Re: James / Excalibur ThreadPool conflict

Peter M. Goldstein wrote:

>Stephen et al,
>
>  
>
>>>I
>>>looked into the sources and CVS history and discovered that
>>>org.apache.avalon.excalibur.thread.ThreadPool.execute() interface was
>>>changed by you about three months ago - retracting Runnable as a valid
>>>argument in favour of Executable.
>>>      
>>>
>
>
>I'm very confused.  This interface is actually marked deprecated.  So why
>would the deprecated interface be changed?
>

Just a quick note - when I mention Pete - I was referring to the changes
committed by Pete Donald back in September - not yourself (I should have 
been
more explicit with the reference to Pete)!

>
>
>Moreover, why is the replacement interface
>(org.apache.excalibur.thread.ThreadPool) not changed?  It still has the
>execute(Runnable) method.  In fact, it has both methods.
>

That was my impression - but I'm wondering about the validity of the 
return type.
The ThreadPool interface from rev. 1.2 (5 months ago) is:

   org.apache.avalon.excalibur.thread.ThreadControl execute( Runnable 
work );

The current interface reflects a revision back in September in
which the return type was modified to the following:

   org.apache.excalibur.thread.ThreadControl execute( Runnable work );

Keep in mind that org.apache.excalibur.thread.ThreadControl extends 
org.apache.avalon.excalibur.thread.ThreadControl so we are talking about
a derived return type.  I'm not an expert on return type constrains in the
Java language - but I do know that I'm getting a NoSuchMethodException so
I figure something is wrong somewhere.

Cheers, Steve.

>
>
>Can anyone shed any light on this?
> 
>--Peter
>
>
>
>--
>To unsubscribe, e-mail:   <mailto:[email protected]>
>For additional commands, e-mail: <mailto:[email protected]>
>
>
>
>  
>

-- 

Stephen J. McConnell
mailto:[email protected]
http://www.osm.net