[ruby-core:44391] Re: [ruby-trunk - Feature #6293][Assigned] new queue / blocking queues

On Mon, Apr 16, 2012 at 06:25:59PM +0900, SASADA Koichi wrote:
> Hi,
> 
> Just an idea.
> 
> (2012/04/14 11:25), Aaron Patterson wrote:
> > I could, but I think the changes I want would add too much
> > complexity to the traditional Queue class.  We have to add
> > branching code for every method that could block / not-block.  Not
> > to mention users must pass the magical true | false to indicate if
> > they want blocking or non-blocking behavior.  For example:
> > 
> > queue.pop(true)  # does "true" mean I enable blocking? 
> > queue.pop(false) # does "false" mean I disable blocking?
> > 
> > It seems confusing to me, especially given the *actual* api with
> > twists your brain with a double negative (from thread.rb):
> > 
> > def pop(non_block=false) ... end
> > 
> > So calling pop() means we're doing a not not blocking call. :(
> 
> How about to add try_pop?

try_pop seems fine, but it still seems strange to combine blocking and
non-blocking queues (but maybe *I* am the one who is strange).

In the case of BlockingQueue#pop in the patch I submitted, it allows a
timeout.  I don't think it's a feature that should be abandoned.

> >>> * include Enumerable
> >> 
> >> The semantics is not trivial.  See the discussion in #4589.
> > 
> > I've read #4589.  I don't think it's that much of a problem.
> > Other languages[1] include enumerable type behavior in their
> > queues, and consistency is not guaranteed.
> > 
> > 1.
> > http://docs.oracle.com/javase/6/docs/api/java/util/Queue.html#methods_inherited_from_class_java.util.Collection
> 
> How
> > 
> about to implement Queue#to_a method that generate array which
> contains queues containing objects?

That seems fine!  Then we can eliminate Enumerable mixed in. :-)

> # I agree the name Thread::Queue because it is clear
> # that this object should be for synchronization.
> # Simple Queue class is ambiguous that we can use it
> # for synchronization or not.  However, in the Ruby world,
> # ::Queue is familiar as a data for synchronization.

Yes.  I think keeping ::Queue is good.  I think ::Queue can be
implemented in terms of Thread::BlockingQueue.  I would have refactored
::Queue to be a subclass of Thread::BlockingQueue, but I didn't want my
patch to be ignored because the diff size. :-)

-- 
Aaron Patterson
http://tenderlovemaking.com/