caml-list - the Caml user's mailing list
 help / color / mirror / Atom feed
From: Philippe Veber <philippe.veber@gmail.com>
To: caml users <caml-list@inria.fr>
Subject: [Caml-list] Re: Compilation issue with interdependent libraries
Date: Wed, 19 Dec 2012 00:52:57 +0100	[thread overview]
Message-ID: <CAOOOohSa6uT+epprzharphxkTHqwYouj6Q-D7=7P7T_kiXsbQw@mail.gmail.com> (raw)
In-Reply-To: <CAOOOohQU7RjWBhnnJeekHSQk44XUZTR-Asr1bNL1ZymYhQ7G=w@mail.gmail.com>

[-- Attachment #1: Type: text/plain, Size: 1778 bytes --]

Answering to myself: the information needed by findlib to sort the cma
correctly is stored in the META file, using "require" directives. I

In fact I have a slightly more complicated structure:
- library A
- library B.B1 which depends on A
- library B.B2 which depends on B.B1 and A
- executable C which depends on all others

and I had forgotten the dependency between B.B1 and A. The error message
was mentionning B.B2 instead so I overlooked the real missing part. Sorry
for the noise!

Cheers,
  Philippe.

2012/12/19 Philippe Veber <philippe.veber@gmail.com>

> Dear all,
>
> I have 2 libraries, say A and B where B depends on A, and an executable C
> which uses both A and B. Each of these components is a separate (oasis)
> project. I have some test executables in B which work fine. However, when I
> compile C, the compiler complains that it has no implementation for a
> module defined in A and used in B. The compilation (linking) command of C
> contains "-package" options for A and B, it is of the form:
>
> ocamlfind ocamlopt ... -package B -package A ... C.cmo
>
> If I permute the two package options, like in
>
> ocamlfind ocamlopt ... -package A -package B ... C.cmo
>
> the compilation works fine. It is not really clear to me, but I thought
> the order of the -package options did not matter, because findlib would
> store the dependencies between packages and produce a compilation command
> where cma or cmxa are correctly sorted. Indeed, if I use the -verbose
> option, I can see that cma are wrongly sorted in the former case, while
> they are fine in the latter.
>
> Admitedly my description is rather abstract, but does anyone see what
> could I (or oasis) have forgotten, to tell ocamlfind that B.cma depends on
> A.cma?
>
> Cheers,
>   Philippe.
>
>

[-- Attachment #2: Type: text/html, Size: 2112 bytes --]

      reply	other threads:[~2012-12-18 23:53 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-12-18 23:21 [Caml-list] " Philippe Veber
2012-12-18 23:52 ` Philippe Veber [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to='CAOOOohSa6uT+epprzharphxkTHqwYouj6Q-D7=7P7T_kiXsbQw@mail.gmail.com' \
    --to=philippe.veber@gmail.com \
    --cc=caml-list@inria.fr \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).