[webkit-reviews] review denied: [Bug 71152] selectors should match attribute name with case sensitivity based on element & document type : [Attachment 207875] Patch

[bugzilla-daemon at webkit.org]

Darin Adler <darin at apple.com> has denied Rob Buis <rwlbuis at gmail.com>'s request
for review:
Bug 71152: selectors should match attribute name with case sensitivity based on
element & document type
https://bugs.webkit.org/show_bug.cgi?id=71152

Attachment 207875: Patch
https://bugs.webkit.org/attachment.cgi?id=207875&action=review

------- Additional Comments from Darin Adler <darin at apple.com>
View in context: https://bugs.webkit.org/attachment.cgi?id=207875&action=review


I’m pretty sure that there is a problem sharing m_localNameUpperLower. The same
QualifiedNameImpl can be used in a context where localNameUpper is called on it
and another where localNameLower needs to be called on it. Not the same
qualified name, but the same QualifiedNameImpl from the global table.

> Source/WebCore/dom/QualifiedName.h:91
> +    // Uppercased/lowercased localName, cached for efficiency

I wish this made it clear that you can only call one or the other, not both, on
any given QualifiedName.

I still don’t understand how we can guarantee that. If the two qualified names
are identical, won’t we share the same underlying QualifiedNameImpl? So can’t
we use the equal qualified name in two separate contexts and make the
assertions fire!?!