[concurrency-interest]yet another proposal for Executor

Tom May tom.may@infospace.com
07 Feb 2002 11:26:43 -0800


Doug Lea <dl@cs.oswego.edu> writes:

> > hopefully DL will step in soon and straighten us all out :).
> 
> Well, I don't think I can do that, but I've been contemplating ways in
> which we might be able to in JSR-166 establish better integration of
> JVM thread interruption and user task cancellation, so as to simplify
> this for programmers.

As a programmer, I would really appreciate that.  IMO, since
interrupt() is pretty much all we've got, I've often said it should be
a silver bullet that causes any blocking operation (other than
obtaining a lock) to throw an Exception.  The obvious problem with
things like close() to interrupt I/O is that it only works after the
object has been constructed, and the construction itself may block.

Tom.