> > In fact, I was thinking of the following algorithm:
> > - the client does
> >   caml-get update http://foo.bar/archive.cga
> > - then caml-get retrieves the archive, and for each element, look at the
> >   url of the license, retrieves it, and store it in the client repository,
> >   making it available to the use even when he/she is offline.
> 
> It should either not download the archive or put a fat warning if the
> licence is not retrievable, but i think this is already your intention.

Yes. "Je dirais même plus", when downloading an archive, only the elements whose
license can be retrieved will be added to the client repository. 
 
> > Caml-get is done to help the developer, by making use of *pieces* of
> > code easier, thus not making his/her software depend on a library if it only
> > uses one or two functions of this library for example. This can be seen
> > as way to distribute a software which otherwise would depend on a
> > hard-to-install or no-packaged software.
> > 
> > But if a software heavily uses a library, it should rather really depend on
> > this lib (that is: link with it) rather than use caml-get to copy-paste
> > all the library.
> > 
> > Does it answer your question ?
> 
> Well, i was thinking of some way of automatically or semi-automatically
> transforming a bit of caml-getted source into a proper debian package.

I don't think so, then. caml-get has no information about how to compile,
where to install,..., and is not intented to, since it is not a replacement
for the "traditional" way of distributing code.

- Maxence

-------------------
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