Create an extra style context using the visited values (if they exist). This
mirrors the logic Gecko performs in nsStyleSet::GetContext for visited support.
MozReview-Commit-ID: EiJQXDgz8tX
Also, add GetParentAllowServo() to whitelist the usage in servo-backed style
system in subsequent parts so that we could revisit them later.
MozReview-Commit-ID: JthU18Ftx02
--HG--
extra : rebase_source : b8840adcc4959f96cd429db65584f3e83c09c597
nsStyleContext::CalcDifference had an optimization where, when we knew
that the old and new style context have the same rule node, we knew that
the only change hints that would need to be handled are those in the
"not handled for descendants" category, generated due to explicit
'inherit' values on reset properties. This was because any changes due
to differences in inherited properties should only have generated
"handled for descendants" change hints (and thus would already have been
handled on an ancestor).
Before bug 931668, this let us avoid calling CalcDifference on structs
that only would have generated hints that we knew we already would have
handled. However, after bug 931668, we compare all structs anyway so
that we can set the aEqualStructs outparam, so we don't gain anything
from this optimization. We can still return these change hints we know
will not need to be handled, and rely on ElementRestyler::CaptureChange
to filter them out.
MozReview-Commit-ID: Ld1s2Js0i6r
We also assert against the Servo traversal because we don't currently test the
parallel traversal on CI (though we will soon).
MozReview-Commit-ID: 9GRNizE44Oi
The idea is to be able to call the right function during style resolution time
without doing a linear walk via nsCSSAnonBoxes::IsNonInheritingAnonBox.
MozReview-Commit-ID: JKt33GggTjz
--HG--
extra : rebase_source : fcf477dcef8529c803bf97b1d3a16f8f4fcea39f
For the reasoning for this change, please see the related bugs and:
http://logs.glob.uno/?c=mozilla%23layout&s=22+Feb+2017&e=22+Feb+2017#c27236
Mainly, before this change, there was nothing forcing style structs computed in
a style context to remain computed for the new style context after a call to
CalcStyleDifference. This can make us skip change hints when a style change
doesn't force to recompute one of these structs.
MozReview-Commit-ID: FoWbLjt97Uu
Signed-off-by: Emilio Cobos Álvarez <emilio@crisal.io>
--HG--
extra : rebase_source : 27772fc67b3c241ca4cef580112c949ad506b167
Since we use the presence of a change hint to determine whether we should
recreate a style context, we can't just strip them out on CalcStyleDifference.
MozReview-Commit-ID: GLhbTc2W3d7
Signed-off-by: Emilio Cobos Álvarez <emilio@crisal.io>
I think there are three advantages of this change:
1. removes some dependencies from layout / painting code to pre-computed
value stuff in the style system;
2. makes it easier to audit usage of specific fields in style structs
(which is probably a side effect of the first one);
3. potentially improves performance since it doesn't go through the
unnecessary general logic in ExtractComputedValue.
Also, combined with the part before, we get a unified list for visited-
dependent properties so that we can ensure the assertion here and the
style difference calc code are consistent.
MozReview-Commit-ID: 5B9aN7CfRgI
--HG--
extra : rebase_source : ac80eaea2474b9ec4b47b1cc9a5bdd2e61f6ec4d
This change avoids lots of false positives for Coverity's CHECKED_RETURN
warning, caused by NS_WARN_IF's current use in both statement-style and
expression-style.
In the case where the code within the NS_WARN_IF has side-effects, I made the
following change.
> NS_WARN_IF(NS_FAILED(FunctionWithSideEffects()));
> -->
> Unused << NS_WARN_IF(NS_FAILED(FunctionWithSideEffects()));
In the case where the code within the NS_WARN_IF lacks side-effects, I made the
following change.
> NS_WARN_IF(!condWithoutSideEffects);
> -->
> NS_WARNING_ASSERTION(condWithoutSideEffects, "msg");
This has two improvements.
- The condition is not evaluated in non-debug builds.
- The sense of the condition is inverted to the familiar "this condition should
be true" sense used in assertions.
A common variation on the side-effect-free case is the following.
> nsresult rv = Fn();
> NS_WARN_IF_(NS_FAILED(rv));
> -->
> DebugOnly<nsresult rv> = Fn();
> NS_WARNING_ASSERTION(NS_SUCCEEDED(rv), "Fn failed");
--HG--
extra : rebase_source : 58788245021096efa8372a9dc1d597a611d45611
It's a shame that assertion can be made a crashing one, but unfortunately it was
hitting me regularly.
MozReview-Commit-ID: 4iZf1rm1fO2
--HG--
extra : rebase_source : 0b9ef55e3036eeefe4b989c284b1aff3e99a906d
Structs in our style system use an arena-style allocation system,
managed by the presshell to which they belong. All of the relevant
overloads that forward allocation requests to the presshell declare
themselves as CPP_THROW_NEW, which indicates that they do not throw
exceptions. The C++ specification states that operator new overloads
that declare themselves to not throw exceptions require a null check on
their return value. However, the relevant presshell allocation method,
AllocateByObjectID, is infallible and will never return a null pointer.
The callers of all of these methods are therefore doing useless
(compiler-generated) null checks. Let's get rid of those useless checks
by removing the CPP_THROW_NEW annotations. This change declares these
methods will return non-null pointers and throw exceptions in case of
errors--but as we don't use exceptions, and AllocateByObjectID will
abort on OOM, everything works out OK.
Structs in our style system use an arena-style allocation system,
managed by the presshell to which they belong. All of the relevant
overloads that forward allocation requests to the presshell declare
themselves as CPP_THROW_NEW, which indicates that they do not throw
exceptions. The C++ specification states that operator new overloads
that declare themselves to not throw exceptions require a null check on
their return value. However, the relevant presshell allocation method,
AllocateByObjectID, is infallible and will never return a null pointer.
The callers of all of these methods are therefore doing useless
(compiler-generated) null checks. Let's get rid of those useless checks
by removing the CPP_THROW_NEW annotations. This change declares these
methods will return non-null pointers and throw exceptions in case of
errors--but as we don't use exceptions, and AllocateByObjectID will
abort on OOM, everything works out OK.
This patch is generated by the following commands (note: if you're running
using OS X's sed, which accepts slightly different flags, you'll have to
specify an actual backup suffix in -i, or use gsed from Homebrew):
hg stat -c \
| cut -c 3- \
| tr '\n' '\0' \
| xargs -0 -P 8 gsed --follow-symlinks 's/\bnsCSSProperty\b/nsCSSPropertyID/g' -i''
Then:
hg mv layout/style/nsCSSProperty.h layout/style/nsCSSPropertyID.h
... and finally, manually renaming nsCSSProperty in the include guard in
nsCSSProperty.h.
MozReview-Commit-ID: ZV6jyvmLfA
--HG--
rename : layout/style/nsCSSProperty.h => layout/style/nsCSSPropertyID.h
Otherwise, the parent style context doesn't have some inherited structs in the
cache, so it avoids comparing them entirely.
For example, the following example:
<p>Hey</p>
p { color: blue }
p:hover { color: red }
Wouldn't work as intended, because when calculating the change hint the
nsStyleColor struct in the element hasn't been accessed by layout (only the
child text frame's has), so it will report a change hint of 0 when hovering over
the paragraph, and we would ignore the nsChangeHint_ReconstructFrame generated
by the text node.
MozReview-Commit-ID: FW7Thhuh7LG
