| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
|
|
|
|
| |
Chromium gl bindings headers get mixed with qt opengl headers.
This results in GLDEBUGPROC missing definition errors.
Fixes compilation for apalis-imx6 on zeus.
Task-number: QTBUG-88219
Change-Id: I0a18ceff3336ab457c2d9f35ef17c28c01bbe035
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Desktop audio capture requires loopback device. If creation of the
loopback device fails, it aborts the screen capture too.
Chromium does not support audio loopback on Linux and macOS:
https://crbug.com/223639
This is still the case in Chrome 88.
The fix is based on: https://codereview.chromium.org/24153018
Change-Id: Ifb13bce3b79193203c0bf52e1f2a1b3936e017c7
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
|
|
|
|
|
|
|
|
| |
Uses percentage encoded utf8.
Task-number: QTBUG-85363
Change-Id: Idabe6ff9a66897792732b9ebb648f54a84e86bb8
Reviewed-by: Jüri Valdmann <juri.valdmann@qt.io>
|
|
|
|
|
|
|
|
|
| |
Add error message when trying to launch QtWebEngineProcess
from network share if sandbox is enabled.
Task-number: QTBUG-84632
Change-Id: Ia7d5df38accf60eafe8fb4d43dab3db4d2d51287
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
|
|
|
|
|
|
|
| |
Do not look up fallback values before they are needed.
Change-Id: I8bb21e261e8d218c658ac2a08e3cdc30b2cea47d
Reviewed-by: Michal Klocek <michal.klocek@qt.io>
|
|
|
|
|
|
|
|
|
| |
Chromium will recreate the original mapping based on the native keycode,
breaking any remapping.
Fixes: QTBUG-86672
Change-Id: I4b7e6f7434049cdf7250bee1ef759a9df4d4987c
Reviewed-by: Jüri Valdmann <juri.valdmann@qt.io>
|
|
|
|
|
|
|
|
| |
The QIcon::pixmap function does internal scaling messing up our
intermediate structures.
Change-Id: I727efd58a4cc948fa3e4b23539d97e8ca755d060
Reviewed-by: Peter Varga <pvarga@inf.u-szeged.hu>
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Amends commit 99a668c324. Task 'swapBuffersOnVizThread' is supposed to
be scheduled only when m_middleBuffer is ready after swap on gpu thread.
Checking for m_taskRunner is not enough to verify this. Call to
updatePaintNode may come from SG backend on expose event, thus breaking
expected sequence of buffers swaps.
Task-number: QTBUG-85817
Fixes: QTBUG-86599
Change-Id: Ia97a7a24b5672a2979b514d3cb4879e7630941eb
Reviewed-by: Jüri Valdmann <juri.valdmann@qt.io>
|
|
|
|
|
|
|
|
| |
Add the missing implementation of
ContentRendererClient::IsOriginIsolatedPepperPlugin().
Change-Id: Ic4157da6b373450e620ff46faedad213a69adfec
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Can be reproduced with quicknanobrowser by
1. Open devtools
2. Switch to OTR mode
During the switch to OTR mode, the inspected WebContentsAdapter is
destroyed and recreated. DevToolsFrontendQt is supposed to call
devToolsFrontendDestroyed on the old WebContentsAdapter, but it
accesses this adapter through QQuickWebEngineViewPrivate's
webContentsAdapter() which has already been changed to the new value,
so devToolsFrontendDestroyed is actually called on the wrong object.
Fix by remembering the WebContentsAdapter pointer in the
DevToolsFrontendQt constructor.
Change-Id: I17004ebb1a392399ef3e189c433d3f4b3f05bfd3
Reviewed-by: Tamas Zakor <ztamas@inf.u-szeged.hu>
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
|
|
|
|
|
|
|
| |
And update another error that is now better in its other description.
Change-Id: I1b35db2196d05bb35d4c2e1e702fadb430e784c1
Reviewed-by: Jüri Valdmann <juri.valdmann@qt.io>
|
|
|
|
|
|
|
| |
Most of the includes were included twice
Change-Id: Ie2967e9d8ef4799c28a856ade7318d24bac391c6
Reviewed-by: Michal Klocek <michal.klocek@qt.io>
|
|
|
|
|
| |
Change-Id: I63ed851426b18623d549ceaf87f1b6eeec527966
Reviewed-by: Jüri Valdmann <juri.valdmann@qt.io>
|
|
|
|
|
| |
Change-Id: Ie422b327025da084bc8038c7a0aadee2db801f29
Reviewed-by: Jüri Valdmann <juri.valdmann@qt.io>
|
|
|
|
|
|
|
| |
Change-Id: I1b056a61e2b56b19441f559e97f785c87df61a0c
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
(cherry picked from commit b00af13974b2c72ae4d397fbdf5925181e7cc1bf)
Reviewed-by: Jüri Valdmann <juri.valdmann@qt.io>
|
|\
| |
| |
| | |
Change-Id: Ieb546ccd35db06e8ad6358b11e3139a839c9b4cf
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Otherwise unchanged intercepted request leads to second call
in the same interceptor but on ui thread after io thread.
Ammends a05bb73747.
Fixes: QTBUG-86267
Change-Id: I4e7c662d24a58be5445f5c8b6d0bf3751f40cc05
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
|
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Suppress duplicated progress values coming from chromium.
Verify this behavior reliably (and not only 0 and 100 value) by loading html
with subresources with minor delay through test http server.
Change-Id: Id034dda9012212d54d12fc95d5939ba301577c1c
Reviewed-by: Jüri Valdmann <juri.valdmann@qt.io>
|
|/
|
|
|
|
|
|
|
|
|
|
| |
The ScopedGLContextChecker, which deactivates the current
QOpenGLContext when executing Chromium tasks, is currently enabled
only if QOpenGLContext::supportsThreadedOpenGL() is false. But the gpu
thread can nowadays be disabled also with an environment variable and
ScopedGLContextChecker needs to take this into account.
Change-Id: I012524538b49b033ed053dbc41ed4350db916542
Reviewed-by: Michal Klocek <michal.klocek@qt.io>
(cherry picked from commit 5988cec1a1b59aa163042a419c4c2e978bb814bb)
|
|
|
|
|
|
|
| |
At least turns our support off on html5test.com
Change-Id: Ic82800b0efb70bb5589f7544b851ebcd64000105
Reviewed-by: Michael Brüning <michael.bruning@qt.io>
|
|
|
|
|
|
|
|
|
| |
Was enabled for all platforms except windows using MSVC, but
compiles just fine for us.
Fixes: QTBUG-86092
Change-Id: Ib15d94c36b8a411dd5eab01843db1a3666ac370f
Reviewed-by: Michal Klocek <michal.klocek@qt.io>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The buffer pointers m_middleBuffer and m_frontBuffer are swapped in
updatePaintNode on the condition that m_middleBuffer is non-null and
it's serviceId non-zero. But neither m_middleBuffer nor it's serviceId
is immediately cleared, leaving the possibility that a subsequent call
to updatePaintNode could swap the buffer pointers again if this call
occurs before m_middleBuffer is cleared in swapBuffersOnVizThread. The
m_taskRunner pointer is however cleared immediately and therefore the
subsequent call to updatePaintNode will trigger a segmentation fault.
Since m_taskRunner precisely tracks the condition when we should or
should not swap, change the swap condition to simply require that
m_taskRunner is non-null.
Fixes: QTBUG-85817
Change-Id: Iad89bb0f4de7c0c151d5c5bd63ac74ab6cdd9087
Reviewed-by: Michael Brüning <michael.bruning@qt.io>
|
|
|
|
|
|
|
|
| |
It appears we are no longer presented with pages using these, so match
Chrome and Chromium 80 defaults.
Change-Id: Ie270c2bee9b81f898c7dab9e6d4dad02636c17cc
Reviewed-by: Michal Klocek <michal.klocek@qt.io>
|
|
|
|
|
|
|
|
|
| |
We were missing the combined test after Chromium introduced the new
enums.
Fixes: QTBUG-85890
Change-Id: Id7bb7f05635c64c271e12a12f2409eedc22a9605
Reviewed-by: Jüri Valdmann <juri.valdmann@qt.io>
|
|
|
|
|
|
|
|
| |
We only need the QSslCertificate which is always available.
Task-number: QTBUG-85117
Change-Id: I32e2034060e610ac049d8e3b2a03660e9bf7e7a4
Reviewed-by: Michal Klocek <michal.klocek@qt.io>
|
|
|
|
|
|
|
|
|
|
|
|
| |
Submodule src/3rdparty 0a4240a9..4a996760:
> Fix building FIDO with gcc5
> [Backport] CVE-2020-6512: Type Confusion in V8 (3/3)
Fixes: QTBUG-85117
Task-number: QTBUG-54720
Change-Id: I074831454b469c17a71f2ca75a075ee48157970d
Reviewed-by: Tamas Zakor <ztamas@inf.u-szeged.hu>
Reviewed-by: Michal Klocek <michal.klocek@qt.io>
|
|
|
|
|
|
| |
Fixes: QTBUG-85661
Change-Id: I7cd8ed534d94d6be06f77b9b2d1779905655e772
Reviewed-by: Kirill Burtsev <kirill.burtsev@qt.io>
|
|
|
|
|
| |
Change-Id: Ib06a0a1ec3372d483008cc0ac0ec211ec8316e0e
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
|
|
|
|
|
| |
Change-Id: I854034fd63c8847867fd7ec01d5c25781e7de5ef
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
|
|
|
|
|
|
|
|
|
|
|
| |
The latter has been the preferred name since Qt 4.7.0.
(Pick back to 5.15 shall need to drop the test change, as the test was
added to dev after it diverged from 5.15.)
Change-Id: I3154ea1203e741cb337782d3374e02e15c5f209e
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
(cherry picked from commit 005a338f169b485077c248358e1cde96d22ae506)
Reviewed-by: Edward Welbourne <edward.welbourne@qt.io>
|
|
|
|
|
|
|
|
| |
The options only need to allow httponly, it doesn't set it, and we
shouldn't even forward canonical cookies that were excluded on creation.
Change-Id: Ieec45734938c07a50ac03aa113a02a907bce689f
Reviewed-by: Michal Klocek <michal.klocek@qt.io>
|
|
|
|
|
|
|
|
| |
Submodule src/3rdparty 809e16e4..840c8e5f:
> Make navigation on back/forward mouse buttons optional
Change-Id: I191f7781cf9b824b743b80a05a3e819a414ee546
Reviewed-by: Michael Brüning <michael.bruning@qt.io>
|
|
|
|
|
|
|
|
|
|
| |
Submodule src/3rdparty c91f4d20..809e16e4:
> Add cookie filter to URL requests cookie headers
> Use audio-manager app-name for MPRIS
Fixes: QTBUG-85526
Change-Id: Icab26cad3cea8e2ee021a3e589f41bf0543d64fa
Reviewed-by: Michael Brüning <michael.bruning@qt.io>
|
|
|
|
|
|
|
|
| |
Cleanup after a34b3a4e
Task-number: QTBUG-85526
Change-Id: I23fd5f504937efa44964546a1f4b095efd9ad65f
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
|
|
|
|
|
|
|
|
|
| |
There was some kind of race between loading as a plugin and this
redirect.
Task-number: QTBUG-84340
Change-Id: I57ac8fcae0c1de98b1fd3013d2b5299c85547cc0
Reviewed-by: Michael Brüning <michael.bruning@qt.io>
|
|
|
|
|
|
|
|
| |
It appears we have support for it on all levels.
Fixes: QTBUG-85360
Change-Id: I68d03931197a3b883a1909e96096000fb84925bf
Reviewed-by: Peter Varga <pvarga@inf.u-szeged.hu>
|
|
|
|
|
|
|
|
|
| |
This line of code just hided illegal cookie store changes from being
observed by our API.
Task-number: QTBUG-85526
Change-Id: I044221f70b7628bd727879f11f46bfefe75ca2ca
Reviewed-by: Jüri Valdmann <juri.valdmann@qt.io>
|
|
|
|
|
|
|
|
| |
It seems this does not work already for a while.
Add missing headers.
Change-Id: I158519f69e2ce87fe6c84c03a6ac2ce178c20206
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
|
|
|
|
|
|
|
|
| |
Used for screensharing on wayland
Fixes: QTBUG-85309
Change-Id: I7ec61611bb9e3f318a6a5bd3a43212f391766628
Reviewed-by: Kirill Burtsev <kirill.burtsev@qt.io>
|
|
|
|
|
|
|
| |
Was not used
Change-Id: Ic87e35c12c8d6454ed10ca83dfba9ea732919be4
Reviewed-by: Michael Brüning <michael.bruning@qt.io>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Chromium has blacklisted OpenGL version prior to 4.1 on macOS.
To re-enable hardware acceleration in WebEngine on macOS, set the
default surface format to a 4.1 CoreProfile before creating the
shared OpenGL context, unless the surface format has been set to
a specific other version.
According to https://support.apple.com/en-us/HT202823 , the newest
Mac that only had support for OpenGL 3.3 was released in Mid 2011,
so this should be okay on the majority of machines.
If the user requests an OpenGL 3.2 profile,
it will also work as the Cocoa QPA plugin enables sharing between
3.2 and 4.x contexts.
If the surface format has been requested to be a version lower than
OpenGL 3.2 or not a core profile, warn the user about the lack of
hardware acceleration.
To ensure that we have actually created native contexts for the
requested profile and not a context backed by the Apple Software
Renderer, we query the context native renderer ID using Cocoa methods.
Task-number: QTBUG-81693
Task-number: QTBUG-75262
Change-Id: I94e2e9913c8481fa8db37eb8381a89769026c919
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
|
|
|
|
|
|
|
|
|
| |
At least Push API is supposed to be disabled in incognito mode, which
results in permission request with empty origin.
Fixes: QTBUG-85116
Change-Id: Ie587013d135bf92a84779e6da87e12374f99bfdd
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
|
|
|
|
|
|
|
|
|
|
|
|
| |
This prevents unicode characters from becoming garbled when pasting the
clipboard content into an application that uses the HTML content
instead of the text data.
This mirrors the behavior of Chromium's clipboard adaptation for macOS
Fixes: QTBUG-75391
Change-Id: I033819a2caf3410509e90c9bc38c9830d184149d
Reviewed-by: Jüri Valdmann <juri.valdmann@qt.io>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This makes it possible to tell the difference between multiple
applications using QtWebEngine.
Submodule src/3rdparty 15a42873..87e5a52b:
> Pass through a new application name argument to utility processes
> Fix libjpeg_turbo for ARM32
> Add missing include in certificate net log
Fixes: QTBUG-85363
Change-Id: Ib5426d2e431eb3032f76270885c9cf2d83a75ac8
Reviewed-by: Kirill Burtsev <kirill.burtsev@qt.io>
|
|
|
|
|
|
|
| |
Simply add some page navigation items.
Change-Id: I6da3ff1c1e56f0c3fbbd6eb5cd3b77f32e850862
Reviewed-by: Jüri Valdmann <juri.valdmann@qt.io>
|
|
|
|
|
|
|
|
|
| |
This preserves unicode characters in the environment variable on
Windows.
Fixes: QTBUG-85054
Change-Id: I3352cc3db4ce8d7b9b352c1634e3239954780618
Reviewed-by: Kirill Burtsev <kirill.burtsev@qt.io>
|
|
|
|
|
|
|
|
|
|
|
| |
The blink::WebInputEvent::kMouseUp event sets the
WebMouseEvent::click_count to zero before the
blink::WebInputEvent::kMouseDown event handling is completed.
We need to keep the click_count value during mouse-up event.
Fixes: QTBUG-84660
Change-Id: I2494f56b51674413d7c8b324c138abcb7c0f9186
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
|
|
|
|
|
|
|
|
| |
Still disabled by default though
Fixes: QTBUG-72497
Change-Id: I81f270f08f9e4941a0afd34a4cb7ac7cb5320a33
Reviewed-by: Michal Klocek <michal.klocek@qt.io>
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Parts of Chromium depends on it for temporary directories it creates.
This also changes the logic of depending on it being empty as indicating
an in memory profile, so it required a bit of cleaning up.
Note this does subtly change the return value of off-the-record profiles.
Task-number: QTBUG-62957
Change-Id: Iec6e24556128c515fce27803171a766b8a9d92e3
Reviewed-by: Michal Klocek <michal.klocek@qt.io>
|
|
|
|
|
|
|
|
|
|
|
| |
This step is done in GLSurfaceEGL but missing in GLSurfaceEGLQt.
The extension bindings in g_driver_egl are checked, among other things,
when initializing DXVAVideoDecodeAccelerator, so we need them to enable
hardware-accelerated video decoding on Windows with ANGLE.
Change-Id: I6a650a0aafe3d4f870101d33d2117427c22304e7
Reviewed-by: Romain Pokrzywka <romain.pokrzywka@gmail.com>
Reviewed-by: Michal Klocek <michal.klocek@qt.io>
|