Am Mit, 2003-08-06 um 03.05 schrieb Jacques Garrigue:
> From: Gerd Stolpmann <info@gerd-stolpmann.de>
> > there is now an improved version of GODI for download:
> > 
> > http://www.ocaml-programming.de/packages/godi-bootstrap-20030805.tar.gz
> > 
> > - The bootstrap script tries to pre-configure labltk. It looks into
> >   much more locations that the O'Caml configure script. I think this
> >   was one of the biggest problems in the first release.
> 
> OK. Nice attempt, and properly splitting the work with ocaml's
> configuration script.
> By the way, I'm not sure this is a good idea to make labltk an
> optional package, for two reasons:
> - since it gets configured with ocaml, you have to recompile ocaml if
>   something goes wrong (your pre-search may help, but since you don't
>   check by explicitly linking this may still be wrong).
> - it includes ocamlbrowser, which is an ocaml tool expected to be
>   there
> 
> If your motivation is to allow binary packages including only part of
> the stuff, then it might be a better idea to make ocaml a metapackage,
> and have packages ocaml-core, ocaml-graphics, ocaml-labltk,
> ocaml-camlp4 as dependencies.

I am more thinking of server installations without X windows, so the
split is X/non-X. camlp4 is part of the main ocaml package, there is
really no need to have it separate.

Maybe a solution of the configuration problem is to run the configure
script again for labltk, but only to use the labltk-specific results
(i.e. tkdefs, tklibs). This is more user-friendly than the current
solution, and there is no danger to configure the various parts in an
incompatible way.

Meta packages: nice idea, but the question is what the users want to
have. Maybe they are more interested in "distribution sets" for tasks,
i.e. a graphics set, a web set, a database set, ...

Gerd
-- 
------------------------------------------------------------
Gerd Stolpmann * Viktoriastr. 45 * 64293 Darmstadt * Germany 
gerd@gerd-stolpmann.de          http://www.gerd-stolpmann.de
------------------------------------------------------------

-------------------
To unsubscribe, mail caml-list-request@inria.fr Archives: http://caml.inria.fr
Bug reports: http://caml.inria.fr/bin/caml-bugs FAQ: http://caml.inria.fr/FAQ/
Beginner's list: http://groups.yahoo.com/group/ocaml_beginners