windowCapture: return 1x1 frame to minimized winodw on Linux.

windowCapture: return 1x1 frame to minimized winodw on Linux.

During window capturing, if the target window is minimized, OSX/Windows
will return a 1x1 frame and then webrtc knows to replace it with a black
frame. Let's do same on Linux too.

BUG=568840
Review-Url: https://codereview.webrtc.org/2989233002
Cr-Commit-Position: refs/heads/master@{#19224}
Committed: https://chromium.googlesource.com/external/webrtc/+/adb161fecca5d46c75582a80bd07b4d7e658793a

[braveyao1, 2017-08-01 22:51:20]

The CQ bit was checked by braveyao@webrtc.org to run a CQ dry run

[commit-bot: I haz the power, 2017-08-01 22:51:27]

Dry run: CQ is trying da patch.

Follow status at:
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.webrtc.org/...

[commit-bot: I haz the power, 2017-08-01 23:43:57]

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

[commit-bot: I haz the power, 2017-08-01 23:43:58]

Dry run: This issue passed the CQ dry run.

[braveyao1, 2017-08-02 16:41:40]

Hi zijiehe@, please take a look.

[braveyao1, 2017-08-02 16:42:10]

braveyao@webrtc.org changed reviewers:
+ zijiehe@chromium.org

[braveyao1, 2017-08-02 16:42:10]

Hi zijiehe@, please take a look.

[Hzj_jie, 2017-08-02 20:36:58]

On 2017/08/02 16:42:10, braveyao1 wrote:
> Hi zijiehe@, please take a look.

This change itself is lgtm.
But I do think returning a 1x1 black DesktopFrame to indicate that the selected
window is minimized is a little bit hacky.

I think we can add a new DesktopCapturer::Result, maybe SOURCE_UNAVAILABLE. So
DesktopCaptureDevice can handle it according to its own requirement, i.e.
generating a 2x2 black DesktopFrame. WDYT?

[braveyao1, 2017-08-02 21:49:40]

On 2017/08/02 20:36:58, Hzj_jie wrote:
> On 2017/08/02 16:42:10, braveyao1 wrote:
> > Hi zijiehe@, please take a look.
> 
> This change itself is lgtm.
> But I do think returning a 1x1 black DesktopFrame to indicate that the
selected
> window is minimized is a little bit hacky.
> 
> I think we can add a new DesktopCapturer::Result, maybe SOURCE_UNAVAILABLE. So
> DesktopCaptureDevice can handle it according to its own requirement, i.e.
> generating a 2x2 black DesktopFrame. WDYT?

This is as same as the hacky thing in windows,
https://cs.chromium.org/chromium/src/third_party/webrtc/modules/desktop_captu...,
which follows the OSX system behavior :)

To add window-iconic checking on all platforms and return a new Result is also
good. (https://chromium-review.googlesource.com/c/596920/ needs to be re-written
then). Just can't see any advantage from this option.
(BTW: In another cl, I preferred to pre-process the input frame, e.g. generate
2x2 black frame and round odd dimensions, in
DesktopCaptureDevice::OnCaptureResult(), and that idea was disliked. And if we
are about to handle a new Result for minimized window, I think we are back to
that way again. Then I still prefer to do the rounding together before
resolution_chooser)
This is kind of big refactoring. What to call?

[Hzj_jie, 2017-08-02 22:23:18]

On 2017/08/02 21:49:40, braveyao1 wrote:
> On 2017/08/02 20:36:58, Hzj_jie wrote:
> > On 2017/08/02 16:42:10, braveyao1 wrote:
> > > Hi zijiehe@, please take a look.
> > 
> > This change itself is lgtm.
> > But I do think returning a 1x1 black DesktopFrame to indicate that the
> selected
> > window is minimized is a little bit hacky.
> > 
> > I think we can add a new DesktopCapturer::Result, maybe SOURCE_UNAVAILABLE.
So
> > DesktopCaptureDevice can handle it according to its own requirement, i.e.
> > generating a 2x2 black DesktopFrame. WDYT?
> 
> This is as same as the hacky thing in windows,
>
https://cs.chromium.org/chromium/src/third_party/webrtc/modules/desktop_captu...,
> which follows the OSX system behavior :)
> 
> To add window-iconic checking on all platforms and return a new Result is also
> good. (https://chromium-review.googlesource.com/c/596920/ needs to be
re-written
> then). Just can't see any advantage from this option.
> (BTW: In another cl, I preferred to pre-process the input frame, e.g. generate
> 2x2 black frame and round odd dimensions, in
> DesktopCaptureDevice::OnCaptureResult(), and that idea was disliked. And if we
> are about to handle a new Result for minimized window, I think we are back to
> that way again. Then I still prefer to do the rounding together before
> resolution_chooser)
> This is kind of big refactoring. What to call?

I just feel that assuming 1x1 == minimized window is too hacky: what if a
platform supports 1x1 window. FYI, on Windows 10 with WinForms, the minimum
window size can be 2x2.
I agree usually this does not matter.

[braveyao1, 2017-08-02 22:34:12]

The CQ bit was checked by braveyao@webrtc.org

[commit-bot: I haz the power, 2017-08-02 22:34:30]

CQ is trying da patch.

Follow status at:
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.webrtc.org/...

[commit-bot: I haz the power, 2017-08-02 22:37:30]

CQ is committing da patch.

Bot data: {"patchset_id": 1, "attempt_start_ts": 1501713252732640, "parent_rev":
"dd7d8f1b609d51bcf39e9585871967a694a856bb", "commit_rev":
"adb161fecca5d46c75582a80bd07b4d7e658793a"}

[commit-bot: I haz the power, 2017-08-02 22:37:36]

Description was changed from

==========
windowCapture: return 1x1 frame to minimized winodw on Linux.

During window capturing, if the target window is minimized, OSX/Windows
will return a 1x1 frame and then webrtc knows to replace it with a black
frame. Let's do same on Linux too.

BUG=568840
==========

to

==========
windowCapture: return 1x1 frame to minimized winodw on Linux.

During window capturing, if the target window is minimized, OSX/Windows
will return a 1x1 frame and then webrtc knows to replace it with a black
frame. Let's do same on Linux too.

BUG=568840

Review-Url: https://codereview.webrtc.org/2989233002
Cr-Commit-Position: refs/heads/master@{#19224}
Committed:
https://chromium.googlesource.com/external/webrtc/+/adb161fecca5d46c75582a80b...
==========

[commit-bot: I haz the power, 2017-08-02 22:37:37]

Committed patchset #1 (id:1) as
https://chromium.googlesource.com/external/webrtc/+/adb161fecca5d46c75582a80b...