RFR: 8167289 : Backport ES6 updates from Graal.js

[Attila Szegedi]

And the patch is now much easier to review! Anyway, apologies for having you make another roundtrip.

That said, I’d prefer if subexpressions (LiteralNode.ArrayLiteralNode)literalNode and literalNode.getElementExpressions() were assigned to some shortish-named locals and used as such to improve the readability of this expression: 

+                if (((LiteralNode.ArrayLiteralNode)literalNode).hasSpread() && ((LiteralNode.ArrayLiteralNode)literalNode).hasTrailingComma()) {
+                    throw error("Rest element must be last", literalNode.getElementExpressions().get(literalNode.getElementExpressions().size() - 1).getToken());
+                }

+1 otherwise. (I’m also fine if you decide to not make this change; it’s admittedly no big deal.)

Attila.


> On 07 Oct 2016, at 10:25, Hannes Wallnöfer <hannes.wallnoefer at oracle.com> wrote:
> 
> This change conflicts with Attila’s add-finals change in Parser.java. I uploaded a new webrev with the merged patch. I double-checked that it keeps the finals and that the code I added uses final too.
> 
> http://cr.openjdk.java.net/~hannesw/8167289/webrev.01/
> 
> Hannes
> 
> 
>> Am 07.10.2016 um 05:27 schrieb Sundararajan Athijegannathan <sundararajan.athijegannathan at oracle.com>:
>> 
>> +1
>> 
>> -Sundar
>> 
>> 
>> On 10/6/2016 8:21 PM, Hannes Wallnöfer wrote:
>>> Please review:
>>> 
>>> Bug: https://bugs.openjdk.java.net/browse/JDK-8167289
>>> Webrev: http://cr.openjdk.java.net/~hannesw/8167289/webrev.00/
>>> 
>>> These is mostly refactoring Andreas did in the Graal.js parser. It also contains minor ES6 parsing fixes. Tests for these will be included together with other additional ES6 tests I’m working on, but I wanted to do a separate commit for the backport, so this issue is labeled noreg-cleanup.
>>> 
>>> Thanks,
>>> Hannes
>> 
>