[webkit-reviews] review granted: [Bug 183523] ClassExprNode evaluation of computed property names differs observably from spec and other implementations : [Attachment 335609] Patch

[bugzilla-daemon at webkit.org]

Keith Miller <keith_miller at apple.com> has granted Caitlin Potter (:caitp)
<caitp at igalia.com>'s request for review:
Bug 183523: ClassExprNode evaluation of computed property names differs
observably from spec and other implementations
https://bugs.webkit.org/show_bug.cgi?id=183523

Attachment 335609: Patch

https://bugs.webkit.org/attachment.cgi?id=335609&action=review




--- Comment #2 from Keith Miller <keith_miller at apple.com> ---
Comment on attachment 335609
  --> https://bugs.webkit.org/attachment.cgi?id=335609
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=335609&action=review

r=me with some nits. 

Also, does this fix any test262 tests? I'm cool with doing some gardening after
this patch lands too.

> Source/JavaScriptCore/parser/Nodes.h:728
> +	   unsigned m_classElementTag: 2;

Can you add:

static_assert(1 << 2 > ClassElementTag::LastTag, "ClassElementTag shouldn't use
more than two bits"); 

You'll also need to add a LastTag value to the ClassElementTag enum too.

> Source/JavaScriptCore/parser/Nodes.h:742
> +	   RegisterID* emitBytecode(BytecodeGenerator& generator, RegisterID*
dst = 0) override

Nit: This should be: RegisterID* dst = nullptr