The Animation(Player) API got implemented (https://developer.mozilla.org/en-US/docs/Web/API/Animation) and may provide a new high resolution timestamp. If so, make sure to reduce its granularity respectively. This is behind a pref, dom.animations-api.core.enabled in Firefox 38 ESR. Thus, deferring to Firefox 45 ESR.
To upload designs, you'll need to enable LFS and have an admin enable hashed storage. More information
Child items 0
Show closed items
No child items are currently assigned. Use child items to break down this issue into smaller parts.
Trac: Summary: Investigate whether Animations API provides new high resolution timestamp to Investigate whether Animations(Player) API provides new high resolution timestamp Description: The Animations API got implemented (https://developer.mozilla.org/en-US/docs/Web/API/Animation) and may provide a new high resolution timestamp. If so, make sure to reduce its granularity respectively. This is behind a pref, dom.animations-api.core.enabled in Firefox 38 ESR. Thus, deferring to Firefox 45 ESR.
to
The Animation(Player) API got implemented (https://developer.mozilla.org/en-US/docs/Web/API/Animation) and may provide a new high resolution timestamp. If so, make sure to reduce its granularity respectively. This is behind a pref, dom.animations-api.core.enabled in Firefox 38 ESR. Thus, deferring to Firefox 45 ESR.
Actually, Element.animate() seems to be available right now already even if the other features of the API are not implemented yet (or not available on the release channel). See: https://bugzilla.mozilla.org/show_bug.cgi?id=1245000. Moving this back on our ESR 52 radar.
Trac: Keywords: ff59-esr deleted, ff52-esr added Sponsor: None to Sponsor4
We don't want to have a ResistFingerprinting-check here? What makes that feature different from, say, the one in #10286 (moved) where you explicitly added one? Or maybe that's just because the other timestamp clamping code does not have such a check either?
We don't want to have a ResistFingerprinting-check here? What makes that feature different from, say, the one in #10286 (moved) where you explicitly added one? Or maybe that's just because the other timestamp clamping code does not have such a check either?
Yes, I left it out to be consistent with the other clamping code we currently have in Tor Browser. Once it gets uplifted to Mozilla, we can put it behind a pref. Jonathan has already created infrastructure for that in https://bugzilla.mozilla.org/show_bug.cgi?id=1217238
This patch looks okay. Do we also need to patch GetCurrentTimeAsDouble() inside dom/animation/AnimationTimeline.h? Kathy and I are not sure how the Animation.timeline object relates to the other timing information.
This patch looks okay. Do we also need to patch GetCurrentTimeAsDouble() inside dom/animation/AnimationTimeline.h? Kathy and I are not sure how the Animation.timeline object relates to the other timing information.
That's a good point. Animation.timeline and document.timeline aren't exposed in TBB/ESR52 because "dom.animations-api.core.enabled" is set to false. But I looked at it again and I found there is a simpler way to implement this patch that covers the *.timeline cases as well. So here's an updated patch. (The document.timeline.currentTime test doesn't run unless the "dom.animations-api.core.enabled" pref has been set to true by default, but I don't think this is a problem for TBB.)