| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
|
|
|
|
| |
In case there is no window system with qml simply use
screen rect. This will also work well with rotated view
(for non windowing system).
Task-number: QTBUG-86034
Task-number: QTBUG-89358
Change-Id: Ic042d766d661c8322422b194c08b58609992961b
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
|
|
|
|
|
|
|
|
|
|
|
| |
In qml when on platforms without windowing system
we need to send show notification.
Note this was not required with old compositor, since
it was hooked up differently than viz.
Task-number: QTBUG-86034
Change-Id: I3ecdc033802d59c15637bc0e48961cd7748b1c6b
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
|
|
|
|
|
| |
Change-Id: Ic7cff2556945272761288fedd271796dbf240cb5
Reviewed-by: Kirill Burtsev <kirill.burtsev@qt.io>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This reverts commit 7af8aa79d76b31f5d881d62f8c8661117274734c.
Reason for revert: appears to always BPASS on CI
after tests for msvc2019 were enabled and 79-based integrated
and only fail with timeout error, which really takes more than
one minute for renderer process to initialize devtools, so
increase also timeout for load.
Fixes: QTBUG-81263
Change-Id: Ia07a62f2e0a6822edb9bb1fb86df1c9fe0536072
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
|
|
|
|
|
|
|
|
|
|
| |
Scrolling is momentum based and with smooth scrolling handling may lead
to multiple scroll position offset changes for series of touch event.
Moreover, it's not certain that offset position fetched through
js after input is final position, so check it on each scrolling step.
Change-Id: I793b9c237b024b5f68cdd5662a91d72b21bb64c4
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
|
|
|
|
|
|
|
|
| |
We were actually using closure compilation, let's disable it for now,
until we have java build-time detection.
Change-Id: Ie5fd0f397a773d38a1eb8a94ac2f186e9073ea6e
Reviewed-by: Michal Klocek <michal.klocek@qt.io>
|
|
|
|
|
|
|
| |
They are just confusing users.
Change-Id: I107b4212e425e77b892c1ce77ee0fbfb89a97c7f
Reviewed-by: Michal Klocek <michal.klocek@qt.io>
|
|
|
|
|
|
|
| |
We now get permission requests for CAMERA_PAN_TILT_ZOOM
Change-Id: I1a86a1aa4d8ec4e8c7f9c1842cd709425de9c954
Reviewed-by: Peter Varga <pvarga@inf.u-szeged.hu>
|
|
|
|
|
|
|
| |
Must be created before RenderFrame.
Change-Id: I4ee121f6ad6a77fc48d12bc56b1fdb82ec97139b
Reviewed-by: Kirill Burtsev <kirill.burtsev@qt.io>
|
|
|
|
|
|
|
|
|
|
|
|
| |
onbeforeunload event is fired when navigating to another page since:
d51e7703e747 Remove InterstitialPage
https://chromium-review.googlesource.com/c/chromium/src/+/2146137
Navigate away from pages with onbeforeunload handler at the end of the
tests.
Change-Id: I79b306620e202e1466a8125c8f8f60a7218b2969
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
|
|
|
|
|
| |
Change-Id: Ic4ffd98e02f986dbaf986405360e727c813e696e
Reviewed-by: Peter Varga <pvarga@inf.u-szeged.hu>
|
|
|
|
|
| |
Change-Id: I7e0ebecdbb68cfff0b574c966f3fa80d28680e1c
Reviewed-by: Peter Varga <pvarga@inf.u-szeged.hu>
|
|
|
|
|
| |
Change-Id: I33c1af7c431055d95e0fb540246765cce684de15
Reviewed-by: Peter Varga <pvarga@inf.u-szeged.hu>
|
|
|
|
|
| |
Change-Id: I359805d0bea84147fca6de2e2c7b17b4dcb17bc7
Reviewed-by: Peter Varga <pvarga@inf.u-szeged.hu>
|
|
|
|
|
| |
Change-Id: Idb41daa68709c769330527a713689241ac073359
Reviewed-by: Kirill Burtsev <kirill.burtsev@qt.io>
|
|
|
|
|
| |
Change-Id: Ia360553626c0277c7ef703504c8a7a76ca792801
Reviewed-by: Michael Brüning <michael.bruning@qt.io>
|
|
|
|
|
|
|
|
|
| |
The test seems to have gotten flakyon macOS recently.
Blacklist it until the cause has been identified and fixed.
Task-number: QTBUG-89627
Change-Id: Ib711a6ee468f4d4416de9f96a32ff3cab51a24fc
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
|
|
|
|
|
| |
Change-Id: Iad51e005cd11d060021a1ea5f1f312b1a0a08985
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Load with client or server http error results in successful navigation,
which leads to 'true' loadFinished result, and subsequent chromium's error
page load and display with second set of loadStarted/loadFinished signals.
This effectively ignores QWebEngineSettings::ErrorPageEnabled. Fixing it
requires submodule change to ask embedder if error pages should also be
suppressed for http errors.
Also update chromium for required change, which pulls in the following changes:
* e71010069b4 Fix embedded builds with printing enabled
* f5a93d251cc Allow the embedder to suppress an error page for http errors
Change-Id: I731678575439a6dad90dfb89e79b0083c63b49c2
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
|
|
|
|
|
|
|
|
|
| |
Disable the InstallApp feature, which we have disabled as it integrates
with the payment subsystem.
Fixes: QTBUG-89740
Change-Id: I212bad785294a70812385e96d5f7e300f5806f87
Reviewed-by: Michael Brüning <michael.bruning@qt.io>
|
|
|
|
|
|
|
|
|
|
|
|
| |
Pulls in the following changes:
* dfe93385f41 [Revert] Don't use harfbuzz-subset
* 9087c921839 Add user script data mojo interface and traits
* 25db271c9b5 Add WebEnginePageRenderFrame mojo interface
Change-Id: I3bf77b63acb3dab3113f70105d3dd304d2d34b95
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
|
|
|
|
|
|
|
|
|
|
|
|
| |
Since mojo interface no longer works on render view, but
on render frame rename the class to be WebEnginePageHost
and WebEnginePageRenderFrame
Keep SetBackgroundColor render message, since it is required
to things work.
Change-Id: I4d6b70dd448926f8ac2d75d2b87f3d490a4cfefb
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
|
|
|
|
|
| |
Change-Id: Ica6b794e16aecc60dc2c39d31750acc6b25410df
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
|
|
|
|
|
| |
Change-Id: If9394dbfe49adf7330558eac194a0ec643eb8ee6
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
|
|
|
|
|
|
|
|
|
|
|
| |
After site isolation we observe render frames and we do not need
render view observer any longer.
Remove RenderViewObserverHelper and track user script per
render frame. This approach is more intuitive since we run those scripts
on frames, subframes and not on render views, which can span over few frames.
Change-Id: Ife02d8a4dbecc884cfdc667b1a16dfbdd866367a
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
|
|
|
|
|
|
|
|
| |
Follow same naming as in UserResourceController, which also implements
RenderThreadObserver.
Change-Id: I8a8386e6fb9fcc52826da1d0eef6efa97fd475e9
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
|
|
|
|
|
| |
Change-Id: If9a6b2ceca41534f28ec1d87093778f16fc31ff0
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
|
|
|
|
|
|
|
|
|
| |
Mojo interface when running in single process was not correctly
destructed, since we used user resource controller as global static object.
Move user resource controller to content render client.
Change-Id: I219510c9bc382545174aa5aae99ac8282a2049e6
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Use mojo instead of old IPC, keep current design and use two
interfaces one global and one per frame for now, also use
in both cases associated interface with ipc channel otherwise
script can be added during the page load, which will radomly
brake tests.
This change moves UserDataScript to chromium since mojo binding
generation did not work correctly. Use StructTraits when serializing
the class.
Change-Id: I7073fb831c96849e47864382188300db3c9137d9
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
|
|
|
|
|
|
|
| |
QString::SkipEmptyParts was replaced and deprecated in 5.15 not 5.14.
Change-Id: I9585250b4d1781f87567aff957f8b35fb6015647
Reviewed-by: Kirill Burtsev <kirill.burtsev@qt.io>
|
|
|
|
|
|
|
|
|
| |
Chromium uses HarfBuzz only for font subsetting since the 83-based
update and it can't be built with older HarfBuzz version than 2.4.0.
Task-number: QTBUG-88976
Change-Id: Ibcea31b487c3f8d1b3f9b8566b95ddf2a35987bb
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
|
|
|
|
|
| |
Change-Id: I4c4bbc75b9f6346a446f8094f669d142f76c164a
Reviewed-by: Kirill Burtsev <kirill.burtsev@qt.io>
|
|
|
|
|
|
|
|
|
|
|
| |
QtZlib is a semi private module, headers are installed
as Qt5Zlib, however all object are in QtCore.
Add support to compile against Qt5Zlib when doing
static build.
Task-number: QTBUG-87154
Change-Id: I74d1b217725b37781808a54902f9acbee41c50c4
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Pulls in the following changes:
* f830b86ef77 [Backport] mac: make find_sdk.py work when the sdk goes to 11
* 9236b21c883 Fix build with system ICU 68
* 138a7203f16 Fix build with msvc2019 16.8.0
* 1ca6cb93ca8 Add qt static dependencies: qtfreetype, qtharfbuzz
* 29b90d82710 Add qt static dependencies: qtlibpng
* da115b70738 Add qt static dependencies: qtlibjpeg
* bb90182aa90 Add qt static dependencies: qtzlib
Change-Id: Ib986dde7db99c7d6663d12f933ddbb4acc4e79e1
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Add qtfreetype, qthurfbuzz, qtlibpng, qtlibpng
static 3rdparty libs.
If qtbase has some 3rdaprty libs, which are also
present in chromium. This leads to duplicated symbols
and can lead to unexpected issues with application.
Currently we do only static builds for qtpdf, therefore
this changes only affects qtpdf builds.
If qtfreetype is used, hurfbuzz dependency is not used,
qtlibpng is only used when pdf compiled with xfa.
Task-number: QTBUG-87154
Change-Id: Ibba528268d9c989a2c63c1a7881ba0c95c325437
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Make the WebContentsDelegateQt::EmitLoadStarted() and the
WebContentsDelegateQt::EmitLoadFinished() independent from
the WebContentsDelegateQt::LoadProgressChanged() by removing
m_lastLoadProgress.
Adapt the WebContentsDelegateQt::LoadProgressChanged() to send
signal only if load is in progress.
Add a new test based on the bugreport.
Fix qmltests::WebEngineViewSource::test_viewSourceURL() flaky tests.
Fixes: QTBUG-65223
Fixes: QTBUG-87089
Change-Id: I90af4d2e85105dba801beb8102991eb4ef14c6a3
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This change addresses following wrong assumptions implemented earlier:
* fix assert condition for custom MotionEventQt: pointer index is only
reserved only for two distinct events POINTER_UP and POINTER_DOWN,
which represent one single non first/last pointer up and down event.
* MotionEvent API doesn't support delivering multiple touch point
changes at once, hence code should send every touch point's up/down
step by step to not confuse gesture detection classes in chromimum
(ultimately fixing arising asserts which old code was triggering before).
* MotionEvent shouldn't contain more touch points then were already
forwarded or reported released, hence every step should contain only
part of all touch points received in handleTouchEvent.
Fixes: QTBUG-86389
Change-Id: I62275dc295494f88a6e44f36fe72f5f8227e37d4
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
Reviewed-by: Michal Klocek <michal.klocek@qt.io>
|
|
|
|
|
|
| |
Fixes: QTBUG-89001
Change-Id: I9f336313b075ad0cb3115975afc410beb415e4a0
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
QQuickPdfDocument passed source QUrl::path() to QPdfDocument::load.
QUrl::path() output gets "/" as prefix on Windows preventing the file
from being found so this needs to be accounted for by changing it to
QUrl::toLocalFile().
Fixes: QTBUG-88938
Change-Id: I52ed44f67c22bea31637050442487e1914011de9
Reviewed-by: Andy Shaw <andy.shaw@qt.io>
Reviewed-by: Shawn Rutledge <shawn.rutledge@qt.io>
(cherry picked from commit 5852716178972dd34df34dabc52a6616eead0aca)
Reviewed-by: Qt Cherry-pick Bot <cherrypick_bot@qt-project.org>
|
|
|
|
|
|
| |
Fixes: QTBUG-88861
Change-Id: I7091aca70aaf87edf0b1e67ec3fa705a59c8192c
Reviewed-by: Michal Klocek <michal.klocek@qt.io>
|
|
|
|
|
|
|
|
|
|
| |
TouchCancel may come with empty points list, so previous one (from TouchBegin)
is used. Therefore it should be cleaned independently on leaving the scope.
Thus unify common code cleanup in the end of handleTouchEvent and ensure
TouchCancel's ids remove from mapping.
Change-Id: I1ef2baa5f920fa9096ddac2ddd5d4ac70ab6e218
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
|
|
|
|
|
|
|
|
|
| |
This api was made private in dev 4e400369c0 and currently has side effects,
which lead to touch malfunction and test failures.
Task-number: QTBUG-88001
Change-Id: Ie84e770a229f58704363448df116aedbc3e8eb82
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
When built against a Qt version different from the module version, all
QtWebEngine libraries will be versioned with MODULE_VERSION.
QtWebEngineProcess as part of the module should be handled in the same
way.
Change-Id: I34073337e813baac46f207443bf54ca78d879ea6
Reviewed-by: Andy Shaw <andy.shaw@qt.io>
(cherry picked from commit c636e88bc49f59548e1ac3401ae547e2ae9d2570)
Reviewed-by: Qt Cherry-pick Bot <cherrypick_bot@qt-project.org>
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
When on macOS, with a frameworkless build, WebEngine resources are
stored in a Resources directory named with a capital R.
This is the standard directory name for resources on macOS, however
Qt WebEngine was expecting to find resources in a directory named
`resources` (no capitalized first letter).
Task-number: QTBUG-72368
Change-Id: I2106a50a63c6d812dc6ad649645e3b6b9e0471e2
Reviewed-by: Kirill Burtsev <kirill.burtsev@qt.io>
|
|\
| |
| |
| | |
Change-Id: I516761e18c03c6cfd4d44a6ee176a2add9a5b022
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
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>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Including security issue under active exploit.
Submodule src/3rdparty 55b3d183..fb6ab5e4:
> [Backport] Security bug 1137608
> [Backport] CVE-2020-16011: Heap buffer overflow in UI on Windows.
> [Backport] CVE-2020-16009: Inappropriate implementation in V8
> [Backport] CVE-2020-16008: Stack buffer overflow in WebRTC
> [Backport] CVE-2020-16005: Insufficient policy enforcement in ANGLE
> Fix assert when running tests build with -no-webengine-webrtc
Change-Id: Ic9ee45fbd78c999881454e5e0e3091489f637aae
Reviewed-by: Michael Brüning <michael.bruning@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>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Pulls in the changes:
2cad1d38b57 [Backport] CVE-2020-6540: Heap buffer overflow in Skia
e1f5ffc2a12 [Backport] CVE-2020-6569: Integer overflow in WebUSB
75b417b86b4 [Backport] CVE-2020-6570: Side-channel information leakage in WebRTC
0785cd83782 [Backport] CVE-2020-6573: Use after free in video
3ddb1e9639a [Backport] CVE-2020-6575: Race in Mojo
e967bb155a0 [Backport] CVE-2020-6576: Use after free in offscreen canvas
40cebfb3051 [Backport] CVE-2020-15959: Insufficient policy enforcement in networking
405e7526583 [Backport] CVE-2020-15965: Out of bounds write in V8
295feb590b1 [Backport] CVE-2020-15963 and CVE-2020-15966
43812fd40d4 [Backport] CVE-2020-15961: Insufficient policy enforcement in extensions
08a25bf9dd2 Only upgrade to WebSocket when WebSocket was requested
2f75c909270 [Backport] CVE-2020-15962: Insufficient policy enforcement in serial
06afd03d826 [Backport] CVE-2020-15960: Out of bounds read in storage
9e5518f06ad [Backport] Security bug 1111149
b627f77e859 [Backport] CVE-2020-15964: Insufficient data validation in media
bc35c7711ad Fix assert when checking if RendererAppContainer feature is enabled
06b27f95d60 [Backport] CVE-2020-6571: Incorrect security UI in Omnibox
8227b9613d1 CVE-2020-15999: Heap buffer overflow in freetype
cc1e87a0d71 [Backport] CVE-2020-15978 Insufficient data validation in navigation
308da5d58b4 [Backport] CVE-2020-15969: Use after free in WebRTC.
9f5fde5b649 [Backport] CVE-2020-15968: Use after free in Blink
97b92535b31 [Backport] CVE-2020-15972: Use after free in audio.
873abc1112c [Backport] CVE-2020-15979: Inappropriate implementation in V8.
c0343ceec47 [Backport] CVE-2020-15976: Use after free in WebXR
e874d48f9e6 [Backport] CVE-2020-15992 Insufficient policy enforcement in networking
956ce06c56c [Backport] CVE-2020-15974: Integer overflow in Blink
4662223422d [Backport] Security bug 1125199
74e26aab318 [Backport] Security bug 1107824
33e9e655166 [Backport] CVE-2020-15977: Insufficient data validation in dialogs.
f6db4999b0e [Backport] CVE-2020—15973: Insufficient policy enforcement in extensions (1/2)
466da17a0cd [Backport] CVE-2020—15973: Insufficient policy enforcement in extensions (2/2)
027c3d7bae7 [Backport] CVE-2020-15989: Uninitialized Use in PDFium
14e9b6f6d3a [Backport] CVE-2020-16003: Use after free in printing
8527c994fe3 [Backport] CVE-2020-16002: Use after free in PDFium
4f461642a79 [Backport] CVE-2020-16001: Use after free in media.
99877493b32 Fixup: [Backport] CVE-2020-16001: Use after free in media
489d6e637e5 [Backport] CVE-2020-15987: Use after free in WebRTC (1/2)
6ef8f4ed829 [Backport] CVE-2020-15987: Use after free in WebRTC (2/2)
9d173d02d5e [Backport] Security bug 1106091
327474aed0e [Backport] CVE-2020-6557: Inappropriate implementation in networking
55b3d183921 [Backport] CVE-2020-15985: Inappropriate implementation in Blink.
Task-number: QTBUG-87967
Change-Id: Ibca52ab97ac407679e4bac5c6b7dc2285f8bcd37
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
|
| |
| |
| |
| |
| |
| |
| |
| | |
Add QTest::qWaitForWindowExposed() to make sure that
the window is visible before the input event is sent.
Change-Id: I72b02abaaf5970a1533f1d6f6dae59e47e891548
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
|