1270154 - ogv video stopped working after one of the updates

Status: VERIFIED FIXED

(Core :: Audio/Video: Playback, defect)

Description

[creativetalking]

User Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:46.0) Gecko/20100101 Firefox/46.0
Build ID: 20160421124000

Steps to reproduce:

Please, open this link http://www.chinesehideout.com/tools/strokeorder.php?c=5ZCX - there you're supposed to see chinese animation.


Actual results:

The animation (ogv videos) are loaded but not displayed. It was fine a long time ago, but after one of the updates it stopped displaying. It is working fine in google Chrome.


Expected results:

The animation should be visible.

Comment 1

[Loic]

Attachment: 1270154.html

Comment 2

[Loic]

Regression range:
https://hg.mozilla.org/integration/mozilla-inbound/pushloghtml?fromchange=8f72e3889e802851d568f7a3cefa0a2979f9ba3d&tochange=32f7dd06ba4c1ecdbb281a0a25df3a08c6401e73

Surely faulty:
Jean-Yves Avenard — Bug 1217304

Comment 3

[Jean-Yves Avenard [:jya]]

[Tracking Requested - why for this release]: it's a regression that appears worth fixing.

Comment 4

[Loic]

For the record, the MP4 video in the testcase is broken if loaded in a tab, but it has been always the case, it's probably poorly encoded (and that's not the main issue of this bug).

Comment 5

[Jean-Yves Avenard [:jya]]

the mp4 isn't broken, it's just mpeg4 part 2 (divx) which we don't support.

The issue here is that following 1217304, loadedmetadata succeeded and it appears that we have an opaque blank frame container located right above the ogv window playing underneath, so you can't see it.

Comment 6

[Jean-Yves Avenard [:jya]]

Attachment: MozReview Request: Bug 1270154: P1. Forget the media element's media-resource-specific tracks when error occurs. r?jwwang

Review commit: https://reviewboard.mozilla.org/r/51945/diff/#index_header
See other reviews: https://reviewboard.mozilla.org/r/51945/

Comment 7

[JW Wang [:jwwang] [:jw_wang]]

Comment on attachment 8751299 [details]
MozReview Request: Bug 1270154: P1. Forget the media element's media-resource-specific tracks when error occurs. r?jwwang

https://reviewboard.mozilla.org/r/51945/#review49001

Can you explain how this change fixes the bug?

Comment 8

[Jean-Yves Avenard [:jya]]

(In reply to JW Wang [:jwwang] from comment #7)
> Can you explain how this change fixes the bug?

TBH, I can't. the elements are created in exactly the same fashion. But if the videoTrack is enabled on the previously tried video then it hides the video that's actually playing.

Comment 9

[JW Wang [:jwwang] [:jw_wang]]

nsVideoFrame::BuildLayer calls |element->GetVideoSize(&videoSizeInPx)| which returns NS_ERROR_FAILURE for |mDisableVideo| is true which is set by |mDisableVideo = !track->Selected()| in HTMLMediaElement::NotifyMediaTrackEnabled().

It makes sense to clear all audio/video tracks when loading a new resource as HTMLMediaElement::AbortExistingLoads() also does that.

However, NotifyMediaTrackEnabled() doesn't handle multiple tracks for now. (https://hg.mozilla.org/mozilla-central/file/3461f3cae78495f100a0f7d3d2e0b89292d3ec02/dom/html/HTMLMediaElement.cpp#l1008)

Loading a 2nd resource without clearing tracks will hit this case.

Comment 10

[Pulsebot]

https://hg.mozilla.org/integration/mozilla-inbound/rev/7043159e6b74

Comment 11

[Carsten Book [:Tomcat]]

https://hg.mozilla.org/mozilla-central/rev/7043159e6b74

Comment 12

[Maire Reavy [:mreavy]]

(In reply to Jean-Yves Avenard [:jya] from comment #3)
> [Tracking Requested - why for this release]: it's a regression that appears
> worth fixing.

Jean-Yves  -- Can you ask for uplift if you still think this fix is a good candidate for uplift to  Fx48 and Fx47?  Thanks.

Comment 13

[Ritu Kothari (:ritu) (Inactive, please n-i to RyanVM, jcristau, or pascal)]

Hello, could you please verify whether this issue is fixed as expected on a latest Nightly build? Thanks!

Comment 14

[Jean-Yves Avenard [:jya]]

Comment on attachment 8751299 [details]
MozReview Request: Bug 1270154: P1. Forget the media element's media-resource-specific tracks when error occurs. r?jwwang

Approval Request Comment
[Feature/regressing bug #]: 1270154
[User impact if declined]: regression, content with multiple media source will not play. 
[Describe test coverage new/current, TreeHerder]: in central for a while, manually tested. 
[Risks and why]: low. We're allowing fallback when things didn't work. 
[String/UUID change made/needed]:none

Comment 15

[Ritu Kothari (:ritu) (Inactive, please n-i to RyanVM, jcristau, or pascal)]

Comment on attachment 8751299 [details]
MozReview Request: Bug 1270154: P1. Forget the media element's media-resource-specific tracks when error occurs. r?jwwang

The fix seems simple enough, seems like a recent regression since Fx46, Aurora48+, Beta47+

Comment 16

[Liz Henry (:lizzard Please n-i to RyanVM, jcristau, or pascal)]

Wontfix for 46, as we don't have another driver for 46.0.2 and are only a couple of weeks away from 47 release.

Comment 17

[Wes Kocher (:KWierso) (Not reading bugmail; email directly if needed)]

https://hg.mozilla.org/releases/mozilla-aurora/rev/6eed58568de0

Comment 18

[Wes Kocher (:KWierso) (Not reading bugmail; email directly if needed)]

https://hg.mozilla.org/releases/mozilla-beta/rev/9b2f7e585d3f

Comment 19

[Mohammad Maruf Islam]

I have reproduced this bug with Firefox Nightly 49.0a1 (Build ID: 20160504043118) on 
Windows 8.1, 64-bit.

Verified as fixed with Firefox release 47.0 (Build ID:20160604131506)
Mozilla/5.0 (Windows NT 6.3; WOW64; rv:47.0) Gecko/20100101 Firefox/47.0

Verified as fixed with Firefox beta 48.0b1 (Build ID:20160606200529)
Mozilla/5.0 (Windows NT 6.3; WOW64; rv:48.0) Gecko/20100101 Firefox/48.0

Verified as fixed with Firefox Developer edition 49.0a2 (Build ID:20160613004107)
Mozilla/5.0 (Windows NT 6.3; WOW64; rv:49.0) Gecko/20100101 Firefox/49.0

Comment 20

[Farhadur Reja Fahim]

I have reproduced this bug with Firefox Nightly 49.0a1 (2016-05-04)on Ubuntu 16.04, 64 bit.

The bug's fix is now verified on latest Aurora 49.0a2, Nightly 50.0a1.

Aurora 49.0a2:
Build ID 	20160722004032
User Agent 	Mozilla/5.0 (X11; Linux x86_64; rv:49.0) Gecko/20100101 Firefox/49.0

Nightly 50.0a1:
Build ID 	20160722030235
User Agent 	Mozilla/5.0 (X11; Linux x86_64; rv:50.0) Gecko/20100101 Firefox/50.0


[testday-20160722]

Comment 21

[Rezaul Huque Nayeem]

As it is verified on both windows(Comment 19) and linux(Comment 20) Marking it as verified!