If the compilation of a subdir was explicitly disabled by default, then
the testing of that subdir should also be disabled by default.
Change-Id: I928e232393e89a7c27813b7a48864d4e0fb687f8
Reviewed-on: http://codereview.qt.nokia.com/2835
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: Kalle Lehtonen <kalle.ju.lehtonen@nokia.com>
This logic should be used for all mac platforms.
Currently it is erroneously skipped for qpa on mac.
Change-Id: Ibc8e0bb38d4c08f48292639c614bc328c703b1bc
Reviewed-on: http://codereview.qt.nokia.com/2836
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: Kalle Lehtonen <kalle.ju.lehtonen@nokia.com>
make users more aware of QLabel guessing the text format.
Reviewed-by: Gabriel de Dietrich <gabriel.dietrich-de@nokia.com>
(cherry picked from commit 31f7ecbdcdbafbac5bbfa693e4d060757244941b)
Change-Id: I681c7daf568a6107b01908fa85d2da2a7e4e228a
Reviewed-on: http://codereview.qt.nokia.com/2817
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: Peter Hartmann <peter.hartmann@nokia.com>
debug/release is always supposed to default on for platforms where
debug/release affects linkage (i.e. mac and windows). Previously, it
did not correctly default on for qpa mac builds.
Change-Id: I8d3c418f3fb787136ab120872fdd4d2ebf82c90e
Reviewed-on: http://codereview.qt.nokia.com/2796
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: Toby Tomkins <toby.tomkins@nokia.com>
The floating point problem which causes this test to fail is not
triggered if Qt is compiled in debug mode. Enable QEXPECT_FAIL more
narrowly.
Task-number: QTBUG-20661
Change-Id: I25f3260b516cd4fda9d8b711aa1140cbda821412
Reviewed-on: http://codereview.qt.nokia.com/2719
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: Kalle Lehtonen <kalle.ju.lehtonen@nokia.com>
As virtually all of our tests which use the live qt-test-server, this
test is inherently unstable.
Change-Id: I27fc07a28b8dd6cd4dd3cb6414389e3cf41d0318
Reviewed-on: http://codereview.qt.nokia.com/2795
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: Kalle Lehtonen <kalle.ju.lehtonen@nokia.com>
QGLXContext might be used from different threads than the gui thread, so
don't trigger any XCB logging from there.
Change-Id: I96a51ebf6987749d209b447c9ae2ac8b7298bd4c
Reviewed-on: http://codereview.qt.nokia.com/2782
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: Jørgen Lind <jorgen.lind@nokia.com>
We had to leave the non-const operator== for binary compatibility.
Remove them all, just leave the const version in there. 100% source
compatible.
Change-Id: Ib7a70fb441fe51d5164d9cbf495cbeda0f48fafe
Reviewed-on: http://codereview.qt.nokia.com/2773
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: Robert Griebl <robert.griebl@nokia.com>
At least QMouseEvent already uses QPointF so there is no reason to
restrict handleMouseEvent to integer positions.
Change-Id: I9c72d2e0e99518b40a1898aec9ad2b3421a3e9e2
Reviewed-on: http://codereview.qt.nokia.com/2739
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: Samuel Rødal <samuel.rodal@nokia.com>
Mark all of these autotests as `insignificant' until the underlying
crash bug is fixed.
Task-number: QTBUG-20756
Change-Id: I18fb7038dd86ef2e9a5f37c8f26d7fe76032754e
Reviewed-on: http://codereview.qt.nokia.com/2714
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: Kalle Lehtonen <kalle.ju.lehtonen@nokia.com>
The rest of this test gives unpredictable results if initTestCase fails.
Normally, a failure in initTestCase skips the rest of the test, but this
is not the case if QEXPECT_FAIL is used. Adjust the test to ensure the
rest of it is still correctly skipped.
Change-Id: I8ba76161e788f10801029ef10693e9ac275a2897
Reviewed-on: http://codereview.qt.nokia.com/2713
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: Kalle Lehtonen <kalle.ju.lehtonen@nokia.com>
Since 833090abb9 these need to be called manually.
Change-Id: Iabc263d499e724d4d39c3d8350563feb57749879
Reviewed-on: http://codereview.qt.nokia.com/2704
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: Jørgen Lind <jorgen.lind@nokia.com>
So that this function instead of the one in QPlatformFontDatabase
will be called.
Change-Id: Ifdb75ada27723c688e42067c7633b08391d66467
Reviewed-on: http://codereview.qt.nokia.com/2696
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: Jørgen Lind <jorgen.lind@nokia.com>
This test unstably fails for qpa, xcb, so mark it as insignificant.
Task-number: QTBUG-20761
Change-Id: I0b4228c2803946ef87d7d11f4167017c2ffd0ad6
Reviewed-on: http://codereview.qt.nokia.com/2658
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: Sergio Ahumada <sergio.ahumada@nokia.com>
First check the usWeightClass in OS/2 font table, then check
the weight byte in panose structure because Nokia Pure Text
fonts only have that set correctly.
Change-Id: Idce2626c8df17ce74ba78b317846cb42c3f1fe84
Reviewed-by: Jørgen Lind
Reviewed-on: http://codereview.qt.nokia.com/2682
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: Jiang Jiang <jiang.jiang@nokia.com>
this is a rather crude hack, but whatever.
Change-Id: I3fa363572842b207202d4aac8526951c0e16bf5b
Reviewed-on: http://codereview.qt.nokia.com/2602
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: Samuel Rødal <samuel.rodal@nokia.com>
QBezier::pointAt() could potentially return values outside the bezier's
bounds, even when the bezier was a straight horizontal line. For
example, with y = 0.5, it would produce y=0.5 or y=0.49999999999999 for
different values of t, which when rounded cause jittering in a QML
PathView.
Task-number: QTBUG-17007
Task-number: QTBUG-18133
Cherry-pick-of: 8b66982ec7b4b5d2071931c288973dce73dc9875
Change-Id: I4ecac7b9085aaaaaaaaaaaaaaaaaaaaaa7d7b0bc
Reviewed-on: http://codereview.qt.nokia.com/2467
Reviewed-by: Samuel Rødal <samuel.rodal@nokia.com>
This test had an invalid circular dependency: the test is in qtbase, and
it depends on lrelease in qttools, which depends on qtbase.
Change the test to skip if lrelease is not available or not working.
Change-Id: I4696f4a178a6f8c0f2e76b5c4ae7f60c4f953334
Reviewed-on: http://codereview.qt.nokia.com/2619
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: Rohan McGovern <rohan.mcgovern@nokia.com>
We need this to avoid syncing certain files in QtDeclarative on
Windows.
Change-Id: Ia9fc95262ac18bafc39efc038ea70a1b5d67cddd
Reviewed-on: http://codereview.qt.nokia.com/2641
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: Marius Storm-Olsen <marius.storm-olsen@nokia.com>
... and add a new method subjectAlternativeNames() instead. This was
a typo in the API.
Change-Id: Id8704c387c9ff8e1af2b9a524ff628f5c053a294
Reviewed-on: http://codereview.qt.nokia.com/2618
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: João Abecasis <joao.abecasis@nokia.com>
This test had an invalid circular dependency: the test is in qtbase, and
it depends on lrelease in qttools, which depends on qtbase.
Change the test to skip if lrelease is not available or not working.
Change-Id: I964ddaf9485a2afed5a1bd4f88d9bdccadf1cfca
Reviewed-on: http://codereview.qt.nokia.com/2472
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: Kalle Lehtonen <kalle.ju.lehtonen@nokia.com>
The wayland plugin's clipboard implementation lacked support for data
set via QMimeData::setImageData, QClipboard::setPixmap, etc. because
everything was treated as a QByteArray. A variant having an image
cannot however be converted to a byte array. Therefore, to make these
convenience functions to work, the image is now encoded via
QImageWriter before sending it. On the receiving side QMimeData is
capable of decoding images from byte arrays, so no problem there.
The patch also renames an internal QWaylandMimeData class to a more
suitable clipboard-specific name as the functionality is clipboard
specific.
Change-Id: I77e7df903b8dbe9731613fdbb9693b2a37b05d18
Reviewed-on: http://codereview.qt.nokia.com/2616
Reviewed-by: Qt Sanity Bot <qt_sanity_bot@ovi.com>
Reviewed-by: Jørgen Lind <jorgen.lind@nokia.com>
