Re: New module system

[Mikael Djurfeldt]

Michael Livshin <address@hidden> writes:

> Mikael Djurfeldt <address@hidden> writes:
> 
> > Note that Dybvig's system is entirely at the syntactic level.  It
> > requires no special support in the evaluator since the syntactic
> > forms of the module system will be expanded away when the evaluator
> > starts building lexical environments.
> 
> so this means that the module system as is is "invisible" after the
> expansion, right?

Right.

> this also means that the module system's "entities" can't be reified
> on the C level, right?

No, it would still be possible to have a C API to the module
system---to create modules, export bindings etc.  Also, note that we
would build a layer on top of the low-level "module" syntax.  This
layer would handle mapping to file names, dynamic loading etc.

(As a demo, Dybvig and Wadell have actually implemented MzScheme's
 unit system on top of the "module" syntax!)

> > The basic problem you are addressing is essentially what to do with
> > already expanded code when a macro definition is changed.  Should we
> > or should we not re-expand code using the new definition?
> > 
> > I've never seen a system which re-expands code, but maybe this is
> > possible?
> 
> FWIW, the Common Lisp systems I've seen don't re-expand code when a
> macro definition changes.

Still, it's an interesting idea to think about.  It might turn out to
heavy to implement, but would definitely assist development.