT-5304.
However the test is also checking basic functionality of head
requests too.
Task-Number: QT-5304
Change-Id: I0d98bd1f1125ffd49d9f3c859a51b2aaee95379a
Reviewed-By: Martin Petersson
Reviewed-on: http://codereview.qt-project.org/6033
Reviewed-by: Martin Petersson <Martin.Petersson@nokia.com>
We can't rely on absolute paths when comparing directories for equality
as these don't take into account symbolic links and may also bypass ../
and ./ simplification.
Instead, canonical paths must be computed and can then be compared
according to the case sensitivity rules for the platform or file engine,
as is done in QFileInfo.
Task-number: QTBUG-20495
Reviewed-by: Prasanth Ullattil
(cherry-picked from dcee6e1371d899eb79717b8e3f3eec08b765db82)
Change-Id: Ib5f2a6ee11311c55782ea5dd0e9c3b45f9231686
Reviewed-on: http://codereview.qt-project.org/5812
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: João Abecasis <joao.abecasis@nokia.com>
Multiplying by zero was not the intention.
Task-number: QTBUG-20482
Change-Id: I3c4dfd69da4d98ed65f318d1fa0c3cf827017688
Reviewed-on: http://codereview.qt-project.org/6048
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: Gunnar Sletta <gunnar.sletta@nokia.com>
Made physicalSize() return QSizeF instead, to prevent rounding errors.
Added logicalSize() as the base to compute font pixel sizes instead, and
added convenience functions in QScreen to access the logical and
physical sizes and DPI metrics.
Task-number: QTBUG-21736
Task-number: QTBUG-21737
Change-Id: Ic705dc98eb3632617659e65a0c9a552673dc0c65
Reviewed-on: http://codereview.qt-project.org/5888
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@nokia.com>
Setting tos to 0 once should be enough for everyone.
Task-number: QTBUG-20482
Change-Id: I7aa46ff5d4ed8064e3fb2cebc3dfcaa4e73741f8
Reviewed-on: http://codereview.qt-project.org/6053
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: Roberto Raggi <roberto.raggi@nokia.com>
The QSKIP is only reachable on Mac, so only compile it for Mac.
Change-Id: I268e87829e01755051f81cdaf856d936b04b3d49
Reviewed-on: http://codereview.qt-project.org/6091
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: Rohan McGovern <rohan.mcgovern@nokia.com>
Passing -no-xcb or -no-wayland was enough to get a build with neither
but the error message says you need to pass both. Fix the logic.
Change-Id: I2d554d371b603e7df7bc07e1b4d5271ba31243f0
Reviewed-on: http://codereview.qt-project.org/6099
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: Rohan McGovern <rohan.mcgovern@nokia.com>
QTestLib already makes a superior version of this functionality
available to all tests, without the need to recompile. Each qtestlib
test program can be passed, via the command line, a list of test
functions and test data tags to be executed instead of executing all
test functions.
Change-Id: I1b54fc70a0c58a17209817afabfb953b5748dac4
Reviewed-on: http://codereview.qt-project.org/6015
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: Rohan McGovern <rohan.mcgovern@nokia.com>
Omit focusColors test function at compile-time in preference to
skipping at run-time. Also removed commented out code.
Change-Id: Ib8955656f5a43b30df95b4c9865e52902dbd6b1e
Reviewed-on: http://codereview.qt-project.org/6010
Reviewed-by: Rohan McGovern <rohan.mcgovern@nokia.com>
Remove empty constructor/destructor. Remove dropDownDeleted test
function which has been entirely skipped since before Qt 4.0, is
obviously incomplete (it attempts to use a QComboBox that does not
exist) and provides no statement of intention that might allow the
incompleteness to be addressed.
Change-Id: I4975b6d5a5dde1e58dab1dd393be92eddeb4c9f8
Reviewed-on: http://codereview.qt-project.org/6009
Reviewed-by: Rohan McGovern <rohan.mcgovern@nokia.com>
The test was a no-op on Mac OS. It's better to avoid building or running
it on Mac OS, rather than spending time compiling and running a test
that does nothing on that platform.
Change-Id: I4aa579036788e404f8a6a8f1a80acd32338fb3cd
Reviewed-on: http://codereview.qt-project.org/5958
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: Rohan McGovern <rohan.mcgovern@nokia.com>
Require the user to pass -no-xcb and -no-wayland in order to produce
a build that has only the minimal plugin available.
This prevents "useless" builds from being accidentally created.
Change-Id: If2f7f516a52b75998f3ae7bf802f89f93f6e8218
Reviewed-on: http://codereview.qt-project.org/6005
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: Rohan McGovern <rohan.mcgovern@nokia.com>
The * was not appearing against the default for the -xcb switch because
of a spelling mistake.
Change-Id: I3491cc9772d02179711b0d474db7ab2ef2398e88
Reviewed-on: http://codereview.qt-project.org/6003
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: Lincoln Ramsay <lincoln.ramsay@nokia.com>
This also uncovered a missing implementation of child for
dock widgets.
Change-Id: Iac4c5a51d4626769af11b277a9a345e1e31dc490
Reviewed-on: http://codereview.qt-project.org/5987
Reviewed-by: Jan-Arve Sæther <jan-arve.saether@nokia.com>
The operator tested by this function (QMatrix * QRect) has not existed
since Qt 2.x.
Change-Id: Id09154de2bdda70c0fe342fe78b82709953e2b56
Reviewed-on: http://codereview.qt-project.org/5965
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: Rohan McGovern <rohan.mcgovern@nokia.com>
The test was being skipped one test function at a time on IRIX. It's
better to avoid building or running it, rather than spending time
running a test that does nothing on that platform.
This commit also turns a silent skip on other platforms into an explicit
QSKIP to make it more obvious that there's an unfixed TODO there.
Change-Id: I7cae4f0f2eaebf5977a781b2ce15d1e71ef13b36
Reviewed-on: http://codereview.qt-project.org/5957
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: Rohan McGovern <rohan.mcgovern@nokia.com>
The unprintablenames test was designed to prevent an old failure in
QDir from returning, but the test can't be run reliably on some
file-systems and certainly isn't going to achieve its goal when it's
commented out.
Change-Id: Ib4cb965d59c291ab2436130b87e865ac21b9b483
Reviewed-on: http://codereview.qt-project.org/5956
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: Rohan McGovern <rohan.mcgovern@nokia.com>
QHttpNetworkReply was waiting for a body to be sent for 401 and 407
responses, whereas with a HTTP HEAD request, there will be no body.
This delayed the authentication signal until after the http channel
is closed by the server after a timeout. For example with the server
used for autotesting, the authentication signal is delayed 15 seconds.
When the server has a very long timeout, the authentication signal may
not be emitted at all.
Task-Number: QT-5304
Reviewed-By: Martin Petersson
(cherry picked from commit 8610ee14b8636641651a8ba6040cca16c4141ed6)
Change-Id: Ie4ce6c598df86ce59910f793fd5ae7c1ccf39f9d
Reviewed-on: http://codereview.qt-project.org/6032
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: Martin Petersson <Martin.Petersson@nokia.com>
Set -platformargument foo[=bar] as a dynamic property
on the native interface.
Change-Id: Ica8b25478da0ce9508b90370f15812fd11ff5d13
Reviewed-on: http://codereview.qt-project.org/6031
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: Oliver Wolff <oliver.wolff@nokia.com>
Reviewed-by: Samuel Rødal <samuel.rodal@nokia.com>
Pair the _next.testAndSetRelease() call in QFreeList<T>::release() with
_next.loadAcquire(), to ensure that the update to the bucket
being released is properly fenced.
QFreeList<T>:next() does not need a release fence, only an acquire fence,
which is placed on the memory bucket address load.
Change-Id: Ib5b9d6ef6107f87aa8e3ea2dd3a7f9116c75da70
Reviewed-by: Thiago Macieira
Reviewed-on: http://codereview.qt-project.org/5802
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: João Abecasis <joao.abecasis@nokia.com>
By default it doesn't do anything special, but for fontconfig font db
it will do some tricks to convert alias like serif to actualy font
family names like Times.
Change-Id: I73058d187a5422e6f93bc1b728cecdc0250d5e1e
Reviewed-on: http://codereview.qt-project.org/5984
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: Pierre Rossi <pierre.rossi@gmail.com>
Renamed QGuiApplication::activeWindow() to
QGuiApplication::focusWindow(), implemented QWindow::isActive() as a
style hint, and added focus in / out events.
Change-Id: I71866e76c5a817def3e17bcc20a4fc32081a0e7a
Reviewed-on: http://codereview.qt-project.org/5811
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: Gunnar Sletta <gunnar.sletta@nokia.com>
QSKIP is intended to be used to skip test functions that are found at
run-time to be inapplicable or unsafe. If a test function can be
determined to be inapplicable at compile-time, the entire test function
should be omitted instead of replacing the body of the test function
with a QSKIP, which only serves to slow down test runs and to inflate
test run-rates with empty, inapplicable tests.
Task-number: QTQAINFRA-278
Change-Id: I582732e3dd657df834f9a98fd52d7ee368f2f29b
Reviewed-on: http://codereview.qt-project.org/5946
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: Rohan McGovern <rohan.mcgovern@nokia.com>
QSKIP is intended to be used to skip test functions that are found at
run-time to be inapplicable or unsafe. If a test function can be
determined to be inapplicable at compile-time, the entire test function
should be omitted instead of replacing the body of the test function
with a QSKIP, which only serves to slow down test runs and to inflate
test run-rates with empty, inapplicable tests.
Task-number: QTQAINFRA-278
Change-Id: Ib2025339422749cf216e87ac414a3056250bf8f9
Reviewed-on: http://codereview.qt-project.org/5942
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: Rohan McGovern <rohan.mcgovern@nokia.com>
qtdeclarative/src/imports/testlib uses the exported variable
QTest::printAvailableFunctions.
As it is a plugin, it has only QT_DLL defined and not QT_SHARED
and thus sees Q_TESTLIB_EXPORT=''.
Change-Id: I34646e920a812fb907b60e0c4c9399499d689a01
Reviewed-on: http://codereview.qt-project.org/5990
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: Charles Yin <charles.yin@nokia.com>
