[concurrency-interest] ConcurrentNavigableMap additional methods: 6415641

Jason Mehrens jason_mehrens at hotmail.com
Mon May 1 10:59:43 EDT 2006

Thanks Josh and Doug for the congrats and the API review.

I haven't received the machine yet.  It would be funny if it arrives the 
week of Java ONE.

Process.waitFor has the same issue (in 5.0 and 6.0) but the bug report 
(503840) I filled for it wasn't accepted. I find that funny since the 
SynchronousQueue bug was a winner.



>From: "Joshua Bloch" <josh at bloch.us>
>To: "Doug Lea" <dl at cs.oswego.edu>
>CC: "Jason Mehrens" <jason_mehrens at hotmail.com>, 
>concurrency-interest at cs.oswego.edu
>Subject: Re: [concurrency-interest] ConcurrentNavigableMap additional 
>methods: 6415641
>Date: Sun, 30 Apr 2006 23:35:03 -0700
>Congrats Jason!  Did you get the machine?  Way cool.
>That said, I'm with Doug and Tim on this API issue.
>        Regards,
>        Josh
>On 4/30/06, Doug Lea <dl at cs.oswego.edu> wrote:
>>Jason Mehrens wrote:
>> > The only use case I can think of would be clearing a global cache and 
>> > the resulting Map to log the entries ejected from the cache at that 
>>point in
>> > time.  ...  Which is why it "might" be useful to have a drainTo 
>>specified in
>> > the interface.
>>Well, I'm afraid I'm with Tim that it's probably not worth adding to API.
>>If there were such a thing as a BlockingNavigableMap
>>(i.e., that extended ConcurrentNavigableMap to support
>>methods like takeFirst), then drainTo should
>>surely be in such an API. But no one has asked us for such
>>a thing yet.
>>(BTW, everyone should congratulate Jason for winning the Sun Mustang
>>Regression contest by pointing out failure to maintain interrupt
>>conventions in our SynchronousQueue update!)
>>Concurrency-interest mailing list
>>Concurrency-interest at altair.cs.oswego.edu

More information about the Concurrency-interest mailing list