RFR (2nd): 8015318: Extend Collector with 'finish' operation

[Mike Duigou]

Looks OK. A few minor non-blocking comments:

DelegatingStream::

- I wonder if an extends wildcard might also be included on R

ReduceOps::

- It is not clear in makeRef why collector.accumulator() wouldn't also benefit from a Objects.requireNonNull

General::

- Is there a constant immutable EnumSet for the Collector.Characteristics.IDENTITY_FINISH ? It would be nice to avoid creating lots of EnumSet instances.

On Aug 5 2013, at 22:40 , Henry Jen wrote:

> Ping on this, the 8012238 fix is already in, thus this webrev is ready
> to go, only if we can have a reviewer review it.
> 
> Cheers,
> Henry
> 
> 
> On 07/09/2013 02:38 PM, Henry Jen wrote:
>> Hi,
>> 
>> An update from earlier request based on some feedbacks.
>> 
>> The webrev is at
>> 
>> http://cr.openjdk.java.net/~henryjen/ccc/8015318.1/webrev
>> 
>> This webrev depends on fix of bug 8012238[1]. Before that bug is fixed,
>> applies this patch[2] to add needed type information.
>> 
>> See the summary of API changes, specdiff is at
>> 
>> http://cr.openjdk.java.net/~henryjen/ccc/8015318.1/specdiff/overview-summary.html
>> 
>> 
>> [1] http://bugs.sun.com/view_bug.do?bug_id=8012238
>> [2] http://cr.openjdk.java.net/~henryjen/ccc/8015318.1/collector-hacks.patch
>> 
>> Cheers,
>> Henry
>> 
>