Introspect data from Python missing return parameters

[John (J5) Palmieri]

I have been convinced to add optional typing of methods and signals.
Actually I think a patch is being written as we speak. However if the
person who has written the service does not use it you still can not
rely on the introspect data having any out parameters.  This is because
if a method simply returns an empty message will be sent in which case
specifying 'v' here would be wrong.

On Fri, 2005-10-07 at 19:45 +0200, Tako Schotanus wrote:
> John (J5) Palmieri wrote:
> 
> >On Fri, 2005-10-07 at 12:12 +0200, Tako Schotanus wrote:
> >  
> >
> >
> >>I also think it would be worthwhile to stimulate people to make their 
> >>introspection data as complete and correct as possible otherwise lazy 
> >>programmers will just leave it out and we would lose a host of 
> >>interesting possibilities with respect to auto-generated apis, discovery 
> >>tools and who knows what. (Personally I'd make introspection data 
> >>mandatory ;-)
> >>    
> >>
> >
> >These are the same arguments typed and weakly typed proponents had had
> >for ages.  My first goal is to make D-Bus a good fit for Python not
> >Python a good fit for D-Bus.
> > 
> >  
> >
> Ah no, not trying to force typing on a weakly typed language, just 
> saying that it would be nice if introspection was forced to be as 
> "up-to-date" and complete as possible in the _language implementing the 
> service_ , nothing more. So be all means have python-based services 
> return v. But if there are (ever) too many services that just have 
> incomplete or worse wrong introspection information a lot of potential 
> for the uses of introspection could be lost. That's the only point I'm 
> trying to make I think.
> 
> Cheers,
>  -Tako
> _______________________________________________
> dbus mailing list
> dbus at lists.freedesktop.org
> http://lists.freedesktop.org/mailman/listinfo/dbus
-- 
John (J5) Palmieri <johnp at redhat.com>