Re: ThreadPool and NetServe Release?

peter-4lf8KW9E9MLMqX/[email protected] wrote:


> It may also be a good idea to add in DNA implementations if we have time. Or
> else we can just do it next release.
> 

now that we've done the refactor it's very straightforward to add in the 
DNA impl.  I would do it.
Anything, left to do add to DNA for release?  I'd like do the impl based 
on a release candidate (in essence if not in name).

> We could just do a simple RC/beta build and if no problems are found just
> release it. Then again it may be easier to just release and re-release on
> problems. I am easy either way.

I would opt for a straight release whenever there are no outstanding 
tasks that do not affect functionality (eg more unit tests, docs etc)
and - if required - do a bugfix release.
Else we can do a beta first - like you suggest for netserve.

> kool. ThreadPool should be easy to get coverage on. However Netserve is
> going to be a real PITA to do. The only way that we can do it as far as I
> can tell is to write MockSocket, MockServerSocket and corresponding
> Mock*putStreams which is a lot of work. It may be worth it though as it will
> allow us to guarentee quality better.
> 
> Maybe we do a final release of ThreadPool and a beta of NetServe and then
> when we convert the unit tests we release a final NetServe?

+1

> Kool! Added a JIRA issue for this.

ok - will assign it.






-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf