Enable <adview> tag for packaged apps.

Initial check-in enabling <adview> tag for packaged apps.

The <adview> tag is available only in packaged apps v2 -- when specifying the "adview" permission. It enables packaged apps to embed ads in a separate process (similar to <webview>), thus enforcing some separation between ad networks and app code for both security and privacy.

BUG=180618
Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=190043

[rpaquay, 2013-03-08 18:09:42]

Fady, Antony,

This is the "real" inital stab at implementing <adview>.

* The meat of the change is in "ad_view.js".
* Two new command line flags: "enable-adview" to enable the use of the
"<adview>" element, and "enable-adview-src-attribute" to enable adding a "src"
attribute to an <adview> element (needed for manual testing while we work on
finalizing the whitelist).
* A new corresponding permission "adview".
* A few tests.

asargent:
	chrome/app/generated_resources.grd
	chrome/browser/about_flags.cc
	chrome/browser/extensions/ad_view_browsertest.cc
	chrome/common/chrome_switches.h
	chrome/common/chrome_switches.cc
	chrome/common/extensions/api/_permission_features.json
	chrome/common/extensions/permissions/api_permission.h
	chrome/common/extensions/permissions/api_permission.cc
	chrome/test/data/*

fsamuel:
	chrome/browser/chrome_content_browser_client.cc
	chrome/chrome_tests.gypi
	chrome/renderer/chrome_content_renderer_client.cc
	chrome/renderer/extensions/dispatcher.cc
	chrome/renderer/resources/extensions/ad_view.js
	chrome/renderer/resources/extensions/ad_view_custom.js
	chrome/renderer/resources/extensions/ad_view_deny.js
	chrome/renderer/resources/extensions/platform_app.css
	chrome/renderer/resources/renderer_resources.grd

[asargent_no_longer_on_chrome, 2013-03-09 00:51:05]

I haven't finished looking at all the test/data files yet, but here are a few
small comments.

https://codereview.chromium.org/12463015/diff/16038/chrome/app/generated_reso...
File chrome/app/generated_resources.grd (right):

https://codereview.chromium.org/12463015/diff/16038/chrome/app/generated_reso...
chrome/app/generated_resources.grd:6323: +        Enables 'adview' element.
nit: I think you want "Enable" instead of "Enables" here - at least if I'm
remembering right that this appears as the title of the option in chrome://flags

https://codereview.chromium.org/12463015/diff/16038/chrome/app/generated_reso...
chrome/app/generated_resources.grd:6326: +        Enables 'adview' HTML element
in packaged apps v2 and above.
nit: you can just say "packaged apps" here. I think externally we're referring
to v1 apps as "legacy packaged apps"

https://codereview.chromium.org/12463015/diff/16038/chrome/app/generated_reso...
chrome/app/generated_resources.grd:6329: +        Enables specifying 'src'
attribute on 'adview' elements.
Enables -> Enable

https://codereview.chromium.org/12463015/diff/16038/chrome/browser/about_flag...
File chrome/browser/about_flags.cc (right):

https://codereview.chromium.org/12463015/diff/16038/chrome/browser/about_flag...
chrome/browser/about_flags.cc:478: "enable-adview",  // FLAGS:RECORD_UMA
Does having this FLAGS:RECORD_UMA comment somehow turn on UMA stats for whether
this flag is turned on? Would we really care or do anything with this data? If
not then probably not worth collecting

https://codereview.chromium.org/12463015/diff/16038/chrome/common/extensions/...
File chrome/common/extensions/permissions/api_permission.cc (right):

https://codereview.chromium.org/12463015/diff/16038/chrome/common/extensions/...
chrome/common/extensions/permissions/api_permission.cc:198: {
APIPermission::kAdView, "adview", kFlagCannotBeOptional },
Any reason not to let this be an optional permission? 

In case you aren't familiar with optional permissions, they are described here:

https://developer.chrome.com/extensions/permissions.html

[asargent_no_longer_on_chrome, 2013-03-11 19:10:13]

https://codereview.chromium.org/12463015/diff/16038/chrome/test/data/extensio...
File chrome/test/data/extensions/platform_apps/ad_view/ad_network/test_sdk.js
(right):

https://codereview.chromium.org/12463015/diff/16038/chrome/test/data/extensio...
chrome/test/data/extensions/platform_apps/ad_view/ad_network/test_sdk.js:20:
return location.href.substr(0, prefix_index) + "/" + ad_path;
FYI, we are usually careful about using randomness in tests because it can cause
flaky failures that might be hard to reproduce. In this case since you're just
picking from a static list of images it should be ok.

But just keep in mind that for using random "functional" behavior, we want to at
a minimum have a way to make tests with randomness repeatable (eg have them pick
a seed for the random number generator and print that out along with failure
information, and a way to pass that seed in as an input so you can diagnose
failures).

https://codereview.chromium.org/12463015/diff/16038/chrome/test/data/extensio...
chrome/test/data/extensions/platform_apps/ad_view/ad_network/test_sdk.js:51:
function sint(value) {
nits: can you use a more meaningful function name than "sint"? Maybe "sizeAsInt"
or something?

Also, right now you're returning a float - does "int" in the name not imply that
you wanted an integer result? 

I'm guessing you're returning a float here because you don't want to lose
precision with arithmetic? Would it be better to have a function that takes in a
list of property names, and have it look up their values and do the addition? 


E.g. 

function getCombinedSize(style, propertyNames) {
  float result;
  for (var i = 0; i < propertyNames.length; i++) {
    var value = style.getPropertyValue(propertyNames[i]);
    result += parseFloat(value.substr(0, value.indexOf("px")));
  }
  return Math.round(result);
}


(again, not sure if you wanted this to remain a float or round to integer)

https://codereview.chromium.org/12463015/diff/16038/chrome/test/data/extensio...
File
chrome/test/data/extensions/platform_apps/ad_view/ad_network_loaded/display_ad.js
(right):

https://codereview.chromium.org/12463015/diff/16038/chrome/test/data/extensio...
chrome/test/data/extensions/platform_apps/ad_view/ad_network_loaded/display_ad.js:1:
(function( window ) {
nit: Any reason you pass in window as a parameter here to this anonymous
function? If you don't, I'm pretty sure references to window will just resolve
properly. 

Additionally, I think (but am not positive) that window is a read-only property
in JS that always refers to the global object and can't be overridden.

https://codereview.chromium.org/12463015/diff/16038/chrome/test/data/extensio...
chrome/test/data/extensions/platform_apps/ad_view/ad_network_loaded/display_ad.js:16:
var timeout_sequence_id = 0;
style nit: we use variableNamesLikeThis for JS code (this variable and other
names here too)

http://google-styleguide.googlecode.com/svn/trunk/javascriptguide.xml#Naming

https://codereview.chromium.org/12463015/diff/16038/chrome/test/data/extensio...
File
chrome/test/data/extensions/platform_apps/ad_view/display_first_ad/chrome_test.js
(right):

https://codereview.chromium.org/12463015/diff/16038/chrome/test/data/extensio...
chrome/test/data/extensions/platform_apps/ad_view/display_first_ad/chrome_test.js:8:
var adview = document.getElementsByTagName("ADVIEW")[0];
WHY ARE  YOU SHOUTING?

I.e. does ADVIEW really need to be all-caps? Generally tag names are lower case.

[rpaquay, 2013-03-12 19:56:48]

Address asargent@ code review feedback.

fsamuel@, please review the following files:
	chrome/browser/chrome_content_browser_client.cc
	chrome/chrome_tests.gypi
	chrome/renderer/chrome_content_renderer_client.cc
	chrome/renderer/extensions/dispatcher.cc
	chrome/renderer/resources/extensions/ad_view.js
	chrome/renderer/resources/extensions/ad_view_custom.js
	chrome/renderer/resources/extensions/ad_view_deny.js
	chrome/renderer/resources/extensions/platform_app.css
	chrome/renderer/resources/renderer_resources.grd

https://codereview.chromium.org/12463015/diff/16038/chrome/app/generated_reso...
File chrome/app/generated_resources.grd (right):

https://codereview.chromium.org/12463015/diff/16038/chrome/app/generated_reso...
chrome/app/generated_resources.grd:6323: +        Enables 'adview' element.
On 2013/03/09 00:51:05, Antony Sargent wrote:
> nit: I think you want "Enable" instead of "Enables" here - at least if I'm
> remembering right that this appears as the title of the option in
chrome://flags

Done.

https://codereview.chromium.org/12463015/diff/16038/chrome/app/generated_reso...
chrome/app/generated_resources.grd:6326: +        Enables 'adview' HTML element
in packaged apps v2 and above.
On 2013/03/09 00:51:05, Antony Sargent wrote:
> nit: you can just say "packaged apps" here. I think externally we're referring
> to v1 apps as "legacy packaged apps"

Done.

https://codereview.chromium.org/12463015/diff/16038/chrome/app/generated_reso...
chrome/app/generated_resources.grd:6329: +        Enables specifying 'src'
attribute on 'adview' elements.
On 2013/03/09 00:51:05, Antony Sargent wrote:
> Enables -> Enable

Done.

https://codereview.chromium.org/12463015/diff/16038/chrome/browser/about_flag...
File chrome/browser/about_flags.cc (right):

https://codereview.chromium.org/12463015/diff/16038/chrome/browser/about_flag...
chrome/browser/about_flags.cc:478: "enable-adview",  // FLAGS:RECORD_UMA
On 2013/03/09 00:51:05, Antony Sargent wrote:
> Does having this FLAGS:RECORD_UMA comment somehow turn on UMA stats for
whether
> this flag is turned on? Would we really care or do anything with this data? If
> not then probably not worth collecting

Done.

https://codereview.chromium.org/12463015/diff/16038/chrome/common/extensions/...
File chrome/common/extensions/permissions/api_permission.cc (right):

https://codereview.chromium.org/12463015/diff/16038/chrome/common/extensions/...
chrome/common/extensions/permissions/api_permission.cc:198: {
APIPermission::kAdView, "adview", kFlagCannotBeOptional },
On 2013/03/09 00:51:05, Antony Sargent wrote:
> Any reason not to let this be an optional permission? 
> 
> In case you aren't familiar with optional permissions, they are described
here:
> 
> https://developer.chrome.com/extensions/permissions.html

<webview> is non optional too. Reading the link above, I would probably argue
that ads should be non-optional too: It makes the developer's life slightly more
difficult, and it is not clear to me it would be a good user experience to
display a "Confirm permission dialog" just before an ad is displayed. It sounds
a bit intrusive.

https://codereview.chromium.org/12463015/diff/16038/chrome/test/data/extensio...
File chrome/test/data/extensions/platform_apps/ad_view/ad_network/test_sdk.js
(right):

https://codereview.chromium.org/12463015/diff/16038/chrome/test/data/extensio...
chrome/test/data/extensions/platform_apps/ad_view/ad_network/test_sdk.js:20:
return location.href.substr(0, prefix_index) + "/" + ad_path;
On 2013/03/11 19:10:14, Antony Sargent wrote:
> FYI, we are usually careful about using randomness in tests because it can
cause
> flaky failures that might be hard to reproduce. In this case since you're just
> picking from a static list of images it should be ok.
> 
> But just keep in mind that for using random "functional" behavior, we want to
at
> a minimum have a way to make tests with randomness repeatable (eg have them
pick
> a seed for the random number generator and print that out along with failure
> information, and a way to pass that seed in as an input so you can diagnose
> failures). 

We actually don't need the randomness for tests, so I removed it.

https://codereview.chromium.org/12463015/diff/16038/chrome/test/data/extensio...
chrome/test/data/extensions/platform_apps/ad_view/ad_network/test_sdk.js:51:
function sint(value) {
On 2013/03/11 19:10:14, Antony Sargent wrote:
> nits: can you use a more meaningful function name than "sint"? Maybe
"sizeAsInt"
> or something?
> 
> Also, right now you're returning a float - does "int" in the name not imply
that
> you wanted an integer result? 
> 
> I'm guessing you're returning a float here because you don't want to lose
> precision with arithmetic? Would it be better to have a function that takes in
a
> list of property names, and have it look up their values and do the addition? 
> 
> 
> E.g. 
> 
> function getCombinedSize(style, propertyNames) {
>   float result;
>   for (var i = 0; i < propertyNames.length; i++) {
>     var value = style.getPropertyValue(propertyNames[i]);
>     result += parseFloat(value.substr(0, value.indexOf("px")));
>   }
>   return Math.round(result);
> }
> 
> 
> (again, not sure if you wanted this to remain a float or round to integer)

Done.

https://codereview.chromium.org/12463015/diff/16038/chrome/test/data/extensio...
File
chrome/test/data/extensions/platform_apps/ad_view/ad_network_loaded/display_ad.js
(right):

https://codereview.chromium.org/12463015/diff/16038/chrome/test/data/extensio...
chrome/test/data/extensions/platform_apps/ad_view/ad_network_loaded/display_ad.js:1:
(function( window ) {
On 2013/03/11 19:10:14, Antony Sargent wrote:
> nit: Any reason you pass in window as a parameter here to this anonymous
> function? If you don't, I'm pretty sure references to window will just resolve
> properly. 
> 
> Additionally, I think (but am not positive) that window is a read-only
property
> in JS that always refers to the global object and can't be overridden. 

Done.

https://codereview.chromium.org/12463015/diff/16038/chrome/test/data/extensio...
chrome/test/data/extensions/platform_apps/ad_view/ad_network_loaded/display_ad.js:16:
var timeout_sequence_id = 0;
On 2013/03/11 19:10:14, Antony Sargent wrote:
> style nit: we use variableNamesLikeThis for JS code (this variable and other
> names here too)
> 
> http://google-styleguide.googlecode.com/svn/trunk/javascriptguide.xml#Naming

Done.

https://codereview.chromium.org/12463015/diff/16038/chrome/test/data/extensio...
File
chrome/test/data/extensions/platform_apps/ad_view/display_first_ad/chrome_test.js
(right):

https://codereview.chromium.org/12463015/diff/16038/chrome/test/data/extensio...
chrome/test/data/extensions/platform_apps/ad_view/display_first_ad/chrome_test.js:8:
var adview = document.getElementsByTagName("ADVIEW")[0];
On 2013/03/11 19:10:14, Antony Sargent wrote:
> WHY ARE  YOU SHOUTING?
> 
> I.e. does ADVIEW really need to be all-caps? Generally tag names are lower
case.

I don't need to use all-caps here. I needed to use all-caps in another file
(ad_view.js, tagWatcher).

[asargent_no_longer_on_chrome, 2013-03-12 20:54:27]

LGTM

https://codereview.chromium.org/12463015/diff/16038/chrome/common/extensions/...
File chrome/common/extensions/permissions/api_permission.cc (right):

https://codereview.chromium.org/12463015/diff/16038/chrome/common/extensions/...
chrome/common/extensions/permissions/api_permission.cc:198: {
APIPermission::kAdView, "adview", kFlagCannotBeOptional },
Note that the *developer* gets to choose if the permission is required or
optional. Listing a permission under the 'permissions' key of their manifest
makes it required, and listing it in the 'optional_permissions' key makes it
optional. Also, the permission request is only a one-time thing if the user
agrees to it, then the system remembers that (I think it's stored in
ExtensionPrefs). 

Anyhow, I don't think this is a big deal either way, and  I wouldn't expect many
developers to choose to make ads optional.

https://codereview.chromium.org/12463015/diff/18003/chrome/renderer/resources...
File chrome/renderer/resources/extensions/ad_view.js (right):

https://codereview.chromium.org/12463015/diff/18003/chrome/renderer/resources...
chrome/renderer/resources/extensions/ad_view.js:64: //console.log('Setting
<adview> "src": ' + newValue);
nit: remove this debugging line

https://codereview.chromium.org/12463015/diff/18003/chrome/renderer/resources...
chrome/renderer/resources/extensions/ad_view.js:148:
AD_VIEW_API_METHODS.forEach(function(apiMethod) {
nit: see kalman's recent mail about Array.forEach - you should either use
utils.forEach, or switch to :

for (var i = 0; i < AD_VIEW_API_METHODS.length; i++) {
  var apiMethod = AD_VIEW_API_METHODS[i];
  ...
}

(here and elsewhere in the CL)

[rpaquay, 2013-03-12 23:05:36]

Addressing asargent@ feedback.

https://codereview.chromium.org/12463015/diff/16038/chrome/common/extensions/...
File chrome/common/extensions/permissions/api_permission.cc (right):

https://codereview.chromium.org/12463015/diff/16038/chrome/common/extensions/...
chrome/common/extensions/permissions/api_permission.cc:198: {
APIPermission::kAdView, "adview", kFlagCannotBeOptional },
On 2013/03/12 20:54:27, Antony Sargent wrote:
> Note that the *developer* gets to choose if the permission is required or
> optional. Listing a permission under the 'permissions' key of their manifest
> makes it required, and listing it in the 'optional_permissions' key makes it
> optional. Also, the permission request is only a one-time thing if the user
> agrees to it, then the system remembers that (I think it's stored in
> ExtensionPrefs). 
> 
> Anyhow, I don't think this is a big deal either way, and  I wouldn't expect
many
> developers to choose to make ads optional. 

Oh I see, I missed that part, sorry. I think you are right then, we should leave
the possibility of making the permission optional.

https://codereview.chromium.org/12463015/diff/18003/chrome/renderer/resources...
File chrome/renderer/resources/extensions/ad_view.js (right):

https://codereview.chromium.org/12463015/diff/18003/chrome/renderer/resources...
chrome/renderer/resources/extensions/ad_view.js:64: //console.log('Setting
<adview> "src": ' + newValue);
On 2013/03/12 20:54:27, Antony Sargent wrote:
> nit: remove this debugging line

Done.

https://codereview.chromium.org/12463015/diff/18003/chrome/renderer/resources...
chrome/renderer/resources/extensions/ad_view.js:148:
AD_VIEW_API_METHODS.forEach(function(apiMethod) {
On 2013/03/12 20:54:27, Antony Sargent wrote:
> nit: see kalman's recent mail about Array.forEach - you should either use
> utils.forEach, or switch to :
> 
> for (var i = 0; i < AD_VIEW_API_METHODS.length; i++) {
>   var apiMethod = AD_VIEW_API_METHODS[i];
>   ...
> }
> 
> (here and elsewhere in the CL)

Done.

[Fady Samuel, 2013-03-18 18:48:05]

Sorry for the delay. LGTM once my two comments are addressed.

https://codereview.chromium.org/12463015/diff/58001/chrome/renderer/chrome_co...
File chrome/renderer/chrome_content_renderer_client.cc (right):

https://codereview.chromium.org/12463015/diff/58001/chrome/renderer/chrome_co...
chrome/renderer/chrome_content_renderer_client.cc:1130:
WebKit::WebString::fromUTF8(kAdViewTagName))) {
Please merge with this https://codereview.chromium.org/12512009/

https://codereview.chromium.org/12463015/diff/58001/chrome/renderer/extension...
File chrome/renderer/extensions/dispatcher.cc (right):

https://codereview.chromium.org/12463015/diff/58001/chrome/renderer/extension...
chrome/renderer/extensions/dispatcher.cc:997: if (context_type ==
Feature::BLESSED_EXTENSION_CONTEXT) {
https://codereview.chromium.org/12496015/patch/3001/4001

Make sure you enable adView only in Platform apps and not extensions.

[commit-bot: I haz the power, 2013-03-19 18:13:40]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/rpaquay@chromium.org/12463015/69001

[commit-bot: I haz the power, 2013-03-19 18:13:57]

Presubmit check for 12463015-69001 failed and returned exit status 1.

INFO:root:Found 37 file(s).

Running presubmit commit checks ...
Running /b/commit-queue/workdir/chromium/PRESUBMIT.py
Running /b/commit-queue/workdir/chromium/chrome/PRESUBMIT.py
Running /b/commit-queue/workdir/chromium/chrome/browser/extensions/PRESUBMIT.py
Running /b/commit-queue/workdir/chromium/chrome/common/extensions/PRESUBMIT.py

** Presubmit ERRORS **
Missing LGTM from an OWNER for these files:
    chrome/renderer/chrome_content_renderer_client.cc
    chrome/renderer/resources/renderer_resources.grd

Presubmit checks took 5.2s to calculate.

Was the presubmit check useful? Please send feedback & hate mail to
maruel@chromium.org!

[rpaquay, 2013-03-19 18:21:21]

Darin, can you look at the following files for OWNERS approval?

chrome/renderer/chrome_content_renderer_client.cc
chrome/renderer/resources/renderer_resources.grd

[darin (slow to review), 2013-03-19 20:12:31]

"Initial checking enabling <adview> tag for packaged apps."

^^^ typo: checking -> check-in

[rpaquay, 2013-03-19 20:19:02]

On 2013/03/19 20:12:31, darin wrote:
> "Initial checking enabling <adview> tag for packaged apps."
> 
> ^^^ typo: checking -> check-in

Done.

[darin (slow to review), 2013-03-19 20:20:03]

https://codereview.chromium.org/12463015/diff/69001/chrome/renderer/chrome_co...
File chrome/renderer/chrome_content_renderer_client.cc (right):

https://codereview.chromium.org/12463015/diff/69001/chrome/renderer/chrome_co...
chrome/renderer/chrome_content_renderer_client.cc:405: if (extension &&
extension->HasAPIPermission(
nit: only need to null check |extension| once, right?

nit: even though it is more code, you might also consider a loop so that it is
easy to grow this in the future.  totally optional.

if (extension) {
  const extensions::APIPermission::ID perms[] = {
    extensions::APIPermission::kWebView,
    extensions::APIPermission::kAdView
  };
  for (size_t i = 0; i < arraysize(perms); ++i) {
    if (extension->HasAPIPermission(perms[i]))
      return false;
  }
}

https://codereview.chromium.org/12463015/diff/69001/chrome/renderer/chrome_co...
chrome/renderer/chrome_content_renderer_client.cc:1137:
WebString::fromUTF8(kWebViewTagName)) ||
nit: save tagName() to a local variable.

[rpaquay, 2013-03-19 20:45:29]

Addressing code review feedback from darin.

https://codereview.chromium.org/12463015/diff/69001/chrome/renderer/chrome_co...
File chrome/renderer/chrome_content_renderer_client.cc (right):

https://codereview.chromium.org/12463015/diff/69001/chrome/renderer/chrome_co...
chrome/renderer/chrome_content_renderer_client.cc:405: if (extension &&
extension->HasAPIPermission(
On 2013/03/19 20:20:03, darin wrote:
> nit: only need to null check |extension| once, right?
> 
> nit: even though it is more code, you might also consider a loop so that it is
> easy to grow this in the future.  totally optional.
> 
> if (extension) {
>   const extensions::APIPermission::ID perms[] = {
>     extensions::APIPermission::kWebView,
>     extensions::APIPermission::kAdView
>   };
>   for (size_t i = 0; i < arraysize(perms); ++i) {
>     if (extension->HasAPIPermission(perms[i]))
>       return false;
>   }
> }

Done.

https://codereview.chromium.org/12463015/diff/69001/chrome/renderer/chrome_co...
chrome/renderer/chrome_content_renderer_client.cc:1137:
WebString::fromUTF8(kWebViewTagName)) ||
On 2013/03/19 20:20:03, darin wrote:
> nit: save tagName() to a local variable.

Done.

[darin (slow to review), 2013-03-20 04:59:30]

LGTM

https://codereview.chromium.org/12463015/diff/56003/chrome/renderer/chrome_co...
File chrome/renderer/chrome_content_renderer_client.cc (right):

https://codereview.chromium.org/12463015/diff/56003/chrome/renderer/chrome_co...
chrome/renderer/chrome_content_renderer_client.cc:1141: return
tagName.equals(WebString::fromUTF8(kWebViewTagName)) ||
nit: use google C++ style variable naming conventions, so tagName -> tag_name

[rpaquay, 2013-03-20 16:24:45]

https://codereview.chromium.org/12463015/diff/56003/chrome/renderer/chrome_co...
File chrome/renderer/chrome_content_renderer_client.cc (right):

https://codereview.chromium.org/12463015/diff/56003/chrome/renderer/chrome_co...
chrome/renderer/chrome_content_renderer_client.cc:1141: return
tagName.equals(WebString::fromUTF8(kWebViewTagName)) ||
On 2013/03/20 04:59:30, darin wrote:
> nit: use google C++ style variable naming conventions, so tagName -> tag_name

Done.

[commit-bot: I haz the power, 2013-03-20 16:25:52]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/rpaquay@chromium.org/12463015/83001

[commit-bot: I haz the power, 2013-03-20 17:22:54]

Retried try job too often on linux_rel for step(s) unit_tests
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=linux_rel&...

[commit-bot: I haz the power, 2013-03-20 21:10:47]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/rpaquay@chromium.org/12463015/112001

[commit-bot: I haz the power, 2013-03-21 16:25:14]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/rpaquay@chromium.org/12463015/89003

[commit-bot: I haz the power, 2013-03-22 18:15:12]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/rpaquay@chromium.org/12463015/89003

[commit-bot: I haz the power, 2013-03-23 14:51:29]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/rpaquay@chromium.org/12463015/89003

[commit-bot: I haz the power, 2013-03-23 17:37:56]

Change committed as 190043