On Sun, Sep 26, 2010 at 02:39:31AM +0900, John Mettraux wrote:
> 
> On Fri, Sep 24, 2010 at 09:15:30AM -0700, Eric Platon wrote:
> > 
> > It is clear that calling #launch initializes a new process and thus
> > produces the expected behavior (meaning no bug), but I was actually
> > wondering whether #launch should not be "resume-aware" through the
> > storage. Not sure what is worth right now. Your extended quickstart is
> > sound and clear to me, but I feel that it could be more compact: The
> > storage could be used as a memory across runs, instead of adding a
> > wfid file. I mean, it looks less elegant than just relying on the
> > storage, notably when using FsStorage.
> > 
> > That's maybe just me, and it would be a significant rewrite. Still
> > learning! Going forward. Thanks again for your prompt support!
> 
> I hope to finish it by tomorrow.
> 
> It's true that the wfid file trick is not adapted to environments where 
> engines in different Ruby runtimes launch such "unique" processes, hence my 
> launch_single() work.

Hello Eric,

thanks for your idea, I've refined the launch_single method into :

  
http://github.com/jmettraux/ruote/commit/b1d1046b60b4a11ef3b0cb4cc83b368870c34854

The tests look like :

  
http://github.com/jmettraux/ruote/blob/ruote2.1/test/functional/ft_46_launch_single.rb

Not super happy with the 'single[s]' appellation, but I didn't want to use 
'singleton'... Well I could have, I already have 'instances'...


Thanks again,

-- 
John Mettraux - http://jmettraux.wordpress.com

-- 
you received this message because you are subscribed to the "ruote users" group.
to post : send email to [email protected]
to unsubscribe : send email to [email protected]
more options : http://groups.google.com/group/openwferu-users?hl=en

Reply via email to