Bug 1633486 Comment 5 Edit History

Note: The actual edited comment in the bug view page will always show the original commenter’s name and original timestamp.

(In reply to Boris Chiou [:boris] from comment #4)
> The test case creates an animation with all properties by implicit keyframes. I tried to make it simpler: only create `opacity: ['initial']`, and I still can reproduce this issue. It seems [`opacity` shouldn't be applied to `::marker` pseudo element](https://drafts.csswg.org/css-pseudo-4/#marker-pseudo), so I suspect that's why we can not get an underlying value for `opacity` on `::marker`.

If changing the property to color, we don't compose animation segment. Perhaps there are other issues. And the `hidden = false` is also necessary.
(In reply to Boris Chiou [:boris] from comment #4)
> The test case creates an animation with all properties by implicit keyframes. I tried to make it simpler: only create `opacity: ['initial']`, and I still can reproduce this issue. It seems [`opacity` shouldn't be applied to `::marker` pseudo element](https://drafts.csswg.org/css-pseudo-4/#marker-pseudo), so I suspect that's why we can not get an underlying value for `opacity` on `::marker`.

If I change the property to `color` or `transform`, we don't compose the animation segment on `color` or `transform`. Perhaps there are other issues. And the `hidden = false` is also necessary.
(In reply to Boris Chiou [:boris] from comment #4)
> The test case creates an animation with all properties by implicit keyframes. I tried to make it simpler: only create `opacity: ['initial']`, and I still can reproduce this issue. It seems [`opacity` shouldn't be applied to `::marker` pseudo element](https://drafts.csswg.org/css-pseudo-4/#marker-pseudo), so I suspect that's why we can not get an underlying value for `opacity` on `::marker`.

If I change the property to `color` or `transform`, we don't compose the animation segment on `color` or `transform`. Perhaps there are other issues. Besides, the `hidden = false` is also necessary.
Anyway, looks like for a normal animation on `::marker`, we should have a valid underlying value. However, this case doesn't have underlying value, so I got a lot of warning or crash on `opacity` or `color` property. Why we don't have an underlying value? Perhaps the change from hidden to unhidden is the reason. We perhaps need to handle this special case.
Anyway, looks like for a normal animation on `::marker`, we should have a valid underlying value. However, this case doesn't have underlying value, so I got a lot of warnings or a moz crash on `opacity` or `color` property. Why we don't have an underlying value? Perhaps the change from hidden to unhidden is the reason. We perhaps need to handle this special case.
Anyway, looks like for a normal animation on `::marker`, we should have a valid underlying value. However, this case doesn't have underlying value, so I got a lot of warnings or a moz crash on `opacity` or `color` property. So where is the underlying value? Perhaps the change from hidden to unhidden is the reason. We perhaps need to handle this special case.

Back to Bug 1633486 Comment 5