This breaks non-reg tests in lutin/test_ok because the exc Polka.overflow
is raised whereas the previous version was silently doing something wrong.

et "let x = i"

highligth the following bug: it is impossible to load from the same
main the same lutin program twice.  This occurs for instance with
lurettetop if 2 different main nodes are defined in the same file.

ZZZ : test4 should be re-activated (in the test rule) when this is
fixed.

Enhance gnuplot-rif with the capability to only show some variables (it was
only possible to hide variables)

Add some non-reg test in
- examples/lutin/lurette-socket to check that lurette can use
  test several sut that uses socket (ok)
- examples/lutin/xlurette to check that  lurette can use
  test several lutin programs (ko!!)

TODO : Make it possible to call several lutin interpreters from the same main.
Indeed, lurettetop needs that.

in examples/lutin/xlurette/Makefile cf test4 rule.

The problem occur in Expand.make. The message is

  Sorry, Internal Error, function CheckEnv.get_binding:
  ---> unbounded ident '='(/home/jahier/lurette/examples/lutin/xlurette/env-2.lut:4:12-12)
  break signal catched

BTW, rename stuff luc4c modules so that it sounds less lucky-centric.

Fix a bug in the lutin toplevel when using the --precision option..

Also, clean-up the distribution and fix some minor pbs while writing
the lutin tutorial.

This is very handy  to see inline what's going on.

The problem was that it was not possible to run 2 test inside the
same lurettetop session.

The symptom was a message that looked like:

    Process 18828 (ecexe) created
    Error, '' contains no node

The was in source/Lutin/mainArg.ml, when I was using Arg.parse_argv
with the optional current parameter:

Cf examples/lutin/xlurette/.

In  order to  do that,  I rely  on the  ocaml Dynlink  support (.cmxs
files).  The idea is to generate caml code that is able to call the C
code  (as  before)  and to  generate  from  that  a .cmxs  file  that
registers the OcamlRM functions.

I should modify gen_stubs to automate that process.

lucky programs by lutin programs.

(combinators).  The bug is that when a controllable becomes
uncontrollable, macros that are defined before still refers to the
controllable var.

nb : this break luckyDrawC.

nb : this break luckyDrawC.

the fact that one can not init a (controllable) variable with an input.

that, I've modified Lutin/mainArg.ml so that all globals are inside a
big structure. Indeed, Lurette has its own Arg.argv, on it may call
several lutin programs.

inherit from pre values of global ones

programs to be used from the command line.

nb : -exe no longer exist. use -old to get the old algo.