Bug 228730 – Replace webkit- prefix logical properties with Standard Properties in LayoutTests/

• Status: RESOLVED
• Resolution: FIXED
• Component: CSS

[Sonia Singla]

Similar to Bug 228697,  we should replace webkit- prefix properties with Standard Properties in Layout/

[Radar WebKit Bug Importer]

<rdar://problem/81729542>

[Sonia Singla]

Created attachment 435558 [details]
Patch

[Sonia Singla]

Created attachment 435559 [details]
Patch

[Frédéric Wang (:fredw)]

Comment on attachment 435559 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=435559&action=review

Thanks Sonia. It looks that you are close to get EWS green. Some comments inline...

> LayoutTests/ChangeLog:144
> +        * platform/gtk/imported/w3c/web-platform-tests/css/cssom/cssstyledeclaration-csstext-expected.txt:

As a reminder, you should not modify any test in LayoutTests/imported/w3c/web-platform-tests/ (these have not been sync). Since you don't modify any code here in this patch, there is no need to update the expectation.

> LayoutTests/fast/css/remove-shorthand-expected.txt:57
> +removes "margin-block-start-collapse, margin-block-end-collapse"

You didn't modify remove-shorthand.html in this patch.

It looks like this test is checking what are the longhand properties removed when removing the shorthand -webkit-margin-collapse.

> LayoutTests/platform/glib/svg/css/getComputedStyle-basic-expected.txt:468
> +rect: style.getPropertyCSSValue(margin-block-start-collapse) : [object CSSPrimitiveValue]

For all these getComputedStyle-basic-expected.txt, you modified the expectation but not the test LayoutTests/svg/css/getComputedStyle-basic.xhtml itself.

I believe we want to continue to test the -webkit prefixed properties here, until they are actually removed. What you should do instead is updating Layout/fast/css/getComputedStyle/resources/property-names.js to add the unprefixed forms. Please check all the tests that are relying on this helper JS file and run locally before uploading a new patch:

	LayoutTests/fast/css/getComputedStyle/computed-style-without-renderer.html
	LayoutTests/fast/css/getComputedStyle/computed-style.html
	LayoutTests/svg/css/getComputedStyle-basic.xhtml

> LayoutTests/platform/ios/fast/css/getComputedStyle/computed-style-expected.txt:236
> +margin-block-start-collapse: collapse;

For these computed-style-expected.txt files, see my comment about property-names.js

> LayoutTests/platform/ios/fast/css/getComputedStyle/computed-style-without-renderer-expected.txt:235
> +margin-block-start-collapse: collapse

For these computed-style-without-renderer-expected.txt files, see my comment about property-names.js

[Sonia Singla]

Created attachment 436106 [details]
Patch

[Frédéric Wang (:fredw)]

Comment on attachment 436106 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=436106&action=review

LGTM, but probably wait that Oriol or Rego are back to comment on this.

> LayoutTests/ChangeLog:4
> +        https://bugs.webkit.org/show_bug.cgi?id=228730

I think it would be nice to have a short statement explaining what we are doing in this patch. In particular, the details of the old/new property names.

[Sonia Singla]

Created attachment 436290 [details]
Patch

[Manuel Rego Casasnovas]

Comment on attachment 436290 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=436290&action=review

Nice work, thanks for the patch. Just some comments inline.

> LayoutTests/ChangeLog:32
> +        Reviewed by NOBODY (OOPS!).

Nit: This line is usually before the description of the change.

> LayoutTests/ChangeLog:56
> +        * fast/css/border-start-end.html:

This test is specifically testing the -webkit- prefixed property, so it doesn't make sense to change it. Please keep it as it is.

There are more like this: margin-start-end.html, padding-start-end.html.

[Sonia Singla]

Created attachment 437013 [details]
Patch

[EWS]

Committed r281908 (241220@main): <https://commits.webkit.org/241220@main>

All reviewed patches have been landed. Closing bug and clearing flags on attachment 437013 [details].