This matches Gecko, and fixes the last reported case in bug 1397619.
MozReview-Commit-ID: 4Z7GeaokKE1
Signed-off-by: Emilio Cobos Álvarez <emilio@crisal.io>
We share value representation, so there's no reason to do this in a more complex
way.
MozReview-Commit-ID: BITfCPqTkal
Signed-off-by: Emilio Cobos Álvarez <emilio@crisal.io>
We only need to handle changes when the display property is changed from 'none'
when we have a restyle hint for SMIL. The only other case where we expect to
see changes to display property during an animation are from using the CSSOM.
However, when the display property is changed from 'none' by the CSSOM, during
the animation-only restyle we can skip all descendants since they will be
traversed in the subsequent normal traversal because at that time we flush
style sheets and traverse all elements in the document. So we don't need to
care about the descendants during animation-only restyle.
`AnimationValue` has its own path for conversion to computed values, so we need
to update `for_non_inherited_property` in the context to match the property
being converted.
I didn't notice any other generic conversions, so hopefully this is the only
missing path.
MozReview-Commit-ID: HskrpMctmYE
Use Atom::from(nsIAtom) to increment reference count in case of dynam…
…ic atom for will-change.
If we don't increment the reference count for the Atom in servo side, it's
possible to try to release the Atom in servo side even if we have already
released in gecko side. When it happens, nsIAtom::mKind is no longer reliable.
https://bugzilla.mozilla.org/show_bug.cgi?id=1401809
<!-- Please describe your changes on the following line: -->
---
<!-- Thank you for contributing to Servo! Please replace each `[ ]` by `[X]` when the step is complete, and replace `__` with appropriate data: -->
- [X] `./mach build -d` does not report any errors
- [X] `./mach test-tidy` does not report any errors
<!-- Reviewable:start -->
---
This change is [<img src="https://reviewable.io/review_button.svg" height="34" align="absmiddle" alt="Reviewable"/>](https://reviewable.io/reviews/servo/servo/18590)
<!-- Reviewable:end -->
stylo: do not handle the fallback discrete animation inside the Animate trait
At present, we do the fallback discrete animation for non-invertible matrices in
ComputedMatrix.animate(). However, according to the spec, we should fallback to
discrete animation for cases like:
1. animation between transform with single non-invertible matrix
2. animation between transform with matched transform functions that have at least
one non-invertible matrix
2. animation between transform with mismatched transform functions that have at
least one non-invertible matrix.
The current implementation only handles the first case.
Moreover, we already have fallback discrete animation procedures in CSS Animation
and Web Animation, so we should be able to not doing any fallback inside the
Animate trait.
In this patch, we let the animation between non-invertible matrices to return Err().
So, we can propagate the Err() to the callers, and let the fallback discrete
animation procedure stay at the Servo_MatrixTransform_Operate, which is ouside
the Animate trait.
Gecko bug: [Bug 1394284](https://bugzilla.mozilla.org/show_bug.cgi?id=1394284)
---
- [X] `./mach build -d` does not report any errors
- [X] `./mach test-tidy` does not report any errors
- [X] These changes fix [Bug 1394284](https://bugzilla.mozilla.org/show_bug.cgi?id=1394284)
- [X] There are wpt tests for these changes, and thet will be landed in [Bug 1394284](https://bugzilla.mozilla.org/show_bug.cgi?id=1394284)
<!-- Reviewable:start -->
---
This change is [<img src="https://reviewable.io/review_button.svg" height="34" align="absmiddle" alt="Reviewable"/>](https://reviewable.io/reviews/servo/servo/18591)
<!-- Reviewable:end -->
At present, we do the fallback discrete animation for non-invertible matrices in
ComputedMatrix.animate(). However, according to the spec, we should fallback to
discrete animation for cases like:
1. animation between transform with single non-invertible matrix
2. animation between transform with matched transform functions that have at least
one non-invertible matrix
2. animation between transform with mismatched transform functions that have at
least one non-invertible matrix.
The current implementation only handles the first case.
Moreover, we already have fallback discrete animation procedures in CSS Animation
and Web Animation, so we should be able to not doing any fallback inside the
Animate trait.
In this patch, we let the animation between non-invertible matrices to return Err().
So, we can propagate the Err() to the callers, and let the fallback discrete
animation procedure stay at the Servo_MatrixTransform_Operate, which is ouside
the Animate trait.
If we don't increment the reference count for the Atom in servo side, it's
possible to try to release the Atom in servo side even if we have already
released in gecko side. When it happens, nsIAtom::mKind is no longer reliable.
Drop RulesChanged
It's no longer used.
<!-- Please describe your changes on the following line: -->
---
<!-- Thank you for contributing to Servo! Please replace each `[ ]` by `[X]` when the step is complete, and replace `__` with appropriate data: -->
- [X] `./mach build -d` does not report any errors
- [X] `./mach test-tidy` does not report any errors
<!-- Either: -->
- [X] These changes do not require tests because this just drops unused code.
<!-- Also, please make sure that "Allow edits from maintainers" checkbox is checked, so that we can help you if you get stuck somewhere along the way.-->
<!-- Pull requests that do not address these steps are welcome, but they will require additional verification as part of the review process. -->
<!-- Reviewable:start -->
---
This change is [<img src="https://reviewable.io/review_button.svg" height="34" align="absmiddle" alt="Reviewable"/>](https://reviewable.io/reviews/servo/servo/18584)
<!-- Reviewable:end -->
ServoStyleSetSizes now has two uses, one for the Stylist, and one for the UA
cache, and so the patch removes 'Stylist' from the field names.
Example output from about:memory:
> +----1,359,608 B (00.55%) -- layout
> | +----756,488 B (00.31%) -- style-sheet-cache [2]
> | +----393,968 B (00.16%) -- servo-ua-cache
> | | +--234,496 B (00.10%) -- element-and-pseudos-maps
> | | +---59,648 B (00.02%) -- revalidation-selectors
> | | +---58,320 B (00.02%) -- invalidation-map
> | | +---30,752 B (00.01%) -- other
> | | +---10,752 B (00.00%) -- precomputed-pseudos
I don't need this per se, but it seems like a footgun for the methods to
return incorrect information depending on the build type. I don't see anywhere
where the overhead would be at all significant.
MozReview-Commit-ID: G1qyUFhI0aB
Make shape-outside animatable.
We decide to make shape-outside animatable on Gecko, so let's do it on Servo.
---
- [X] `./mach build -d` does not report any errors
- [X] `./mach test-tidy` does not report any errors
- [X] These changes fix [Bug 1390026](https://bugzilla.mozilla.org/show_bug.cgi?id=1390026).
- [X] These changes do not require tests because we could reuse the animation tests on Gecko for now.
<!-- Reviewable:start -->
---
This change is [<img src="https://reviewable.io/review_button.svg" height="34" align="absmiddle" alt="Reviewable"/>](https://reviewable.io/reviews/servo/servo/18559)
<!-- Reviewable:end -->
