GitLab

If the product sets PRODUCT_PREBUILT_WEBVIEWCHROMIUM, do not attempt to
build the WebView from source.

Bug: 10824652
Change-Id: I33c34b768a5f051e530d1a67303b831284e1edb5

BUG: 10892941
Change-Id: Ie8dbd5c441573b0bbbd75fb9926813556af93add

Skip calling glBindRenderbufferEXT if it's not changed from 0.

Android WebView for now do not use this feature since we do not support
WebGL and such. This is a workaround for what is suspected is a qualcomm
driver bug.

BUG: 11061206
Change-Id: Ic6d8b20810cfe9f24e85cac94e86cdc7a631865b

Chromium stores the extra headers passed to loadUrl in the
NavigationState and applies them to all requests initiated in that frame
while that's the topmost navigation. This is different to the old WebView's
behaviour where these headers are only applied to the load of the main
frame's URL. Modify RenderFrameImpl::willSendRequest to add the headers
more selectively pending a better way to do this (upstream bug
http://crbug.com/306873).

Bug: 10566204
Change-Id: Ie51f00a1739dcb03adf7d6c744ddd912df6a3965

Cherry-pick of Chromium crrev.com/r228366

BUG: 11063885

Original description:

Using getDownTime is better in most cases because it would not be
affected by tiny delays in delivering the event to Chrome.

However, there are situations in Android WebView when the input event
stream can be arbitrarily delayed, in which case using getDownTime is
not appropriate since it can be long in the past. So getEventTime is
the time when the event is delivered. This also matches with how the
rest of android views handle long press events.

Change-Id: I1de24e5d7d2a40fe97fccb4039b09450bba539d2

Bug: 11174127

Keep a ref to WebView in AwPdfExporter to prevent Java side
being deleted while PrintDocumentAdapter is in use.

Change-Id: I950fa532859b4428efeaf87c74b77e97bc0342ac

Cherry-pick of Chromium crrev.com/r228287

Note the conflict is in Google TV code and does not apply to webview.

Conflicts:
	content/renderer/media/android/webmediaplayer_android.cc

BUG: 11158022

Original Description:

In android webview, StreamTextureProxy cannot be created until the page
has entered hardware accelerated mode. This can usually be racy between
when the WebMediaPlayerAndroid is first create. In this patch, retry to
create the StreamTextureProxy in play, to help reduce the possiblility
of this race.

Change-Id: I6e9b166f21979885056bd467838cb0406ae3a1a9

Cherry-pick of Chromium crrev.com/r228222

Removed the new test file to avoid unnecessary gyp change.

Conflicts:
	cc/trees/layer_tree_host_impl.cc

BUG: 11164012

Original description:

Currently will NPE in HeadsUpDisplayLayerImpl::UpdateHudTexture
immediately after a DidLoseOutputSurface call that throws away the
resource. Resource loss can happen quite frequently under normal usage
in Android WebView.

Also fixed a condition WillDraw as writing the test for issue above.

Fix for internal bug b/11164012.

TEST=HeadsUpDisplayLayerImplTest.UpdateAfterResourceLoss

Change-Id: Ia0c12f8357f4e5e0ed6e63a5058a940241696466

This is to support mobile frameworks that dynamically add several
<meta viewport> tags. Instead of discarding the old tag,
we should merge their values.

Bug: 11141251

Upstream issue: http://crbug.com/306548

Change-Id: I63b2bad271bdfb3ff2b195d42cf1b8ad5151e6b4

This commit was generated by merge_to_master.py.

Change-Id: Ie9e9e034f3e5d9a3fe3edf87595c7be467d3136b

This commit was generated by merge_to_master.py.

Change-Id: I9152e5618c61fb2d560d0d0cbd38f8639e477793

Cherry-pick http://crrev.com/26705002

> We should update the fixedLayoutSize even if the physical size of
> the AwContents doesn't change. This is because the fixedLayoutSize
> is directly derived from physical size and the pageScaleFactor, so
> if either of these change the fixedLayoutSize should be updated.
> Currently the AwLayoutSizer code assumes that every requestLayout
> results in an onSizeChanged, but that's not always the case.
>
> BUG=internal b/10917278
> TEST=AndroidWebViewTest
> R=joth@chromium.org
>
> Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=227932

BUG: 10917278
Change-Id: If4bee85eae233770ad994a62242bfcb1baeda7ad

Cherry-pick of http://crrev.com/227928

Bug: 10813938

Original description:

Instead of disabling 2d canvas antialiasing for Android in
content_startup_flags (which can't easily be overridden by content
embedders), disable it in chrome_startup_flags (which disables it for
Chrome and Chromium TestShell on Android, which is what was desired) and
also in shell_main_delegate (which disables it for ContentShell on
Android for the purposes of consistency in manual/automated testing).
Other content embedders (such as android_webview) are now free to
control this flag themselves.

BUG=213238
R=pfeldman@chromium.org, senorblanco@chromium.org, yfriedman@chromium.org

Review URL: https://codereview.chromium.org/26667005

Change-Id: I00a2e3f13a7fed7c1f82135dddf573a7f48dafde
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@227928 0039d316-1c4b-4281-b951-d872f2087c98

Cherry-pick http://crrev.com/26705002

> We should update the fixedLayoutSize even if the physical size of
> the AwContents doesn't change. This is because the fixedLayoutSize
> is directly derived from physical size and the pageScaleFactor, so
> if either of these change the fixedLayoutSize should be updated.
> Currently the AwLayoutSizer code assumes that every requestLayout
> results in an onSizeChanged, but that's not always the case.
>
> BUG=internal b/10917278
> TEST=AndroidWebViewTest
> R=joth@chromium.org
>
> Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=227932

BUG: 10917278
Change-Id: If4bee85eae233770ad994a62242bfcb1baeda7ad

Cherry-pick of chromium http://crrev.com/26460004

Bug: 11026947

Original description:

This enables the browser to return screen.width / availWidth (and height),
@media device-width (and height), window.outerWidth (and height),
but not window.innerWidth (and height) in physical rather than CSS
pixels.

BUG=305236

Change-Id: I8050131f5ba39aaa25efcce900d227e1da81da01

bug 10569530

Making a klp workaround for apps impacted by these invalid
URL resolve. Taking this upstream is TBD

Change-Id: I658b9d9caf0b90d644b2291953c43220b12143b8

Cherry-pick of http://crrev.com/26274003.

> Use total page scale in the LayerScrollOffsetDelegate.
>
> In order for the LayerScrollOffsetDelegate to see the most up to
> date pageScale we need to use the total page scale rather than
> just the value that was last synced with the main thread.
>
> BUG=None
> TEST=cc_unittests

BUG: 10806014
Change-Id: I6d0c7dad789bf799e3122e19dc70d522974cfb42

Cherry-pick of uncommitted https://codereview.chromium.org/24676002/ PS2

Depends on Change-Id: Ica4cdeaa35b971b75ef1e664cd54f3fc7f95507a in
third_party/WebKit, which is also a fork from upstream Chromium, so this
is also a fork.

Conflicts:
	content/renderer/web_preferences.cc

BUG: 10750978

Change-Id: I8cab4b5ac209c55c2ac5a498cd4dc72f6e0a4229

If the first CookieManager API called by the app was
SetAcceptFileSchemeCookies or AllowFileSchemeCookies, the app would
crash with a segfault as the CookieMonster has not yet been created. Add
the call to ensure that it exists (creating the temporary one if needed)
before accessing it.

Bug: 11141036
Change-Id: I7a8a5960b159d34adfbd9aa00adf05896ea207d0

Merge "Cherry-pick http://crrev.com/25233002 "Make Android WebView filtering depend on scrolling status."" into klp-dev

This fixes an image quality regression which affects to some degree
almost every software WebView using an image, by introducing the same
workaround as classic WebView.

BUG: 10706494
Change-Id: I758257c5046b167780e5fcd07918f1855f08760c

Bug 11070366

Taken from
https://src.chromium.org/viewvc/chrome?revision=227511&view=revision and
straight-forward reolve (new setting was added amonst other settings
that didn't exist back in M30)

Original description:

Plumb the flag to blink and the compositor.
Disable by default in Android webview for now.

Depends on blink CL: https://src.chromium.org/viewvc/blink?view=rev&revision=158956

BUG=303945

Review URL: https://codereview.chromium.org/25904004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@227511 0039d316-1c4b-4281-b951-d872f2087c98

Conflicts:
	cc/base/switches.cc
	cc/base/switches.h
	cc/trees/layer_tree_settings.cc
	content/renderer/web_preferences.cc
	webkit/common/webpreferences.cc
	webkit/common/webpreferences.h

Change-Id: I140594e9b280f408c7b237ee5f17c569911f5b85