Bug 164886 – [CSS Parser] Hook up InspectorStyleSheet to the new CSS parser.

• Status: RESOLVED
• Resolution: FIXED
• Component: New Bugs

[Dave Hyatt]

[CSS Parser] Hook up InspectorStyleSheet to the new CSS parser.

[Dave Hyatt]

Created attachment 295087 [details]
Patch

[Dean Jackson]

Comment on attachment 295087 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=295087&action=review

> Source/WebCore/inspector/InspectorStyleSheet.cpp:199
> +    m_currentRuleDataStack.last()->ruleHeaderRange.end = listEndOffset;
> +    if (!m_currentRuleDataStack.last()->selectorRanges.isEmpty())
> +        m_currentRuleDataStack.last()->selectorRanges.last().end = listEndOffset;

Can m_currentRuleDataStack ever be empty?

> Source/WebCore/inspector/InspectorStyleSheet.cpp:225
> +    // Skip the rule body opening brace.
> +    if (m_parsedText[offset] == '{')
> +        ++offset;

Should we skip any spaces as well?

> Source/WebCore/inspector/InspectorStyleSheet.cpp:371
> +    bool parsedOk = propertyData.parsedOk || propertyData.name.startsWith("-moz-") || propertyData.name.startsWith("-o-") || propertyData.name.startsWith("-webkit-") || propertyData.name.startsWith("-ms-");

I don't understand this. All -moz- properties are ok? Shouldn't they be marked as not OK?

> Source/WebCore/inspector/InspectorStyleSheet.cpp:378
> +    m_currentRuleDataStack.last()->styleSourceData->propertyData.append(CSSPropertySourceData(propertyData.name, propertyData.value, false, true, true, SourceRange(startOffset - topRuleBodyRange.start, endOffset - topRuleBodyRange.start)));

Is this the place where we set disabled to true?

[Dave Hyatt]

Fixed in r208886.