Re: [xavier.leroy@inria.fr: [Caml-list] Objective Caml 3.07 final release]

[Sven Luther <sven.luther@wanadoo.fr>]

On Tue, Sep 30, 2003 at 05:49:00PM +0200, Claudio Sacerdoti Coen wrote:
> > Mmm, Xavier had told me that he had hopped to have it fixed for the
> > release, or maybe it was another issue. What does this bug affect, does
> > it break other packages or is it mostly transparent ? 
> > 
> > I suppose that this does not affect binary compatibility,
> 
>  Well, it is a camlp4 bug that changes the associativity of the arrow
>  when the first argument is optional ;-(((((
>  Thus every type expression is parsed incorrectly and being able to compile
>  the sources becomes a real miracle. Moreover, the .cmi files changes
>  (since the declared type is utterly incompatible) and the same happens
>  to the MD5 sum.

So, this mean that the binaries before or after the fix will be
incompatible, and that we have to rebuild at least some of them after
the fix is applied :(((

>  Anyway, to my knowledge no major debian package uses both camlp4 and
>  optional arguments (Coq does use camlp4 heavily, but no optional args;
>  the other packages do use optional arguments, but no camlp4). [But there
>  are a few ocaml debian packages I have never used...]

Mmm, so there may be binary incompatibility, but not in all packages.
This means that we may upload, and upload fixed packages with a
versioned dependency on ocaml once the fix is in (It would need a
dependency on ocaml-3.07 and on ocaml (>>3.07-2), since versioned
virtual dependencies are not supported.

>  In case some package becomes unbuildable, it is possible to manually
>  insert parentheses to disambiguate the parsing. The parentheses should
>  not affect MD5 sums w.r.t. the files without the parentheses in a bug
>  fixed version of camlp4.

Mmm again, i am not fluent in camlp4, so someone more knowledgeable than
me will have to fix this.

Friendly,

Sven Luther