| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
|
| |
This test suffers for out of order handling of network
requests. To speed up release blacklist flaky tests.
Change-Id: Ifafc4f429b7a2bb8cd4b5c786143b689657cc753
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
|
|
|
|
|
|
|
| |
Accept the two requests might come out of order
Change-Id: I313506d975ebf007fce3cd857baa742866b8f38b
Reviewed-by: Michael Brüning <michael.bruning@qt.io>
|
|
|
|
|
|
|
| |
Convert it to QtWebEngine.
Change-Id: I7a453cd7544a20a9c461c3bb29b3a2fb78016d44
Reviewed-by: Joerg Bornemann <joerg.bornemann@qt.io>
|
|\
| |
| |
| | |
refs/staging/5.10
|
| |\
| | |
| | |
| | |
| | |
| | |
| | | |
Conflicts:
src/core/download_manager_delegate_qt.cpp
Change-Id: Ica57e582a323c6bd014a64bb615dd9454e656548
|
| | |
| | |
| | |
| | |
| | |
| | | |
Task-number: QTBUG-62640
Change-Id: I2b16f24533b38c20a7071319723382ba240e35f3
Reviewed-by: Peter Varga <pvarga@inf.u-szeged.hu>
|
| | |
| | |
| | |
| | |
| | | |
Change-Id: I7b7c4a6501387b1d657e55f222908365990b8fef
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
|
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
This test makes little sense for QtWebEngine
Change-Id: I0177fb59ea23383a3c8135853f7e62b26ce5708a
Reviewed-by: Peter Varga <pvarga@inf.u-szeged.hu>
|
|/ /
| |
| |
| |
| |
| |
| | |
We have no plans of implement QtWebKit still plugins.
Change-Id: I9f7bb94cc240e2f0eda8c85c03fa5860435d52a3
Reviewed-by: Peter Varga <pvarga@inf.u-szeged.hu>
|
| |
| |
| |
| |
| |
| |
| | |
Have been skipped for years but appears to work.
Change-Id: I461459b64fe3731b2fb50a3cc9c9d4ab75b135d4
Reviewed-by: Alexandru Croitor <alexandru.croitor@qt.io>
|
| |
| |
| |
| |
| |
| |
| |
| |
| | |
It was added for a single test failure that was caused by a screen locker on
the CI, a problem that has since been solved. This CI configuration is the only one
that's running qmltests on osx.
Change-Id: Ia6f11649ce2f63000488aa18080c5d149d04eac0
Reviewed-by: Alexandru Croitor <alexandru.croitor@qt.io>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Triple click selects the entire paragraph and it should be
deselected on a quadra click.
Task-number: QTBUG-65649
Change-Id: I87c9405a202d8b6eacd7c19dbbcb051756c41220
Reviewed-by: Viktor Engelmann <viktor.engelmann@qt.io>
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
|
|\|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Conflicts:
src/3rdparty
src/core/web_contents_adapter.cpp
src/webengine/api/qquickwebengineprofile.cpp
src/webenginewidgets/api/qwebengineprofile.cpp
tests/auto/widgets/qwebenginedownloads/tst_qwebenginedownloads.cpp
tests/auto/widgets/qwebenginepage/tst_qwebenginepage.cpp
Change-Id: I56c093ebab5ee8b577783ce71761719159cd3ddd
|
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Unlike QNetworkProxy, Chromium's net::ProxyServer expects the hostname and port
to be at default values for special schemes (DIRECT and INVALID). Otherwise, a
DCHECK is triggered at proxy_server.cc:73.
Change-Id: I1ac6c425ea03fcbfe084d25c2fd05bf174c753d6
Reviewed-by: Viktor Engelmann <viktor.engelmann@qt.io>
|
| |
| |
| |
| |
| |
| |
| |
| | |
It was vulnerable to race conditions if two requests were send
before the first was caught.
Change-Id: Ia0b68594337028675678ad190f261202372c2ad6
Reviewed-by: Viktor Engelmann <viktor.engelmann@qt.io>
|
| |
| |
| |
| |
| |
| |
| |
| | |
Use in-memory profile and do not use qwebengineviews when all we need
are qwebenginepages.
Change-Id: I5798c761bccfc0b4bf2cf623e7db35f033bdfd84
Reviewed-by: Peter Varga <pvarga@inf.u-szeged.hu>
|
|\|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Conflicts:
.qmake.conf
src/core/api/qwebengineurlrequestjob.cpp
src/core/browser_context_adapter.cpp
src/core/renderer/user_resource_controller.cpp
src/core/web_engine_context.cpp
src/webenginewidgets/api/qwebenginepage.cpp
Change-Id: I5278e5e22e1776d42975fc94d70ff8ca4f81fb9a
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
It's used from another thread, so it shouldn't be deleted immediately,
and it's not deleted by the QtWebEngine code, so tell people to do it
themselves at the right time.
Change-Id: I54786be320f5fe82f144e7b1c2e6137260d9ceab
Reviewed-by: Leena Miettinen <riitta-leena.miettinen@qt.io>
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
|
| |
| |
| |
| |
| |
| | |
Task-number: QTBUG-64967
Change-Id: Ida79d6cdb682d510e9a2e91e3e3ca263acf34a99
Reviewed-by: Viktor Engelmann <viktor.engelmann@qt.io>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Continiously failed 5 times on Windows when integrating unrelated
patches.
Task-number: QTBUG-61887
Change-Id: Ibc3e1563355364302f7ee430c97209798e86e3e7
Reviewed-by: Michael Brüning <michael.bruning@qt.io>
(cherry picked from commit f1003c50652c1f5ef9ae0e16036b175fe3e8f215)
Reviewed-by: Liang Qi <liang.qi@qt.io>
|
| |
| |
| |
| |
| |
| |
| |
| |
| | |
The code was not correctly giving a function as argument to the
addEventHandler, and the whole logic was more complicated than it
needed to be.
Change-Id: I21e4e6bb6adf1071ae7eb2798d129224af8ef0fb
Reviewed-by: Peter Varga <pvarga@inf.u-szeged.hu>
|
| |
| |
| |
| |
| |
| | |
Change-Id: If18bb163744b9064c0e4d97a9476851b6dee39ad
Reviewed-by: Qt CI Bot <qt_ci_bot@qt-project.org>
Reviewed-by: Peter Varga <pvarga@inf.u-szeged.hu>
|
| |
| |
| |
| |
| | |
Change-Id: I685ee99eedfc47065ed01301943802d7b89ffa2a
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
|
| |
| |
| |
| |
| | |
Change-Id: I32a92913f3dc80fc409069e624259eecc50c357b
Reviewed-by: Michael Brüning <michael.bruning@qt.io>
|
| |
| |
| |
| |
| |
| | |
Task-number: QTBUG-65206
Change-Id: Idc85c71ff56cb24764625877c8f54c978661a044
Reviewed-by: Michael Brüning <michael.bruning@qt.io>
|
| |
| |
| |
| |
| |
| |
| | |
Files to test binary compatibility added for 5.10
Change-Id: I97940a92003084ed0c5c92eb62ce1be94fdd0715
Reviewed-by: Sergio Ahumada <sahumada@texla.cl>
|
| |
| |
| |
| |
| |
| |
| |
| |
| | |
We have no qwebenginehistoryinterface and the test does basically
nothing, however since it exposes problems at shutdown we can keep the
test as an very fast test just to check fast shutdown work.
Change-Id: I612ac79fad788fe3835789fcfba11647f43a8796
Reviewed-by: Alexandru Croitor <alexandru.croitor@qt.io>
|
| |
| |
| |
| |
| |
| |
| |
| | |
Most of them are unsupported but verify that loading them doesn't cause
crash or assert.
Change-Id: I81f219e8d56923dbe6df16f3412e384803808245
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
|
| |
| |
| |
| |
| |
| |
| |
| |
| | |
- Reset loadFinishedSpy counter after initialization
- Do not expect loadFinished signal after "same document" navigation
Change-Id: If78304d8fa67ec96b28811c148b4a23f664987fb
Reviewed-by: Viktor Engelmann <viktor.engelmann@qt.io>
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Avoid false positive results by waiting for a value change when it is not
expected to be changed and make sure the event is processed before check
for unexpected change. Moreover, verify that the composition events are
sent.
Change-Id: Ie7a53237b90cda433502e7f376fbe2ae1964385d
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
|
| |
| |
| |
| |
| | |
Change-Id: I4743c69202443c3ddb128ae89efc93cd524e775b
Reviewed-by: Michal Klocek <michal.klocek@qt.io>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Occasionally the mouse pointer might be moved above the testDiv during
the html load and this changes its text. This seems to happen on CI
from time to time.
Text of testDiv is expected to be empty in the beginning to make sure
an explicit mouse move event changes its text.
Change-Id: Ib8488b1e2dd7cc80b78cc40adecbf04c12dbec1b
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
|
| |
| |
| |
| |
| |
| |
| |
| | |
Giving replies their job as parent is unstable currently.
Change-Id: Ifd4598b199c8409c15b13e362df383b11051fb7a
Reviewed-by: Jüri Valdmann <juri.valdmann@qt.io>
Reviewed-by: Alexandru Croitor <alexandru.croitor@qt.io>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
The fix for bic tests for qtqa is now in. Therefore,
readd the missing bic test data.
This reverts commit fe547d51e8f3feef433bfde77b17048c8aa9ed72.
Task-number: QTBUG-63005
Change-Id: If8af66320786a07f1a79dd3eca5925ce952efcfe
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@qt.io>
|
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Continiously failed 5 times on Windows when integrating unrelated
patches.
Task-number: QTBUG-61887
Change-Id: I8055673e04c3bdfddfcfe05c3905ef5a86e8d1ec
Reviewed-by: Michael Brüning <michael.bruning@qt.io>
|
|\|
| |
| |
| |
| |
| |
| |
| |
| | |
Conflicts:
src/3rdparty
tests/auto/quick/qmltests/qmltests.pro
tests/auto/widgets/qwebengineview/tst_qwebengineview.cpp
Change-Id: I2a1e3a83e29a1d6709f3e98fda3f7b4ab7ab297e
|
| |
| |
| |
| |
| |
| |
| | |
Remove strange options for call to QUrl::path and add test for "qrc" protocol.
Change-Id: I6528d858b7661832852c333a7f932d4714f953f2
Reviewed-by: Alexandru Croitor <alexandru.croitor@qt.io>
|
| |
| |
| |
| |
| |
| |
| |
| | |
Forward QEvent::Leave for Widget and QEvent::HoverLeave for Quick.
Task-number: QTBUG-64265
Change-Id: Ide32768902956476d24b1d4115e305392b62feb3
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Consider the scenario
1. user enters url "http://localhost:8000/"
A new navigation entry is created and committed in the
NavigationController.
2. user enters url "http://localhost:8000/download.bin"
A new navigation entry is created and a download is triggered, but the
pending navigation entry in the NavigationController is neither committed
nor discarded (since our WebContentsDelegate's ShouldPreserveAbortedURLs()
returns true).
3. user enters url "http://localhost:8000/download.bin"
At this point the NavigationController will have "http://localhost:8000/"
as the committed navigation entry and "http://localhost:8000/download.bin"
as the pending entry. NavigateToPendingEntry will see that the user is
trying to navigate again to the same URL as the last pending entry and will
therefore identify this new navigation as a reload.
However Blink interprets 'reload' to mean reloading the last committed
entry, i.e. "http://localhost:8000/", and so we end up trying to download
"http://localhost:8000/" instead of "http://localhost:8000/download.bin" as
the user might have expected.
The patch removes the ShouldPreserveAbortedURLs override and relies on the
default implementation which always returns false. As a result the pending
navigation entry in step 2 above is discarded once the download has been
triggered and the unexpected behavior in step 3 is no longer triggered.
Removing the override resurrects QTBUG-48995 where, for example, calling
QWebEnginePage::setUrl triggers first a urlChanged signal for the *old* URL. The
patch adds url and title properties to WebContentsDelegateQt so that property
change signals are triggered only if the properties have actually changed. A
consequence of this fix is that the first urlChanged signal is delivered
directly from the setUrl/load method and not asynchronously once the loading
starts (this is also how Chrome's URL bar is updated).
Task-number: QTBUG-63388
Change-Id: Icfa300b165e5e56f1fbc8978a00a237c263df183
Reviewed-by: Peter Varga <pvarga@inf.u-szeged.hu>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Document the somewhat strange behavior of the signals
QWebEnginePage::urlChanged,
QWebEnginePage::titleChanged, and
QWebEnginePage::loadFinished
when trying to load invalid URLs via QWebEnginePage::setUrl.
Change-Id: I5218a49105ad6187f87e7091589b5175f4349593
Reviewed-by: Peter Varga <pvarga@inf.u-szeged.hu>
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
|
| |
| |
| |
| |
| |
| |
| |
| | |
This is done to make sure there are no conflicts with features in
other modules, because they all share a global namespace.
Change-Id: I95b3b7fadd8ffc2979ee3aad2234ee543d57c7d8
Reviewed-by: Michal Klocek <michal.klocek@qt.io>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
A 5 second timeout might not be enough for our underpowered CI VMs.
Also, if the page isn't loaded (to slow or for other reasons), we
must abort, to prevent the JavaScripts from being run on different
pages (first scripts to the old page, then to the new page if it
eventually finishes loading).
The first JavaScript in particular is rejected on the old page,
because "about:blank" doesn't have a secure origin. This could
even make the test accept wrongfully if all scripts are run on the
old page, because the request (which is supposed to be rejected)
can be rejected for being run on the wrong page, instead of for
requesting forbidden permissions.
Task-number: QTBUG-63347
Change-Id: Ib63d55477d35001b174900b73dc3dd23d33c7263
Reviewed-by: Jüri Valdmann <juri.valdmann@qt.io>
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Moreover, extend the list of supported editor shortcuts and stabilize
the corresponding auto test.
Task-number: QTBUG-54692
Task-number: QTBUG-54812
Task-number: QTBUG-54221
Task-number: QTBUG-59053
Change-Id: I4dd8230519639ea6e3340992dbb54a609ecfcd91
Reviewed-by: Alexandru Croitor <alexandru.croitor@qt.io>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
This is done because qtConfig(spellchecker) is present
in an example .pro file, and thus "spellchecker" should be
a public feature.
Change-Id: Idff977b8d0835b049c4c19dc42c1475d2c55c323
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@qt.io>
Reviewed-by: Michal Klocek <michal.klocek@qt.io>
|
| |
| |
| |
| |
| |
| |
| | |
Only projects which are built before the actual module need to do that.
Change-Id: Ia8b071a627bd39c5df4600877b6873f308cbf690
Reviewed-by: Michael Brüning <michael.bruning@qt.io>
|
|\|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Conflicts:
.qmake.conf
src/core/core_module.pro
src/core/delegated_frame_node.cpp
src/core/surface_factory_qt.cpp
src/webenginewidgets/api/qwebengineprofile.cpp
tests/auto/widgets/widgets.pro
Change-Id: I92f3ef4eee779afef6c5381a7aa8b551417c1b17
|
| |
| |
| |
| |
| |
| |
| |
| | |
Original patch by Daniel Mack <daniel@zonque.org>
Task-number: QTBUG-64015
Change-Id: I7dd0a1aa9bff08c6d19e818acf0e1b8b4b701f5b
Reviewed-by: Alexandru Croitor <alexandru.croitor@qt.io>
|
| |
| |
| |
| |
| |
| |
| |
| | |
We didn't handle the case where the QIODevice does not have all the data
available all the time.
Change-Id: I6aea8ed48ba9ed297efb907b8f6e5c5fc2a18abd
Reviewed-by: Michal Klocek <michal.klocek@qt.io>
|
| |
| |
| |
| |
| |
| |
| |
| | |
It was only reading window.mousePressReceived once, which can sometimes
be before the click is processed.
Change-Id: Ia0a7067a202afa563696708e6df5aa9445335eb8
Reviewed-by: Peter Varga <pvarga@inf.u-szeged.hu>
|
| |
| |
| |
| |
| | |
Change-Id: I5cb198445f283c6cea8187a1a2883896df0ee918
Reviewed-by: Michal Klocek <michal.klocek@qt.io>
|