Add --enable-speech-input switch and set ENABLE_INPUT_SPEECH=1

Add --enable-speech-input switch and set ENABLE_INPUT_SPEECH=1

Flip the compile-time enable switch for speech input to on and add a
command-line switch to put the functionality behind. The command-line
flag will be used both on the Chromium side and in WebKit via
WebRuntimeFeatures. 

This CL depends on the webkit patch https://bugs.webkit.org/show_bug.cgi?id=43146

BUG=none
TEST=browser_tests --gtest_filter=SpeechInputEnableSwitchTest.*

Originally Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=54235, but reverted due to tests failure.

Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=54389

[jorlow, 2010-07-29 10:58:30]

LGTM

http://codereview.chromium.org/3064017/diff/1/2
File build/features_override.gypi (right):

http://codereview.chromium.org/3064017/diff/1/2#newcode28
build/features_override.gypi:28: 'ENABLE_INPUT_SPEECH=1',
You'll need to add the flag to the third_party/WebKit feature defines as well.

[jorlow, 2010-07-29 10:59:18]

Joth, did you do the WebKit side for Geolocation?

Hans, maybe take a quick peak as well.

[Satish, 2010-07-29 11:04:55]

http://codereview.chromium.org/3064017/diff/1/2
File build/features_override.gypi (right):

http://codereview.chromium.org/3064017/diff/1/2#newcode28
build/features_override.gypi:28: 'ENABLE_INPUT_SPEECH=1',
On 2010/07/29 10:58:30, Jeremy Orlow wrote:
> You'll need to add the flag to the third_party/WebKit feature defines as well.

I have it defined in WebKit/WebKit/chromium/features.gypi
(http://trac.webkit.org/browser/trunk/WebKit/chromium/features.gypi) a while
ago. Is this enough or were you indicating something else?

[jorlow, 2010-07-29 11:08:36]

that's fine

On Thu, Jul 29, 2010 at 12:04 PM, <satish@chromium.org> wrote:

>
> http://codereview.chromium.org/3064017/diff/1/2
> File build/features_override.gypi (right):
>
> http://codereview.chromium.org/3064017/diff/1/2#newcode28
> build/features_override.gypi:28: 'ENABLE_INPUT_SPEECH=1',
> On 2010/07/29 10:58:30, Jeremy Orlow wrote:
>
>> You'll need to add the flag to the third_party/WebKit feature defines
>>
> as well.
>
> I have it defined in WebKit/WebKit/chromium/features.gypi
> (http://trac.webkit.org/browser/trunk/WebKit/chromium/features.gypi) a
> while ago. Is this enough or were you indicating something else?
>
>
> http://codereview.chromium.org/3064017/show
>

[hans, 2010-07-29 11:08:53]

On 2010/07/29 10:59:18, Jeremy Orlow wrote:
> Joth, did you do the WebKit side for Geolocation?
> 
> Hans, maybe take a quick peak as well.

Have you checked if any layout tests change their output depending on the
compile-time switch? In that case, suppressions for those should probably go in
with this patch to avoid tree redness. (This was the case for me.)

Looks good otherwise.

[joth, 2010-07-29 11:24:54]

LGTM

On 29 July 2010 11:59, <jorlow@chromium.org> wrote:

> Joth, did you do the WebKit side for Geolocation?
>
> Hans, maybe take a quick peak as well.
>
> http://codereview.chromium.org/3064017/show
>