SpeechInputExtensionManager now interface with SpeechRecognitionManagerDelegate

SpeechInputExtensionManager now interface (exclusively) with SpeechRecognitionManagerDelegate (Speech CL1.11).

- The tray icon and balloon handing has been moved from speech_input_extensions_manager to chrome_speech_recognition_manager_delegate, since that code (tray icon) will be used also for continuous recognition.
- Removed the SpeechRecognizer interface from /content/public (thus de-virtualized and refcounted the SpeechRecognizerImpl)


BUG=116954
TEST=none

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=138498

[Primiano Tucci (use gerrit), 2012-05-10 10:55:08]

Low priority on this CL, since depends on CL1.8 and CL1.10, pending for owner
approval. Anyway, this CL should not block CL2.x serie.

+CC leandrogarcia@ for speech input extensions mod.

Note for myself: a little conflict will arise on /content/public interfaces with
CL2.0. Depending on which lands first, rebase the other.

[hans, 2012-05-15 13:35:17]

http://codereview.chromium.org/10377082/diff/5001/chrome/browser/speech/chrom...
File chrome/browser/speech/chrome_speech_recognition_manager_delegate.cc
(right):

http://codereview.chromium.org/10377082/diff/5001/chrome/browser/speech/chrom...
chrome/browser/speech/chrome_speech_recognition_manager_delegate.cc:50: bool
IsRequiredTrayIcon(int session_id) {
these names (this and IsRequiredBubble) aren't great. Maybe "RequiresTrayIcon"
or "ShouldUseTrayIcon" or something like that?

http://codereview.chromium.org/10377082/diff/5001/chrome/browser/speech/chrom...
chrome/browser/speech/chrome_speech_recognition_manager_delegate.cc:306: //  
been closed by InfoBubbleFocusChanged(), which triggered an AbortSession.
so could we add a DCHECK that the bubble is closed?

http://codereview.chromium.org/10377082/diff/5001/chrome/browser/speech/chrom...
File chrome/browser/speech/chrome_speech_recognition_manager_delegate.h (right):

http://codereview.chromium.org/10377082/diff/5001/chrome/browser/speech/chrom...
chrome/browser/speech/chrome_speech_recognition_manager_delegate.h:74: // Lazy
initializers for bubble and tray icon controller;
s/;/./

http://codereview.chromium.org/10377082/diff/5001/chrome/browser/speech/speec...
File chrome/browser/speech/speech_input_extension_manager.cc (right):

http://codereview.chromium.org/10377082/diff/5001/chrome/browser/speech/speec...
chrome/browser/speech/speech_input_extension_manager.cc:330:
prefs::kSpeechInputTrayNotificationShown, true);
this looks like a simplification of the old code..
but I can't figure out what the old code actually did?

it first does GetBoolean(prefs::kSpeechInputTrayNotificationShown), and then  if
that's true, it later calls SetBoolean(prefs::kSpeechInputTrayNotificationShown,
true) ???

http://codereview.chromium.org/10377082/diff/5001/chrome/browser/speech/speec...
File chrome/browser/speech/speech_input_extension_manager.h (right):

http://codereview.chromium.org/10377082/diff/5001/chrome/browser/speech/speec...
chrome/browser/speech/speech_input_extension_manager.h:11: #include
"base/string16.h"
sort order

http://codereview.chromium.org/10377082/diff/5001/content/browser/speech/spee...
File content/browser/speech/speech_recognizer_impl.h (right):

http://codereview.chromium.org/10377082/diff/5001/content/browser/speech/spee...
content/browser/speech/speech_recognizer_impl.h:29: // TODO(primiano) Next CL:
Remove the Impl suffix.
good, I was just about to suggest that :)

http://codereview.chromium.org/10377082/diff/5001/content/content_browser.gypi
File content/content_browser.gypi (right):

http://codereview.chromium.org/10377082/diff/5001/content/content_browser.gyp...
content/content_browser.gypi:143: 'public/browser/speech_recognizer.h',
this should be removed now?

[Primiano Tucci (use gerrit), 2012-05-16 10:16:00]

https://chromiumcodereview.appspot.com/10377082/diff/5001/chrome/browser/spee...
File chrome/browser/speech/chrome_speech_recognition_manager_delegate.cc
(right):

https://chromiumcodereview.appspot.com/10377082/diff/5001/chrome/browser/spee...
chrome/browser/speech/chrome_speech_recognition_manager_delegate.cc:50: bool
IsRequiredTrayIcon(int session_id) {
On 2012/05/15 13:35:17, hans wrote:
> these names (this and IsRequiredBubble) aren't great. Maybe "RequiresTrayIcon"
> or "ShouldUseTrayIcon" or something like that?

Done.

https://chromiumcodereview.appspot.com/10377082/diff/5001/chrome/browser/spee...
chrome/browser/speech/chrome_speech_recognition_manager_delegate.cc:306: //  
been closed by InfoBubbleFocusChanged(), which triggered an AbortSession.
On 2012/05/15 13:35:17, hans wrote:
> so could we add a DCHECK that the bubble is closed?
I am afraid not. See point 2 in the code comment.

https://chromiumcodereview.appspot.com/10377082/diff/5001/chrome/browser/spee...
File chrome/browser/speech/chrome_speech_recognition_manager_delegate.h (right):

https://chromiumcodereview.appspot.com/10377082/diff/5001/chrome/browser/spee...
chrome/browser/speech/chrome_speech_recognition_manager_delegate.h:74: // Lazy
initializers for bubble and tray icon controller;
On 2012/05/15 13:35:17, hans wrote:
> s/;/./

Done.

https://chromiumcodereview.appspot.com/10377082/diff/5001/chrome/browser/spee...
File chrome/browser/speech/speech_input_extension_manager.cc (right):

https://chromiumcodereview.appspot.com/10377082/diff/5001/chrome/browser/spee...
chrome/browser/speech/speech_input_extension_manager.cc:330:
prefs::kSpeechInputTrayNotificationShown, true);
On 2012/05/15 13:35:17, hans wrote:
> this looks like a simplification of the old code..
> but I can't figure out what the old code actually did?
> it first does GetBoolean(prefs::kSpeechInputTrayNotificationShown), and then 
if
> that's true, it later calls
SetBoolean(prefs::kSpeechInputTrayNotificationShown,
> true) ???

The previous code was doing 3 things:
 (a) checking if a security balloon is needed (only the first time ever for an
extension)
 (b) extracting the extension name for the security balloon
 (c) setting the flag, which avoids showing the balloon in future for the
profile.

(a) and (b) have been moved to SpeechInputExtensionManager::Start. With this CL
the extension name and the need of the security balloon are put into the
SRSessionContext upon Start.

The only thing left here is (c).

https://chromiumcodereview.appspot.com/10377082/diff/5001/chrome/browser/spee...
File chrome/browser/speech/speech_input_extension_manager.h (right):

https://chromiumcodereview.appspot.com/10377082/diff/5001/chrome/browser/spee...
chrome/browser/speech/speech_input_extension_manager.h:11: #include
"base/string16.h"
On 2012/05/15 13:35:17, hans wrote:
> sort order

Done.

https://chromiumcodereview.appspot.com/10377082/diff/5001/content/browser/spe...
File content/browser/speech/speech_recognizer_impl.h (right):

https://chromiumcodereview.appspot.com/10377082/diff/5001/content/browser/spe...
content/browser/speech/speech_recognizer_impl.h:29: // TODO(primiano) Next CL:
Remove the Impl suffix.
On 2012/05/15 13:35:17, hans wrote:
> good, I was just about to suggest that :)
Yep, just trying to avoid too much confusion in this CL.

https://chromiumcodereview.appspot.com/10377082/diff/5001/content/content_bro...
File content/content_browser.gypi (right):

https://chromiumcodereview.appspot.com/10377082/diff/5001/content/content_bro...
content/content_browser.gypi:143: 'public/browser/speech_recognizer.h',
On 2012/05/15 13:35:17, hans wrote:
> this should be removed now?

Ah, right I forgot to remove it (but I didn't receive any error during build,
despite the fact that the file does not exist anymore !)

[hans, 2012-05-17 10:32:48]

On 2012/05/16 10:16:00, Primiano Tucci wrote:
>
https://chromiumcodereview.appspot.com/10377082/diff/5001/chrome/browser/spee...
> chrome/browser/speech/speech_input_extension_manager.cc:330:
> prefs::kSpeechInputTrayNotificationShown, true);
> On 2012/05/15 13:35:17, hans wrote:
> > this looks like a simplification of the old code..
> > but I can't figure out what the old code actually did?
> > it first does GetBoolean(prefs::kSpeechInputTrayNotificationShown), and then

> if
> > that's true, it later calls
> SetBoolean(prefs::kSpeechInputTrayNotificationShown,
> > true) ???
> 
> The previous code was doing 3 things:
>  (a) checking if a security balloon is needed (only the first time ever for an
> extension)
>  (b) extracting the extension name for the security balloon
>  (c) setting the flag, which avoids showing the balloon in future for the
> profile.
> 
> (a) and (b) have been moved to SpeechInputExtensionManager::Start. With this
CL
> the extension name and the need of the security balloon are put into the
> SRSessionContext upon Start.
> 
> The only thing left here is (c).

Ah, I misread the code a little. Thanks for clarifying.


>
https://chromiumcodereview.appspot.com/10377082/diff/5001/content/content_bro...
> content/content_browser.gypi:143: 'public/browser/speech_recognizer.h',
> On 2012/05/15 13:35:17, hans wrote:
> > this should be removed now?
> 
> Ah, right I forgot to remove it (but I didn't receive any error during build,
> despite the fact that the file does not exist anymore !)

Yeah, I think we include the .h files here just for dependency tracking.


lgtm

[Primiano Tucci (use gerrit), 2012-05-17 14:27:22]

+jam@ for OWNERS approval on content/content_browser.gypi and content/public/

[jam, 2012-05-17 16:45:16]

http://codereview.chromium.org/10377082/diff/26/content/public/browser/speech...
File content/public/browser/speech_recognition_session_context.cc (right):

http://codereview.chromium.org/10377082/diff/26/content/public/browser/speech...
content/public/browser/speech_recognition_session_context.cc:9:
SpeechRecognitionSessionContext::SpeechRecognitionSessionContext()
just to be clear, this file is now needed because of clang errors?

http://codereview.chromium.org/10377082/diff/26/content/public/browser/speech...
File content/public/browser/speech_recognition_session_context.h (right):

http://codereview.chromium.org/10377082/diff/26/content/public/browser/speech...
content/public/browser/speech_recognition_session_context.h:43: bool
show_security_balloon;
this is a chrome concept and doesn't belong in content, i.e. different embedders
of content may treat security UI differently

[Satish, 2012-05-20 21:40:25]

http://codereview.chromium.org/10377082/diff/26/chrome/browser/speech/chrome_...
File chrome/browser/speech/chrome_speech_recognition_manager_delegate.cc
(right):

http://codereview.chromium.org/10377082/diff/26/chrome/browser/speech/chrome_...
chrome/browser/speech/chrome_speech_recognition_manager_delegate.cc:162: //
Note, the session might have been destroyed (see InfoBubbleButtonClicked).
Yes the session might have been destroyed, so what should or should not be done?
I expect the comment to mention that (like you did in the comment in the
previous method)

http://codereview.chromium.org/10377082/diff/26/chrome/browser/speech/chrome_...
chrome/browser/speech/chrome_speech_recognition_manager_delegate.cc:297: if
(RequiresTrayIcon(session_id))
are there race conditions where one session with tray icon is going on now
(either in recording or waiting-for-result state), then another session is
requested so that takes over the tray icon and when the first one ends in the
background that hides the tray icon of the active second session?

http://codereview.chromium.org/10377082/diff/26/chrome/browser/speech/chrome_...
chrome/browser/speech/chrome_speech_recognition_manager_delegate.cc:299: // No
need to remove the bubble here, since:
it isn't clear that the following are actually either-or. Could clarify

http://codereview.chromium.org/10377082/diff/26/chrome/browser/speech/chrome_...
File chrome/browser/speech/chrome_speech_recognition_manager_delegate.h (right):

http://codereview.chromium.org/10377082/diff/26/chrome/browser/speech/chrome_...
chrome/browser/speech/chrome_speech_recognition_manager_delegate.h:75:
SpeechRecognitionBubbleController* BubbleController();
I'd suggest naming these as
  GetBubbleController()
because they are not simple getters which are of the form
  bubble_controller()

http://codereview.chromium.org/10377082/diff/26/content/browser/speech/speech...
File content/browser/speech/speech_recognition_manager_impl.cc (right):

http://codereview.chromium.org/10377082/diff/26/content/browser/speech/speech...
content/browser/speech/speech_recognition_manager_impl.cc:332: // an
EndRecognition event) between a request (e.g. Abort) and its dispatch.
In any of these cases, does the caller need to be informed by a callback that
their request has completed (even though the session is already gone) ?

http://codereview.chromium.org/10377082/diff/26/content/browser/speech/speech...
File content/browser/speech/speech_recognizer_impl.h (right):

http://codereview.chromium.org/10377082/diff/26/content/browser/speech/speech...
content/browser/speech/speech_recognizer_impl.h:93: virtual
~SpeechRecognizerImpl();
do you need this to be virtual anymore?

http://codereview.chromium.org/10377082/diff/26/content/public/browser/speech...
File content/public/browser/speech_recognition_session_context.h (right):

http://codereview.chromium.org/10377082/diff/26/content/public/browser/speech...
content/public/browser/speech_recognition_session_context.h:35: bool
use_bubble_on_element;
similar to jam's comment below, the concept of bubble should really be confined
to the chrome/ layer. I'm guessing in this case webkit is setting this flag to
true for x-webkit-speech api and false for JS API - if this is correct the
variable should probably be named accordingly (e.g. invoked_by_page_element) and
the chrome/ layer decides that it maps to whether it should show a bubble
attached to the element or not

[Primiano Tucci (use gerrit), 2012-05-21 16:03:13]

http://codereview.chromium.org/10377082/diff/26/chrome/browser/speech/chrome_...
File chrome/browser/speech/chrome_speech_recognition_manager_delegate.cc
(right):

http://codereview.chromium.org/10377082/diff/26/chrome/browser/speech/chrome_...
chrome/browser/speech/chrome_speech_recognition_manager_delegate.cc:162: //
Note, the session might have been destroyed (see InfoBubbleButtonClicked).
On 2012/05/20 21:40:26, Satish wrote:
> Yes the session might have been destroyed, so what should or should not be
done?
> I expect the comment to mention that (like you did in the comment in the
> previous method)

Done.

http://codereview.chromium.org/10377082/diff/26/chrome/browser/speech/chrome_...
chrome/browser/speech/chrome_speech_recognition_manager_delegate.cc:297: if
(RequiresTrayIcon(session_id))
On 2012/05/20 21:40:26, Satish wrote:
> are there race conditions where one session with tray icon is going on now
> (either in recording or waiting-for-result state), then another session is
> requested so that takes over the tray icon and when the first one ends in the
> background that hides the tray icon of the active second session?
Yes, it was wrong. Now the tray icon is shown only during audio capture (which
cannot be interleaved among different sessions).

http://codereview.chromium.org/10377082/diff/26/chrome/browser/speech/chrome_...
chrome/browser/speech/chrome_speech_recognition_manager_delegate.cc:299: // No
need to remove the bubble here, since:
On 2012/05/20 21:40:26, Satish wrote:
> it isn't clear that the following are actually either-or. Could clarify

Done.

http://codereview.chromium.org/10377082/diff/26/chrome/browser/speech/chrome_...
File chrome/browser/speech/chrome_speech_recognition_manager_delegate.h (right):

http://codereview.chromium.org/10377082/diff/26/chrome/browser/speech/chrome_...
chrome/browser/speech/chrome_speech_recognition_manager_delegate.h:75:
SpeechRecognitionBubbleController* BubbleController();
On 2012/05/20 21:40:26, Satish wrote:
> I'd suggest naming these as
>   GetBubbleController()
> because they are not simple getters which are of the form
>   bubble_controller()

Done.

http://codereview.chromium.org/10377082/diff/26/content/browser/speech/speech...
File content/browser/speech/speech_recognition_manager_impl.cc (right):

http://codereview.chromium.org/10377082/diff/26/content/browser/speech/speech...
content/browser/speech/speech_recognition_manager_impl.cc:332: // an
EndRecognition event) between a request (e.g. Abort) and its dispatch.
On 2012/05/20 21:40:26, Satish wrote:
> In any of these cases, does the caller need to be informed by a callback that
> their request has completed (even though the session is already gone) ?

As a general reply, every caller (delegate) receives always a notification,
through, the OnRecognitionEnd event, when a session terminates (thus is
deleted).
This comment, however, refers to the cases of double Aborts, e.g., one internal
(issued by the caller) and one external. Some cases in which this can happen
are, for instance:

- Browser shutdown: the AbortAllSessionsForListener method is called to abort
the pending sessions, however, the session narually terminates in the meanwhile
(so at the time in which this posted function is executed the session has
already gone)

- A user clicks on a x-webkit-speech input while a session is going on for an
extension (so the extension's session is aborted) but the session ends in the
meanwhile.

http://codereview.chromium.org/10377082/diff/26/content/browser/speech/speech...
File content/browser/speech/speech_recognizer_impl.h (right):

http://codereview.chromium.org/10377082/diff/26/content/browser/speech/speech...
content/browser/speech/speech_recognizer_impl.h:93: virtual
~SpeechRecognizerImpl();
On 2012/05/20 21:40:26, Satish wrote:
> do you need this to be virtual anymore?
Yes, due to the base classes dtor (even if they are empty). Clang, otherwise,
would complain about: "[chromium-style] Overriding method must have "virtual"
keyword."

http://codereview.chromium.org/10377082/diff/26/content/public/browser/speech...
File content/public/browser/speech_recognition_session_context.cc (right):

http://codereview.chromium.org/10377082/diff/26/content/public/browser/speech...
content/public/browser/speech_recognition_session_context.cc:9:
SpeechRecognitionSessionContext::SpeechRecognitionSessionContext()
On 2012/05/17 16:45:16, John Abd-El-Malek wrote:
> just to be clear, this file is now needed because of clang errors?
Yes.

http://codereview.chromium.org/10377082/diff/26/content/public/browser/speech...
File content/public/browser/speech_recognition_session_context.h (right):

http://codereview.chromium.org/10377082/diff/26/content/public/browser/speech...
content/public/browser/speech_recognition_session_context.h:35: bool
use_bubble_on_element;
On 2012/05/20 21:40:26, Satish wrote:
> similar to jam's comment below, the concept of bubble should really be
confined
> to the chrome/ layer. I'm guessing in this case webkit is setting this flag to
> true for x-webkit-speech api and false for JS API - if this is correct the
> variable should probably be named accordingly (e.g. invoked_by_page_element)
and
> the chrome/ layer decides that it maps to whether it should show a bubble
> attached to the element or not
Right, renamed to requested_by_page_element

http://codereview.chromium.org/10377082/diff/26/content/public/browser/speech...
content/public/browser/speech_recognition_session_context.h:43: bool
show_security_balloon;
On 2012/05/17 16:45:16, John Abd-El-Malek wrote:
> this is a chrome concept and doesn't belong in content, i.e. different
embedders
> of content may treat security UI differently
Renamed and added a TODO, since it will be removed very soon (but in the
meanwhile I need it in this CL to keep the behavior of speech input extensions
consistent).

[Satish, 2012-05-22 09:15:51]

lgtm

[jam, 2012-05-22 15:29:48]

content/public lgtm

[commit-bot: I haz the power, 2012-05-22 15:41:11]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/primiano@chromium.org/10377082/17001

[commit-bot: I haz the power, 2012-05-22 16:10:00]

Try job failure for 10377082-17001 (retry) on linux_chromeos for step "compile"
(clobber build).
It's a second try, previously, step "compile" failed.
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=linux_chro...

[commit-bot: I haz the power, 2012-05-22 16:51:33]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/primiano@chromium.org/10377082/17001

[commit-bot: I haz the power, 2012-05-22 18:33:16]

Try job failure for 10377082-17001 (retry) on win for step "compile" (clobber
build).
It's a second try, previously, step "compile" failed.
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=win&number...

[commit-bot: I haz the power, 2012-05-22 20:04:33]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/primiano@chromium.org/10377082/17001

[commit-bot: I haz the power, 2012-05-23 04:03:22]

Try job failure for 10377082-17001 (retry) (retry) (retry) on win_rel for step
"sync_unit_tests".
It's a second try, previously, step "sync_unit_tests" failed.
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=win_rel&nu...

[commit-bot: I haz the power, 2012-05-23 07:21:23]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/primiano@chromium.org/10377082/17001

[commit-bot: I haz the power, 2012-05-23 07:21:30]

Failed to apply patch for
chrome/browser/speech/chrome_speech_recognition_manager_delegate.cc:
While running patch -p1 --forward --force;
patching file
chrome/browser/speech/chrome_speech_recognition_manager_delegate.cc
Hunk #1 succeeded at 13 with fuzz 2.
Hunk #2 FAILED at 36.
Hunk #3 succeeded at 109 (offset 1 line).
Hunk #4 succeeded at 127 (offset 1 line).
Hunk #5 succeeded at 139 (offset 1 line).
Hunk #6 succeeded at 150 (offset 1 line).
Hunk #7 succeeded at 175 (offset 1 line).
Hunk #8 succeeded at 220 (offset 1 line).
Hunk #9 succeeded at 243 (offset 1 line).
Hunk #10 succeeded at 276 (offset 1 line).
Hunk #11 succeeded at 324 (offset 1 line).
Hunk #12 succeeded at 377 (offset 3 lines).
1 out of 12 hunks FAILED -- saving rejects to file
chrome/browser/speech/chrome_speech_recognition_manager_delegate.cc.rej

[commit-bot: I haz the power, 2012-05-23 08:04:29]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/primiano@chromium.org/10377082/17001

[commit-bot: I haz the power, 2012-05-23 08:04:43]

Failed to apply patch for
chrome/browser/speech/chrome_speech_recognition_manager_delegate.cc:
While running patch -p1 --forward --force;
patching file
chrome/browser/speech/chrome_speech_recognition_manager_delegate.cc
Hunk #1 succeeded at 13 with fuzz 2.
Hunk #2 FAILED at 36.
Hunk #3 succeeded at 109 (offset 1 line).
Hunk #4 succeeded at 127 (offset 1 line).
Hunk #5 succeeded at 139 (offset 1 line).
Hunk #6 succeeded at 150 (offset 1 line).
Hunk #7 succeeded at 175 (offset 1 line).
Hunk #8 succeeded at 220 (offset 1 line).
Hunk #9 succeeded at 243 (offset 1 line).
Hunk #10 succeeded at 276 (offset 1 line).
Hunk #11 succeeded at 324 (offset 1 line).
Hunk #12 succeeded at 377 (offset 3 lines).
1 out of 12 hunks FAILED -- saving rejects to file
chrome/browser/speech/chrome_speech_recognition_manager_delegate.cc.rej

[commit-bot: I haz the power, 2012-05-23 08:05:54]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/primiano@chromium.org/10377082/17001

[commit-bot: I haz the power, 2012-05-23 08:06:01]

Failed to apply patch for
chrome/browser/speech/chrome_speech_recognition_manager_delegate.cc:
While running patch -p1 --forward --force;
patching file
chrome/browser/speech/chrome_speech_recognition_manager_delegate.cc
Hunk #1 succeeded at 13 with fuzz 2.
Hunk #2 FAILED at 36.
Hunk #3 succeeded at 109 (offset 1 line).
Hunk #4 succeeded at 127 (offset 1 line).
Hunk #5 succeeded at 139 (offset 1 line).
Hunk #6 succeeded at 150 (offset 1 line).
Hunk #7 succeeded at 175 (offset 1 line).
Hunk #8 succeeded at 220 (offset 1 line).
Hunk #9 succeeded at 243 (offset 1 line).
Hunk #10 succeeded at 276 (offset 1 line).
Hunk #11 succeeded at 324 (offset 1 line).
Hunk #12 succeeded at 377 (offset 3 lines).
1 out of 12 hunks FAILED -- saving rejects to file
chrome/browser/speech/chrome_speech_recognition_manager_delegate.cc.rej

[commit-bot: I haz the power, 2012-05-23 09:11:13]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/primiano@chromium.org/10377082/24003

[commit-bot: I haz the power, 2012-05-23 12:01:27]

Try job failure for 10377082-24003 (retry) on win_rel for step "runhooks"
(clobber build).
It's a second try, previously, step "compile" failed.
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=win_rel&nu...

[commit-bot: I haz the power, 2012-05-23 12:35:07]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/primiano@chromium.org/10377082/24003

[commit-bot: I haz the power, 2012-05-23 12:37:21]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/primiano@chromium.org/10377082/24003

[commit-bot: I haz the power, 2012-05-23 16:00:42]

Try job failure for 10377082-24003 (retry) on win_rel for step
"sync_unit_tests".
It's a second try, previously, step "sync_unit_tests" failed.
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=win_rel&nu...

[Joao da Silva, 2012-05-23 18:56:45]

On 2012/05/23 16:00:42, I haz the power (commit-bot) wrote:
> Try job failure for 10377082-24003 (retry) on win_rel for step
> "sync_unit_tests".
> It's a second try, previously, step "sync_unit_tests" failed.
>
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=win_rel&nu...

This change causes crashes in some browser_tests on chromeos. Unfortunately the
linux_chromeos runs only compiled and didn't run the suites.

https://code.google.com/p/chromium/issues/detail?id=129408

[Primiano Tucci (use gerrit), 2012-05-24 09:34:34]

This CL was reverted.
The new CL with the fix is here:
http://codereview.chromium.org/10440011