Better timeout handling

[mmcc]

I've left this comment on a few issues, so it's probably time to create a new catch-all for this one.

@heff and I talked this over, and we think it makes sense to have a setTimeout utility method that is identical to window.setTimeout, but adds the timeout to an array on the player instance so we can dispose all of them at once when the player is disposed.

This would clean up all the instances of having to create a dispose listener each time a timeout is created, and help us avoid accidentally missing one (which seems to happen pretty often) and allows plugins to utilize the same functionality.

Other (potential) references: #1521, #1484, #1505, #1431

[deweydb]

Is there any work around for these issues in the interim? I'm having a pretty big problem with this in my application as users switch between states after i .dispose() the player and try to re-create one later.

[ntodd]

Same here. I'm trying to use video.js in an Ember app and running into this as I dynamically create and remove players.

[deweydb]

I realized that the problem is mostly caused by timers that don't get cleaned up, so a simple fix is just to pause the video first before dispose.

player.pause();
setTimeout(function() {
player.dispose();
}, 0);

Found here: http://stackoverflow.com/questions/25493887/videojs-return-vdata-error-when-dispose-is-call

[ntodd]

@deweydb Interesting. I tried pause(), but didn't think to defer dispose(). I'll give it a shot. Thanks.

[ppaul]

In a context worked for me. Now again I had the issues. I could fix it by forcing in vjs lib to exit, like this:

vjs.trigger = function(elem, event) {
//there is a bug after disposing video
if (!elem) {
return false;
}
....

and also:

vjs.trigger = function(elem, event) {
// Fetches element data and a reference to the parent (for bubbling).
// Don't want to add a data object to cache for every parent,
// so checking hasData first.
var elemData = (elem && vjs.hasData(elem)) ? vjs.getData(elem) : {};
...

[mmcc]

@toru-sean The original context of this issue was resolved with #1642. If you're seeing this issue again, it could be a regression or an entirely different problem altogether, but we need some context here. Please use the issue template so folks can take a look. Thanks!

[willpiam]

Hey I'm here 5 years after the last activity, with at the very least a very similar issue. I'll try deferring dispose but a) that's disgusting and b) my error is coming from the line where I call videojs().

[willpiam]

yeah it didn't help

[mmcc]

It's really difficult to determine whether this is the same, new, or implementation issue without seeing a reduced test case that shows the problem. Think you could put something together there and we can take a look?