<html>
    <head>
      <base href="https://bugs.webkit.org/" />
    </head>
    <body>
      <p>
        <div>
            <b><a class="bz_bug_link 
          bz_status_NEW "
   title="NEW - [ES6] Arrow function syntax. Arrow function specific features. Lexical bind &quot;super&quot; property"
   href="https://bugs.webkit.org/show_bug.cgi?id=149615#c6">Comment # 6</a>
              on <a class="bz_bug_link 
          bz_status_NEW "
   title="NEW - [ES6] Arrow function syntax. Arrow function specific features. Lexical bind &quot;super&quot; property"
   href="https://bugs.webkit.org/show_bug.cgi?id=149615">bug 149615</a>
              from <span class="vcard"><a class="email" href="mailto:sbarati&#64;apple.com" title="Saam Barati &lt;sbarati&#64;apple.com&gt;"> <span class="fn">Saam Barati</span></a>
</span></b>
        <pre>Comment on <span class=""><a href="attachment.cgi?id=267103&amp;action=diff" name="attach_267103" title="Patch">attachment 267103</a> <a href="attachment.cgi?id=267103&amp;action=edit" title="Patch">[details]</a></span>
Patch

View in context: <a href="https://bugs.webkit.org/attachment.cgi?id=267103&amp;action=review">https://bugs.webkit.org/attachment.cgi?id=267103&amp;action=review</a>

Mostly looks good to me but I have a question and some basic comments.

<span class="quote">&gt; Source/JavaScriptCore/ChangeLog:9
&gt; +        inside of the arrow function in case if arrow function is nested in method, </span >

Might be worth explicitly adding constructor to this list of available contexts inside a class.

<span class="quote">&gt; Source/JavaScriptCore/ChangeLog:11
&gt; +        class, lead to wrong type of error, should be SyntaxError and this will be fixed in separete patch.</span >

Do you have a bug open for this? If you don't, you should make one.
You should link to it here

<span class="quote">&gt; Source/JavaScriptCore/bytecode/ExecutableInfo.h:39
&gt; +    ExecutableInfo(bool needsActivation, bool usesEval, bool isStrictMode, bool isConstructor, bool isBuiltinFunction, ConstructorKind constructorKind, GeneratorThisMode generatorThisMode, SuperBinding superBinding, SourceParseMode parseMode, DerivedContextType _derivedContextType, bool _isArrowFunctionContext, bool isClassContext)</span >

style: Remove the &quot;_&quot; prefixes.

<span class="quote">&gt; Source/JavaScriptCore/bytecode/UnlinkedFunctionExecutable.cpp:70
&gt; +    // method, getter &amp; setter are part of the class, so this functionCodeBlock is part of the class
&gt; +    bool isClassContext = executable-&gt;parseMode() == SourceParseMode::MethodMode || executable-&gt;parseMode() == SourceParseMode::GetterMode || executable-&gt;parseMode() == SourceParseMode::SetterMode;</span >

Is this always correct?
We could have code like this:
```
let x = {
    get y() { return 20; }
};
x.y
```
I'm not sure if this would count as SourceParseMode::GetterMode.

<span class="quote">&gt; Source/JavaScriptCore/bytecompiler/BytecodeGenerator.cpp:578
&gt; +</span >

style: revert newline.

<span class="quote">&gt; Source/JavaScriptCore/bytecompiler/BytecodeGenerator.h:799
&gt; +            bool newisDerivedConstructorContext = constructorKind() == ConstructorKind::Derived || (derivedContextType() == DerivedContextType::DerivedConstructorContext &amp;&amp; metadata-&gt;parseMode() == SourceParseMode::ArrowFunctionMode);</span >

This calculation doesn't seem right to me. Shouldn't we be interested in the fact that we're creating an arrow function or not?

<span class="quote">&gt; Source/JavaScriptCore/bytecompiler/BytecodeGenerator.h:806
&gt; +                bool isArrowFunctionInClassContext = m_codeBlock-&gt;isClassContext() || (m_codeBlock-&gt;isArrowFunction() &amp;&amp; m_derivedContextType == DerivedContextType::DerivedMethodContext);
&gt; +                derivedContextType = isArrowFunctionInClassContext ? DerivedContextType::DerivedMethodContext : DerivedContextType::None;</span >

same here.

<span class="quote">&gt; Source/JavaScriptCore/bytecompiler/NodesCodegen.cpp:189
&gt; +</span >

style: revert newline.

<span class="quote">&gt; Source/JavaScriptCore/tests/stress/arrowfunction-lexical-bind-superproperty.js:1
&gt; +var testCase = function (actual, expected, message) {</span >

Style: 4-space indent this file.

<span class="quote">&gt; Source/JavaScriptCore/tests/stress/arrowfunction-lexical-bind-superproperty.js:158
&gt; +// Fixme: should by check if e instanceof SyntaxError <a class="bz_bug_link 
          bz_status_NEW "
   title="NEW - [ES6] Arrow function syntax. Using 'super' in arrow function that declared out of the class should lead to Syntax error"
   href="show_bug.cgi?id=150893">https://bugs.webkit.org/show_bug.cgi?id=150893</a></span >

Style: &quot;Fixme&quot; =&gt; &quot;FIXME&quot; and also indented properly.

<span class="quote">&gt; Source/JavaScriptCore/tests/stress/arrowfunction-lexical-bind-superproperty.js:215
&gt; +// FIXME: Problem with access to the super before super in constructor
&gt; +// <a class="bz_bug_link 
          bz_status_NEW "
   title="NEW - ES6 Class syntax. Invoking method of parent class in constructor before super() lead to crash"
   href="show_bug.cgi?id=152108">https://bugs.webkit.org/show_bug.cgi?id=152108</a>
&gt; +//  let j2 = new J(true);
&gt; +//  testCase(j2._value, testValue, 'Error: Some problem with using &quot;super&quot; inside of the constructor');</span >

Style: Indent this comment and all others to be at the level of block they're in.

<span class="quote">&gt; LayoutTests/js/script-tests/arrowfunction-superproperty.js:1
&gt; +description('Tests for ES6 arrow function, access to the super property in arrow function');</span >

Style: 4-space indent this file.

<span class="quote">&gt; LayoutTests/js/script-tests/arrowfunction-superproperty.js:80
&gt; +//shouldThrow('(new C(false))', 'ReferenceError');</span >

This should be &quot;new C(true)&quot;</pre>
        </div>
      </p>
      <hr>
      <span>You are receiving this mail because:</span>
      
      <ul>
          <li>You are the assignee for the bug.</li>
      </ul>
    </body>
</html>