On Wed, Jun 29, 2005 at 02:31:11AM +0200, Christophe TROESTLER wrote:
> Let me start with a little story that happened to me recently.
> [...]
> 
> The problem was that the attached information (an additional field in
> a record) was a cyclic data structure.  From there on, all equality
> tests became deadly! (I would have preferred to have the exception
> Invalid_argument "equal: abstract value".)  What made matters worse is
> that the compiler could not help me to find the locations of such
> problems -- which can be hidden e.g. in List.mem.  Not a nice job to
> do...

A harrowing and tragic tale indeed!  While not really a solution to your
problems, you *can* obtain the "abstract value" exception behavior with a
little trick; Jacques Garrigue posted about it to the list a couple of
months ago:

    http://caml.inria.fr/pub/ml-archives/caml-list/2005/04/66b52e5c944a3bd4c45e8b68c450461a.en.html

I say it's no solution because (a) it doesn't keep you from having to write
your own equality predicate with a different name, and (b) the compiler
gives you no static warnings about invalid uses -- might as well program in
Scheme! ;)

cheers,
William