[Webkit-unassigned] [Bug 139313] REGRESSION (173394): Support for webcam is broken

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=139313

Alexey Proskuryakov <ap at webkit.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
 Attachment #242662|review?                     |review+
              Flags|                            |

--- Comment #3 from Alexey Proskuryakov <ap at webkit.org> ---
Comment on attachment 242662
  --> https://bugs.webkit.org/attachment.cgi?id=242662
patch

View in context: https://bugs.webkit.org/attachment.cgi?id=242662&action=review

> Source/WebKit2/NetworkProcess/NetworkResourceLoader.cpp:200
> +    // For multipart/x-mixed-replace didReceiveResponseAsync gets called multiple times and buffering would require special handling.
> +    if (response.isMultipart() && !isSynchronous())
> +        m_bufferedData = nullptr;

I'm not sure if I understand what happens (or what should happen) for sync loads.

Anyway, this would probably read a little better if the new code was inside an else clause below - there is no need to check isSynchronous twice.

> LayoutTests/ChangeLog:13
> +            The last image is repeated due to a WebCore side bug that causes last part to not render.

This may be appropriate to have in an HTML comment inside the test, as that's where one would be looking when they notice this strangeness.

> LayoutTests/http/tests/multipart/multipart-image.html:10
> +    if (window.testRunner)
> +        setTimeout("testRunner.notifyDone()", 1000);

Can we watch image size changes to finish faster and more reliably? One second is a long time to waste, but may be not enough to avoid random failures.

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.webkit.org/pipermail/webkit-unassigned/attachments/20141205/6c16c49c/attachment-0002.html>