It is bogus, because it depends on the display property as it's cascaded, but
the display property can change afterwards, for example, if we get blockified
because we're the root element or a flex item.
Replace it with a normal field instead.
Also, it carries some weight, because it's the last property that uses this
concept of "derived" property, and "custom cascade". So we can remove some code
after this.
Compute it after the cascade process in StyleAdjuster.
style: Remove -servo-display-for-hypothetical-box from longhand
This is a sub-PR of #19015
r? emilio
For the `fn set_original_display` inside `properties.mako.rs`, I removed `is_item_or_root` first to see how the tests result is. If it's needed, I'll add it back.
---
- [x] `./mach build -d` does not report any errors
- [x] `./mach test-tidy` does not report any errors
- [x] These changes fix#19697
- [x] These changes do not require tests
<!-- 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/19709)
<!-- Reviewable:end -->
style: Adjust the writing-mode fixup to call set_adjusted_display on Servo.
This is functionally equivalent right now (set_adjusted_display will just do
set_display), but won't be after #19709.
<!-- 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/19758)
<!-- Reviewable:end -->
This reduces time spent in NoteDirtyElement by about 15% in the testcase. Even
though the subsequent patch will cause us to call Servo_Element_IsDisplayNone less
for this particular testcase, it could still be hot on other testcases, and so
it's worth optimizing.
MozReview-Commit-ID: 3F3Zfp48dDW
style: Use Python's True instead of string for arbitrary Python code in mako
As discussion in IRC with @nox and @gootorov , it's more correct to use Python's `True` for `${arbitrary Python code}` in mako.
Following is related IRC logs
```
15:39 nox
As I said, that's because there are 2 ways to invoke the things with Mako.
15:40 nox
${arbitrary Python code} vs <%named_helper ..>...</%named_helper>
15:40 nox
IIRC <%foo bar=True></%foo> is not valid Mako syntax,
15:40 nox
but with bar="True" it is.
15:42 cybai
gootorov: I did meet some issues when moving makos if I set "True" instead of Python's `True`. Thus, like nox mentioned above, it's related to how the code works
15:44 gootorov
nox: Alright, I got it now, thanks. There are a couple ${function} blocks that still use `boxed="True" though. They aren't <%named_helpers>, shouldn't they use Python's True too, then?
15:45 nox
gootorov: I think there is an issue because this specific attribute is also used as a string through other <%helpers>,
15:46 nox
gootorov: Seems like it's not an issue anymore because the boxed attribute is passed to arg_to_bool,
15:46 nox
which either returns the boolean as is or check whether it's a "True" or "False" string.
15:47 nox
gootorov: So indeed, you can probably make those use booleans now.
15:47 gootorov
nox: Okay, got it, thanks for the explanation :)
```
---
- [x] `./mach build -d` does not report any errors
- [x] `./mach test-tidy` does not report any errors
- [x] These changes do not require tests
<!-- 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/19752)
<!-- Reviewable:end -->
Update WR (new transaction API).
In the future we'll want to start making use of this in Servo
to get atomic updates of resources with display list changes etc.
<!-- 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/19750)
<!-- Reviewable:end -->
style: Add a document state invalidator.
Don't use it yet (since I was working from a Servo tree). Will hook it up and improve in the Gecko bug.
Right now it takes a `StyleRuleCascadeData`, which means that if all the origins in the document have state selectors we could do just one walk over the tree and not multiple, that will be improved.
Other than that, this is completely untested of course, but I prefer to land it, given I don't think it's complex, and work on the Gecko integration separately. The reason for this is that I also plan to fix the `<slot>` bugs, which will change `StyleRuleCascadeData` and such, and I want the two bugs to conflict as little as possible.
<!-- 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/19747)
<!-- Reviewable:end -->
Allow warnings during mutation tests.
Sometimes mutations include removing the only use of a variable. When we deny warnings globally, we inhibit these mutations that trigger new warnings.
<!-- 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/19740)
<!-- Reviewable:end -->
