Restrict extensionview to chrome-extension://

Restrict extensionview to chrome-extension:// only.

To set an extension id for the extensionview, there must be a  connect(extensionid, src); call. After an extension has been set, it cannot be reset without the guestview being destroyed/recreated. Once an extension has been set, the src can be modified to navigate to different pages for that extension.

The extensionview will navigate to pages with format of chrome-extension://extensionid/page.html, where:
extension="extensionid" and src="page.html"

BUG=459680
Committed: https://crrev.com/19c01ac0d65560febf3d2ea5e843ff8a94afb694
Cr-Commit-Position: refs/heads/master@{#316884}

[apacible, 2015-02-12 22:42:37]

Patchset #1 (id:1) has been deleted

[apacible, 2015-02-13 16:38:48]

Patchset #1 (id:20001) has been deleted

[apacible, 2015-02-13 17:41:08]

Patchset #1 (id:40001) has been deleted

[apacible, 2015-02-13 18:30:30]

Patchset #1 (id:60001) has been deleted

[apacible, 2015-02-13 18:32:45]

Patchset #1 (id:80001) has been deleted

[apacible, 2015-02-13 19:33:20]

Patchset #1 (id:100001) has been deleted

[apacible, 2015-02-13 23:15:42]

Patchset #1 (id:120001) has been deleted

[apacible, 2015-02-13 23:15:47]

Patchset #1 (id:140001) has been deleted

[apacible, 2015-02-13 23:47:43]

Patchset #1 (id:160001) has been deleted

[apacible, 2015-02-14 00:16:38]

Patchset #1 (id:180001) has been deleted

[apacible, 2015-02-14 00:34:27]

Patchset #1 (id:200001) has been deleted

[apacible, 2015-02-14 01:40:18]

Patchset #1 (id:220001) has been deleted

[apacible, 2015-02-14 01:42:08]

Patchset #1 (id:240001) has been deleted

[apacible, 2015-02-14 01:44:44]

apacible@chromium.org changed reviewers:
+ fsamuel@chromium.org

[apacible, 2015-02-14 01:45:51]

fsamuel, PTAL. Thanks!

[Fady Samuel, 2015-02-16 19:26:28]

https://codereview.chromium.org/913393003/diff/260001/extensions/browser/gues...
File extensions/browser/guest_view/extension_view/extension_view_guest.cc
(right):

https://codereview.chromium.org/913393003/diff/260001/extensions/browser/gues...
extensions/browser/guest_view/extension_view/extension_view_guest.cc:44: GURL
url(GetExtensionViewPage(src));
I would follow WebViewGuest's code to navigate to about:blank if this is a bad
src.

Also, I would plumb loadcommit up to update the src attribute.

https://codereview.chromium.org/913393003/diff/260001/extensions/browser/gues...
extensions/browser/guest_view/extension_view/extension_view_guest.cc:74: GURL
extension_url_from_id =
I would keep things as a GURL rather than as a string.

https://codereview.chromium.org/913393003/diff/260001/extensions/browser/gues...
extensions/browser/guest_view/extension_view/extension_view_guest.cc:156: return
extension_url_ + src;
I would model this after WebViewGuest::ResolveURL. There's a bunch of validity
checks that are built into GURL that are not captured here.

https://codereview.chromium.org/913393003/diff/260001/extensions/renderer/res...
File extensions/renderer/resources/guest_view/extension_view.js (right):

https://codereview.chromium.org/913393003/diff/260001/extensions/renderer/res...
extensions/renderer/resources/guest_view/extension_view.js:57:
console.log(this.attributes[i].name);
Remove this.

https://codereview.chromium.org/913393003/diff/260001/extensions/renderer/res...
File extensions/renderer/resources/guest_view/extension_view_api_methods.js
(right):

https://codereview.chromium.org/913393003/diff/260001/extensions/renderer/res...
extensions/renderer/resources/guest_view/extension_view_api_methods.js:22:
'getProcessId',
Do you need this?

https://codereview.chromium.org/913393003/diff/260001/extensions/renderer/res...
extensions/renderer/resources/guest_view/extension_view_api_methods.js:41:
return this.processId;
What's this? How is this populated?

https://codereview.chromium.org/913393003/diff/260001/extensions/renderer/res...
File extensions/renderer/resources/guest_view/extension_view_attributes.js
(right):

https://codereview.chromium.org/913393003/diff/260001/extensions/renderer/res...
extensions/renderer/resources/guest_view/extension_view_attributes.js:70:
ExtensionViewAttribute.prototype.reset = function() {
Rebase?

[apacible, 2015-02-17 16:18:49]

Patchset #2 (id:280001) has been deleted

[apacible, 2015-02-17 18:30:05]

Patchset #2 (id:300001) has been deleted

[apacible, 2015-02-17 18:31:26]

fsamuel, PTAL. I have a couple of questions as well. Thanks!

https://codereview.chromium.org/913393003/diff/260001/extensions/browser/gues...
File extensions/browser/guest_view/extension_view/extension_view_guest.cc
(right):

https://codereview.chromium.org/913393003/diff/260001/extensions/browser/gues...
extensions/browser/guest_view/extension_view/extension_view_guest.cc:44: GURL
url(GetExtensionViewPage(src));
On 2015/02/16 19:26:28, Fady Samuel wrote:
> I would follow WebViewGuest's code to navigate to about:blank if this is a bad
> src.
> 
> Also, I would plumb loadcommit up to update the src attribute.

Navigates to about:blank, and followed WebViewGuest's method of checking src.

Haven't plumbed loadcommit yet. Should I also plumb loadabort?

https://codereview.chromium.org/913393003/diff/260001/extensions/browser/gues...
extensions/browser/guest_view/extension_view/extension_view_guest.cc:74: GURL
extension_url_from_id =
On 2015/02/16 19:26:28, Fady Samuel wrote:
> I would keep things as a GURL rather than as a string.

Changed extension_url_ to a GURL.

https://codereview.chromium.org/913393003/diff/260001/extensions/browser/gues...
extensions/browser/guest_view/extension_view/extension_view_guest.cc:156: return
extension_url_ + src;
On 2015/02/16 19:26:28, Fady Samuel wrote:
> I would model this after WebViewGuest::ResolveURL. There's a bunch of validity
> checks that are built into GURL that are not captured here.

Done.

https://codereview.chromium.org/913393003/diff/260001/extensions/renderer/res...
File extensions/renderer/resources/guest_view/extension_view.js (right):

https://codereview.chromium.org/913393003/diff/260001/extensions/renderer/res...
extensions/renderer/resources/guest_view/extension_view.js:57:
console.log(this.attributes[i].name);
On 2015/02/16 19:26:28, Fady Samuel wrote:
> Remove this.

Done.

https://codereview.chromium.org/913393003/diff/260001/extensions/renderer/res...
File extensions/renderer/resources/guest_view/extension_view_api_methods.js
(right):

https://codereview.chromium.org/913393003/diff/260001/extensions/renderer/res...
extensions/renderer/resources/guest_view/extension_view_api_methods.js:22:
'getProcessId',
On 2015/02/16 19:26:28, Fady Samuel wrote:
> Do you need this?

No. I kept this around to make a note of it but forgot to remove it.

https://codereview.chromium.org/913393003/diff/260001/extensions/renderer/res...
extensions/renderer/resources/guest_view/extension_view_api_methods.js:41:
return this.processId;
On 2015/02/16 19:26:28, Fady Samuel wrote:
> What's this? How is this populated?

Removed.

https://codereview.chromium.org/913393003/diff/260001/extensions/renderer/res...
File extensions/renderer/resources/guest_view/extension_view_attributes.js
(right):

https://codereview.chromium.org/913393003/diff/260001/extensions/renderer/res...
extensions/renderer/resources/guest_view/extension_view_attributes.js:70:
ExtensionViewAttribute.prototype.reset = function() {
On 2015/02/16 19:26:28, Fady Samuel wrote:
> Rebase?

Not sure what this comment means. This is used in onElementDetached to set all
the attribute values to empty.

https://codereview.chromium.org/913393003/diff/320001/extensions/browser/gues...
File extensions/browser/guest_view/extension_view/extension_view_guest.cc
(right):

https://codereview.chromium.org/913393003/diff/320001/extensions/browser/gues...
extensions/browser/guest_view/extension_view/extension_view_guest.cc:62: GURL
validated_url(url);
WebViewGuest creates a validated_url here, but is there any difference between
using this and just passing in the earlier url? Is it just for readability?

[Fady Samuel, 2015-02-17 18:48:29]

I would also verify that the main frame never navigates outside of the
extension:

https://code.google.com/p/chromium/codesearch#chromium/src/extensions/browser...

Is an example of how to do that.

https://codereview.chromium.org/913393003/diff/260001/extensions/browser/gues...
File extensions/browser/guest_view/extension_view/extension_view_guest.cc
(right):

https://codereview.chromium.org/913393003/diff/260001/extensions/browser/gues...
extensions/browser/guest_view/extension_view/extension_view_guest.cc:44: GURL
url(GetExtensionViewPage(src));
On 2015/02/17 18:31:25, apacible wrote:
> On 2015/02/16 19:26:28, Fady Samuel wrote:
> > I would follow WebViewGuest's code to navigate to about:blank if this is a
bad
> > src.
> > 
> > Also, I would plumb loadcommit up to update the src attribute.
> 
> Navigates to about:blank, and followed WebViewGuest's method of checking src.
> 
> Haven't plumbed loadcommit yet. Should I also plumb loadabort?

I would plumb loadcommit at least. loadcommit should probably update the src
attribute as it does in <webview>.

https://codereview.chromium.org/913393003/diff/320001/extensions/browser/gues...
File extensions/browser/guest_view/extension_view/extension_view_guest.cc
(right):

https://codereview.chromium.org/913393003/diff/320001/extensions/browser/gues...
extensions/browser/guest_view/extension_view/extension_view_guest.cc:50:
(!content::ChildProcessSecurityPolicy::GetInstance()->IsWebSafeScheme(
Wow, this works? chrome-extension:// is not a WebSafeScheme I think?

https://codereview.chromium.org/913393003/diff/320001/extensions/browser/gues...
extensions/browser/guest_view/extension_view/extension_view_guest.cc:62: GURL
validated_url(url);
On 2015/02/17 18:31:26, apacible wrote:
> WebViewGuest creates a validated_url here, but is there any difference between
> using this and just passing in the earlier url? Is it just for readability?

That's probably some old code that didn't get refactored well somewhere along
the way.

https://codereview.chromium.org/913393003/diff/320001/extensions/browser/gues...
extensions/browser/guest_view/extension_view/extension_view_guest.cc:162: GURL
default_url(base::StringPrintf("%s://%s/%s",
This doesn't seem necessary to me.

return extension_url_.Resolve(src);

should do the trick I think.

Now that I think about it, you can probably do this inline in NavigateGuest.

https://codereview.chromium.org/913393003/diff/320001/extensions/renderer/res...
File extensions/renderer/resources/guest_view/extension_view_api_methods.js
(right):

https://codereview.chromium.org/913393003/diff/320001/extensions/renderer/res...
extensions/renderer/resources/guest_view/extension_view_api_methods.js:18:
'connect',
nit: remove comma.

[apacible, 2015-02-17 20:07:30]

Plumbing loadcommit in next diff, but wanted to get a comment response out.

https://codereview.chromium.org/913393003/diff/320001/extensions/browser/gues...
File extensions/browser/guest_view/extension_view/extension_view_guest.cc
(right):

https://codereview.chromium.org/913393003/diff/320001/extensions/browser/gues...
extensions/browser/guest_view/extension_view/extension_view_guest.cc:50:
(!content::ChildProcessSecurityPolicy::GetInstance()->IsWebSafeScheme(
On 2015/02/17 18:48:29, Fady Samuel wrote:
> Wow, this works? chrome-extension:// is not a WebSafeScheme I think?

Yeah, it does! I know chrome:// isn't a WebSafeScheme, and checked that
something like chrome://settings gets blocked. Since we really just want
chrome-extension://extensionid/pagetonavigateto.html URLs, it may make more
sense to check for that.. something like
https://code.google.com/p/chromium/codesearch#chromium/src/chrome/browser/bro...
?

https://codereview.chromium.org/913393003/diff/320001/extensions/browser/gues...
extensions/browser/guest_view/extension_view/extension_view_guest.cc:62: GURL
validated_url(url);
On 2015/02/17 18:48:29, Fady Samuel wrote:
> On 2015/02/17 18:31:26, apacible wrote:
> > WebViewGuest creates a validated_url here, but is there any difference
between
> > using this and just passing in the earlier url? Is it just for readability?
> 
> That's probably some old code that didn't get refactored well somewhere along
> the way.

Acknowledged.

https://codereview.chromium.org/913393003/diff/320001/extensions/browser/gues...
extensions/browser/guest_view/extension_view/extension_view_guest.cc:162: GURL
default_url(base::StringPrintf("%s://%s/%s",
On 2015/02/17 18:48:29, Fady Samuel wrote:
> This doesn't seem necessary to me.
> 
> return extension_url_.Resolve(src);
> 
> should do the trick I think.
> 
> Now that I think about it, you can probably do this inline in NavigateGuest.

Done, added inline.

https://codereview.chromium.org/913393003/diff/320001/extensions/renderer/res...
File extensions/renderer/resources/guest_view/extension_view_api_methods.js
(right):

https://codereview.chromium.org/913393003/diff/320001/extensions/renderer/res...
extensions/renderer/resources/guest_view/extension_view_api_methods.js:18:
'connect',
On 2015/02/17 18:48:29, Fady Samuel wrote:
> nit: remove comma.

Done.

[Fady Samuel, 2015-02-17 20:17:04]

https://codereview.chromium.org/913393003/diff/340001/extensions/browser/gues...
File extensions/browser/guest_view/extension_view/extension_view_guest.cc
(right):

https://codereview.chromium.org/913393003/diff/340001/extensions/browser/gues...
extensions/browser/guest_view/extension_view/extension_view_guest.cc:53: if
(scheme_is_blocked || !url.is_valid()) {
I think a simpler check might be:

// If the URL is not valid or it's not about:blank and not the same origin as
the extension, then navigate to about:blank.
if (!url.is_valid() || ((url != GURL(url::kAboutBlankURL) && (url.GetOrigin() !=
extension_url_.GetOrigin()))) {
  NavigateGuest(url::kAboutBlankURL, true /* force_navigation */);
  return;
}

[apacible, 2015-02-17 22:57:47]

Patchset #4 (id:360001) has been deleted

[apacible, 2015-02-17 23:02:23]

Patchset #4 (id:380001) has been deleted

[apacible, 2015-02-17 23:02:58]

PTAL -- Added plumbing for loadcommit. Thanks!

https://codereview.chromium.org/913393003/diff/340001/extensions/browser/gues...
File extensions/browser/guest_view/extension_view/extension_view_guest.cc
(right):

https://codereview.chromium.org/913393003/diff/340001/extensions/browser/gues...
extensions/browser/guest_view/extension_view/extension_view_guest.cc:53: if
(scheme_is_blocked || !url.is_valid()) {
On 2015/02/17 20:17:04, Fady Samuel wrote:
> I think a simpler check might be:
> 
> // If the URL is not valid or it's not about:blank and not the same origin as
> the extension, then navigate to about:blank.
> if (!url.is_valid() || ((url != GURL(url::kAboutBlankURL) && (url.GetOrigin()
!=
> extension_url_.GetOrigin()))) {
>   NavigateGuest(url::kAboutBlankURL, true /* force_navigation */);
>   return;
> }

Sounds good. Changed.

[Fady Samuel, 2015-02-17 23:20:36]

https://codereview.chromium.org/913393003/diff/400001/extensions/browser/gues...
File extensions/browser/guest_view/extension_view/extension_view_guest.cc
(right):

https://codereview.chromium.org/913393003/diff/400001/extensions/browser/gues...
extensions/browser/guest_view/extension_view/extension_view_guest.cc:48: bool
urlNotAllowed = (url != GURL(url::kAboutBlankURL)) &&
nit: url_not_allowed

https://codereview.chromium.org/913393003/diff/400001/extensions/renderer/res...
File extensions/renderer/resources/guest_view/extension_view.js (right):

https://codereview.chromium.org/913393003/diff/400001/extensions/renderer/res...
extensions/renderer/resources/guest_view/extension_view.js:69: if (oldValue !=
newValue) {
No need for this. If this is in webview then it's cruft. I'd just call:

this.attributes[ExtensionViewConstants.ATTRIBUTE_SRC].
    setValueIgnoreMutation(newValue);

https://codereview.chromium.org/913393003/diff/400001/extensions/renderer/res...
File extensions/renderer/resources/guest_view/extension_view_events.js (right):

https://codereview.chromium.org/913393003/diff/400001/extensions/renderer/res...
extensions/renderer/resources/guest_view/extension_view_events.js:16:
customHandler: function(handler, event, extensionViewEvent) {
just call this handler instead of customHandler.

Pass in ExtensionViewEvents.prototype.handleLoadCommitEvent

https://codereview.chromium.org/913393003/diff/400001/extensions/renderer/res...
extensions/renderer/resources/guest_view/extension_view_events.js:20: fields:
['url', 'isTopLevel']
These fields don't do anything here. Please remove.

https://codereview.chromium.org/913393003/diff/400001/extensions/renderer/res...
extensions/renderer/resources/guest_view/extension_view_events.js:43: var
extensionViewEvent = new Event(name, details);
Delete the two lines above?

https://codereview.chromium.org/913393003/diff/400001/extensions/renderer/res...
extensions/renderer/resources/guest_view/extension_view_events.js:44: if
(info.customHandler) {
if (info.handler)
  info.handler.call(this, e);

[apacible, 2015-02-18 00:04:12]

PTAL. Also added DidNavigateMainFrame. Thanks!

https://codereview.chromium.org/913393003/diff/400001/extensions/browser/gues...
File extensions/browser/guest_view/extension_view/extension_view_guest.cc
(right):

https://codereview.chromium.org/913393003/diff/400001/extensions/browser/gues...
extensions/browser/guest_view/extension_view/extension_view_guest.cc:48: bool
urlNotAllowed = (url != GURL(url::kAboutBlankURL)) &&
On 2015/02/17 23:20:36, Fady Samuel wrote:
> nit: url_not_allowed

Done.

https://codereview.chromium.org/913393003/diff/400001/extensions/renderer/res...
File extensions/renderer/resources/guest_view/extension_view.js (right):

https://codereview.chromium.org/913393003/diff/400001/extensions/renderer/res...
extensions/renderer/resources/guest_view/extension_view.js:69: if (oldValue !=
newValue) {
On 2015/02/17 23:20:36, Fady Samuel wrote:
> No need for this. If this is in webview then it's cruft. I'd just call:
> 
> this.attributes[ExtensionViewConstants.ATTRIBUTE_SRC].
>     setValueIgnoreMutation(newValue);

Done.

https://codereview.chromium.org/913393003/diff/400001/extensions/renderer/res...
File extensions/renderer/resources/guest_view/extension_view_events.js (right):

https://codereview.chromium.org/913393003/diff/400001/extensions/renderer/res...
extensions/renderer/resources/guest_view/extension_view_events.js:16:
customHandler: function(handler, event, extensionViewEvent) {
On 2015/02/17 23:20:36, Fady Samuel wrote:
> just call this handler instead of customHandler.
> 
> Pass in ExtensionViewEvents.prototype.handleLoadCommitEvent

Done.

https://codereview.chromium.org/913393003/diff/400001/extensions/renderer/res...
extensions/renderer/resources/guest_view/extension_view_events.js:20: fields:
['url', 'isTopLevel']
On 2015/02/17 23:20:36, Fady Samuel wrote:
> These fields don't do anything here. Please remove.

Done.

https://codereview.chromium.org/913393003/diff/400001/extensions/renderer/res...
extensions/renderer/resources/guest_view/extension_view_events.js:43: var
extensionViewEvent = new Event(name, details);
On 2015/02/17 23:20:36, Fady Samuel wrote:
> Delete the two lines above?

Done.

https://codereview.chromium.org/913393003/diff/400001/extensions/renderer/res...
extensions/renderer/resources/guest_view/extension_view_events.js:44: if
(info.customHandler) {
On 2015/02/17 23:20:36, Fady Samuel wrote:
> if (info.handler)
>   info.handler.call(this, e);

Done.

[Fady Samuel, 2015-02-18 00:40:58]

I think this is the last round. So lgtm + nits.

https://codereview.chromium.org/913393003/diff/420001/extensions/browser/gues...
File extensions/browser/guest_view/extension_view/extension_view_constants.cc
(right):

https://codereview.chromium.org/913393003/diff/420001/extensions/browser/gues...
extensions/browser/guest_view/extension_view/extension_view_constants.cc:14:
const char kAttributeExtension[] = "extension";
nit: sort alphabetically

https://codereview.chromium.org/913393003/diff/420001/extensions/browser/gues...
File extensions/browser/guest_view/extension_view/extension_view_constants.h
(right):

https://codereview.chromium.org/913393003/diff/420001/extensions/browser/gues...
extensions/browser/guest_view/extension_view/extension_view_constants.h:15:
extern const char kAttributeExtension[];
nit: sort alphabetically.

https://codereview.chromium.org/913393003/diff/420001/extensions/browser/gues...
File extensions/browser/guest_view/extension_view/extension_view_guest.cc
(right):

https://codereview.chromium.org/913393003/diff/420001/extensions/browser/gues...
extensions/browser/guest_view/extension_view/extension_view_guest.cc:75:
create_params.GetString(extensionview::kAttributeExtension, &extension_id_);
I don't see any point in storing the extension_id_ in a member variable any
more. I would keep it in a local here, and instead just store the
extension_url_.

[apacible, 2015-02-18 00:55:43]

New patchsets have been uploaded after l-g-t-m from fsamuel@chromium.org

[apacible, 2015-02-18 00:56:15]

Thanks!

https://codereview.chromium.org/913393003/diff/420001/extensions/browser/gues...
File extensions/browser/guest_view/extension_view/extension_view_constants.cc
(right):

https://codereview.chromium.org/913393003/diff/420001/extensions/browser/gues...
extensions/browser/guest_view/extension_view/extension_view_constants.cc:14:
const char kAttributeExtension[] = "extension";
On 2015/02/18 00:40:58, Fady Samuel wrote:
> nit: sort alphabetically

Done.

https://codereview.chromium.org/913393003/diff/420001/extensions/browser/gues...
File extensions/browser/guest_view/extension_view/extension_view_constants.h
(right):

https://codereview.chromium.org/913393003/diff/420001/extensions/browser/gues...
extensions/browser/guest_view/extension_view/extension_view_constants.h:15:
extern const char kAttributeExtension[];
On 2015/02/18 00:40:58, Fady Samuel wrote:
> nit: sort alphabetically.

Done.

https://codereview.chromium.org/913393003/diff/420001/extensions/browser/gues...
File extensions/browser/guest_view/extension_view/extension_view_guest.cc
(right):

https://codereview.chromium.org/913393003/diff/420001/extensions/browser/gues...
extensions/browser/guest_view/extension_view/extension_view_guest.cc:75:
create_params.GetString(extensionview::kAttributeExtension, &extension_id_);
On 2015/02/18 00:40:58, Fady Samuel wrote:
> I don't see any point in storing the extension_id_ in a member variable any
> more. I would keep it in a local here, and instead just store the
> extension_url_.

Done.

[apacible, 2015-02-18 00:58:09]

apacible@chromium.org changed reviewers:
+ asvitkine@chromium.org, rockot@chromium.org

[apacible, 2015-02-18 00:59:06]

Adding OWNERS:

rockot@:
- extensions/renderer/dispatcher.cc
- extensions/renderer/resources/extensions_renderer_resources.grd

asvitkine@:
- tools/metrics/actions/actions.xml

PTAL. Thanks!

[Ken Rockot(use gerrit already), 2015-02-18 16:31:08]

On 2015/02/18 00:59:06, apacible wrote:
> Adding OWNERS:
> 
> rockot@:
> - extensions/renderer/dispatcher.cc
> - extensions/renderer/resources/extensions_renderer_resources.grd
> 
> asvitkine@:
> - tools/metrics/actions/actions.xml
> 
> PTAL. Thanks!

said files lgtm

[Alexei Svitkine (slow), 2015-02-18 18:11:03]

lgtm

[Alexei Svitkine (slow), 2015-02-18 18:11:32]

However, this is a non-trivial enough change that it should have a BUG= and
corresponding crbug

[apacible, 2015-02-18 19:24:11]

Thanks!

On 2015/02/18 18:11:32, Alexei Svitkine wrote:
> However, this is a non-trivial enough change that it should have a BUG= and
> corresponding crbug

Added crbug.

[apacible, 2015-02-18 19:24:19]

The CQ bit was checked by apacible@chromium.org

[commit-bot: I haz the power, 2015-02-18 19:25:27]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/913393003/440001

[commit-bot: I haz the power, 2015-02-18 20:32:40]

Committed patchset #6 (id:440001)

[commit-bot: I haz the power, 2015-02-18 20:33:29]

Patchset 6 (id:??) landed as
https://crrev.com/19c01ac0d65560febf3d2ea5e843ff8a94afb694
Cr-Commit-Position: refs/heads/master@{#316884}