Re: Re: Registries [2]

At 13:40 2003-07-27, David Abrahams wrote:
>Daniel Wallin <[email protected]> writes:
>
> > At 01:26 2003-07-27, David Abrahams wrote:
> >>Daniel Wallin <[email protected]> writes:
> >> >>Hmm, if there was really a mutual dependency we might have an issue,
> >> >>but we could still handle it with a simpler mechanism that simply
> >> >>deferred imports until initialization was completed.
> >> >
> >> > I guess, but it would be nice to handle it without introducing code in
> >> > the dispatch functions. Which is the only place I can think of where
> >> > we could defer importing to.
> >>
> >>In Boost.Python, the module initialization function is special; we
> >>actually generate it with a macro and it calls what the user wrote
> >>(for exception-handling purposes), so we can easily generate code
> >>there to handle any deferred imports.
> >
> > Right, but how do you call this code after all other modules has been
> > initialized?
>
>You keep a count of modules being initialized in each thread and when
>it falls back to zero you do the imports.
>
> > Also, at least in lua. Modules could be loaded at any time, so there isn't
> > really a point where we can assume everything is loaded and import
> > converters.
>
>Can't you do something similar to the above?

I don't think so. When I said "at any time" I didn't mean concurrently.
Consider (lua code):

loadmodule "X" -- exposes f(std::vector<double>)
loadmodule "Y" -- exposes some convenience converters
                -- lua -> C++ std::vector<double>

We would really like for module X to be able to use the converters
introduced by module Y here, but module X is already initialized
when module Y's converters are created.

> >> >> > How are extensions to be registered in this central registry? In my
> >> >> > previous post I proposed that they should be registered with a
> >> >> > string literal identifier.
> >> >> >
> >> >> >    register_extension("my_extension");
> >> >> >
> >> >> > How does this fit in with pythons module system? (Note that we are
> >> >> > really registering the extension, and not modules).
> >> >>
> >> >>Every extension is a module (not nececssarily the other way around),
> >> >>so I see no reason not to register the extension by its module name
> >> >
> >> > I guess in python you can ask the interpreter state for the current
> >> > module name?
> >>
> >>Erm, I'm not sure you can.  It is supplied as part of the module
> >>initialization function; in fact, it is stored in the module object as
> >>its __name__ attribute, so we can get it thence, assuming we can get
> >>to the module.
> >
> > Ok.
> >
> >> > In lua this isn't possible though, so we would need some user defined
> >> > identifier instead.
> >>
> >>You've gotta have the module or its name stored somewhere, dontcha?
> >
> > Other than in the .so filename? No.
>
>Why not just store it then?  When you create it, stick a string in its
>__name__ attribute.

I don't know what this means, create what? And what string? And how? :)

---
Daniel Wallin



-------------------------------------------------------
This SF.Net email sponsored by: Free pre-built ASP.NET sites including
Data Reports, E-commerce, Portals, and Forums are available now.
Download today and enter to win an XBOX or Visual Studio .NET.
http://aspnet.click-url.com/go/psa00100003ave/direct;at.aspnet_072303_01/01