Only return a string for the attribute value
(an empty string if the attribute is not set).
The previous implementation was based on the incorrect
signature in the XML spec for the AT-SPI Text interface from
before the AT-SPI commit that removed the extra out params [1]:
commit 8786849ce6e9914383aa766ff9ce7e00f5b2178d
Author: Patryk Kaczmarek <patryk.k@samsung.com>
Date: Mon Sep 28 14:23:15 2015 +0200
Fixed atspi_text_ functions
* atspi_text_get_text_attribute_value
Fixed dbus signature in _atspi_dbus_call function
and add missing argument for string.
* atspi_text_get_default_attributes
Receiving return value by reference from hash table
https://bugzilla.gnome.org/show_bug.cgi?id=755731
[1] 8786849ce6
Fixes: QTBUG-105752
Pick-to: 6.4
Change-Id: I66d3c484ecc4b469684635723242c915e4365e6a
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
This adds support for setting parent window for GTK dialogs on Wayland
using the new API added for portals
Change-Id: I29085c926ce4338ff2ad75728d566ec843d3aa5a
Reviewed-by: Liang Qi <liang.qi@qt.io>
We currently build Qt for simulator using X86_64, even
on ARM based macs. This results in the simulator running
on ARM, while the app is running inside it using Rosetta.
And with this combination, the event.timestamp, which is
documented to be in seconds, looks to be something else, and
is not progressing in sync with a normal clock.
Sending out mouse events with a timestamp that doesn't follow
normal clock time will cause problems for mouse-, and pointer
handlers that uses them to e.g calculate the time between a
press and release, and to decide if the user is performing a tap
or a drag.
For that reason, we choose to ignore UIEvent.timestamp under
the mentioned condition, and instead rely on NSProcessInfo.
Note that if we force the whole simulator to use Rosetta
(and not only the Qt app), the timestamps will progress normally.
Fixes: QTBUG-105810
Pick-to: 6.4 6.3 6.2
Change-Id: Ib4e60593cac3230567ebc53d634f434fcefc98d0
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
Clarify that the hyphen is the actual character that one can use.
The way the documentation was written might have been misleading.
Pick-to: 6.4
Change-Id: I55a9ede9903a8ac5bd33d8e138b3a0a21a820406
Reviewed-by: Paul Wicking <paul.wicking@qt.io>
The default camera to plane distance is 1024, when rotating a big image
along the x or y axis, some areas of the screen may move above the
camera, causing the rotation to fail. A new rotation interface has been
added to allow users to specify the distance from the camera to the
plane themselves when rotating the QImage. Also, this support has been
added to QMatrix4x4::projectedRotate.
[ChangeLog][QtGui][QTransform] Added overloads to rotate() and
rotateRadians() that allow specifying of the distance to the rotation
plane.
Fixes: QTBUG-105088
Change-Id: I81f629916ddd9b6ab84e0282191e4284a88a85f5
Reviewed-by: Eirik Aavitsland <eirik.aavitsland@qt.io>
Not just the ones we added to the pending sockets list
Pick-to: 6.4
Change-Id: I0a0016fe39df7ca2fc3f0c4e4111195bc6d90198
Reviewed-by: Konrad Kujawa <konrad.kujawa@qt.io>
Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
If a client doesn't send any data then we would leave the socket open
for as long as it needed, wasting resources. Add timeouts to limit the
amount of time this can happen for.
Since there is a limit on number of sockets that the server will have
queued, having idle sockets stick around forever is a vector for ddos.
Pick-to: 6.4
Change-Id: Ida6251c92c625eeadf2065861b840b14255654b8
Reviewed-by: Ievgenii Meshcheriakov <ievgenii.meshcheriakov@qt.io>
Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
Reviewed-by: Volker Hilsheimer <volker.hilsheimer@qt.io>
SecureTransport ignores any content that comes in until it is large
enough to be a handshake. So a plaintext client may be left hanging
while it is waiting for a response.
Pick-to: 6.4
Change-Id: I501ae61d89d516765c7ba5f0d916d9246fde5d4d
Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
Then we don't have to do a 'global' disconnect() on the socket object
just to disconnect the QSslServer from the socket.
Pick-to: 6.4
Change-Id: Ie3c9680db2717e21a0c548c74374a58d533534fe
Reviewed-by: Konrad Kujawa <konrad.kujawa@qt.io>
Reviewed-by: Timur Pocheptsov <timur.pocheptsov@qt.io>
Uses CMake's add_test and fixtures to call qmake and make,
thus allowing to create tests that ensure that building qmake projects
works.
Pick-to: 6.4
Task-number: QTBUG-96058
Change-Id: I6f03c4389896f9571bb4f9551f93295075c3d8e1
Reviewed-by: Jörg Bornemann <joerg.bornemann@qt.io>
QAccessibleTableCellInterface provides everything
needed to support the AT-SPI TableCell interface [1].
Therefore, expose that AT-SPI interface and implement
handling of the corresponding methods.
[1] https://gitlab.gnome.org/GNOME/at-spi2-core/-/blob/master/xml/TableCell.xml
Fixes: QTBUG-104793
Change-Id: Ie7068c029eaf911186daf3956f11cfd4eb2800a6
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
... fixing a int/qsizetype mismatch in the process.
As a drive-by, replace SESE with an early return.
Pick-to: 6.4 6.3 6.2
Task-number: QTBUG-103525
Change-Id: I961728ba818c4cb43c0e2c9056c70c8f37ad042e
Reviewed-by: Sona Kurazyan <sona.kurazyan@qt.io>
It takes qint64, not int.
This is not public documentation, so not picking to older branches.
Task-number: QTBUG-103525
Change-Id: I4303af326bb6054ae0cccd87aa337354ad11209b
Reviewed-by: Sona Kurazyan <sona.kurazyan@qt.io>
And use the new overloads in examples and tests.
[ChangeLog][QtXml][QDomDocument] Deprecated the old setContent()
overloads in favor of the new ones that take ParseOptions and
ParseError.
Task-number: QTBUG-104507
Change-Id: I61b37eba2fe3002c03bddc90f6877676d539f7ec
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>
Esp. on Windows, it can be desirable to deploy an application and its
DLLs directly in the install prefix, without a bin directory. To do
that, one must set CMAKE_INSTALL_BINDIR to ".", but that resulted in a
faulty prefix entry in the generated qt.conf.
Check for this case when generating qt.conf to write the correct prefix.
Fixes: QTBUG-105583
Pick-to: 6.3 6.4
Change-Id: I0e8295c70b48b991c19f58f6b3f2ed132112dd29
Reviewed-by: Alexandru Croitor <alexandru.croitor@qt.io>
The focused state exists in both, Qt
(QAccessible::State::focused) and AT-SPI, so forwarding
changes to that state is straightforward.
However, so far, changes to the focused state in a
QAccessible::StateChanged event were just ignored
and not forwarded to the AT-SPI layer.
While announcing focus changes to Qt's own widgets works
without that because it uses events of type QAccessible::Focus
instead of QAccessibleStateChangeEvent with the focused
state set, this is not exactly the same and just ignoring the
focused state in QAccessible::StateChanged events e.g. turned out to
be one cause for the Orca screen reader not announcing UI elements
when they received focus in the Qt-based UI variant of LibreOffice,
which uses QAccessible::StateChanged with the focused flag
set (s.a. LibreOffice commit [1]).
[1] https://git.libreoffice.org/core/commit/8c3e8af0e60865ec6d38e2117efdb4ed2f10a20c
Change-Id: If226af28960048c56af32979477605d6860e30e9
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
It seems QWindow here is only for making the dialog modal,
but QDialog already handles that and this makes two modals
block each other depending on the order they created with
Task-number: QTBUG-98988
Pick-to: 6.4 6.3 6.2 5.15
Change-Id: I6847cfab480395f62eaa0ebf79acf8b024192178
Reviewed-by: David Edmundson <davidedmundson@kde.org>
Reviewed-by: Liang Qi <liang.qi@qt.io>
Reviewed-by: Dmitry Shachnev <mitya57@gmail.com>
With NTLM/Negotiate we delete the context used to generate replies once
we get SEC_E_OK. Due to some faulty logic in the http backend we could
end up trying to generate another response. Qt would then pass
references to some offsets of nullptr into the API calls causing it to
crash. Add some sanity checks before the "sspi continue" calls to make
sure this won't happen, and update the condition in the http
backend to check that we have not already sent our credentials.
As a drive-by: correct the initialization of the handles to use
SecInvalidateHandle instead of memset to 0.
Pick-to: 6.4 6.3 6.2 5.15
Fixes: QTBUG-102359
Change-Id: I884ff8fc70609fe8746b99a1d56eeafcda9d2620
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
When the xsettings value of relevant style hints is changed
we propagate it to QGuiApplication via handleThemeChange.
Change-Id: I316c90e776f52e92c1249aa4e1fcb3ced331f8f5
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
Just like for the "GetSummary" case, the argument in the
reply for this AT-SPI table method needs to
be wrapped in a variant.
Fixes this warning from the output of the client side
using libatspi:
> (accerciser:181933): dbind-WARNING **: 11:50:36.394: atspi_dbus_get_property:
> expected a variant when fetching Caption from interface org.a11y.atspi.Table; got (so)
Fixes: QTBUG-105520
Pick-to: 6.4 6.3 6.2
Change-Id: I18167359e4cd3bb14a94289ac7481f9e39a18ad0
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
So far, only screen coordinates (`ATSPI_COORD_TYPE_SCREEN`)
and coordinates relative to the widget's top-level window
(`ATSPI_COORD_TYPE_WINDOW`) were supported.
In at-spi 2.30, a third coordinate type, describing
coordinates relative to the widget's immediate parent,
was added: `ATSPI_COORD_TYPE_PARENT` (commit: [1])
This commit adds the handling to convert to and
from that coord type as well and unifies the
existing handling to convert to/from screen coordinates
by introducing two static helper methods
(`AtSpiAdaptor::translateToScreenCoordinates` and
`AtSpiAdaptor::translateFromScreenCoordinates`) and
making use of those wherever conversion needs to
be done.
In addition, also add a check to only handle
requests using a coordinate type that is one of those
that is actually supported instead of returning
incorrect values (for the case that new coord types
should be introduced in AT-SPI in the future).
[1] 737c9853b6
Fixes: QTBUG-105313
Change-Id: Ia8752bd6a35328cc2de33ecad08f2964735f41ae
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
State the problem and suggest possible solutions for the
app developer.
Pick-to: 6.4
Change-Id: I0484c4eb31020ec0016005f43f910f32a319b850
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
Emscripten's option for enabling asyncify (-sASYNCIFY) is a link-time
option, which means there is no requirement to have a separate asyncify
build, at least for static builds.
Replace the current QT_HAVE_EMSCRIPTEN_ASYNCIFY compile-time option
with a run-time option which checks if the asyncify API is available.
Keep support for configuring with "-device-option QT_EMSCRIPTEN_ASYNCIFY=1"
for backwards compatibility and for the use case where want asyncify
support to be on by default for a given Qt build.
Enable asyncify for the asyncify_exec example.
Pick-to: 6.4
Change-Id: I301fd7e2d3c0367532c886f4e34b23e1093646ad
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
The call to this function was behind a broken #define and
was never enabled. It also does not work, so remove it.
This means that calling processEvents() in a loop (as opposed
to calling exec()) most likely does not work either, which
is something we want to revisit.
Pick-to: 6.4
Change-Id: I90802a69d3a48e7f2b5d0b657d89452c09d9571a
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
The coordinate type for "GetOffsetAtPoint" describes the
coordinate type of the input coordinates and
`QAccessibleTextInterface::offsetAtPoint` expects the
coordinates in screen coordinates.
Therefore, if window-relative coordinates are given,
the screen coordinates of the window need to be added
(not subtracted) to the window-relative coordinates of the point
in order to calculate the screen coordinates of the point.
Fixes: QTBUG-105281
Pick-to: 6.4 6.3 6.2
Change-Id: I5517335c0c0d6a47b8c45c7e4f8f8b5a91f0fcd7
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
Since a dialog is also a top-level window, it should be taken
into account when trying to find the accessible's top-level
window, relative to which the position will be calulated for
`ATSPI_COORD_TYPE_WINDOW`.
Fixes: QTBUG-105042
Pick-to: 6.4 6.3 6.2
Change-Id: I74fae096fd886bab04187c122f268c26c91b86ab
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
The minimal platform plugin does not have any native APIs it could offer
through that interface, but needs to provide an instance anyway. A lot
of code within Qt assumes that a native interface is always available
and does not check the pointer value, which leads to segfaults when used
with the minimal plugin.
Pick-to: 6.4 6.3 6.2 5.15
Change-Id: I625bd95afd49872cff3a34b29ceb0ffbbcd39db7
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
error: right shift count >= width of type [-Werror=shift-count-overflow]
Pick-to: 6.4
Fixes: QTBUG-105687
Change-Id: Ic6547f8247454b47baa8fffd170bd9cd0e2a8ca3
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
This led to an infinite recursion in case the annotation was completely
missing. Instead of trying to fix that, I'm simply implementing the
"### Qt6" request from c62f717226 .
[ChangeLog][qdbusxml2cpp] Removed the old compatibility code that
accepted "In" annotations for signal argument names, introduced in Qt
5.8.
Pick-to: 6.4
Fixes: QTBUG-104722
Change-Id: Ie4bb662dcb274440ab8bfffd1709bfc3daf0846d
Reviewed-by: David Faure <david.faure@kdab.com>
Some lines were suggesting the method name as context, while most of the
others were the parameter or property name.
Pick-to: 6.4
Change-Id: Ie4bb662dcb274440ab8bfffd1709c0a71abc27dc
Reviewed-by: David Faure <david.faure@kdab.com>
The function name was easy to mistake for a function you should call
after adding fonts via e.g. registerFont or registerFontFamily, but
in reality it's meant to be used when the platform font database
knows that fonts have changed in a way that needs a full populate.
Pick-to: 6.4
Change-Id: I4a897944aaa5df891562f729440b5d98b54f61e0
Reviewed-by: Eskil Abrahamsen Blomfeldt <eskil.abrahamsen-blomfeldt@qt.io>
The old stack structure used to keep track of windows has been improved
to conform to the actual windowing assumptions: there shall be one root
window, which is always at the bottom. The first created window
immediately becomes the root window. Should the root window be removed,
all windows are non-root, i.e. any of them can become the top-level window
Fixes: QTBUG-105094
Pick-to: 6.4
Change-Id: Ic553244fa9f5bc3ee590b702935e66cfc62d5f8f
Reviewed-by: Morten Johan Sørvig <morten.sorvig@qt.io>
Comparing the backends revealed that one cannot expect that
exposing a storage buffer or a texture to a shader with
bufferLoad, imageLoad, bufferStore, imageStore, etc. will
always be functional in any graphics and compute pipeline
stage. In fact the only place where this is universally
supported are compute shaders.
In other shaders some backends will not expose the resources
at all, e.g. because in D3D11 unordered access views (for a
buffer or image used with load/store) have limitations on the
API level, whereas others may have runtime limits, e.g. with
OpenGL ES an implementation may just say
GL_MAX_VERTEX_SHADER_STORAGE_BLOCKS is 0 meaning no SSBOs in
vertex shaders. (apparently the case on some embedded GPUs,
presumably due to the tiled architecture, e.g. on Mali)
So for now add a note to all the relevant
QRhiShaderResourceBinding functions.
It should be possible to make it work with fragment shaders
to a degree at least, but for D3D we then have to deal with
OMSetRenderTargetsAndUnorderedAccessViews and some remapping
of binding points, whereas elsewhere they may be issues with
missing or incorrect barriers. So do not go there now.
Change-Id: Ib18949e0184626a9abf5bb72c6ef72bc1cb2e1fa
Reviewed-by: Andy Nichols <andy.nichols@qt.io>
No matter what the surface type was. We only support VulkanSurface, so
if the type is something else, then a a qWarning is in order, but instead
of returning nullptr (which will almost certainly cause a fail or crash
in some form), we can just carry on normally. The resulting window
will not be renderable from the application of course, but that is
not always needed in fact, plus failing gracefully is nicer.
The tst_qrhi autotest can now be run on the RPi4 also with
-platform vkkhrdisplay. While there are some odd messages occurring
from Mesa and co., the run now completes normally, with no failures.
Change-Id: I77bd249096aede8e05458559458581d702054540
Pick-to: 6.4
Reviewed-by: Andy Nichols <andy.nichols@qt.io>
Looks like something inherited from the Qt Quick code that got
moved into the helper. Setting a request of API version 1.2 is
harmless in practice, but to be correct the version should be
1.1 when 1.2 and newer are not reported by the instance.
Change-Id: I62d2c143972d6753e4311c6907ccf846ae0e9088
Pick-to: 6.4
Reviewed-by: Andy Nichols <andy.nichols@qt.io>
Other platforms name it singular 'qt.qpa.window'.
Change-Id: I668ed67e1686605fe5f77313c7a01c31fd574c32
Reviewed-by: Friedemann Kleint <Friedemann.Kleint@qt.io>
This also fixes 0-used-as-nullptr warnings from clang-tidy.
Pick-to: 6.4 6.3 6.2 5.15
Change-Id: Ibf10f1719a6c8cf22046cf30298dadb307a853d5
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
There's no good replacement yet for our use of GetIconRef
and ReleaseIconRef, so for now ignore the deprecation.
Pick-to: 6.2 6.3 6.4 5.15
Change-Id: Iffcaa2af3c9e2ee053303a2272e8874913dd74d4
Reviewed-by: Alexandru Croitor <alexandru.croitor@qt.io>
If a font family has several instances from different foundries,
we disambiguate this by adding the foundry name in brackets behind
the family. But QFontDatabase::hasFamily() would only check for
families().contains(familyName). So if the database contains e.g.
Foo [Bar] and Foo [Baz] then a check for hasFamily("Foo") would
fail.
So we need to actually check for the family name instead. In
doing this, we also skip the extra step of building the list
and then searching it, but just go directly to the source.
This removes the BLACKLISTing of Ubuntu and also introduces a
QSKIP on Unix-based platforms without fontconfig, since there
is no way to know which default fonts are acceptable on those
platforms.
Pick-to: 6.4
Fixes: QTBUG-86967
Change-Id: Id8ad80a1671daf1c14fbad8bb8f4c51ee1c59709
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
The exact version doesn't add much information, and will most likely
go stale again; so just remove the mentioning.
Change-Id: Ifd87f7cf655a610aa45e341cf5a3491ae7d91171
Reviewed-by: Jörg Bornemann <joerg.bornemann@qt.io>
Zero reason to do this query, but the validation layer complains otherwise.
Pick-to: 6.4 6.3 6.2
Fixes: QTBUG-105527
Change-Id: I7a72e4d6d89f41279db94efe663a696302f05ded
Reviewed-by: Andy Nichols <andy.nichols@qt.io>
The caveat being having to manually create HLSL versions of the hull,
domain, and geometry shaders in parallel with the Vulkan GLSL ones,
while keeping the interfaces intact (stage inputs and outputs, cbuffer
layouts, binding points/registers). This is not always trivial but
typically doable in not very complicated case after inspecting the
SPIRV-Cross-generated vertex/fragment code in the .qsb files. Once
written, the HLSL files can be injected into a .qsb file with qsb -r.
or the corresponding CMake syntax. Conceptually this is no different
from how samplerExternalOES support is implemented for Multimedia.
(there the problem is that the shaders cannot be compiled to SPIR-V
to begin with, here it is that we cannot translate from SPIR-V, but
in the end the workaround for both problems is effectively the same)
The manual tests demonstrate this, both the tessellation and geometry
apps work now with D3D out of the box.
On the bright side, the implementation here in the the D3D backend of
QRhi does not need to know about how the shaders got there in the
QShader. So none of the implementation is dependent on this manual
process. If some day qsb would start translating to these kind of
shaders as well, it would all still work as-is.
Change-Id: I32d9ab94e00174e4bd5b59ac814dfedef9f93ad1
Reviewed-by: Andy Nichols <andy.nichols@qt.io>
Michael Weghorn
Ilya Fedin
Tatiana Borisova
Richard Moe Gustavsen
Mikolaj Boc
JiDe Zhang
Mårten Nordheim
Alexandru Croitor
Sona Kurazyan
Marc Mutz
Joerg Bornemann
Morten Sørvig
Jaroslaw Kubik
Thiago Macieira
Tor Arne Vestbø
Laszlo Agocs
Sze Howe Koh
Eskil Abrahamsen Blomfeldt
Kai Köhne