[Webkit-unassigned] [Bug 162744] Mutating styleSheet in shadow tree doesn't update the style

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=162744

--- Comment #15 from Ryosuke Niwa <rniwa at webkit.org> ---
Comment on attachment 290846
  --> https://bugs.webkit.org/attachment.cgi?id=290846
patch

View in context: https://bugs.webkit.org/attachment.cgi?id=290846&action=review

It looks like two test cases are failing on WK1?

> Source/WebCore/css/CSSStyleSheet.cpp:173
> +    auto* scope = styleSheetScope();

I'm not sure scope is the most descriptive name of AuthorStyleSheets objects but perhaps you're planning to rename it?

> Source/WebCore/css/CSSStyleSheet.cpp:-205
> -    owner->authorStyleSheets().didChangeCandidatesForActiveSet();

Could you clarify in the change log why it's okay to remove this call?

> Source/WebCore/css/CSSStyleSheet.cpp:410
> +{
> +    return const_cast<CSSStyleSheet&>(*this).rootStyleSheet();
> +}

Maybe we can just inline this in the header?

> Source/WebCore/dom/AuthorStyleSheets.cpp:88
> +AuthorStyleSheets& AuthorStyleSheets::forNode(Node& node)

It's pretty neat that we just needed three functions to abstract Document/ShadowRoot difference!

> Source/WebCore/dom/AuthorStyleSheets.cpp:337
> +    if (m_shadowRoot)
> +        updateType = UpdateType::ContentsOrInterpretation;
> +

Why do we want to override this? It seems that we do want to analyzeStyleSheetChange even inside a shadow tree?
Please clarify that in the change log.

> Source/WebCore/html/HTMLLinkElement.cpp:-333
> -
> -    if (document().hasLivingRenderTree())
> -        document().authorStyleSheets().didChangeCandidatesForActiveSet();

Could you clarify why it's okay to remove this in the change log?

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.webkit.org/pipermail/webkit-unassigned/attachments/20161006/7aef7d5e/attachment-0001.html>