Add support to enable EmbeddedSearch API in Android Chrome search results page.

Add support to enable EmbeddedSearch API in Android Chrome search results page.

This CL,
- Adds a new EmbeddedSearchPageVerion variant to enable EmbeddedSearch API in Android Chrome search results page.
- Adds an experiment flag to enable "EmbeddedSearch API" for Android Chrome.
- Creates SearchTabHelper for TabAndroid WebContents.
- Uses EmbeddedSearch API to submit search queries if the underlying search results page supports Instant.
- Assigns Instant search URL renderers to a priviledged Instant process.
- Uses "chrome-search://" scheme for Instant search URLs.
- Creates new test to verify the embedded search page version.

To test: Enable EmbeddedSearch API flag from about://flags page and run Android Chrome against demo server URL (cl/64511863).
Server side bug: b/10402468

BUG=none
TEST=none

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=267069

[kmadhusu, 2014-04-23 23:25:56]

Please review.

Known Issue:
When we navigate from web search mode to "Maps" mode, the location bar contents
are updated to display "/maps" url instead of retaining the search query. This
is because of the default policy changes that's happening when we change from an
Instant process to a non-Instant process. I will submit a follow up CL to
address this issue.

Thanks.

[kmadhusu, 2014-04-23 23:29:16]

+tedchoc@

PTAL. Thanks.

[Ted C, 2014-04-24 16:24:13]

lgtm - Awesome!

Glad to see this wasn't a gigantic change.  And yay for sharing more code.

https://codereview.chromium.org/215733002/diff/40001/chrome/browser/search/se...
File chrome/browser/search/search.cc (right):

https://codereview.chromium.org/215733002/diff/40001/chrome/browser/search/se...
chrome/browser/search/search.cc:372: switches::kEnableEmbeddedSearchAPI)) {
I believe this should be indented 4 from the if not from CommandLine

https://codereview.chromium.org/215733002/diff/40001/chrome/browser/search/se...
File chrome/browser/search/search_android_unittest.cc (right):

https://codereview.chromium.org/215733002/diff/40001/chrome/browser/search/se...
chrome/browser/search/search_android_unittest.cc:19: EXPECT_EQ(2ul,
EmbeddedSearchPageVersion());
Maybe add checks for IsInstantExtendedAPIEnabled being false above and true now
as well?

[kmadhusu, 2014-04-24 17:38:48]

Addressed comments.

Thanks.

https://codereview.chromium.org/215733002/diff/40001/chrome/browser/search/se...
File chrome/browser/search/search.cc (right):

https://codereview.chromium.org/215733002/diff/40001/chrome/browser/search/se...
chrome/browser/search/search.cc:372: switches::kEnableEmbeddedSearchAPI)) {
On 2014/04/24 16:24:13, Ted C wrote:
> I believe this should be indented 4 from the if not from CommandLine

Done.

https://codereview.chromium.org/215733002/diff/40001/chrome/browser/search/se...
File chrome/browser/search/search_android_unittest.cc (right):

https://codereview.chromium.org/215733002/diff/40001/chrome/browser/search/se...
chrome/browser/search/search_android_unittest.cc:19: EXPECT_EQ(2ul,
EmbeddedSearchPageVersion());
On 2014/04/24 16:24:13, Ted C wrote:
> Maybe add checks for IsInstantExtendedAPIEnabled being false above and true
now
> as well?

Done.

[samarth, 2014-04-25 06:50:20]

lgtm

https://codereview.chromium.org/215733002/diff/120001/chrome/browser/android/...
File chrome/browser/android/tab_android.cc (right):

https://codereview.chromium.org/215733002/diff/120001/chrome/browser/android/...
chrome/browser/android/tab_android.cc:473: return DEFAULT_PAGE_LOAD;
Is this the right return type here?

https://codereview.chromium.org/215733002/diff/120001/chrome/browser/search/s...
File chrome/browser/search/search.cc (right):

https://codereview.chromium.org/215733002/diff/120001/chrome/browser/search/s...
chrome/browser/search/search.cc:61: const uint64 kEmbeddedPageVersionVariant =
2;
kEmbeddedSearchEnabledVersion or something

https://codereview.chromium.org/215733002/diff/120001/chrome/browser/search/s...
chrome/browser/search/search.cc:267: bool check_renderer_process = false;
Instead of adding this boolean, just factor out the check into a helper. Then,
you can just call it when you need to.

[kmadhusu, 2014-04-25 17:23:18]

Addressed comments. 

Thanks.

https://codereview.chromium.org/215733002/diff/120001/chrome/browser/android/...
File chrome/browser/android/tab_android.cc (right):

https://codereview.chromium.org/215733002/diff/120001/chrome/browser/android/...
chrome/browser/android/tab_android.cc:473: return DEFAULT_PAGE_LOAD;
On 2014/04/25 06:50:20, samarth wrote:
> Is this the right return type here?

Ted@, Is this okay? Do you want me to create a new status?

https://codereview.chromium.org/215733002/diff/120001/chrome/browser/search/s...
File chrome/browser/search/search.cc (right):

https://codereview.chromium.org/215733002/diff/120001/chrome/browser/search/s...
chrome/browser/search/search.cc:61: const uint64 kEmbeddedPageVersionVariant =
2;
On 2014/04/25 06:50:20, samarth wrote:
> kEmbeddedSearchEnabledVersion or something

Done. s/kEmbeddedPageVersionVariant/kEmbeddedSearchEnabledVersion/g

https://codereview.chromium.org/215733002/diff/120001/chrome/browser/search/s...
chrome/browser/search/search.cc:267: bool check_renderer_process = false;
On 2014/04/25 06:50:20, samarth wrote:
> Instead of adding this boolean, just factor out the check into a helper. Then,
> you can just call it when you need to.

Done. Called IsInstantExtendedAPIEnabled() here.

[kmadhusu, 2014-04-25 17:25:47]

avi@: Please review tab_helpers.cc.
sky@: Please review chrome/renderer/resources/renderer_resources.grd.

Thanks.

[sky, 2014-04-25 20:17:36]

LGTM

[Avi (use Gerrit), 2014-04-29 16:50:39]

On 2014/04/25 20:17:36, sky wrote:
> LGTM

tab_helpers.cc LGTM

[Ted C, 2014-04-29 16:53:13]

https://codereview.chromium.org/215733002/diff/120001/chrome/browser/android/...
File chrome/browser/android/tab_android.cc (right):

https://codereview.chromium.org/215733002/diff/120001/chrome/browser/android/...
chrome/browser/android/tab_android.cc:473: return DEFAULT_PAGE_LOAD;
On 2014/04/25 17:23:19, kmadhusu wrote:
> On 2014/04/25 06:50:20, samarth wrote:
> > Is this the right return type here?
> 
> Ted@, Is this okay? Do you want me to create a new status?

This should be fine.  This isn't a prerendered page, so I think this is the
"most" correct.  We might want to add a new type at some point if we need to
differentiate, but until we do this is ok.

The one thing this made me think of is whether the loading progress bar is shown
when going through this path?  We used these load value types to simulate load
progress in the past, so if this isn't happening for these pages then we might
want to add a new type to handle that.

[kmadhusu, 2014-04-29 17:01:30]

https://codereview.chromium.org/215733002/diff/120001/chrome/browser/android/...
File chrome/browser/android/tab_android.cc (right):

https://codereview.chromium.org/215733002/diff/120001/chrome/browser/android/...
chrome/browser/android/tab_android.cc:473: return DEFAULT_PAGE_LOAD;
On 2014/04/29 16:53:13, Ted C wrote:
> On 2014/04/25 17:23:19, kmadhusu wrote:
> > On 2014/04/25 06:50:20, samarth wrote:
> > > Is this the right return type here?
> > 
> > Ted@, Is this okay? Do you want me to create a new status?
> 
> This should be fine.  This isn't a prerendered page, so I think this is the
> "most" correct.  We might want to add a new type at some point if we need to
> differentiate, but until we do this is ok.
> 
> The one thing this made me think of is whether the loading progress bar is
shown
> when going through this path?  We used these load value types to simulate load
> progress in the past, so if this isn't happening for these pages then we might
> want to add a new type to handle that.

Loading progress bar is shown while going through this path. Looks like it works
as expected.

[kmadhusu, 2014-04-29 17:17:27]

The CQ bit was checked by kmadhusu@chromium.org

[commit-bot: I haz the power, 2014-04-29 17:17:48]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/kmadhusu@chromium.org/215733002/180001

[commit-bot: I haz the power, 2014-04-30 04:52:57]

Change committed as 267069