This attempts to reconcile a minor difference between Qt 5 and Qt 6:
Running Qt Quick applications with a platform plugin such as vnc,
led to an automatic fallback to the 'software' backend based on
the OpenGL capability reported from the platform plugin. In Qt 6.0
this logic is gone from Qt Quick, because we do not have, and wish
not to have, individual flags for each and every 3D API on this
level.
Therefore in Qt 6.0 a Qt Quick application running with the vnc
(or linuxfb, or minimal) platform needs an explicit selection of
the software backend via QT_QUICK_BACKEND or
QQuickWindow::setGraphicsApi().
To keep migration for users of such platform plugins easier, we
can still reintroduce a Qt 5-like logic: by having a high level
Is-QRhi-Supported type of flag, we can make Qt Quick query that,
and trigger the fallback to the software backend when it is
reported as false by the platform plugin. As this is the minority
case and a conscious choice by platform plugins, the flag can be
opt-out (i.e. true by the default hasCapability implementation).
When it comes to the existing OpenGL flag, that needs to stay for
compatibility reasons, but it is worth noting that the new flag
semantically falls in the exact same category: it does not indicate
things will really work at run time, but rather serves as an opt-out,
"do not even try" type of declaration the platform plugin can make,
which then allows modules like Qt Quick to make early, upfront
decisions about which rendering paths/backends to take.
Change-Id: I8d6fddeb82ca6eece7b7abc1a5b64ebe6d8af29d
Task-number: QTBUG-89561
Reviewed-by: Andy Nichols <andy.nichols@qt.io>
Even though there is no D3D-specific logic in the windows platform
plugin, meaning a QWindow with either OpenGLSurface or VulkanSurface
(or anything really) is DXGI/D3D-compatible, it now looks like it is
beneficial, and more future proof, if there is a dedicated surface
type.
As the linked report shows, there are OpenGL-specific workarounds
accumulated in the platform plugin, while not being clear if these
are relevant to non-OpenGL content, or if they are relevant at all
still. (and some of these can be difficult/impossible to retest and
verify in practice)
When D3D-based windows use the same surface type, all these are
active for those windows as well, while Vulkan-based windows have
their own type and so some of these old workarounds are not active
for those. To reduce confusion, having a dedicated surface type for
D3D as well allows the logic to skip the old OpenGL workarounds,
giving us (and users) a more clear overall behavior when it comes
to OpenGL vs. Vulkan vs. D3D.
The change is compatible with any existing code in other modules
because any code that uses OpenGLSurface for D3D will continue to
work, using the new type can be introduced incrementally.
Task-number: QTBUG-89715
Change-Id: Ieba86a580bf5a3636730952184dc3a3ab7669b26
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
qt_plugins target is needed by qt_internal_add_plugin function, but when
standalone building sqldrivers, this target does not exist, which causes
configure error. So we need to add this target manually.
Fixes: QTBUG-89714
Pick-to: 6.0
Change-Id: Ia7d613faabb82399c2f3e5771aab24302b3c8f56
Reviewed-by: Joerg Bornemann <joerg.bornemann@qt.io>
And renamed PeekRemove* to PeekConsumed to document
better that nodes are not removed when peeking.
Change-Id: I7349f57fcfc3287d1a1309a31ee7f481f8d18755
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
The PostgreSQL driver can change it's connection status
after the first established connection.
In this case, the function "isOpen()" returns false
and the "close()" function would not free all resources.
With this fix, the "close()" function
always frees any allocated resource
independent of the connection status.
Fixes: QTBUG-88984
Pick-to: 6.0 5.15 5.12
Change-Id: I7c9add6a183bf46a8573952ab39f8cb1f728c00c
Reviewed-by: Andy Shaw <andy.shaw@qt.io>
The configure tests are all in place for CMake as well, and this one
has no special library depdendencies. Generate the CMakeLists and
remove the special case.
Change-Id: I55817045f9e259515fb3aad33d1446d9413111d9
Reviewed-by: Joerg Bornemann <joerg.bornemann@qt.io>
Make sure we use the new QXcbScrollingDevice class for the "core
pointer" as well.
Change-Id: I97575e8125cbb2aa280662aa675e61021c21ea59
Reviewed-by: Volker Hilsheimer <volker.hilsheimer@qt.io>
Change ac210c73e4 introduced the requirement that all input devices with
Scroll capability must have a QXcbScrollingDevicePrivate as their d_ptr.
However, this was not enforced, and would fail for the "Virtual core
pointer".
To fix this, always use qobject_cast to verify that the device is of the
correct type.
Change-Id: I4a6b1d4d79308eb04e9f52dda00294fffe377bdf
Reviewed-by: Volker Hilsheimer <volker.hilsheimer@qt.io>
Commit f4a66e5c accidentally made it so that the offscreen
plugin would produce a null QPlatformIntegration on
X11 if QT_QPA_OFFSCREEN_NO_GLX is set.
Restore previous logic where the generic QOffscreenIntegration
is created if the X11 code path did not create an object.
Change-Id: Ic71fb04bbd824f4121480dbaf8fa0571ebb39f42
Pick-to: 5.15
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
Alignment problems occur for pushbuttons because they appear smaller then their
actual widget geometry, which is used for their alignment in layouts. To fix,
shift the pushbutton's rect to the left, adjust SE_PushButtonLayoutItem
accordingly and use QFocusFrame to render the focus frame ring outside
the widget's paintable area.
Fixes: QTBUG-89133
Fixes: QTBUG-81452
Pick-to: 5.15 6.0
Change-Id: Iee885a4fb3674d966e5ff3b5c04a0845521b2d72
Reviewed-by: Volker Hilsheimer <volker.hilsheimer@qt.io>
KEYCODE_MEDIA_PLAY_PAUSE was incorrectly translated to Qt.Key_MediaPlay.
This patch changes it to Qt.Key_MediaTogglePlayPause.
Fixes: QTBUG-66448
Pick-to: 5.15 5.12 6.0
Change-Id: Icd2747857bc479228bfdffcd8ab8fdef7b79e918
Reviewed-by: Andy Shaw <andy.shaw@qt.io>
This patch amends a41701904e
If peeking into the event queue looking for a clipboard event fails,
QXcbClipboard::waitForClipboardEvent() calls queue->peek for the second
time to "process other clipboard events, since someone is probably
requesting data from us". QXcbEventQueue::peek() in turn calls
QXcbEventQueue::flushBufferedEvents(). This second flushing can acquire
a waited-for clipboard event. The issue was that the code in
waitForNewEvents() ignored this possibility and assumed that there were
no clipboard events before or at its current m_flushedTail. If there
were no more events on the X11 connection after tailBeforeFlush,
the waitForNewEvents() in waitForClipboardEvent() blocked execution
for 5 seconds and eventually timed out.
The fix is to remember QXcbEventQueue::m_flushedTail just after looking
for and not finding a clipboard event in the queue. And then wait for
more events via QWaitCondition in waitForNewEvents() only if there were
no more events after the remembered m_flushedTail.
Fixes: QTBUG-75319
Pick-to: 5.15
Pick-to: 5.12
Change-Id: I4919c5b9b9227b3a8a29a11e7094f97960b3a121
Reviewed-by: Gatis Paeglis <gatis.paeglis@qt.io>
When accessibility was not active it was possible to force its
update, causing a null pointer exception in java.
Pick-to: 6.0 5.15
Change-Id: I60f65612d6e9f712ba3c591ffcc6803f06f09efb
Reviewed-by: Assam Boudjelthia <assam.boudjelthia@qt.io>
The pro2cmake script doesn't handle static scopes correctly, and the
generated "CONDITION static" must read "CONDITION NOT
QT_BUILD_SHARED_LIBS".
Change-Id: Ic5a5763e9fd5f5d04fb4c12227cbe85fac17b826
Reviewed-by: Cristian Adam <cristian.adam@qt.io>
It can happen that updateApplicationState() is called before
m_androidPlatformIntegration is set. In such case, new applicationState
will be remembered and propagated later.
Before this change, pending application state was passed only to
QWindowSystemInterface.
ResumePauseListeners that are subscribed for changes to
ApplicationActive/ApplicationInactive state also should be informed
about new application state.
Fixes: QTBUG-84737
Pick-to: 6.0 5.15
Change-Id: I67e79860b340ee5de2d13d148a222e9f1c942b93
Reviewed-by: Rami Potinkara <rami.potinkara@qt.io>
Reviewed-by: Assam Boudjelthia <assam.boudjelthia@qt.io>
Until now, all wheel events came from one "core pointer". It's useful
in Qt Quick to tell the devices apart, because some support smooth scrolling
and some don't.
Also remove the QHash storing legacy ScrollingDevice structs, and use
a subclass of QPointingDevicePrivate instead.
Task-number: QTBUG-46412
Task-number: QTBUG-72167
Task-number: QTBUG-69433
Change-Id: Ie6a3d8dd494f981e8706b9a66a1021dfb51baec4
Reviewed-by: Liang Qi <liang.qi@qt.io>
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: Volker Hilsheimer <volker.hilsheimer@qt.io>
QXcbVirtualDesktop::m_workArea is not updated after screen rotation
if _NET_WORKAREA is not updated, e.g. when a window manager is not
running.
Use width of -1 for m_workArea to indicate that _NET_WORKAREA is
not set and thus there is no need to find its intersection with
screens' geometries.
Pick-to: 5.15 6.0
Change-Id: Ic1f3ec23576ed8407a76cc2b9f84a145c67b1d4b
Reviewed-by: Liang Qi <liang.qi@qt.io>
Amends 4d943846128118e1b9932a17ce6f977a0f4127a5:
- The macOS style must now elide the text by it's own since it is
no longer done by QHeaderView
- Add documentation of QStyleOptionHeader::textElideMode
- Remove unused variables from QHeaderView::initStyleOptionForIndex()
Task-number: QTBUG-86426
Change-Id: I98fc6771c0cd56d6002390125ffbab1269f6dd39
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
NSTitleBarView in Big Sur will observe the window it is in, but makes
sure to remove these observers in viewWillMoveToWindow, as it should.
To avoid spamming user logs with this warning we remove it.
It was dubious what cases it would have caught anyways, as anyone
trying to KVO observe properties Z of a child object Y of X.Y would
need to track whether X changes its Y between adding and removing
the observer. There are no guarantees that Y is stable, and this
applies to an NSView's window property as well.
Pick-to: 6.0
Pick-to: 5.15
Pick-to: 5.12
Change-Id: I27a0d04625f96c4c5d382c39ac068721ce1e8f9d
Reviewed-by: Alexandru Croitor <alexandru.croitor@qt.io>
- Accessibility focus can follow the position of the widget
(for example when swiping on a scrollview)
- controls are clickable directly after appearing on the screen after
scroll (previously you had to click somewhere else on the screen,
and after that you could focus the newly appeared control)
- checkbox and switch react correctly on click action
- fixed combobox behavior with accessibility enabled
Task-number: QTBUG-79611
Pick-to: 6.0 5.15
Change-Id: If36914ab0165f33593e68fd7ecf168693f8538a7
Reviewed-by: Assam Boudjelthia <assam.boudjelthia@qt.io>
The `restartCount` variable on line 245 is mutating
global scope. The PR makes it consistent with the
rest ob the code (`self.restartCount`).
It was observed when importing the qtloader in a
typical Webpack/Babel build environment.
Change-Id: I338285f4f6bcb80df0c16d80cc3ebfec944a8be7
Reviewed-by: Morten Johan Sørvig <morten.sorvig@qt.io>
Amends 1535fc9fb9 : when high-dpi scaling
is enabled, there was an offset from the cursor position to the event
position, because QWindow::mapFromGlobal() works in device-independent
pixels, but we are using actual screen pixels here.
Pick-to: 6.0
Pick-to: 5.15
Pick-to: 5.12
Task-number: QTBUG-77826
Change-Id: Ic8743b9e5c4041065f530ed1d9d6c49337b0207a
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@qt.io>
Reviewed-by: Morten Johan Sørvig <morten.sorvig@qt.io>
f4889e63c7 changed the worksWhenModal logic for QNSWindow to
be based on the transient parent relationship of the child window
to the modal session window, to fix many issues where windows that
should be blocked were not.
Unfortunately, some window types do not maintain a transient parent
relationship (e.g. QCompleter, which is itself just a QObject), or
are not common for users to create with a QWidget parent (such as
a context QMenu).
This change restores part of the special-casing that was removed in
f4889e, so that all popup windows are always allowed to be interacted
with during modal sessions. This includes popup windows that were
opened as part of a parent modal session, which would normally be
fully blocked, but we assume that popup windows are intermittent
enough that this will not be a problem.
For now we leave out the other two special casings from f4889e,
namely tool windows and dialogs. The former should in most cases
be created with a parent window, while the latter definitely should.
Fixes: QTBUG-88188
Fixes: QTBUG-88985
Fixes: QTBUG-87849
Fixes: QTBUG-86845
Pick-to: 6.0
Pick-to: 5.15
Change-Id: I005a402b21e8dc16c3b18bcd7e67d12b94a66f44
Reviewed-by: Volker Hilsheimer <volker.hilsheimer@qt.io>
While it worked as expected with beta, with final release this thing
looks wrong again, especially in the dark mode.
Pick-to: 5.15
Pick-to: 5.12
Change-Id: I6857344de897e0f39d9d8e66431635dbe2f4d077
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
QXcbConnection::getTimestamp uses dummy events to get timestamp from
X server. However, in some cases, X server shuts down while client tries
to get timestamp. In this case, QXcbConnection::getTimestamp keeps
getting null event and thus falls into indefinite loop.
This fix checks if xcb connection is still valid and use a special
xcb_timestamp_t value, CurrentTime (0L), as returned value.
CurrentTime should not be generated by X server and if getTimestamp
returns this value, it means an "exception" case is triggered.
This fix is introduced because in kwin_x11 (KDE project), X server can
exit on logout. kwin_x11 should handle disconnection from X server.
But the indefinite loop prevents kwin_x11 to process disconnection
event and therefore kwin_x11 cannot quit properly.
Fixes: QTBUG-88435
Pick-to: 5.12 5.15 6.0
Change-Id: Iaf7ef3f8a35fa8389d22a608e3c49041bf90e1b9
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: Liang Qi <liang.qi@qt.io>
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
Q_MOVABLE_TYPE was conceived before C++ had move semantics. Now, with
move semantics, its name is misleading. Q_RELOCATABLE_TYPE was
introduced as a synonym to Q_MOVABLE_TYPE. Usage of Q_MOVABLE_TYPE
is discouraged now. This patch replaces all usages of Q_MOVABLE_TYPE
by Q_RELOCATABLE_TYPE in QtBase. As the two are synonymous, this
patch should have no impact on users.
Pick-to: 6.0
Change-Id: Ie653984363198c1aeb1f70f8e0fa189aae38eb5c
Reviewed-by: Lars Knoll <lars.knoll@qt.io>
Reviewed-by: Andrei Golubev <andrei.golubev@qt.io>
Add the overloads for mouse events with device/without timestamp
and pass the active tablet or touch device.
Task-number: QTBUG-88678
Task-number: QTBUG-46412
Pick-to: 6.0
Change-Id: I8695b493540d0cbf50e9c72afe870a7633de3ab9
Reviewed-by: Oliver Wolff <oliver.wolff@qt.io>
The scope of the variable 'currBind' can be reduced if the variable 'r'
is not 0. So declare the variable when the variable 'r' is 0. The local
variable 'i' shadows outer variable in fetchNext(), so move it to the
front of switch.
Don't declare 'res' until we need and initialize it.
Change-Id: Idfb220b96cfbcd4088fd7858ed9392d0a3e10aea
Reviewed-by: Volker Hilsheimer <volker.hilsheimer@qt.io>
Two unit tests fail at the moment with the offscreen theme because
the font function of the offscreen theme does not return a monospace
font when asked for a monospace font. This patch fixes it.
Change-Id: Id682c75e2a94b246cffe97b342b1b562bac82a68
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: Andy Shaw <andy.shaw@qt.io>
Reviewed-by: Shawn Rutledge <shawn.rutledge@qt.io>
The scope of the variables 'bind', 'field', 'i' can be
reduced if the variable 'meta' is a null pointer. So
declare the variables when the variable 'meta' is valid.
Change-Id: I599406c034168dd73799c69dfb9c5e676861cdf2
Reviewed-by: Andy Shaw <andy.shaw@qt.io>
This also fixes the cursor changing during resize mode
Fixes: QTBUG-85361
Pick-to: 5.15
Change-Id: Ic85a5c31a8a2dc4265a84f16fd5fcdc231062c6d
Reviewed-by: Morten Johan Sørvig <morten.sorvig@qt.io>
Two issues here, timestamp value was getting smashed, and
MouseButtonDblClick was not getting sent.
Fixes: QTBUG-85712
Pick-to: 5.15
Change-Id: I912e968f1eb0fb9c6f4cf8548b114e23d182396e
Reviewed-by: Morten Johan Sørvig <morten.sorvig@qt.io>
If a target inherits precompile headers from a different target
that has more languages enabled (CXX, OBJCXX), the target will
depend on PCH artifacts for a language that is not actually
present.
Ammends d9d137c8f3
Fixes: QTBUG-88609
Pick-to: 6.0.0 6.0
Change-Id: I9735317e634b62732df0fce84fe7cf229cfdea1b
Reviewed-by: Alexandru Croitor <alexandru.croitor@qt.io>
When preparing a system tray icon on a system with a retina screen,
we end up creating a full-height icon of height 44. If there's also
a 1x screen available, macOS will scale down this image for us when
presenting the icon on the 1 screen, but the downscale will fail to
preserve the aspect ratio of the original image on Big Sur.
Telling macOS which device-pixel-ratio the image has seems to fix
this, and is the right thing to do in any case.
Pick-to: 5.15
Pick-to: 5.12
Fixes: QTBUG-88600
Change-Id: Ic31def94d073da1b256bbfcaf5905f92c169f43d
Reviewed-by: Alexandru Croitor <alexandru.croitor@qt.io>
By depending on setFamilies() then we can be sure that font names with
spaces, commas, quotes and so on are correctly handled without being
misinterpreted. For now it will split on the comma when a string
containing one is passed to setFamily. But from Qt 6.2 this will be
removed to preserve the family string as a convenience function.
[ChangeLog][QtGui][QFont] Indicated that setFamilies/families is
preferred over setFamily/family to ensure that font family names are
preserved when spaces, commas and so on are used in the name.
Change-Id: Id3c1a4e827756a4c928fed461a4aafa5a0f06633
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: Eskil Abrahamsen Blomfeldt <eskil.abrahamsen-blomfeldt@qt.io>
Polymorphic classes should not be copied.
However, we do rely on event copying in our propagation logic. So, make the
members protected, don't delete them, using a dedicated macro.
This way, QMutable*Event classes can be used to make copies.
Remove some last usage of copying of QInputMethod(Query)Events.
Change-Id: Ia0a8ae4ca9de97dcd7788ca3c6ed930b6460c43a
Reviewed-by: Lars Knoll <lars.knoll@qt.io>
Extend the scope of the resources variable to avoid its
destruction while it's still in use.
Fixes: QTBUG-88512
Pick-to: 5.15 5.12
Change-Id: I3298aabc871ff455bd1203ec276e7600d3e151ef
Reviewed-by: Liang Qi <liang.qi@qt.io>
Reviewed-by: Shawn Rutledge <shawn.rutledge@qt.io>
To build with eglfs module and gbm device integration module,
some headers should consider global path, not local.
This covers eglfs_kms and eglfs_kms_egldeivce device integration.
Other device integrations might be future work if needed.
Task-number: QTBUG-85268
Change-Id: I1bad5fbac99aa79d146e90f88b53519b09254e13
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: Laszlo Agocs <laszlo.agocs@qt.io>
Reviewed-by: Elvis Lee <kwangwoong.lee@lge.com>
A QWindow created without an explicit geometry needs to pick up
a default geometry in the platform plugin. If the window has a
maximized of fullscreen window state, it will fill the entire
available geometry of the parent window (or the screen if there's
not parent window).
Fixes: QTBUG-69159
Fixes: QTBUG-69156
Fixes: QTBUG-69154
Pick-to: 5.15
Change-Id: If8565d92a97bb4b3fa44757e68969d54d0bc7ebe
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
These look like leftovers (API flaws).
Construction of QFileInfo from QString (or similar) should be not
implicit, as QFileInfo construction is expensive (might hit the file
system), and this may have users overlook APIs (for instance build a
QFileInfo out of QDirIterator::next(), instead of using ::fileInfo();
using QDir::entryList instead of entryInfoList; etc.).
Leave an opt-out mechanism to ease porting.
Fix a handful of usages around qtbase, with at least a couple of them
likely to be actual "sloppy" code.
[ChangeLog][Potentially Source-Incompatible Changes][QFileInfo] Most
QFileInfo constructors are now explicit. The
QT_IMPLICIT_QFILEINFO_CONSTRUCTION macro is provided to keep old code
working.
Change-Id: Ic580e6316e67edbc840aa0c60d98c7aaabaf1af6
Reviewed-by: Volker Hilsheimer <volker.hilsheimer@qt.io>
QChar should not be convertible from any integral type except from
char16_t, short and possibly char (since it's a direct superset).
David provided the perfect example:
if (str == 123) { ~~~ }
compiles, with 123 implicitly converted to QChar (str == "123"
was meant instead). But similarly one can construct other
scenarios where QString(123) gets accidentally used (instead of
QString::number(123)), like QString s; s += 123;.
Add a macro to revert to the implicit constructors, for backwards
compatibility.
The breaks are mostly in tests that "abuse" of integers (arithmetic,
etc.). Maybe it's time for user-defined literals for QChar/QString,
but that is left for another commit.
[ChangeLog][Potentially Source-Incompatible Changes][QChar] QChar
constructors from integral types are now by default explicit.
It is recommended to use explicit conversions, QLatin1Char,
QChar::fromUcs4 instead of implicit conversions. The old behavior
can be restored by defining the QT_IMPLICIT_QCHAR_CONSTRUCTION
macro.
Change-Id: I6175f6ab9bcf1956f6f97ab0c9d9d5aaf777296d
Reviewed-by: Volker Hilsheimer <volker.hilsheimer@qt.io>
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
Laszlo Agocs
Li Xinwei
Gatis Paeglis
Matthias Möller
Paul Olav Tvete
Lu Zhen
Morten Johan Sørvig
Friedemann Kleint
Doris Verria
Piotr Mikolajczyk
Igor Kushnir
Joerg Bornemann
Bartlomiej Moskal
Shawn Rutledge
Alexander Volkov
Andy Shaw
Christian Ehrlicher
Tor Arne Vestbø
Avindra Goolcharan
Timur Pocheptsov
Sheng Mao
Andreas Buhr
Zou Ya
Huang Jie
Lorn Potter
Cristian Adam
Volker Hilsheimer
Alexander Volkov
Elvis Lee
Zhang Sheng
Giuseppe D'Angelo
Allan Sandfeld Jensen