Re: persistent name server option coming in 4.30

Morten Lind <[email protected]> Fri, 7 Nov 2014 17:45:13 +0100
Newsgroups gmane.comp.python.pyro
Organization SINTEF Manufacturing AS
Message-ID <1794591.cuCABStTCE@mly11>
Agreed. Very flexible and reasonable!

Best regards,
Morten Lind
Senior Researcher
Dept. of Production Technology
SINTEF Raufoss Manufacturing
Mobile: +47 9266 2194

On Friday 7. November 2014 16.29.57 Irmen de Jong wrote:
> On 2014-11-07 08:49, Morten Lind wrote:
> > Very nice! I have considered this myself, and would have used MongoDB
> > for it. Maybe you could throw that in as an option?
> 
> This would introduce a dependency to a 3rd party library to Pyro,
> something which I'm not a fan of. So I won't build this backend myself.
> 
> The way I'm implementing the feature however, already allows you to
> construct a name server object yourself and provide your own storage
> backend class. You can easily build a mongodb version of this yourself
> by looking at the methods of the dbm-storage class.
> 
> You cannot use the nameserver command line tool though to launch your
> own storage type, you'll have to do this in code yourself. But this is
> not rocket science: Pyro4.naming contains  helper methods to do this
> (startNs/startNsLoop). In the next version these helper methods gain a
> new parameter to pass in your own storage backend object.
> 
> 
> Irmen
> 
> 
> ----------------------------------------------------------------------------
> -- _______________________________________________
> Pyro-core mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/pyro-core

------------------------------------------------------------------------------