Bug 229542 – [CSS Cascade Layers] Initial support

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

[Antti Koivisto]

Implement initial support

[Antti Koivisto]

Created attachment 436483 [details]
patch

[Antti Koivisto]

Created attachment 436491 [details]
patch

[Simon Fraser (smfr)]

Comment on attachment 436491 [details]
patch

View in context: https://bugs.webkit.org/attachment.cgi?id=436491&action=review

> Source/WebCore/css/StyleRule.h:265
> +    Variant<CascadeLayerName, Vector<CascadeLayerName>> m_nameVariant;

Is this preferable to just always using a Vector<>? Or is a name list with one item different from a simple name?

> Source/WebCore/css/StyleRuleType.h:40
>      Namespace = 10,

What happened to 9? Why do we number these from here on?

> Source/WebCore/css/parser/CSSAtRuleID.h:51
> +    CSSAtRuleLayer = 13,

Not sure why we need exploit values here?

> Source/WebCore/style/RuleSet.cpp:510
> +            // FIXME: This is not correct when adding a sublayer to an already registered layer after it has gained siblings.

Reference a bug url?

> Source/WebCore/style/RuleSet.h:196
> +    // FIXME: These should be in stack.

in a stack? on the stack?

[Antti Koivisto]

> Is this preferable to just always using a Vector<>? Or is a name list with
> one item different from a simple name?

One for the statement case, another for the block case. See the different constructor variants. I just used Vector initially but this ended up reading better. Otherwise I also need a separate bool to differentiate between the cases.

> What happened to 9? Why do we number these from here on?

 Needs some historical digging.

> Not sure why we need exploit values here?

We don't, I was just sticking with the existing style.

> Reference a bug url?
> in a stack? on the stack?

I'll be fixing both shortly.

[EWS]

Committed r281701 (241051@main): <https://commits.webkit.org/241051@main>

All reviewed patches have been landed. Closing bug and clearing flags on attachment 436491 [details].

[Radar WebKit Bug Importer]

<rdar://problem/82445409>