| Commit message (Collapse) | Author | Age | Files | Lines |
|\
| |
| |
| | |
Change-Id: I24c41d6ab08dd3594edd4ff62a1c823c904b9851
|
| |
| |
| |
| |
| |
| |
| |
| |
| | |
JavaString to a private implementation class
See http://trac.webkit.org/changeset/53449
This is required to sync the Android tree with webkit.org to allow unforking in WebCore/bridge.
Change-Id: I61cfea194d679d506d92ca3951032531d6908965
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
WebCore/bridge/Bridge
See http://trac.webkit.org/changeset/53464
This is required to sync the Android tree with webkit.org to allow unforking in WebCore/bridge.
Note that this change required an update to ScriptControllerAndroid.cpp and a manual fix in
PluginView.cpp to add a temporary ANDROID guard.
Change-Id: Ib2fe7793f7b3212e3b41324ff7dc5bb15c0e14b4
|
| |
| |
| |
| |
| |
| | |
the Qt team.
Change-Id: I2953472cee68aadf18f9dd740e9b3f69ad729cf0
|
|\ \
| |/
| |
| |
| |
| |
| |
| |
| | |
to JavaClassJSC
Merge commit 'f83060e89e8ba30d0321f33164a0c4819f837a03'
* commit 'f83060e89e8ba30d0321f33164a0c4819f837a03':
Cherry-pick WebKit change 53443 to move jni_class to JavaClassJSC
|
| |
| |
| |
| |
| |
| |
| | |
See http://trac.webkit.org/changeset/53443
This is required to sync the Android tree with webkit.org to allow unforking in WebCore/bridge.
Change-Id: I0f2b0581e0850694953f79d45242c44d16bbf5fa
|
|\ \
| |/
| |
| | |
Change-Id: If8bb4335848f51919608d28a30ee610acb120ae7
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
See http://trac.webkit.org/changeset/53436
This is required to sync the Android tree with webkit.org to allow unforking in WebCore/bridge.
Note that changes to the following were required as a result of this cherry-pick.
- WebCoreFrameBridge.cpp - updated to use JavaInstance::m_instance
- V8Binding/jni/jni_instance - Updated to rename the V8 version of JavaInstance::_instance to JavaInstance::m_instance
to allow the same code path to be used in WebCoreFrameBridge.
Change-Id: I6884f7424c8a0917095f828bda4ca62452e527b5
|
|\ \
| |/
| |
| | |
Change-Id: I02d6b5e94724091decfe0c26317554c86dcfb2da
|
| |\
| | |
| | |
| | | |
JavaInstanceJSC" into eclair-mr2
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
See http://trac.webkit.org/changeset/53412
Note that changes to WebCoreFrameBridge.cpp cpp were required as a result of this cherry-pick.
Change-Id: I0a6f6fafc5b0d5d9337ef4e339c6a0c5dd188eb9
|
|\ \ \
| |/ /
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
started. Otherwise the plugin won\'t get the rect later as it thinks it is not changed.
Merge commit '1a318d002f30c34f2a43bbd33897398ea2966f9c'
* commit '1a318d002f30c34f2a43bbd33897398ea2966f9c':
Don't call setNPWindowRect unless the plugin has
|
| | |
| | |
| | |
| | | |
started. Otherwise the plugin won't get the rect
later as it thinks it is not changed.
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
This is required to ease the process of upstreaming the Geolocation maximumAge handling code.
Note that changes to WebCoreFrameBridge.cpp and WebCoreJniOnLoad.cpp were required as a result of this.
Change-Id: Icd8524da9050acda3b991051877c3b5cf262b2b3
|
| | |
| | |
| | |
| | |
| | |
| | | |
The online event is fired when the networkType changes.
Bug: 2368650
|
|\ \ \
| | | |
| | | |
| | | | |
Geolocation calls back with a position fix."
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
calls back with a position fix.
This call was removed in https://android-git.corp.google.com/g/#change,21293
but inadvertantly added back in https://android-git.corp.google.com/g/#change,21677 See https://android-git.corp.google.com/w/?p=platform/external/webkit.git;a=commitdiff;h=e0330ac957b8434cd2c9c7b5447aaa0faabe77ec
This call is not present in webkit.org.
Change-Id: I712b5d29871e8bda54340e510c2bd7bc411728ee
|
| | | | |
|
|\ \ \ \
| | | | |
| | | | |
| | | | | |
the XHR responseXML method is being called on a document with meta tags. http://b/issue?id=2346028"
|
| |/ / /
| | | |
| | | |
| | | |
| | | |
| | | | |
This crashes in the situation where the XHR responseXML method is being
called on a document with meta tags.
http://b/issue?id=2346028
|
|\ \ \ \
| | |/ /
| |/| |
| | | |
| | | |
| | | |
| | | | |
Merge commit '93f808abccaf6f09c5c0226c01942c3be445031a'
* commit '93f808abccaf6f09c5c0226c01942c3be445031a':
Add proper support for fixed position elements
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
this is two-part change, see http://android-git.corp.google.com/g/#change,37092 for the java counterpart.
This CL implements fixed position elements (css position: fixed) using layers, supporting pixels and percent lengths.
It addresses http://b/1914365
|
|\ \ \ \
| |/ / /
| | / /
| |/ /
|/| |
| | |
| | | |
Merge commit '459f6c347256a4d9fe815793302130fe06bd8f83'
* commit '459f6c347256a4d9fe815793302130fe06bd8f83':
Fix 'disappearing layers' bug (see http://b/2371053)
|
| | |
| | |
| | |
| | |
| | |
| | | |
The removed code was needed in the previous implementation, but not so anymore as we
recopy the layers to the UI. It causes the bug, as webkit deallocate the tree asynchronously,
which ends up resetting the root layer to zero after the new page was loaded...
|
|\ \ \
| |/ /
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
changes to appendClassName." into eclair-mr2
Merge commit '594a347941318556ae40d521e6f5c23fd9cf66b4'
* commit '594a347941318556ae40d521e6f5c23fd9cf66b4':
Update jni_runtime.cpp with the latest changes to appendClassName.
|
| |\ \
| | | |
| | | |
| | | | |
into eclair-mr2
|
| | |/
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
WebKit change http://trac.webkit.org/changeset/53193 updated appendClassName to use
JSC::StringBuilder, rather than JSC-specific types. This allows this method to be
used with V8 as well as JSC.
A clean cherry-pick of this change would require changes 52026, 52028, 52075 and
52329 to be cherry-picked also. To avoid this churn, we simply update jni_runtime.cpp
to reflect the version at change 53193.
Change-Id: I6602469aeea59d8146ab1e72f312888b77522406
|
|\ \ \
| |/ /
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
extension.
Merge commit 'e4771f1b95589eff8479702dfbf5ef03c267c5c3'
* commit 'e4771f1b95589eff8479702dfbf5ef03c267c5c3':
Call into MimeTypeMap to get the mime type for an extension.
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Use the default method to obtain the content type of a plugin. Implement
extension to mime type by calling into the java map.
Requires a small change to the webkit package.
Bug: 2368893
|
|\ \ \
| | | |
| | | |
| | | | |
Change-Id: I6efcacf082170506048d0a2caf3d669796fc4253
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
not already paused.
This fixes the following scenario ...
- The browser back stack contains a page which calls Geolocation::watchPosition form its onload handler.
- User presses the back button quickly and repeatedly until the browser goes to the background.
- The browser calls WebViewCore::Pause when it goes into the background, which suspends any Geolocation
services in use. However, this call is made before the page which calls Geolocation::watchPosition has
been loaded. WebKit later loads this page, which creates a new Geolocation object which is never paused.
With this fix, the new Geolocation object is not started when it is first created. It does nothing until
it is resumed when the Browser is brought back to the foreground.
Bug: 2363338
|
|\ \ \ \ |
|
| | | | |
| | | | |
| | | | |
| | | | | |
Fix http://b/issue?id=2210105
|
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
https://bugs.webkit.org/show_bug.cgi?id=31539.
Restore the default state of a textarea correctly.
Change-Id: Id182817b7cd229e12676fd5ef55657ff6a963a4f
|
|\ \ \ \ \
| |/ / / /
|/| | / /
| | |/ /
| |/| | |
|
| | |/
| |/|
| | |
| | |
| | |
| | | |
context.
There is a companion commit in frameworks/base.
|
|\ \ \
| | | |
| | | |
| | | | |
DOMData now officially has a virtual dtor."
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
that DOMData now officially has a virtual dtor.
Changes upstreamed in https://bugs.webkit.org/show_bug.cgi?id=33390
|
|\ \ \ \
| |/ / /
|/| / /
| |/ /
| | |
| | |
| | |
| | |
| | | |
script-to-Java bridge for V8." into eclair-mr2
Merge commit '43342a276dcdb9e1f8c677f5077d151cd5d6ad07'
* commit '43342a276dcdb9e1f8c677f5077d151cd5d6ad07':
Updates the use of weak references in the script-to-Java bridge for V8.
|
| |\ \
| | | |
| | | |
| | | | |
V8." into eclair-mr2
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
Currently, V8 uses weak references to the Java object owned by JObjectWrapper.
This is in contrast to JSC, which uses strong references in JObjectWrapper.
However, for the special-case where JObjectWrapper is used in JavaInstance in
WebCoreFrameBridge, JSC uses WeakJavaInstance to swap the strong references for
weak references.
This change updates the V8 version of JObjectWrapper to use strong references
to match the JSC version. To maintain the weak reference behavior where
JObjectWrapper is used in JavaInstance in WebCoreFrameBridge, V8 now make use of
WeakJavaInstance too. This requires changes to jni_npobject to call the
necessary methods on JavaInstance to swap the references when the object is
accessed.
Change-Id: I3724d7e6437588feb0268a2670b02a93b52e54f0
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
https://bugs.webkit.org/show_bug.cgi?id=32871
https://bugs.webkit.org/show_bug.cgi?id=32873
This will be submitted once the above changes land upstream.
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
scrolls initiated by script.
Bug: 2360542
Change-Id: I4a60840dfb89753b4f56f96b5017269bfcf99392
|
|\ \ \ \
| | | | |
| | | | |
| | | | | |
webkit.org."
|
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
webkit.org.
setExtraLayoutDelay was an Android addition and was upstreamed to webkit.org in
https://bugs.webkit.org/show_bug.cgi?id=32875
Removes Document::extraLayoutDelay by incorporating the extra delay in Document::minimumLayoutDelay.
Removes ANDROID_MOBILE guards, as this extra code is not inhernetly specific to Android, and is a
no-op if Document::setExtraLayoutDelay is not called. This removes the last uses of ANDROID_MOBILE.
Change-Id: I1e18a994f32b29558dcf9a5c76aefab62e9fc13d
|
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
Bug: 2289376
modified: WebCore/rendering/RenderBlockLineLayout.cpp
|
|/ / / /
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
Skia is used by Chromium, which defines PLATFORM(SKIA).
Android also uses Skia, but currently defines PLATFORM(SGL) to account for
platform-specific differences between Android and Chromum.
We now use PLATFORM(SKIA) on Android and remove PLATFORM(SGL).
Platform-specific differences are achieved with PLATFORM(ANDROID) guards.
This reduces our diff with webkit.org.
Change-Id: Ibf920b1929e7c3cdabb025ae61ceae3df67258bb
|
|\ \ \ \
| |/ / /
| | | |
| | | |
| | | |
| | | |
| | | | |
Merge commit '3703e41d68a5747ef178cd5fc4361ee7c9e38249'
* commit '3703e41d68a5747ef178cd5fc4361ee7c9e38249':
Revert to ENABLE_ORIENTATION_EVENTS=1
|
| | | | |
|
| |/ / |
|