1496383 - MediaRecorder state error cases does not match its W3C spec

Status: RESOLVED FIXED

(Core :: Audio/Video: Recording, defect, P2)

Description

[Bumsik Kim]

User Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/69.0.3497.100 Safari/537.36

Steps to reproduce:

In the MediaRecorder implementation (dom/media/MediaRecorder.cpp), specifically MediaRecorder.pause() and MediaRecorder.resume(), conditions of NS_ERROR_DOM_INVALID_STATE_ERR does not match to the W3C MediaRecorder spec:  https://w3c.github.io/mediacapture-record/#dom-mediarecorder-pause . I am currently writing a patch to resolve it.


Actual results:

In Firefox implementation (in dom/media/MediaRecorder.cpp):

pause(): if (state != Recording) => NS_ERROR_DOM_INVALID_STATE_ERR
resume(): if (state != Paused) => NS_ERROR_DOM_INVALID_STATE_ERR


Expected results:

W3C standard states:

pause(): if (state == inactive) => NS_ERROR_DOM_INVALID_STATE_ERR
resume(): if (state == inactive) => NS_ERROR_DOM_INVALID_STATE_ERR

Comment 1

[Paul Adenot (:padenot)]

Make sure to let us know when you have a patch ready, thanks!

Comment 2

[Bumsik Kim]

Attachment: Bug 1496383 - Fix state error cases of MediaRecorder to match the W3C spec

Fix a bug that the current MediaRecorder's state error cases does not match the W3C spec.
pause() and resume() should throw an INVAILD_STATE_ERR only when it is inactive state, making them 
idempotent.
Simply changing if statements are enough because the underlying encoder object (TrackEncoder) will 
ignore Suspend/Resume calls when it is already suspended/recording so there won't be side-effects by 
multiple pause()/resume() calls.

Comment 3

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/autoland/rev/a3a767ad55fb
Fix state error cases of MediaRecorder to match the W3C spec r=jya

Comment 4

[Stefan Hindli [:stefan_hindli]]

https://hg.mozilla.org/mozilla-central/rev/a3a767ad55fb