[Caml-list] Why not 'include' in an .mli file?

[Caml-list] Why not 'include' in an .mli file?

[Richard Jones]

I have a library which for various reasons is split into two modules,
called them Analysis_p and Analysis.

I want end-user programs to be able to 'open Analysis', and get all
the symbols from both modules.  This is simple enough using the
'include' directive:

analysis.ml:	include Analysis_p
		(* plus a few extra functions *)

However when I added an analysis.mli file, I found that I could not
automatically 'include' symbols from Analysis_p.  The following
doesn't work:

analysis.mli:	include Analysis_p
		(* plus a few extra prototypes *)

ocamlc -w s -g -I +pcre -I +postgres -I +dbi -c analysis.mli
File "analysis.mli", line 1, characters 8-24:
Unbound module type Analysis_p

Instead it seems, as far as I can see, that I have to list out all the
prototypes again in analysis.mli.  Is this so?  If so, it's a pain for
me, and means that prototypes for one function exist in two different
places (also I don't control Analysis_p directly, so it requires extra
coordination).

Rich.

-- 
Richard Jones. http://www.annexia.org/ http://www.j-london.com/
Merjis Ltd. http://www.merjis.com/ - improving website return on investment
MAKE+ is a sane replacement for GNU autoconf/automake. One script compiles,
RPMs, pkgs etc. Linux, BSD, Solaris. http://www.annexia.org/freeware/makeplus/

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

Re: [Caml-list] Why not 'include' in an .mli file?

[Markus Mottl]

On Wed, 14 Apr 2004, Richard Jones wrote:
> However when I added an analysis.mli file, I found that I could not
> automatically 'include' symbols from Analysis_p.  The following
> doesn't work:
> 
> analysis.mli:	include Analysis_p
> 		(* plus a few extra prototypes *)
> 
> ocamlc -w s -g -I +pcre -I +postgres -I +dbi -c analysis.mli
> File "analysis.mli", line 1, characters 8-24:
> Unbound module type Analysis_p

Note that this is a module type, not a module.  You could put the
module type (interface) of "Analysis_p" into the submodule "S" of the
extra-file "analysis_p_intf.mli". Use it then to restrict the interface
of "Analysis_p" in "analysis_p.mli", and include it in "analysis.mli"
using "include Analysis_p_intf.S".

Regards,
Markus

-- 
Markus Mottl          http://www.oefai.at/~markus          markus@oefai.at

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

Re: [Caml-list] Why not 'include' in an .mli file?

[Christopher Quinn]

it strikes me that a cmi file can be thought of as comprising a module 
type, and the compiler might be extended to allow for Richard's usage, 
which imho is a natural one.
i believe this is in the wish list db from way back and not dismissed 
or otherwise classified.

chris

> 
> Note that this is a module type, not a module.  You could put the
> module type (interface) of "Analysis_p" into the submodule "S" of the
> extra-file "analysis_p_intf.mli". Use it then to restrict the interface
> of "Analysis_p" in "analysis_p.mli", and include it in "analysis.mli"
> using "include Analysis_p_intf.S".
> 
> Regards,
> Markus
> 

-- 
thanks.
- chris

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