| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
|
| |
Delete the FilePickerControl object after we're done with it, analog to
what the QtQuick UIDelegatesManager does.
Change-Id: Id5ef6666c8536b9e5c6877cd522b4c20adae37e5
Reviewed-by: Michal Klocek <michal.klocek@qt.io>
|
|
|
|
|
|
| |
Task-number: QTBUG-56645
Change-Id: I16ca76d3a26d0142846b068cc1ba52bee71cf106
Reviewed-by: Leena Miettinen <riitta-leena.miettinen@qt.io>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The position of the IME window is computed using the widget input item
transform. When a regular QWidget gets a focusIn event, the input item
transform is recomputed inside the
QWidgetPrivate::updateWidgetTransform method. This did not happen
for the QWebEngineView, because the focus event is handled internally
and not passed down to QWidget::event.
Fix consists in calling updateWidgetTransform manually whenever the
view receives focus.
The other cases when updateWidgetTransform should be called (namely
resize and move events) are handled properly by delegating to
QWidget::event.
Task-number: QTBUG-55634
Change-Id: Ic93662929e169d860f8ca567f1955da4dc45f9fe
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
|
|
|
|
|
|
|
|
| |
We do not share context between "processes", but between
"threads"
Change-Id: I20f558f913c1f19fc469ade6faab45762f42d528
Reviewed-by: Kai Koehne <kai.koehne@qt.io>
|
|
|
|
|
|
|
| |
Pulls in changes to build debug+release correctly and build with Xcode 8
Change-Id: I38d6bd50048b332939416b0fae3f20978995d789
Reviewed-by: Michael Brüning <michael.bruning@qt.io>
|
|
|
|
|
|
|
|
|
| |
Based on widget solution: 1e83a2d1b61b13323163dfe8cac64dad397cb202
Task-number: QTBUG-55650
Change-Id: I6df45e7e018fa201c50fe81e8679c36f97ddeb1e
Reviewed-by: Leena Miettinen <riitta-leena.miettinen@qt.io>
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
|
|
|
|
|
|
| |
Change-Id: I9c571c823b6fa2411e003799f7ff8627ef6e1d5e
Reviewed-by: Kai Koehne <kai.koehne@qt.io>
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
|
|\
| |
| |
| | |
Change-Id: I3e152ee134ed38628d1bf9571df58f469b29e74b
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
When a link triggers a NewViewRequest, and the latter creates
a new window with a new WebEngineView inside, the "adapter" pointer
can still be NULL when the adoptWebContents is called by openIn.
Therefore is it necessary to test the adapter pointer before using it.
(cherry picked from commit bfc2683ab4ca3f56a2effd1b2f7e68850acb72ba)
Task-number: QTBUG-55765
Change-Id: Iaa7cb4e8c7780a2e3f1a8c85b7b5da0ec541b2f3
Reviewed-by: Jani Heikkinen <jani.heikkinen@qt.io>
|
| |
| |
| |
| |
| | |
Change-Id: Ie355f72a0bf575e66ab465fb2fcada0430ed8c7c
Reviewed-by: Leena Miettinen <riitta-leena.miettinen@qt.io>
|
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Do not rely on "include/QtWebEngineWidgets" or "include/QtCore"
being in the default include path.
Task-number: QTBUG-56107
Change-Id: I9535f723814efdef7bf8290a891978aabfa899bb
Reviewed-by: Michael Brüning <michael.bruning@qt.io>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
When a link triggers a NewViewRequest, and the latter creates
a new window with a new WebEngineView inside, the "adapter" pointer
can still be NULL when the adoptWebContents is called by openIn.
Therefore is it necessary to test the adapter pointer before using it.
Task-number: QTBUG-55765
Change-Id: Ia1a299dd65c229705462c9444b81d7b26567fc09
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Turn the raw pointer
ResourceDispatcherHostLoginDelegateQt::m_authInfo
into a scoped_refptr, to prevent chromium from freeing the memory,
which caused this pointer to dangle and SEGFAULT upon later usage.
Task-number: QTBUG-55828
Change-Id: Ib57e89ca042a4494e2ab77ea10328495e6fc1431
Reviewed-by: Peter Varga <pvarga@inf.u-szeged.hu>
Reviewed-by: Michael Brüning <michael.bruning@qt.io>
|
|\|
| |
| |
| | |
Change-Id: Ie075563e90ca184f8bd38be4a9a9bd49b4640769
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
In line 282f of
chromium/content/browser/frame_host/render_frame_host_manager.cc,
RenderFrameHostManager::Navigate passes nullptr to
WebContentsImpl::NotifyViewSwapped.
In line 3833f of
chromium/content/browser/web_contents/web_contents_impl.cc,
this is passed on to the observers, including
UserResourceControllerHost::WebContentsObserverHelper::RenderViewHostChanged
which dereferenced it unchecked, causing a crash.
Task-number: QTBUG-55254
Change-Id: Ibdb6645f63957d28a89c50b51faeb3aea086a8b3
Reviewed-by: Kai Koehne <kai.koehne@qt.io>
|
| |
| |
| |
| |
| |
| |
| |
| | |
Chromium defaults to arm_neon=1 and then sets -mfpu=neon if the arm
architecture is 7.
Change-Id: Ib144dd4188ba4221ed35367026de9f9a04c69792
Reviewed-by: Michal Klocek <michal.klocek@qt.io>
|
| |
| |
| |
| |
| | |
Change-Id: I44105c768a958714590b979d3877724f0db659ee
Reviewed-by: Michael Brüning <michael.bruning@qt.io>
|
|/
|
|
|
|
|
|
|
|
|
| |
Add a lightweight "fake" QWebEngineView class that is returned when Qt Designer
queries the default property values by calling
QDesignerCustomWidgetInterface::createWidget() with 0 parent,
preventing crashes during QWebEngine initialization.
Task-number: QTBUG-53984
Change-Id: Iced64b7d8af4c958fe7e29fa2f64bb9b681fbab2
Reviewed-by: Kai Koehne <kai.koehne@qt.io>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
There was introcuded a new logic which proves that
no mouse / keyboard events are forwarded to
Chromium if the view has no focus, and activeFocusOnPress is set to false.
The selection box get focus when the user click on it,
but the popup window never get (focus is on QuickRootItem)
in this case these mouse events are ignored to forward to Chromium.
Task-number: QTBUG-54795
Change-Id: Id6e81ee39dcde21a6c5c46e302888b9e9478352f
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
|
|
|
|
|
|
|
| |
Pulls in fixes for webcursors and building against FFMPEG3
Change-Id: Ic852604eea1a15cbf96decd536278f01a7f9dcf3
Reviewed-by: Michael Brüning <michael.bruning@qt.io>
|
|
|
|
|
|
|
|
|
|
|
|
| |
Fix for QTBUG-48969 was only half-baked patch and
introduced massive memory leak on fbo and texture
alloctaions. Delete fbo and extra allocated
textures.
Task-number: QTBUG-52575
Task-number: QTBUG-48969
Change-Id: I2148f37cd27dab9e40ab72caeb6857752b69379f
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
|
|
|
|
|
|
|
|
| |
Ensure that if QWebEnginePage::createWindow returns 'this' that we fall
back to navigating in current tab.
Change-Id: Idffe25dcafaaf3c824815b3cf1f0e400eaec2923
Reviewed-by: Michal Klocek <michal.klocek@qt.io>
|
|
|
|
|
|
|
| |
Pulls in security updates from Chromium 52
Change-Id: I4fb5183a321b45fdf5515f0a88100b48697ee3aa
Reviewed-by: Michael Brüning <michael.bruning@qt.io>
|
|
|
|
|
| |
Change-Id: I86eea3a64a38f6ec9984f663499abffa03cd9eb2
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
|
|
|
|
|
|
|
| |
This list is shown in qtdoc/qmlmodules.html
Change-Id: Ib51f0405e7581571f8a124ea6806e7bc4982dba3
Reviewed-by: Leena Miettinen <riitta-leena.miettinen@qt.io>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Previously, when a WebEngineView's RenderWidgetHostViewQtDelegateQuick
had focus, but not activeFocus, Chromium was told that it had keyboard
focus. That is not correct, because having focus inside the FocusScope
(the WebEngineView), does not mean it has keyboard focus, it just
means the RWHVQDQ will receive active focus, when the WebEngineView
does.
The call path for that erronous check was done when a new page was
loaded, specifically going through
NavigatorImpl::DidNavigate() ->
RenderFrameHostManager::CommitPending() ->
render_frame_host_->GetView()->HasFocus() ->
RenderWidgetHostViewQtDelegateQuick::hasKeyboardFocus(),
which resulted in incorrect display of a blinking caret in an
input, when the view actually wasn't focused.
The fix consists in checking for the activeFocus rather than the
focus property.
Change-Id: I97002be3fc8adfec2228a1b5ad62696f8872d3cc
Reviewed-by: Michael Brüning <michael.bruning@qt.io>
|
|
|
|
|
|
| |
Task-number: QTBUG-54114
Change-Id: I63ae6c47eb76a4acf85cd5b0c7adb21399a985dd
Reviewed-by: Kai Koehne <kai.koehne@qt.io>
|
|
|
|
|
|
|
|
|
| |
The docs do not currently get built, because docs for a QML type
need to reside in one file. Also edited the docs for grammar and
style.
Change-Id: Ic62f293e113d0e4bfe0497ce4a665a2a42971eb3
Reviewed-by: Peter Varga <pvarga@inf.u-szeged.hu>
|
|
|
|
|
|
|
|
|
|
|
| |
Chromium does not exclude localhost connection from its proxy rules by
default.
When using a qt proxy with any other form of qt-based connections,
connections to localhost are excluded by default.
This patch adds localhost connections to the proxy-bypass rules.
Change-Id: I76c43a2ae0de8d8fad455445a64a739c6c6b40f0
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
|
|
|
|
|
|
|
|
|
| |
Changes core_generated.gyp from being generated in the source directory
to being generated in the build directory.
Task-number: QTBUG-43014
Change-Id: Ia67df47bfadbf5dfca6e60a613dcf7b162b468fd
Reviewed-by: Michael Brüning <michael.bruning@qt.io>
|
|
|
|
|
|
|
|
|
|
|
| |
Since the profile owns the QWebEngineDownloadItem then it is likely that
it will still be around when the QWebEngineProfile is deleted. As the
QWebEngineDownloadItem is a child of the QWebEngineProfile then it will
try to delete it when deleting the QWebEngineProfile which means it cannot
trigger a function call into QWebEngineProfilePrivate.
Change-Id: I51077a7857fb49a6708224a9e9942d17de6f6778
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
|
|
|
|
|
|
| |
Task-number: QTBUG-54644
Change-Id: Ief2e92e6baf3057cb4c1835557c51c4728c34ec0
Reviewed-by: Michael Brüning <michael.bruning@qt.io>
|
|
|
|
|
| |
Change-Id: I51562b2dff1ffe4359c6db4a802c406c3706de84
Reviewed-by: Leena Miettinen <riitta-leena.miettinen@qt.io>
|
|
|
|
|
| |
Change-Id: Idf18d236816aab12fb01e1b5725e5ad1c73dbaad
Reviewed-by: Leena Miettinen <riitta-leena.miettinen@qt.io>
|
|
|
|
|
|
|
| |
Qt WebEngine and Qt WebEngine Core also have C++ API.
Change-Id: I4fa4219cd9c0a984f736585caf84676e64f1155a
Reviewed-by: Leena Miettinen <riitta-leena.miettinen@qt.io>
|
|
|
|
|
|
|
|
|
|
| |
Currently it was placed under WebEngineProcess, which is confusing.
That the scene graph is used for the QtWebEngine module is probably
not really interesting, but it's certainly suprising for Qt WebEngine
Widgets, hence mention it there.
Change-Id: I1a87b0b0d32cc2695c4dfa31aae56c517e39124e
Reviewed-by: Leena Miettinen <riitta-leena.miettinen@qt.io>
|
|
|
|
|
| |
Change-Id: Ie82c3b34974b391ca69f82d00165c0d46ad6b4af
Reviewed-by: Leena Miettinen <riitta-leena.miettinen@qt.io>
|
|
|
|
|
|
| |
Change-Id: I75d3f5641ccd402ea1a167d14533b70df5744ff7
GPush-Base: 8ca4a41dfe6887c2637fe2e562f5314b56facd20
Reviewed-by: Leena Miettinen <riitta-leena.miettinen@qt.io>
|
|
|
|
|
|
| |
Change-Id: I40d9c30dd33eb38d1d33c68df4cb4bbc3704c647
GPush-Base: 8ca4a41dfe6887c2637fe2e562f5314b56facd20
Reviewed-by: Leena Miettinen <riitta-leena.miettinen@qt.io>
|
|
|
|
|
|
|
|
|
|
| |
- Fix typo
- Add dependencies between submodules
- Add sources for diagrams
Task-number: QTBUG-54382
Change-Id: I88aa1dfd0ba16f3d3ca26d8c9cfa05b0ec1e9c83
Reviewed-by: Kai Koehne <kai.koehne@qt.io>
|
|
|
|
|
|
|
|
|
|
|
|
| |
Consolidate the description of related modules in a separate section,
and place them less prominently at the end of the overview.
Also
- remove mentioning of (ambiguous) 'web runtime'
- fix spelling of supersede
Change-Id: Iccac376a6b602ee9d23efd4b856e65b9292d6382
Reviewed-by: Leena Miettinen <riitta-leena.miettinen@qt.io>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Channel-ids are only supposed to be memory only when running in off
the record profiles. We have just never initialized the sql-based
backend.
We follow the cookie-settings, because channel-ids are used together
with cookies, have similar implications, and newer Chromium versions
will assert that cookie-store and channel-id store have matching
storage models.
Change-Id: I0a64146f0ed36a8913706bfc3fcadd7404894745
Reviewed-by: Michael Brüning <michael.bruning@qt.io>
|
|
|
|
|
|
|
|
| |
A copy-paste error meant we only updated user-agent settings when
cookie store settings changed on a profile.
Change-Id: I173ea7dfc309a30a3d3b98ee5ccab74e2abec456
Reviewed-by: Michal Klocek <michal.klocek@qt.io>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Qt WebEngine uses `isOpenGLES` to know whether the OpenGL implementation is
EGL with OpenGL ES 2.0. However, some non-ES OpenGL implementation such as
eglfs_x11 uses EGL and are compatible with OpenGL ES 2.0. This change
allows to use them.
Also the change will allow to detect incompatible combinations.
(i.e. EGL + ES-incompatible OpenGL, API other than EGL + OpenGL ES)
Change-Id: I0abea253696d06ec365bde2176663700e8567f45
Reviewed-by: Alexandru Croitor <alexandru.croitor@qt.io>
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Suppose having a QWebEngineView as one of many child widgets in a
layout. Resize some child widget such that the position of the
QWebEngineView changes (without changing the position of the top-level
window). Chromium must be informed of the changed global position,
otherwise popups will be opened at the old position.
Also see commit 7f941a34, which originally introduced the coordinate
propagation for top-level window moves.
Task-number: QTBUG-51244
Change-Id: Ieb372e8d7554700d5e8d1e2148ab778094ea3878
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Consider a QWebEnginePage subclass that does "return this;" in
createWindow. Commit 1f07d2929a made this a no-op to prevent QtWebEngine
from crashing. The reason for the crash was access to deleted memory
after destroying the current adapter in adoptNewWindow.
Defer the adoption in this case to whenever we hit the event loop
again.
Change-Id: I9674d80ef8b2f301c1446ff505b2486649451ba6
Task-number: QTBUG-42216
Reviewed-by: Michal Klocek <michal.klocek@qt.io>
|
|
|
|
|
|
|
|
|
| |
While 'button' should officially be none on mouse move events, the
aura and windows events set 'button' on mouse move, and selection code
appears to depend on it.
Change-Id: I49f84e6f9178c3b2cb0f2c2c8a7b1d30141d0b4e
Reviewed-by: Michael Brüning <michael.bruning@qt.io>
|
|
|
|
|
|
|
|
| |
The division needs to be done in float, otherwise steps smaller than
one standard tick will be rounded to zero.
Change-Id: Id939be062f1575104ca6cd66f05892841ec27569
Reviewed-by: Michael Brüning <michael.bruning@qt.io>
|
|
|
|
|
|
|
|
|
| |
QExplicitSharedDataPointer is meant for value objects, not for shared
objects. Instead switch to using QSharedPointer.
Change-Id: Ib3791bbcfde627a67508f2819e141d8c538a4a50
Reviewed-by: Michael Brüning <michael.bruning@qt.io>
Reviewed-by: Michal Klocek <michal.klocek@theqtcompany.com>
|
|
|
|
|
|
|
|
|
| |
While the script collection always follows a page, it is safer to have
both hold a reference to the web-contents adapter, so it doesn't get
deleted while adapters are being changed.
Change-Id: I21e268f7228f13702468df61e0032bdf2f99873a
Reviewed-by: Michael Brüning <michael.bruning@qt.io>
|