[concurrency-interest] ConcurrentHashMapV8

Kasper Nielsen kasper at kav.dk
Thu Sep 1 16:35:39 EDT 2011


On 01-09-2011 21:07, Mike Duigou wrote:
> On Aug 28 2011, at 19:53 , concurrency-interest-request at cs.oswego.edu wrote:
>>
>>   4. Re: ConcurrentHashMapV8 (Joe Bowbeer)
>
>> From: Joe Bowbeer<joe.bowbeer at gmail.com>
>> Subject: Re: [concurrency-interest] ConcurrentHashMapV8
>> First impressions:
>>
>> 1. I like the MappingFunction name.  I hope the JDK8 name is as clear *and*
>> specialized.
>
>
> The current proposed name for Java SE 8 is 'Mapper' and it has the same signature as Doug's placeholder.
>
> http://hg.openjdk.java.net/lambda/lambda/jdk/file/38969e64b60e/src/share/classes/java/util/functions/Mapper.java
>
> One key difference is that the proposed Java SE 8 Mapper does not say anything about null handling and it is anticipated that Mapper will be used in places where null must be accommodated. The CHM may choose to just use the Mapper and document it's handling of null or could alternately create a specialized sub-interface to document the extended semantics.
>
> If there are any concerns with the current proposed Mapper or the other basic functional interfaces proposed for Java SE 8 we'd be very interested to hear them on the lambda-dev at openjdk.java.net list.
>
> Thanks!
>
> Mike

Thanks for bringing this up. I must admit I really don't like the name 
Block. I might be damage by working to much with concurrency and IO but 
I really thought it had something to do with blocking/non-blocking.

I really like the name Procedure as used in Doug's ParallelArray classes 
much better. 
(http://gee.cs.oswego.edu/cgi-bin/viewcvs.cgi/jsr166/src/extra166y/Ops.java?revision=1.3&view=markup)

cheers
   Kasper




More information about the Concurrency-interest mailing list