Fix ridiculous DOM proxy getter performance
This implements the missing shadowing checks that were causing us to take many slow paths when dealing with proxy objects. Verified by running `tests/html/binding_perf.html` before and after and observing a 12x improvement.
---
- [X] `./mach build -d` does not report any errors
- [X] `./mach test-tidy` does not report any errors
- [X] These changes fix#12357
- [X] These changes do not require tests because we can't test performance yet.
<!-- 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/12980)
<!-- Reviewable:end -->
Support pair and value iterable WebIDL bindings
The actual iterator implementation and JSAPI calls related to setting up the interface are ported directly from Gecko's Codegen.py, IterableIterator.h, and IterableIterator.webidl. The changes to support multiple interfaces in one file are required because the internal iterator interface the parser generates gets associated with the original interface's WebIDL file. It seemed like a good time to address #571 in that case.
---
- [X] `./mach build -d` does not report any errors
- [X] `./mach test-tidy` does not report any errors
- [X] These changes fix#12628 and fix#571.
- [X] There are tests for these changes
<!-- 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/12819)
<!-- Reviewable:end -->
Each interface gets its own module named ${Interface}Binding. Structs, enums, and callbacks
continue to use the root module of the binding file. If there is only one interface in the
file, we generate reexports for several public APIs and types so that existing DOM implementations
don't need any modifications. When multiple interfaces exist, the reexported names get the interface
name prepended (eg. FooWrap instead of Wrap).
As part of this work, stop glob-importing all DOM types in every generated binding and start generating
more targeted lists of relevant types based on the methods, members, etc. of WebIDL types that are in use.
This commit adds new files related to implementing the [Request
API](https://fetch.spec.whatwg.org/#request-class). This commit also
changes the expected web platform tests results. It also modifies the
following files:
components/net_traits/request.rs
HeapSizeOf is implemented in net_traits/request so that dom::request can
be used as a wrapper around net_traits::request::Request.
components/script/dom/headers.rs
Several methods are added to Headers so that request can access and
modify some of the headers fields.
This currently breaks Servo on Android, because there are a number of
interdependent changes that cannot easily land serially in a way that
keeps it working throughout. We expect to fix this in the near future.
Union types now allow ByteString
<!-- Please describe your changes on the following line: -->
This PR addresses issue #12527 to allow ByteStrings in union types.
---
<!-- 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
- [X] These changes fix#12527 (github issue number if applicable).
<!-- Either: -->
- [X] There are tests for these changes OR
- [ ] These changes do not require tests because _____
<!-- 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/12546)
<!-- Reviewable:end -->
style: Rewrite the restyle hints code to allow different kinds of element snapshots.
<!-- 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 refactoring.
<!-- Pull requests that do not address these steps are welcome, but they will require additional verification as part of the review process. -->
This is a rewrite for how style interfaces with its consumers in order to allow
different representations for an element snapshot.
This also changes the requirements of an element snapshot, requiring them to
only implement MatchAttr, instead of MatchAttrGeneric. This is important for
stylo since implementing MatchAttrGeneric is way more difficult for us given the
atom limitations. This also allows for more performant implementations in the
Gecko side of things.
I don't want to get this merged just yet, mainly because the stylo part is not
implemented, but I'd like early feedback from @bholley and/or @heycam: How do
you see this approach? I don't think we'll have much problem to implement
MatchAttr for our element snapshots, but... worth checking.
r? @heycam
<!-- 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/12469)
<!-- Reviewable:end -->
This is a rewrite for how style interfaces with its consumers in order to allow
different representations for an element snapshot.
This also changes the requirements of an element snapshot, requiring them to
only implement MatchAttr, instead of MatchAttrGeneric. This is important for
stylo since implementing MatchAttrGeneric is way more difficult for us given the
atom limitations. This also allows for more performant implementations in the
Gecko side of things.
Support sequences of sequences in generated bindings.
This fixes a blocker for #11897. `unroll` recursively gets the inner type of any sequence type encountered, so it's inappropriate for codegen that only wants the immediate inner type. However, if a type identifies as a sequence and is nullable, we need to reach through the nullable wrapper first. Gecko does very similar things.
---
- [X] `./mach build -d` does not report any errors
- [X] `./mach test-tidy` does not report any errors
- [X] These changes fix#12528 (github issue number if applicable).
- [X] There are tests for these changes
<!-- 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/12541)
<!-- Reviewable:end -->
unroll recursively gets the inner type of any sequence type encountered, so it's inappropriate for codegen that only wants the immediate inner type. However, if a type identifies as a sequence and is nullable, we need to reach through the nullable wrapper first. Gecko does very similar things.
