Use typed coordinates more
Requires https://github.com/servo/servo/pull/19895
We use Size2D and Point2D across compositing, constellation and script, loosing the type of pixels we use (DevicePixel, DeviceIndepententPixel or CSSPixel) along the way, which might lead to bugs like `window.outerHeight` not taking into account the page zoom (using DeviceIndepententPixel instead of CSSPixel).
This should make the situation a bit better.
---
<!-- 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
- [ ] These changes fix #__ (github issue number if applicable).
<!-- Either: -->
- [ ] There are tests for these changes OR
- [x] These changes do not require tests because we can't zoom in a test
<!-- 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/20071)
<!-- Reviewable:end -->
We use Size2D and Point2D across compositing, constellation and script,
losing the type of pixels we use (DevicePixel, DeviceIndepententPixel
or CSSPixel) along the way, which might lead to bugs like
window.outerHeight not taking into account the page zoom (using
DeviceIndepententPixel instead of CSSPixel).
This fixes a regression caused by the glutin update.
We now are creating EGL contexts in Linux Wayland, instead of X context, so the
GLContextFactory assumption of one GL back-end per platform is broken.
This just works around it, for now, but in general I think not relying on
available WebGL state is a good thing, and we do that already for WebVR anyway.
WR now has a concept of font templates and font instances. This
makes the WR font interfaces closer to Cairo and Gecko, and also
makes some future performance optimizations possible.
A font template is the font family, and data backing the font.
A font instance is a reference to a font template and per-instance
options, such as font size, anti-aliasing settings etc.
To update Servo in a minimally invasive way, I added a new font
cache call, that creates a font instance. This means that when
a font is created, and doesn't exist in the cache there are now
two calls to the font cache thread. We could refactor the font
cache to make this work in one call, which we should do in the
future. However, refactoring the font cache is a large chunk of
work by itself. The extra call is only when a font doesn't already
exist in the font context cache, so it should have minimal
performance impact.
cleanup embedder/compositor/constellation/script messages
Fix: #17226#17200#17201
This is work in progress. Some tests still fail.
I'd like to get early feedback as it's a pretty large PR.
There is nothing fundamentally new. Basically, I added TopLevelBrowsingContrextId to the relevant messages between the embedder, the compositor and the constellation, and enforced the PipelineId to be attached to each ScriptMsg (see #17201).
I unaliased all the ScriptMsg. It was getting difficult to understand the nature of the message as ScriptMsg was used aliased CompositorMsg sometimes (CompositorMsg is an actually type of message already). I renamed constellation_chan to script_to_constellation_chan, again, for clarification.
This cleanup code is necessary for #15934 and for tabs support.
/cc @asajeffrey can I ask you to look at this? No need for a formal review, I need feedback at this stage.
<!-- 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/17425)
<!-- Reviewable:end -->
remove the ability to request for the title
If we offer the ability to request info like these, we need to expose a lot of other getters, like url, loading state, favicon, …
It's up to the embedder to keep track of the state of a browser. So let's keep that consistent.
<!-- 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
- [ ] These changes fix #__ (github issue number if applicable).
<!-- Either: -->
- [ ] There are tests for these changes OR
- [ ] These changes do not require tests because _____
<!-- 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/17686)
<!-- Reviewable:end -->
Use Option instead of ServoUrl for url in opts
Cleaner mapping from parse_url to url_opt
Add comment about url parsing error
Print reason for parsing error
Remove comment about warning
Add home url when openning new browser window in CEF
Fix merge error
In support of this goal, the layout thread collects information about
CSS images that are missing image data and hands it off to the script
thread after layout completes. The script thread stores a list of
nodes that will need to be reflowed after the associated network
request is complete. The script thread ensures that the nodes are
not GCed while a request is ongoing, which the layout thread is
incapable of guaranteeing.
The image cache's API has also been redesigned in support of this
work. No network requests are made by the new image cache, since it
does not possess the document-specific information necessary to
initiate them. Instead, there is now a single, synchronous
query operation that optionally reserves a slot when a cache
entry for a URL cannot be found. This reserved slot is then
the responsibility of the queryer to populate with the contents
of the network response for the URL once it is complete. Any
subsequent queries for the same URL will be informed that the
response is pending until that occurs.
The changes to layout also remove the synchronous image loading
code path, which means that reftests now test the same code
that non-test binaries execute. The decision to take a screenshot
now considers whether there are any outstanding image
requests for layout in order to avoid intermittent failures in
reftests that use CSS images.
