Creating native widgets such as QGLWidget and then calling move()
on top level widget before doing show() on it caused framestrut to
be incorrect, as native widgets force the creation of the
native windows also on their ancestors when they are constructed,
rather than waiting for window to be shown.
Change-Id: I3e8d60547b3b683178ba059c5d92526a1d447442
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@nokia.com>
XQMakeSpec is not available after configure
but the QMakeSpec contains the correct value
Change-Id: I6cd4da8b0d6c95565f31842c17611ffd361bc010
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@nokia.com>
it produces way too many false positives to be useful.
Change-Id: Iefa423f96fa5574267b1468abb5efc8454ab54a3
Reviewed-by: Joerg Bornemann <joerg.bornemann@nokia.com>
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@nokia.com>
Added QCocoaScreen::grabWindow() and copied Qt 4 implimentation.
examples/desktop/screenshot works but WId window is not handled yet.
Change-Id: If228bb59d98a5166788ab38293d7e6a6300d1f85
Reviewed-by: Morten Johan Sørvig <morten.sorvig@nokia.com>
This is necesary but not sufficient to make the small configuration
build.
Task-number: QTBUG-24816
Change-Id: I1a06555e6f54f4d3c09a34fd50bf76e7b47d469f
Reviewed-by: Jørgen Lind <jorgen.lind@nokia.com>
Reviewed-by: Samuel Rødal <samuel.rodal@nokia.com>
Reviewed-by: Tasuku Suzuki <tasuku.suzuki@nokia.com>
Add more test data, merging operator!= tests in with operator==
to take advantage of added data.
Change-Id: If0426a3d01b8800cb7363385dbf3bcb21af5ed8f
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Embedding native widgets creates window before ActiveQt has a chance
to set the _q_embedded_native_parent_handle property on the window,
so embedded flag doesn't get set during window creation.
Update embedded flag also when setting window flags.
Task-number: QTBUG-26438
Change-Id: Id29d7123de81d6542dfd6471d6dba3d31bbfe2df
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@nokia.com>
The code should be there. Also removed
a few obsolete #defines.
Change-Id: Id63418b344157a99ac502329eea4f0bc9dc77849
Reviewed-by: Casper van Donderen <casper.vandonderen@nokia.com>
Previously the URL relative to the installdir was used, now we use a
relative URL to the index file.
Change-Id: Ia2485aa49da8240901d0db1f102eb48da689ef2a
Reviewed-by: Martin Smith <martin.smith@nokia.com>
On Windows Xp we can not connect to the pseudo interfaces used for
Teredo, so do not add these to the tests.
Change-Id: I4e20c880fa2d18f266ffcef2f640d8b2e6d0cd21
Reviewed-by: Shane Kearns <shane.kearns@accenture.com>
exists() resolves the path relative to qmake's current working
directory, which is mkspecs/features. Explicitly refer to the full
path.
Change-Id: I04e9f377528bffc1ebeeed69c86d73b85a117423
Reviewed-by: Toby Tomkins <toby.tomkins@nokia.com>
When used in the SceneGraph we're expected to disable the arrays
after we're done, otherwise we'll get a warning about memory corruption
and crashes.
Change-Id: Ie8e426309716bef4b75d79039e8ca0b4943c79e7
Reviewed-by: Jiang Jiang <jiang.jiang@nokia.com>
This adds:
- QT_QPA_EGLFS_WIDTH
- QT_QPA_EGLFS_HEIGHT
If both are set, fb0 detection is completely bypassed, otherwise, the
environment variables act as an override for the fb0 values.
Change-Id: Iaf38c8b0d40ce0a921bb6c5dbf31af788802d3fa
Reviewed-by: Girish Ramakrishnan <girish.1.ramakrishnan@nokia.com>
It seemingly doesn't include many of the required defines for evdev, so do what
everyone else seems to do, and work around it.
Change-Id: I4a6ffb548c7370f675c736177b63220f034d06b8
Reviewed-by: Girish Ramakrishnan <girish.1.ramakrishnan@nokia.com>
Android doesn't define a number of items in its input.h for MT, so don't test
for them here. They will still work, they just need manual defines.
Change-Id: I471f55b7c7f8950d2048e525119b2b36c8e9743b
Reviewed-by: Girish Ramakrishnan <girish.1.ramakrishnan@nokia.com>
otherwise we get lots of nasty warning messages about missing pri files.
it is arguable whether it is a good idea to do that, but we preserve
mkspecs/modules, too, and the previously built libraries don't just
vanish, either.
Change-Id: Ieded8d8858f1b0135bc3bea894b4a676024ac8ca
Reviewed-by: Girish Ramakrishnan <girish.1.ramakrishnan@nokia.com>
Reviewed-by: Joerg Bornemann <joerg.bornemann@nokia.com>
Reviewed-by: Marius Storm-Olsen <marius.storm-olsen@nokia.com>
tools like idc and lrelease are outside qtbase/bin and depend on libraries
which are not in any search path, so we need some way to let the modules
announce the locations (qt_tool.prf) and use it (in qtPrepareTool()).
Change-Id: I98d5109cbee5e745d86dde94e3dc791d42edc3ec
Reviewed-by: Marius Storm-Olsen <marius.storm-olsen@nokia.com>
a prefix build needs to be installed first anyway
Change-Id: I798861fddad2be289e9e4f182d0373e389b57437
Reviewed-by: Marius Storm-Olsen <marius.storm-olsen@nokia.com>
i have no idea why that was done (the commit message says "not sure why
it works elsewhere"), but it makes no sense whatsoever, specifically
doing it only on mingw. probably some workaround, as usual. the
de-duplication is broken by design anyway.
This reverts commit 7a6302c2baf6861fdaf65992b71a7676859860c2.
Change-Id: I6edecaa062570e59eccd24d50919ba132e65a403
Reviewed-by: Marius Storm-Olsen <marius.storm-olsen@nokia.com>
Screen events should come through QEventDispatcherBlackberry by default.
Change-Id: I64d17b99e9dd8edf23f68e1186e9318849f1a398
Reviewed-by: Sean Harmer <sean.harmer@kdab.com>
Reviewed-by: Thomas McGuire <thomas.mcguire@kdab.com>
Do not try to connect to the Terdo Tunneling Pseudo-Interface as
this will fail for Windows Xp.
Change-Id: I6dcd8369ba1e8642224cd4ac53f4032ed46d050d
Reviewed-by: Shane Kearns <shane.kearns@accenture.com>
This test is nice to have regardless though the main reason for it
is a refactor of hiddenSections in QHeaderView.
Change-Id: Id41a1d5edda2ef75bf432a78cdb3e952303eae92
Reviewed-by: Stephen Kelly <stephen.kelly@kdab.com>
This patch improves the manual resizeSection a bit. Before we didn't
consider that the program could maybe resize other sections when the
user was resizing one section.
The main issue with that is that setOffset is so smart that it helps
moving the mouse cursor - however it really shouldn't do if the
program is trying to change something too.
Maybe this won't solve all (possible) problems at once - but it is
a fixed needed just to make something work - trying to make anything
work without this fix is horrible....
Change-Id: I3cefa375a9b8ee4c1ef1e08ba0900025c671e4c6
Reviewed-by: Stephen Kelly <stephen.kelly@kdab.com>
Currently, only one bps event can be handled per event loop iteration,
bringing about a huge performance penalty. This patch fixes this issue by
bulk processing events whenever they are available.
Change-Id: Iecce1e86730bd90c3c81a7c6ebdf9ed83189e58e
Reviewed-by: Giuseppe D'Angelo <giuseppe.dangelo@kdab.com>
Reviewed-by: Sean Harmer <sean.harmer@kdab.com>
When m_blitProgram is created the constructor will add it to
QOpenGLContext as a child of that QObject, so it will be deleted
when that QOpenGLContext is deleted as QObject will delete its
children in destructor. Delete it here will cause crash in
thread termination.
Change-Id: If9d3287d159cc3276b6a840a584a1b212b9c9fd0
Reviewed-by: Eskil Abrahamsen Blomfeldt <eskil.abrahamsen-blomfeldt@nokia.com>
On Mac the framework include paths were generated wrong.
The $${MODULE_NAME} was missing.
This patch fixes the framework include path generation.
Change-Id: Ic0e8d69ac7ac63be755302dc822c28240c9bc3d2
Reviewed-by: Jason McDonald <jason.mcdonald@nokia.com>
Reviewed-by: Miikka Heikkinen <miikka.heikkinen@digia.com>
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@nokia.com>
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@nokia.com>
On Lion scroll bars are within the scroll area itself and are not being
shown as long as the user is not scrolling.
This patch draws the new scroll bars and makes them fade away. Further
it introduces a pixel metric checking for this behaviour. It's used by
QAbstractScrollArea to put the viewport to the correct place.
Task-number: QTBUG-21673
Change-Id: Id530265043549318ac420b392de6b8642deaa4c6
Reviewed-by: Sean Harmer <sean.harmer@kdab.com>
Reviewed-by: Gabriel de Dietrich <gabriel.dietrich-de@nokia.com>
The \br macro, which is used to insert a line break in the output
text, is now promoted to be a qdoc command. For html, it outputs a
<br/>. For DITA XML it does nothing.
Task Nr: QTBUG-26435
Change-Id: Ie0542c9053f68473ccaa2f50ace74baa30f78dd0
Reviewed-by: Casper van Donderen <casper.vandonderen@nokia.com>
Create a native implementation of QCocoaDrag, using the 10.6 (and earlier) Cocoa
dragging API. This matches the implementation in Qt4 closely for the moment.
In the future it may be desirable to create an alternative implementation using
the new (non-blocking) drag API introduced in 10.7, but that will require deeper
changes to the mime-data handling.
This changes makes one more method on QPlatformDrag virtual, since the
Cocoa behaviour diverges from the base version: ::defaultAction is customised.
Change-Id: I1843293a62b2b4973a07b5e75ea3c312dc064018
Reviewed-by: Christoph Schleifenbaum <christoph.schleifenbaum@kdab.com>
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@nokia.com>
Reviewed-by: Morten Johan Sørvig <morten.sorvig@nokia.com>
Call the corresponding NSWindow size functions.
Add special cases where Qt returns null or empty sizes.
Change-Id: I880c50c04e92271dc342ad587191c8cb2c006dba
Reviewed-by: Bradley T. Hughes <bradley.hughes@nokia.com>
In 96f1fe8855 we agreed that sections
with negative sizes did not make sense. Of the same reason default
section sizes and minimum section sizes should be not negative.
Change-Id: I6a770e7f510d8e2bb90bfd8f38b4fa0566fc137b
Reviewed-by: Stephen Kelly <stephen.kelly@kdab.com>
The value is currently hard-coded to 1024, this is a problem since
any image painted with a size higher that this will be scaled down before
being uploaded.
This patch copies the implementation from QGLContext which works correctly.
Change-Id: Ia2bda60cf21d9adf13c91cea4854a2b20e4041f2
Reviewed-by: Kim M. Kalland <kim.kalland@nokia.com>
This fixes a bug on QEventDispatcherBlackberry, where registerSocketNotifier()
and unregisterSocketNotifier() wrongly assumed that a file descriptor was
being watched exclusively by one QSocketNotifier, while in fact there can be
more than one QSocketNotifier associated with a single file descriptor.
Change-Id: I0ce54bf809df109ad97f4a7f170f448d5d04d453
Reviewed-by: Sean Harmer <sean.harmer@kdab.com>
Now that Q_PROPERTY with a QObject derived type is more powerful.
This property can be used in QML so that wrappers for proxy models
do not need to be created, such as in the example
at https://codereview.qt-project.org/#change,13007
Change-Id: I6ba676549d2135585d429a28e214fef0b2a6b1f9
Reviewed-by: Olivier Goffart <ogoffart@woboq.com>
