1473985 - Promises are not considered part of user-interaction, breaking Twitch embedded replay button.

Status: RESOLVED DUPLICATE of bug 1469200

(Core :: DOM: Core & HTML, enhancement, P2)

Description

[Thomas Wisniewski [:twisniewski]]

In bug 1469200, I found that Twitch's embedded player is calling a video's play() function not directly as part of the `click` event, but in a promise that is made while handling the click event:

>key: 'play',
>value: function () {
>  var e = this;
>  return this._srcReady.then(function () {
>    return new Promise(function (t) {
>      if (e._video.readyState >= 3)
>        t(e._video.play());
>      else {
>        var n = function n() {
>          e._video.removeEventListener(o.c, n),
>          t(e._video.play())
>        };
>        e._video.addEventListener(o.c, n)
>      }
>    })
>  }).catch (function (t) {
>    'NotAllowedError' === t.name && (e._video.dispatchEvent(new CustomEvent(o.m)), e._video.dispatchEvent(new CustomEvent(o.k)))
>  })
>}

Chrome allows the play promise to resolve, Firefox does not (it never resolves or rejects it).

If I do not wait for the _srcReady or other promises, then it plays fine in Firefox (but is missing Twitch's error-handling):

>key: 'play',
>value: function () {
>  var e = this;
>  return e._video.play();
>}

This makes me suspect that we should carry over permission to call play through into promise callbacks (and possibly requestAnimationFrame as well).

Comment 1

[Olli Pettay [:smaug][[e-mail]]]

(Promises have nothing to do with event handling)

Comment 2

[Firefox Bug Husbandry Bot]

See bug 1547409. Migrating webcompat priority whiteboard tags to project flags.

Comment 3

[Mike Taylor [:miketaylr]]

Tom, can we re-test to see if these reproduces?

Comment 4

[Hsin-Yi Tsai (she/her) [:hsinyi]]

(In reply to Mike Taylor [:miketaylr] from comment #3)

Tom, can we re-test to see if these reproduces?

Just tested it - still reproducible.

Comment 5

[Thomas Wisniewski [:twisniewski]]

Confirmed, the test case in https://bugzilla.mozilla.org/show_bug.cgi?id=1469200#c3 still reproduces.

Comment 6

[Hsin-Yi Tsai (she/her) [:hsinyi]]

(In reply to Thomas Wisniewski [:twisniewski] from comment #5)

Confirmed, the test case in https://bugzilla.mozilla.org/show_bug.cgi?id=1469200#c3 still reproduces.

https://bugzilla.mozilla.org/show_bug.cgi?id=1469200#c12 pointed out that this looks the website issue.

Should we just close this as duplicate of bug 1469200?

Comment 7

[Thomas Wisniewski [:twisniewski]]

Yes, based on that we might we well just close this as a dupe, thanks.