DesktopCaptureDevice is available when webrtc is enabled.

DesktopCaptureDevice is available when webrtc is enabled.
    
ENABLED_SCREEN_CAPTURE is not flag to enable DesktopCaptureDevice.
It enables DesktopCaptureDeviceAura (if use_aura=1).
    
Use BUILDFLAG(ENABLE_WEBRTC) instead of ENABLED_SCREEN_CAPTURE to
use DesktopCaptureDevice in VidepCaptureManager.


BUG=internal b/34176576

Review-Url: https://codereview.chromium.org/2625553003
Cr-Commit-Position: refs/heads/master@{#442957}
Committed: https://chromium.googlesource.com/chromium/src/+/c164a25463bbe2aee8e6dd4672b77afe16559ba7

[byungchul, 2017-01-10 00:53:02]

The CQ bit was checked by byungchul@chromium.org to run a CQ dry run

[byungchul, 2017-01-10 00:53:06]

The CQ bit was unchecked by byungchul@chromium.org

[byungchul, 2017-01-10 00:53:25]

The CQ bit was checked by byungchul@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2017-01-10 00:54:07]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[byungchul, 2017-01-10 01:02:24]

Description was changed from

==========
Disable screen capture for cast audio devices.

No DesktopCaptureDevice is available for cast audio device because use_aura=0.

BUG=internal b/34176576
==========

to

==========
Disable screen capture for cast audio devices.

No DesktopCaptureDevice is available for cast audio device because use_aura=0.

BUG=internal b/34176576
==========

[byungchul, 2017-01-10 01:02:24]

byungchul@chromium.org changed reviewers:
+ boliu@chromium.org, slan@chromium.org

[commit-bot: I haz the power, 2017-01-10 02:56:08]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2017-01-10 02:56:09]

Dry run: Try jobs failed on following builders:
  chromeos_amd64-generic_chromium_compile_only_ng on
master.tryserver.chromium.linux (JOB_TIMED_OUT, no build URL)
  linux_chromium_compile_dbg_ng on master.tryserver.chromium.linux
(JOB_TIMED_OUT, no build URL)

[byungchul, 2017-01-10 17:14:38]

byungchul@chromium.org changed reviewers:
+ chfremer@chromium.org, mcasas@chromium.org

[byungchul, 2017-01-10 17:15:20]

The CQ bit was checked by byungchul@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2017-01-10 17:15:44]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[chfremer, 2017-01-10 17:27:21]

lgtm

[boliu, 2017-01-10 17:31:51]

I'll stamp after a chromecast owner approves

https://codereview.chromium.org/2625553003/diff/20001/content/browser/rendere...
File content/browser/renderer_host/media/video_capture_manager.cc (right):

https://codereview.chromium.org/2625553003/diff/20001/content/browser/rendere...
content/browser/renderer_host/media/video_capture_manager.cc:665: #if
defined(OS_LINUX) || defined(OS_MACOSX) || defined(OS_WIN)
don't need the inner define anymore

[byungchul, 2017-01-10 17:35:36]

byungchul@chromium.org changed reviewers:
+ halliwell@chromium.org

[byungchul, 2017-01-10 17:40:38]

https://codereview.chromium.org/2625553003/diff/20001/content/browser/rendere...
File content/browser/renderer_host/media/video_capture_manager.cc (right):

https://codereview.chromium.org/2625553003/diff/20001/content/browser/rendere...
content/browser/renderer_host/media/video_capture_manager.cc:665: #if
defined(OS_LINUX) || defined(OS_MACOSX) || defined(OS_WIN)
On 2017/01/10 17:31:51, boliu wrote:
> don't need the inner define anymore

Done.

[halliwell, 2017-01-10 17:45:44]

https://codereview.chromium.org/2625553003/diff/40001/content/browser/BUILD.gn
File content/browser/BUILD.gn (right):

https://codereview.chromium.org/2625553003/diff/40001/content/browser/BUILD.g...
content/browser/BUILD.gn:1514: if ((is_linux && !(is_chromecast &&
is_cast_audio_only)) || is_mac || is_win) {
From the CL description, it sounds like any linux build with use_aura = 0 could
have this problem.  Did I understand that right?  If so, would it be better to
generalise this to (is_linux && use_aura) ?

(I'm always wary of sprinkling is_chromecast too wide)

[byungchul, 2017-01-10 17:52:25]

On 2017/01/10 17:45:44, halliwell wrote:
> https://codereview.chromium.org/2625553003/diff/40001/content/browser/BUILD.gn
> File content/browser/BUILD.gn (right):
> 
>
https://codereview.chromium.org/2625553003/diff/40001/content/browser/BUILD.g...
> content/browser/BUILD.gn:1514: if ((is_linux && !(is_chromecast &&
> is_cast_audio_only)) || is_mac || is_win) {
> From the CL description, it sounds like any linux build with use_aura = 0
could
> have this problem.  Did I understand that right?  If so, would it be better to
> generalise this to (is_linux && use_aura) ?
> 
> (I'm always wary of sprinkling is_chromecast too wide)

It seems like linux instantiates WebContentsVideoCaptureDevice for web contents.
So, aura is not mandatory for linux.

[halliwell, 2017-01-10 18:04:20]

On 2017/01/10 17:52:25, byungchul wrote:
> On 2017/01/10 17:45:44, halliwell wrote:
> >
https://codereview.chromium.org/2625553003/diff/40001/content/browser/BUILD.gn
> > File content/browser/BUILD.gn (right):
> > 
> >
>
https://codereview.chromium.org/2625553003/diff/40001/content/browser/BUILD.g...
> > content/browser/BUILD.gn:1514: if ((is_linux && !(is_chromecast &&
> > is_cast_audio_only)) || is_mac || is_win) {
> > From the CL description, it sounds like any linux build with use_aura = 0
> could
> > have this problem.  Did I understand that right?  If so, would it be better
to
> > generalise this to (is_linux && use_aura) ?
> > 
> > (I'm always wary of sprinkling is_chromecast too wide)
> 
> It seems like linux instantiates WebContentsVideoCaptureDevice for web
contents.
> So, aura is not mandatory for linux.

If the bug is that non-aura linux builds don't link, then the change should only
mention use_aura (no need to mention is_chromecast).

[byungchul, 2017-01-10 18:14:55]

On 2017/01/10 18:04:20, halliwell wrote:
> On 2017/01/10 17:52:25, byungchul wrote:
> > On 2017/01/10 17:45:44, halliwell wrote:
> > >
> https://codereview.chromium.org/2625553003/diff/40001/content/browser/BUILD.gn
> > > File content/browser/BUILD.gn (right):
> > > 
> > >
> >
>
https://codereview.chromium.org/2625553003/diff/40001/content/browser/BUILD.g...
> > > content/browser/BUILD.gn:1514: if ((is_linux && !(is_chromecast &&
> > > is_cast_audio_only)) || is_mac || is_win) {
> > > From the CL description, it sounds like any linux build with use_aura = 0
> > could
> > > have this problem.  Did I understand that right?  If so, would it be
better
> to
> > > generalise this to (is_linux && use_aura) ?
> > > 
> > > (I'm always wary of sprinkling is_chromecast too wide)
> > 
> > It seems like linux instantiates WebContentsVideoCaptureDevice for web
> contents.
> > So, aura is not mandatory for linux.
> 
> If the bug is that non-aura linux builds don't link, then the change should
only
> mention use_aura (no need to mention is_chromecast).

It seems like same problem will happen when linux=1 && use_aura=0 &&
enable_webrtc=0.

@Christian, is the condition "linux && (use_aura || enabled_webrtc)" be better?

[byungchul, 2017-01-10 19:52:46]

Description was changed from

==========
Disable screen capture for cast audio devices.

No DesktopCaptureDevice is available for cast audio device because use_aura=0.

BUG=internal b/34176576
==========

to

==========
DesktopCaptureDevice is available when webrtc is enabled.
    
ENABLED_SCREEN_CAPTURE is not flag to enable DesktopCaptureDevice.
It enables DesktopCaptureDeviceAura (if use_aura=1).
    
Use BUILDFLAG(ENABLE_WEBRTC) instead of ENABLED_SCREEN_CAPTURE to
use DesktopCaptureDevice in VidepCaptureManager.


BUG=internal b/34176576
==========

[byungchul, 2017-01-10 19:52:46]

byungchul@chromium.org changed reviewers:
+ emircan@chromium.org

[byungchul, 2017-01-10 19:53:55]

On 2017/01/10 18:14:55, byungchul wrote:
> On 2017/01/10 18:04:20, halliwell wrote:
> > On 2017/01/10 17:52:25, byungchul wrote:
> > > On 2017/01/10 17:45:44, halliwell wrote:
> > > >
> >
https://codereview.chromium.org/2625553003/diff/40001/content/browser/BUILD.gn
> > > > File content/browser/BUILD.gn (right):
> > > > 
> > > >
> > >
> >
>
https://codereview.chromium.org/2625553003/diff/40001/content/browser/BUILD.g...
> > > > content/browser/BUILD.gn:1514: if ((is_linux && !(is_chromecast &&
> > > > is_cast_audio_only)) || is_mac || is_win) {
> > > > From the CL description, it sounds like any linux build with use_aura =
0
> > > could
> > > > have this problem.  Did I understand that right?  If so, would it be
> better
> > to
> > > > generalise this to (is_linux && use_aura) ?
> > > > 
> > > > (I'm always wary of sprinkling is_chromecast too wide)
> > > 
> > > It seems like linux instantiates WebContentsVideoCaptureDevice for web
> > contents.
> > > So, aura is not mandatory for linux.
> > 
> > If the bug is that non-aura linux builds don't link, then the change should
> only
> > mention use_aura (no need to mention is_chromecast).
> 
> It seems like same problem will happen when linux=1 && use_aura=0 &&
> enable_webrtc=0.
> 
> @Christian, is the condition "linux && (use_aura || enabled_webrtc)" be
better?

PTAL

I reverted BUILD.gn and modified video_capture_manager.cc only. I think this
approach is better.

[chfremer, 2017-01-10 22:08:08]

chfremer@chromium.org changed reviewers:
+ miu@chromium.org

[chfremer, 2017-01-10 22:08:09]

Hmm, I feel this is somewhat going in the wrong direction.
I felt okay about the slim changes to video_capture_manager.cc in the earlier 3
PatchSets, 
assuming that someone else would sign off on the flag usage in BUILD.gn.

Moving more of the build-time logic to #ifdefs in the .cc as in PatchSet#4 seems
dangerous to me. The build-time logic is 
getting uncomfortably scattered and hard to follow (and maybe redundant with
other places that depend on the same conditions?). Ideally, we should have only
_one_ feature 
flag ENABLE_SCREEN_CAPTURE being used in video_capture_manager.cc, and should
have 
_one_ place in some BUILD.gn file that determines its value. If we need more
fine-grained 
variations, we can introduce additional feature flags, and should give them
revealing names and
set their value in the same place as the ENABLE_SCREEN_CAPTURE flag.

I feel my understanding of the existing flags is too limited to give any better
advice.
Therefore I would like to defer to miu@ to help us out with his knowledge of the
desktop capture components.

miu@ could you please help us with your input on this?

[byungchul, 2017-01-10 22:15:25]

On 2017/01/10 22:08:09, chfremer wrote:
> Hmm, I feel this is somewhat going in the wrong direction.
> I felt okay about the slim changes to video_capture_manager.cc in the earlier
3
> PatchSets, 
> assuming that someone else would sign off on the flag usage in BUILD.gn.
> 
> Moving more of the build-time logic to #ifdefs in the .cc as in PatchSet#4
seems
> dangerous to me. The build-time logic is 
> getting uncomfortably scattered and hard to follow (and maybe redundant with
> other places that depend on the same conditions?). Ideally, we should have
only
> _one_ feature 
> flag ENABLE_SCREEN_CAPTURE being used in video_capture_manager.cc, and should
> have 
> _one_ place in some BUILD.gn file that determines its value. If we need more
> fine-grained 
> variations, we can introduce additional feature flags, and should give them
> revealing names and
> set their value in the same place as the ENABLE_SCREEN_CAPTURE flag.
> 
> I feel my understanding of the existing flags is too limited to give any
better
> advice.
> Therefore I would like to defer to miu@ to help us out with his knowledge of
the
> desktop capture components.
> 
> miu@ could you please help us with your input on this?

I agree that the less flags, the better. However, this is compatible to what
BUILD.gn specifies. Otherwise, you will encounter unresolved symbols with some
configuration, for example, like cast audio devices which is linux=1,
use_aura=1, enable_webrtc=0. So, if you feel the flags are not correct, BUILD.gn
should be updated accordingly, I think.

[chfremer, 2017-01-10 22:36:31]

On 2017/01/10 22:15:25, byungchul wrote:
> On 2017/01/10 22:08:09, chfremer wrote:
> > Hmm, I feel this is somewhat going in the wrong direction.
> > I felt okay about the slim changes to video_capture_manager.cc in the
earlier
> 3
> > PatchSets, 
> > assuming that someone else would sign off on the flag usage in BUILD.gn.
> > 
> > Moving more of the build-time logic to #ifdefs in the .cc as in PatchSet#4
> seems
> > dangerous to me. The build-time logic is 
> > getting uncomfortably scattered and hard to follow (and maybe redundant with
> > other places that depend on the same conditions?). Ideally, we should have
> only
> > _one_ feature 
> > flag ENABLE_SCREEN_CAPTURE being used in video_capture_manager.cc, and
should
> > have 
> > _one_ place in some BUILD.gn file that determines its value. If we need more
> > fine-grained 
> > variations, we can introduce additional feature flags, and should give them
> > revealing names and
> > set their value in the same place as the ENABLE_SCREEN_CAPTURE flag.
> > 
> > I feel my understanding of the existing flags is too limited to give any
> better
> > advice.
> > Therefore I would like to defer to miu@ to help us out with his knowledge of
> the
> > desktop capture components.
> > 
> > miu@ could you please help us with your input on this?
> 
> I agree that the less flags, the better. However, this is compatible to what
> BUILD.gn specifies. Otherwise, you will encounter unresolved symbols with some
> configuration, for example, like cast audio devices which is linux=1,
> use_aura=1, enable_webrtc=0. So, if you feel the flags are not correct,
BUILD.gn
> should be updated accordingly, I think.

I am unable to say anything about the correctness of the flags and need to refer

to miu@ for this.

My concern is that having flag logic in the .cc file that has to be "compatible"

or needs to "match" build logic in some BUILD.gn file sounds like redundancy,
and 
as such a potential source of bugs when changing one site requires changing the
other 
accordingly. Would it not be possible to have the flag logic in just one place,
preferably
in a BUILD.gn file?.

[byungchul, 2017-01-10 22:52:05]

On 2017/01/10 22:36:31, chfremer wrote:
> On 2017/01/10 22:15:25, byungchul wrote:
> > On 2017/01/10 22:08:09, chfremer wrote:
> > > Hmm, I feel this is somewhat going in the wrong direction.
> > > I felt okay about the slim changes to video_capture_manager.cc in the
> earlier
> > 3
> > > PatchSets, 
> > > assuming that someone else would sign off on the flag usage in BUILD.gn.
> > > 
> > > Moving more of the build-time logic to #ifdefs in the .cc as in PatchSet#4
> > seems
> > > dangerous to me. The build-time logic is 
> > > getting uncomfortably scattered and hard to follow (and maybe redundant
with
> > > other places that depend on the same conditions?). Ideally, we should have
> > only
> > > _one_ feature 
> > > flag ENABLE_SCREEN_CAPTURE being used in video_capture_manager.cc, and
> should
> > > have 
> > > _one_ place in some BUILD.gn file that determines its value. If we need
more
> > > fine-grained 
> > > variations, we can introduce additional feature flags, and should give
them
> > > revealing names and
> > > set their value in the same place as the ENABLE_SCREEN_CAPTURE flag.
> > > 
> > > I feel my understanding of the existing flags is too limited to give any
> > better
> > > advice.
> > > Therefore I would like to defer to miu@ to help us out with his knowledge
of
> > the
> > > desktop capture components.
> > > 
> > > miu@ could you please help us with your input on this?
> > 
> > I agree that the less flags, the better. However, this is compatible to what
> > BUILD.gn specifies. Otherwise, you will encounter unresolved symbols with
some
> > configuration, for example, like cast audio devices which is linux=1,
> > use_aura=1, enable_webrtc=0. So, if you feel the flags are not correct,
> BUILD.gn
> > should be updated accordingly, I think.
> 
> I am unable to say anything about the correctness of the flags and need to
refer
> 
> to miu@ for this.
> 
> My concern is that having flag logic in the .cc file that has to be
"compatible"
> 
> or needs to "match" build logic in some BUILD.gn file sounds like redundancy,
> and 
> as such a potential source of bugs when changing one site requires changing
the
> other 
> accordingly. Would it not be possible to have the flag logic in just one
place,
> preferably
> in a BUILD.gn file?.

I see your point. This is the typical case of your concern, breaking something
by inconsistent in BUILD.gn and .cc. Usually, the pattern how to resolve this
kind of issue is factory functions/classes with same signature. Unfortunately,
this is not the case. Weirdly, DesktopCaptureDeviceAura is not a subclass of
DesktopCaptureDevice, but both inherit media::VideoCaptureDevice.

[miu, 2017-01-10 23:22:01]

Hi all. This change may have been necessitated by my recent change
(https://codereview.chromium.org/2605973002/). There are reasons for the
confusion in this file:

1. DesktopCaptureDevice's implementation depends on WebRTC (it uses code from
this 3rd party lib).

2. On some platforms (Win/Linux-nonCrOS), there are situations where
DesktopCaptureDeviceAura could be used, but others where DesktopCaptureDevice
will have to be the fall-back.

3. I had originally envisioned that none/all devices are built into Chrome
depending on the ENABLE_SCREEN_CAPTURE flag. If that flag is not being set in
some circumstances where DesktopCaptureDevice can be used, then we should fix
the content/browser/BUILD.gn file. I see that we should move these lines
https://cs.chromium.org/chromium/src/content/browser/BUILD.gn?rcl=0&l=1503 to
here: https://cs.chromium.org/chromium/src/content/browser/BUILD.gn?rcl=0&l=1542
(wrapped in a "if (enable_webrtc) {...}").

I'll make more-specific comments on the patch in just a few minutes...

[miu, 2017-01-10 23:27:50]

https://codereview.chromium.org/2625553003/diff/60001/content/browser/rendere...
File content/browser/renderer_host/media/video_capture_manager.cc (right):

https://codereview.chromium.org/2625553003/diff/60001/content/browser/rendere...
content/browser/renderer_host/media/video_capture_manager.cc:43: #if
BUILDFLAG(ENABLE_WEBRTC) && !defined(OS_ANDROID)
Both (per prior e-mail on this CR):

  #if defined(ENABLE_SCREEN_CAPTURE) && BUILDFLAG(ENABLE_WEBRTC) &&
!defined(OS_ANDROID)

But, let's consolidate all of lines 43-53 as:

#if defined(ENABLE_SCREEN_CAPTURE)

#if BUILDFLAG(ENABLE_WEBRTC) && !defined(OS_ANDROID)
#include "content/browser/media/capture/desktop_capture_device.h"
#endif

#if defined(USE_AURA)
#include "content/browser/media/capture/desktop_capture_device_aura.h"
#endif

#if defined(OS_ANDROID)
#include "content/browser/media/capture/screen_capture_device_android.h"
#endif

#endif  // defined(ENABLE_SCREEN_CAPTURE)

https://codereview.chromium.org/2625553003/diff/60001/content/browser/rendere...
content/browser/renderer_host/media/video_capture_manager.cc:47: #if
defined(ENABLE_SCREEN_CAPTURE) && !defined(OS_ANDROID) && defined(USE_AURA)
!defined(OS_ANDROID) is not needed (since AURA is not built on Android).

https://codereview.chromium.org/2625553003/diff/60001/content/browser/rendere...
content/browser/renderer_host/media/video_capture_manager.cc:693: #if
defined(ENABLE_SCREEN_CAPTURE)
Let's move this back up to where it was, given previous comments (and prior
e-mails).

https://codereview.chromium.org/2625553003/diff/60001/content/browser/rendere...
content/browser/renderer_host/media/video_capture_manager.cc:715: #if
BUILDFLAG(ENABLE_WEBRTC) && !defined(OS_ANDROID)
...and move this up so that it is within the ENABLE_SCREEN_CAPTURE block.

[byungchul, 2017-01-10 23:48:33]

PTAL.

https://codereview.chromium.org/2625553003/diff/60001/content/browser/rendere...
File content/browser/renderer_host/media/video_capture_manager.cc (right):

https://codereview.chromium.org/2625553003/diff/60001/content/browser/rendere...
content/browser/renderer_host/media/video_capture_manager.cc:43: #if
BUILDFLAG(ENABLE_WEBRTC) && !defined(OS_ANDROID)
On 2017/01/10 23:27:49, miu wrote:
> Both (per prior e-mail on this CR):
> 
>   #if defined(ENABLE_SCREEN_CAPTURE) && BUILDFLAG(ENABLE_WEBRTC) &&
> !defined(OS_ANDROID)
> 
> But, let's consolidate all of lines 43-53 as:
> 
> #if defined(ENABLE_SCREEN_CAPTURE)
> 
> #if BUILDFLAG(ENABLE_WEBRTC) && !defined(OS_ANDROID)
> #include "content/browser/media/capture/desktop_capture_device.h"
> #endif
> 
> #if defined(USE_AURA)
> #include "content/browser/media/capture/desktop_capture_device_aura.h"
> #endif
> 
> #if defined(OS_ANDROID)
> #include "content/browser/media/capture/screen_capture_device_android.h"
> #endif
> 
> #endif  // defined(ENABLE_SCREEN_CAPTURE)

Done.

https://codereview.chromium.org/2625553003/diff/60001/content/browser/rendere...
content/browser/renderer_host/media/video_capture_manager.cc:693: #if
defined(ENABLE_SCREEN_CAPTURE)
On 2017/01/10 23:27:49, miu wrote:
> Let's move this back up to where it was, given previous comments (and prior
> e-mails).

Done.

https://codereview.chromium.org/2625553003/diff/60001/content/browser/rendere...
content/browser/renderer_host/media/video_capture_manager.cc:715: #if
BUILDFLAG(ENABLE_WEBRTC) && !defined(OS_ANDROID)
On 2017/01/10 23:27:49, miu wrote:
> ...and move this up so that it is within the ENABLE_SCREEN_CAPTURE block.

Done.

[miu, 2017-01-10 23:58:15]

PS6 LGTM % a minor thing:

https://codereview.chromium.org/2625553003/diff/100001/content/browser/BUILD.gn
File content/browser/BUILD.gn (right):

https://codereview.chromium.org/2625553003/diff/100001/content/browser/BUILD....
content/browser/BUILD.gn:1505: # Desktop screen capture implementations that are
not dependent on WebRTC.
This comment needs a little adjusting now. Maybe it could be just:

  # Desktop screen capture implementations, conditionally built depending on the
available implementations for each platform.

[boliu, 2017-01-11 01:02:02]

content/browser/BUILD.gn lgtm

[byungchul, 2017-01-11 17:13:04]

https://codereview.chromium.org/2625553003/diff/100001/content/browser/BUILD.gn
File content/browser/BUILD.gn (right):

https://codereview.chromium.org/2625553003/diff/100001/content/browser/BUILD....
content/browser/BUILD.gn:1505: # Desktop screen capture implementations that are
not dependent on WebRTC.
On 2017/01/10 23:58:15, miu wrote:
> This comment needs a little adjusting now. Maybe it could be just:
> 
>   # Desktop screen capture implementations, conditionally built depending on
the
> available implementations for each platform.

Done.

[byungchul, 2017-01-11 17:13:25]

The CQ bit was checked by byungchul@chromium.org

[byungchul, 2017-01-11 17:13:25]

The patchset sent to the CQ was uploaded after l-g-t-m from
chfremer@chromium.org, boliu@chromium.org, miu@chromium.org
Link to the patchset: https://codereview.chromium.org/2625553003/#ps120001
(title: "Fixed the comment in BUILD.gn")

[commit-bot: I haz the power, 2017-01-11 17:13:50]

CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2017-01-11 18:18:29]

CQ is committing da patch.

Bot data: {"patchset_id": 120001, "attempt_start_ts": 1484154805056230,
"parent_rev": "3ae6c6b59c0c9a5609fb220d7c494649ad1e826a", "commit_rev":
"c164a25463bbe2aee8e6dd4672b77afe16559ba7"}

[commit-bot: I haz the power, 2017-01-11 18:19:00]

Description was changed from

==========
DesktopCaptureDevice is available when webrtc is enabled.
    
ENABLED_SCREEN_CAPTURE is not flag to enable DesktopCaptureDevice.
It enables DesktopCaptureDeviceAura (if use_aura=1).
    
Use BUILDFLAG(ENABLE_WEBRTC) instead of ENABLED_SCREEN_CAPTURE to
use DesktopCaptureDevice in VidepCaptureManager.


BUG=internal b/34176576
==========

to

==========
DesktopCaptureDevice is available when webrtc is enabled.
    
ENABLED_SCREEN_CAPTURE is not flag to enable DesktopCaptureDevice.
It enables DesktopCaptureDeviceAura (if use_aura=1).
    
Use BUILDFLAG(ENABLE_WEBRTC) instead of ENABLED_SCREEN_CAPTURE to
use DesktopCaptureDevice in VidepCaptureManager.


BUG=internal b/34176576

Review-Url: https://codereview.chromium.org/2625553003
Cr-Commit-Position: refs/heads/master@{#442957}
Committed:
https://chromium.googlesource.com/chromium/src/+/c164a25463bbe2aee8e6dd4672b7...
==========

[commit-bot: I haz the power, 2017-01-11 18:19:02]

Committed patchset #7 (id:120001) as
https://chromium.googlesource.com/chromium/src/+/c164a25463bbe2aee8e6dd4672b7...