Re: guile-ffi

[Gordon Matzigkeit <gord@trick.fig.org>]

Hi!

>>>>> Marius Vollmer writes:

 MV> That's true.  But I'd say that it's the usual tradeoff: quick
 MV> development with an interpretive approach, and good performance
 MV> with a compiler.

Yep, but my machine is already fast enough that performance isn't an
issue any more (since the main hit is in initialization), and it's two
years old.

 MV> Furthermore, I'm not convinced that we can get reliable bindings
 MV> to C code without the help of a C compiler.

Anything a C compiler can do, Scheme can do better. ;)

 MV> Does libffi know everything it needs to know about the ABI of a
 MV> platform.

The libffi sources imply that it does.

 MV> Can we extract all necessary information from header
 MV> files, without reimplementing much of the compiler frontend?

I believe so.  However, as with all wrappers, we need some human
intervention:

  char **

might be a vector of strings, or else a modifiable string pointer.


I've been thinking long and hard about this issue, not in the context
of Guile, but in the context of a function call interface that is
portable to all platforms (and which doubles as a database format).

My task is to wrap the entire GNU C library in Guile, and that is much
too hard to write and maintain static C bindings.  Basically, for
every public /include/some/file.h, I want to create a
/share/guile/gnulib/some/file.scm module that provides Scheme
bindings.

This is a reasonable task, with a little work.

 MV> You, I'd like to help where I can.  My current focus is on the
 MV> Guile to Gtk+ bindings, which is developing into a quite general
 MV> glue generator.  It is currently very much on the static side,
 MV> but putting some libffi into it to make it more dynamic would be
 MV> cool.

I want to take the same approach with GTK+, so I'll fetch your work
and see what I can contribute.  Benchmarking between the two
approaches will reveal a lot, IMO.

 MV> Tom Lord has interfaced Systas (which is an offspring of Guile, I
 MV> think) to a C parser.  That could be very helpfull for parsing
 MV> header files.

Do you know where I could find this?

-- 
 Gordon Matzigkeit <gord@fig.org> //\ I'm a FIG (http://www.fig.org/)
    Lovers of freedom, unite!     \// I use GNU (http://www.gnu.org/)

Copyright (C) 1998 FIG.org; the creator offers you this gift and wants it
to remain free.  See http://www.fig.org/freedom.html for details.
  This work may be copied, modified and distributed under the GNU General
  Public License (GPL).  See http://www.gnu.org/copyleft/gpl.html.