Otherwise we can have a situation like this:
<style>
.notmatching > #foo {}
</style>
<span id="foo"></span>
<span></span>
and the style sharing cache lookup for the second <span> would try to revalidate
against the cached value for the first <span>, but end up failing asserts about
the two elements matching lists of revalidation selectors that have the same
length.
This ensures that we can pass a node address as part of the asynchronous
transition end notification, making it safe to fire the corresponding
DOM event on the node from the script thread. Without explicitly rooting
this node when the transition starts, we risk the node being GCed before
the transition is complete.
style: Slim down SharedStyleContext, and do various other cleanups around the style crate.
This slims down SharedStyleContext, in preparation for a few things.
First, I would like to eventually move the stylist to the document in Servo, in
order for it to hold the StyleSheetSet.
Also, this gets rid of a fair amount of overhead while creating it in stylo.
Fixes bug 1363245.
<!-- 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/16802)
<!-- Reviewable:end -->
This slims down SharedStyleContext, in preparation for a few things.
First, I would like to eventually move the stylist to the document in Servo, in
order for it to hold the StyleSheetSet.
Also, this gets rid of a fair amount of overhead while creating it in stylo.
Fixes bug 1363245.
I've chosen this approach mainly because there's no other good way to guarantee
the model is correct than holding the snapshots alive until a style refresh.
What I tried before this (storing them in a sort of "immutable element data") is
a pain, since we call into style from the frame constructor and other content
notifications, which makes keeping track of which snapshots should be cleared an
which shouldn't an insane task.
Ideally we'd have a single entry-point for style, but that's not the case right
now, and changing that requires pretty non-trivial changes to the frame
constructor.
MozReview-Commit-ID: FF1KWZv2iBM
Signed-off-by: Emilio Cobos Álvarez <emilio@crisal.io>
Text style is always resolved via ServoStyleSet::ResolveStyleForText, either
from the frame constructor initially, or from
ServoRestyleManager::ProcessPostTraversalForText.
So text-only adjustments should go there instead. Since that doesn't call into
cascade(), all the code that passes `pseudo` there is dead code we can remove.
MozReview-Commit-ID: jpbBYpLlUL
Signed-off-by: Emilio Cobos Álvarez <emilio@crisal.io>
It's unfortunate, but it's a correctness issue. I was looking at the
expectations update here:
* https://hg.mozilla.org/integration/autoland/rev/659cddddd434
And investigating it I realised that it's wrong to coalesce selectors like that,
because we keep the bloom filter flags.
So in the test cases disabled, we have a selector that looks like this:
msub > :not(:first-child),
msup > :not(:first-child),
msubsup > :not(:first-child),
mmultiscripts > :not(:first-child) {
-moz-script-level: +1;
-moz-math-display: inline;
}
And an element that looks like this:
<msubsup><mi></mi><mi></mi></msubsup>
We're only inserting the first selector msub > :not(:first-child) into the set,
so when we're going to match the <mi> elements we fast-reject it in both cases
due to the bloom filter, so they share style.
I can't see an easy way to fix this keeping the deduplication. If we keep it, we
need to remove the bloom filter optimization, which means that we'd trash the
cache for every first-child in the document (the :not(:first-child) effectively
becomes a global rule).
MozReview-Commit-ID: 9VPkmdj9zDg
Signed-off-by: Emilio Cobos Álvarez <emilio@crisal.io>
Ports the Gecko fixup for text-combine-upright writing mode to Servo. In
addition, this passes the current pseudo element (if any) down to the cascade
for use during the fixup process.
MozReview-Commit-ID: BkHd4AvSsOt
Now this method covers SMIL override rules, CSS animation / script
animation rules, and CSS transition rules so we should just use
"animation" in the generic sense.
All elements end up with AFFECTED_BY_STATE right now, presumably due to the UA
sheet, and some non-TS pseudo-classes we've added, which means that we currently
never insert anything into the cache.
MozReview-Commit-ID: 5IU4qrjeJFy
1. We need to call get_after_change_style, which is the computed styles
without transition rules, while process_animations.
2. If we have after-change style, we may replace the new computed values with
after-change style, according to whether we really need to update
transitions.
3. There are some cases we don't update transitions, so we need to early
return. might_needs_transitions_update() will check it first and it
will filter out most common cases.
4. needs_transitions_update() will check each property and existing running
transitions to make sure we really don't need to update transitions.
The logic of this function is similar with that of
nsTransitionManager::DoUpdateTransitions().
MozReview-Commit-ID: 2ccdPjgrxKz
It is possible to call get_after_change_style if there is no transition
rule. In order to avoid cloning the token computed values, we just return None.
The caller can use this Option to know which computed values should be
used.
MozReview-Commit-ID: 7fcgSVEtXWh
We will add another function, needs_update_transitions, to check if we need
to update transitions, so factor this out.
MozReview-Commit-ID: 5LYkyi4aDri
I wanted to add an environmental variable to disable the style sharing
cache for gecko, but the current pattern involves lazy_static!, which
involves an atomic operation on lookup, which is a bit hot to do each
time we try to share styles. This makes that work happen once per
process.
style: Refactor the selectors crate to remove ad-hoc logic.
Also refactors a lot of the restyle hint code to use these new hooks, and removes a lot of hacks in servo that shouldn't be needed because of correct handling of restyles due to attribute changes.
I just smoketested this, so going for a full try run 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/16293)
<!-- Reviewable:end -->
Also, simplify all the pre-snapshot attribute hacks in the script and style
code.
MozReview-Commit-ID: 6c9ipeb7Tnr
Signed-off-by: Emilio Cobos Álvarez <emilio@crisal.io>
The current mechanism requires threading a lot of this state through a bunch of
callsites that are several layers of abstraction above the code that actually
uses this vector (which is only compiled for servo). Putting it in
CurrentElementInfo gets it nicely out of the way.
MozReview-Commit-ID: 9PLBKcJreN0
