There was a cycle amoung a window object -> a HTMLMediaElement -> a MediaDecoder -> a Promise (-> back to the window object).
And we have no way to break the cycle since the MediaDecoder does not participate into the collection.
By moving the Promise form MediaDecoder to HTMLMediaElement, we will be able to break the cycle since the HTMLMediaElement is in the collection.
We'll implement the cycle collection in the next patch.
MozReview-Commit-ID: CyVXBl6IMI3
On Fennec, we use external video player to play unsupported type media.
In this bug, we can't execute play() successfully because the spec said that
we should abort the play if the value of media element's error is
MEDIA_ERR_SRC_NOT_SUPPORTED [1].
Therefore, we do a hack for unsupported type media, we don't set its error, but
we would still remember the src is unsupported type. When media element starts
playing, we would use an external video player to play that src.
[1] https://goo.gl/GlLbeA
MozReview-Commit-ID: 42HxQtXdWDe
Since the audible state would depend on pause and suspend state, we should do
the check after finishing play() or pause().
MozReview-Commit-ID: 1yhxuQDw067
This mMediaTracksConstructed flag should belong to a MediaDecoder,
every time a HTMLMediaElement switches its MediaDecoder, the flag should be reset to false again.
So, we move the mMediaTracksConstructed flag back to MediaDecoder, by this way,
HTMLMediaElement provides only the mechanism to construct and remove media tracks,
and MediaDecoder uses the flag, mMediaTracksConstructed, to provide policy.
MozReview-Commit-ID: L7mMAmLjQCy
We should remove the media control interface immediately when the document is inacitve, even it was paused by media control before.
MozReview-Commit-ID: GBDhzEFOTXE
Move the creation of MediaElementGMPCrashHelper out from MediaDecoder.cpp
which reduces the dependency of MediaDecoder to HTMLMediaElement.
MozReview-Commit-ID: E60aMfcFr7V
Open a GetOwnerDoc() method to the MediaDecoderOwner interface and then we can get the
owner document via a pointer to MediaDecoderOwner in MediaDecoder.
MozReview-Commit-ID: JCzQDLx1MsU
We never suspend videos that is NOT in-tree because we found that, according to the Telemetry data, most (>70%) videos
which are used as the argument of drawImage() are not in-tree. So, by preventing suspending not-in-tree videos, we should
be able to alleviate the pain of not able to resume video decoders synchronously.
MozReview-Commit-ID: 8eqs0pHZLIt
Initialize the MediaDecoder::mIsElementVisible to be "!aOwner->IsHidden()" at the MediaDecoder's constructor is wrong.
Insted, we initialize both MediaDecoder::mIsDocumentVisible and MediaDecoder::mIsElementVisible to be false at the construtor,
and then assign the HTMLMediaElement's real values to them at HTMLMediaElement::FinishDecoderSetup() via the the MediaDecoder::SetActiviyChangesToDecoder().
The initialization values of MediaDecoder::mIsDocumentVisible and MediaDecoder::mIsElementVisible (in the constructor) do not matter because the valuse are
not read untile the first MediaDecoder::SetActiviyChangesToDecoder() method call.
MozReview-Commit-ID: Cdovq5pG9Nv
Mark video element as tainted (stored on the decoder owned by video element) when the video is used as source to drawImage() on canvas.
MozReview-Commit-ID: HdciVwhqPu3
Mark video element as tainted (stored on the decoder owned by video element) when the video is used as source to drawImage() on canvas.
MozReview-Commit-ID: HdciVwhqPu3
If the media starts after muting the tab, IsOwnerAudible() would always return
"eNotAudible". It causes that the "play tab" icon can't be displayed because we
only show the "play tab" icon for the "eMaybeAudible" or "eAudible".
We should check whether owner owns the audio track to decide the return value.
MozReview-Commit-ID: DGwkArx0a4R
In bug1319771, we found that the tab would become visible unexpectly in short
period in some situations. We don't want to resume the tab in this kind of
situation, so we check whether there is any alive media component in the tab
using IsServiceStarted(). However, since we have lots different ways to create
the service, this function is not accurate at all.
Therefore, we can add media element directly to the document when it binds to
tree so that we can really know whether there is any alive media component.
MozReview-Commit-ID: FvZFg91IqgE
The root cause of the intermittent fail is because "DOMAudioPlaybackStopped" has no directly relationship with browser.mute()/unmute().
In [1], the "DOMAudioPlaybackStopped" is caused by audio stop playing, not by calling the browser.mute(). If the audio stops playing before calling the wait_for_event(), the test would be time-out. I guess the bug 1302280 is also caused by same reason.
So this patch would do two thinngs,
1. dispatch "DOMAudioPlaybackStopped" when we mute tab
2. loop the audio in test file, to make sure the "DOMAudioPlaybackStopped" is
dispatched when muting the audio, not the file ended.
[1] https://goo.gl/ymUv8P
MozReview-Commit-ID: 5RnyBRE73lQ
The root cause of the intermittent fail is because "DOMAudioPlaybackStopped" has no directly relationship with browser.mute()/unmute().
In [1], the "DOMAudioPlaybackStopped" is caused by audio stop playing, not by calling the browser.mute(). If the audio stops playing before calling the wait_for_event(), the test would be time-out. I guess the bug 1302280 is also caused by same reason.
So this patch would do two thinngs,
1. dispatch "DOMAudioPlaybackStopped" when we mute tab
2. loop the audio in test file, to make sure the "DOMAudioPlaybackStopped" is
dispatched when muting the audio, not the file ended.
[1] https://goo.gl/ymUv8P
MozReview-Commit-ID: 703JHj9dICT
