Re: About modules

[Clark McGrew <mcgrew at ale dot physics dot sunysb dot edu>]

>>>>> "Jost" == Jost Boekemeier <jostobfe@calvados.zrz.TU-Berlin.DE> writes:

    Jost> Clark McGrew <mcgrew@ale.physics.sunysb.edu> writes:

    Jost> You can create a new environment and start a repl in it.
    Jost> But there are no global environment pointers.

Unless it's obvious what the gh_ interface to environments should look
like maybe it's best to use the scm_ interface for a while.  I hope
the the gh_ stuff stays relatively small and high level so that it's
easy to learn and use.
 
    Jost> I must create a state variable for use with the gh_ API.
    Jost> But that means that the gh_ functions will somehow live
    Jost> their own life.

I think that any gh_ function which *needs* an environment argument,
but doesn't have one should be deprecated.  I'd just like to see older
libraries work with the newer guile.  If that's not possible, then I
think it's better to completely break things and get gh_ right.

Thanks,

Clark

Off topic:

There are lot's of little things I wonder about in gh_:

   Why is there gh_vector_length and gh_uniform_vector_length when a
   vector is just a uniform vector of SCM's?

   Why have gh_vector_ref and gh_uniform_vector_ref?  

   How do I set an element in a uniform vector using gh_?

   Why gh_call0, gh_call1, &c instead of gh_call(SCM proc, ...) that
   takes a SCM_UNDEFINED to terminate the list of arguments (same
   applies to gh_append)?