Bug 295694 – Changing ruby-overhang values does not trigger layout

• Status: RESOLVED
• Resolution: FIXED
• Component: Layout and Rendering

The `ruby-overhang` property can have values of `none` or `auto`.  Switching between these values requires a DOM re-render or other recalculations for it to be applied.

Steps to reproduce:

1. Navigate to https://developer.mozilla.org/en-US/docs/Web/CSS/ruby-overhang#try_it
2. Switch between the options `ruby-overhang: auto;` and `ruby-overhang: none;` in the interactive example

The changes are not applied, but if I use the following declarations in the interactive example (the options are editable in the "Try it" section):

```
ruby-overhang: auto;
ruby-position: over;
```

And:

```
ruby-overhang: none;
ruby-position: under;
```

Changing the ruby-overhang and the ruby-position properties applies the styles as expected.

[Radar WebKit Bug Importer]

<rdar://problem/156038239>

[fantasai]

I believe the problem is in rareInheritedDataChangesRequiresLayout()
https://searchfox.org/wubkat/source/Source/WebCore/rendering/style/RenderStyle.cpp#925

We need to return true if ruby-overhang changes, just like we do for ruby-position (see the line reading `first.rubyPosition != second.rubyPosition`; we need to add one for `first.rubyOverhang != second.rubyOverhang`).

Brian, if you want to take a stab at submitting a WPT test for this, I can do the code changes for you so you don't have to set up a development environment. Alternatively if you want to try submitting a patch yourself, this seems like a good first bug. :)

The WPT repo is at
https://github.com/web-platform-tests/wpt
Probably the easiest thing would be to search the /css directory for a few tests that use the `reftest-wait` class and copy how they're written. You need to trigger layout with the old value, then change it, then trigger layout with the new value, and then remove the reftest-wait.

Thanks a lot for the hints. I was looking at creating a `reftest-wait` test for this, but I'm a bit lost, so I'd prefer to leave that to someone else who'd be faster at this if that's okay! I appreciate the encouragement, though.

[alan]

Created attachment 476496 [details]
Patch

[Tim Nguyen (:ntim)]

Comment on attachment 476496 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=476496&action=review

> COMMIT_MESSAGE:1
> +changing ruby-overhang values does not trigger style recalculation 

Changing ruby-overhang values does not trigger layout

is probably more accurate

> LayoutTests/imported/w3c/web-platform-tests/css/css-ruby/ruby-overhang-dynamic-ref.html:10
> +    font-size: 8px; /* annotation -> 50% */

(here and the -expected.html): you could use 50% directly unless there's something preventing that :)

[alan]

Created attachment 476497 [details]
[fast-cq]Patch

[EWS]

Committed 299091@main (33300160d33d): <https://commits.webkit.org/299091@main>

All reviewed patches have been landed. Closing bug and clearing flags on attachment 476497 [details].

[Tim Nguyen (:ntim)]

Submitted web-platform-tests pull request: https://github.com/web-platform-tests/wpt/pull/54489