1. 24 Apr, 2019 1 commit
  2. 13 Mar, 2019 1 commit
  3. 25 Jan, 2016 1 commit
  4. 17 Oct, 2012 1 commit
  5. 14 Sep, 2011 1 commit
  6. 15 Jul, 2011 1 commit
  7. 15 Apr, 2011 1 commit
  8. 12 Apr, 2011 1 commit
  9. 25 Nov, 2010 1 commit
  10. 30 Jun, 2010 1 commit
  11. 23 Jun, 2010 1 commit
  12. 17 Mar, 2010 29 commits
    • Erwan Jahier's avatar
      lurette 1.45 Mon, 22 Sep 2008 18:20:01 +0200 by jahier · 89aedffe
      Erwan Jahier authored
      Parent-Version:      1.44
      Version-Log:
      
      Fix an embarassing bug in the integer solver (in dimension 1).
      
      Project-Description: Lurette
      89aedffe
    • Erwan Jahier's avatar
      lurette 1.42 Fri, 20 Jan 2006 15:06:27 +0100 by jahier · 04e06981
      Erwan Jahier authored
      Parent-Version:      1.41
      Version-Log:
      
      source/luc_exe.ml
      source/command_line_luc_exe.ml
      source/lurette.ml
      source/graph.ml
      source/lucky.mli
      source/lucky.ml
      source/util.ml
      source/formula_to_bdd.ml
      source/formula_to_bdd.mli
      source/fair_bddd.ml
      source/fair_bddd.mli
      source/bddd.ml
      source/bddd.mli
      source/solver.mli
      source/solver.ml
      source/polyhedron.ml
      source/polyhedron.mli
      source/store.mli
      source/store.ml
      source/parse_luc.mli
      source/parse_luc.ml
      source/show_env.ml
      source/env_state.mli
      source/env_state.ml
      source/run_aut.mli
      source/run_aut.ml
      source/lurettetop.ml
      source/draw.mli
      source/draw.ml
      source/gen_stubs.ml
      source/parse_c_scade.ml
      source/ne.ml
      source/prevar.ml
      source/prevar.mli
      source/show_luc.ml
      utils/lucky.el
      install/configure.in
      Makefile.common.source
      install/Makefile.lurette.in
      user-rules
      source/Makefile.lurettetop
      source/Makefile.gen_fake_lutin
      source/Makefile.show_luc
      source/Makefile.lucky
      source/Makefile.gen_stubs
      source/Makefile.lurette_lib
      source/Makefile
      RUN_ME
      TODO
      test/time-joly.exp
      test/time-joly.res
      test/time-ecrins.exp
      ihm/xlurette/xlurette_glade_main.ml
      ihm/xlurette/xlurette_glade_interface.ml
      ihm/xlurette/makefile
      polka/caml/Makefile
      polka/Makefile.config
      source/Makefile.gen_fake_lucky
      install/xlurette_sh.in
      install/xlurette.sh.in
      install/xlurette.bat.in
      install/set_env_var.in
      install/lurettetop_sh.in
      install/lurettetop.sh.in
      install/lurettetop.bat.in
      install/gen_stubs.sh.in
      source/poly_draw.ml
      source/poly_draw.mli
      source/lustreExp.ml
      source/lustreExp.mli
      source/exp.ml
      source/var.ml
      source/type.mli
      examples/xlurette/fault-tolerant-heater/degradable-sensors.luc
      examples/xlurette/fault-tolerant-heater/heater_control_env.luc
      examples/xlurette/fault-tolerant-heater/sensors.luc
      source/parser.mly
      doc/lurette-man/lurette-man.tex
      source/parse_sildex.ml
      test/time-asti.exp
      RELEASE-NOTES
      FAQ
      source/gnuplot-socket.ml
      source/luc4ocaml.ml
      source/luc4ocaml.mli
      source/Makefile.gnuplot-rif
      source/gnuplot-rif.ml
      source/luckyDraw.mli
      source/luckyDraw.ml
      source/luc4ocaml_nolbl.mli
      source/luc4ocaml_nolbl.ml
      source/lucky2lus.ml
      source/Makefile.luc4ocaml
      source/Makefile.lurette_ocaml_lib
      source/Makefile.lurette_debug
      source/Makefile.luckyDraw
      source/Makefile.lucky2lus
      xlurette/Scade/GenMake_l4sim.tcl
      examples/lucky/external_code/call_external_c_code.luc
      examples/lucky/external_code/Makefile
      VERSION
      install/configure
      install/set_env_var.bat.in
      
      Quite a lot of changes this time... far too much...
      
      Fix a at least 5 bugs, and re-organize directories, in particular
      the test dir. too much to say...
      
       * pre pre x was always equal to pre x (bugs introduced in 1.39 or something
      I think...).
       * ...
      
      One big improvment is that I have completely changed the way runtime automaton
      are handled, using continuations instead of maitaining an explicit
      data-structure.
      This let me reduce the run_aut module by 75 percent, resulting in much clearer
      code, and less buggy, and more efficient. ouaw!
      
      thanks to that, I support DAG now (was rejected before).
      
      The executable that do not need env var are no more called via a script that
      call set_env_var, so that they are fully stand-alone: show_luc, lucky,
      gen_fake_lucky. This is why util was splitted (util+util2)
      
      source/var.ml
      source/lucaml.ml  -> luc4ocaml.ml
      source/lucaml.mli -> luc4ocaml.mli
      
      source/run_aut.ml:
         Fix a bug during the backtraking in the dynamic tree.
      
      source/env_state.ml
      
      source/formula_to_bdd.ml
         sort the list of free index, because the first order of variable is
      generally
         better.
      
      source/bddd.ml
      source/bddd.mli
      source/solver.mli
      source/solver.ml
      source/polyhedron.ml
      source/polyhedron.mli
      source/store.mli
         Add the verbosity level in a few functions.
      
      source/store.ml
         FIX A BUG in the bdd traversal. Some branches were never visited because
         I was raising a No_numeric_solution exception instead od returning an empty
         store when adding a constraint... This embarassing bug is there from the
         begining... Ouille!
      
         Also, when a valid integer cannot be found, return a wrong one, which is a
         bug really. But is it better than the current state were raising the no_num
         expt would cut some branches in the bdd. I will fix that problem later.
      
      source/parse_luc.ml
      source/lustreExp.ml
      source/lustreExp.mli
         The list of pre vars was computed with a ref, which is bad. Moreover,
         the same pre var was created with different indexes, which did not lead
         to incorrect behavior, but still...
      
      source/exp.ml
      test/bj.luc
      examples/lucky/external_code/call_external_c_code.luc
      examples/lucky/external_code/Makefile
      
      Project-Description: Lurette
      04e06981
    • Erwan Jahier's avatar
      lurette 1.41 Mon, 26 Sep 2005 13:51:19 +0200 by jahier · 98380cec
      Erwan Jahier authored
      Parent-Version:      1.40
      Version-Log:
      
      source/luc_exe.ml
      source/lurette.ml
      source/util.ml
      source/formula_to_bdd.ml
      source/bddd.ml
      source/solver.ml
      source/store.ml
      source/parse_luc.mli
      source/parse_luc.ml
      source/env_state.mli
      source/env_state.ml
      source/run_aut.ml
      source/gne.mli
      source/gne.ml
      source/lurettetop.ml
      source/gen_stubs_poc.ml
      source/ne.ml
      source/ne.mli
      share/lucky.el
      share/configure.in
      Makefile.common.source
      OcamlMakefile
      user-rules
      source/Makefile.show_luc
      source/Makefile.lucky
      source/Makefile.gen_stubs
      source/Makefile.lurette_lib
      source/Makefile
      test/time-joly.res
      test/temp_int.luc
      ihm/xlurette/xlurette.glade
      ihm/xlurette/makefile
      source/Makefile.gen_fake_lucky
      source/lustreExp.ml
      source/lustreExp.mli
      source/exp.ml
      source/exp.mli
      source/parser.mly
      doc/lucky-man/lucky-man.tex
      source/gnuplot-socket.ml
      source/gnuplot-rif.ml
      source/luckyDraw.mli
      source/luckyDraw.ml
      source/essai-maitre.ml
      source/essai-esclave.ml
      source/Makefile.ocaml
      source/Makefile.luckyDraw
      
      Add the possibility to call external code via dyn lib.
      
      The generation of default lucky files from Lurette was broken.
      
      Fix warning generated by gcc4.
      
      Fix a bug where variables with default values were generated twice
      (bug only visible in assert mode ; in the other mode, one of the value
      was taken, but I am not sure which one...)
      
      Project-Description: Lurette
      98380cec
    • Erwan Jahier's avatar
      lurette 1.40 Fri, 22 Jul 2005 17:06:41 +0200 by jahier · b1a40b56
      Erwan Jahier authored
      Parent-Version:      1.39
      Version-Log:
      
      source/luc_exe.ml
      source/command_line_luc_exe.ml
      source/command_line_luc_exe.mli
      source/lurette.ml
      source/command_line.ml
      source/command_line.mli
      source/lucky.ml
      source/rif.ml
      source/env_state.mli
      source/env_state.ml
      source/run_aut.mli
      source/run_aut.ml
      source/lurettetop.ml
      ihm/xlurette/xlurette_glade_main.ml
        Add a reactive mode where the output of the previous cycle is returned
        when the Lucky environment is blocked (because no transition from
        the current node(s) is labelled by a satisfiable constraint.
      
      Remove the dependence on cygwin to run it under windows.
      
      In particular, to be able to do that, I now use socket instead of pipes
      in order to communicate between lurettetop and xlurette.
      
      Also, define proper, cp, rm, etc instead of doing them by sys calls.
      
      A lot of changes to make it work with the scade gui
      
      Project-Description: Lurette
      b1a40b56
    • Erwan Jahier's avatar
      lurette 1.32 Wed, 30 Jun 2004 14:03:33 +0200 by jahier · da9a1255
      Erwan Jahier authored
      Parent-Version:      1.31
      Version-Log:
      
      source/fair_bddd.ml:
         Fix a bug where the distribution was not correct according
        to the solution number
      
      Project-Description: Lurette
      da9a1255
    • Erwan Jahier's avatar
      lurette 1.29 Fri, 30 Apr 2004 16:43:39 +0200 by jahier · 1c8385d0
      Erwan Jahier authored
      Parent-Version:      1.28
      Version-Log:
      
      source/ocaml2c.idl:
      source/lurette.ml:
      source/gen_stubs_poc.ml:
      source/gen_stubs_scade.ml:
        Fix a bug where the step and the tries in the sut were completely
        wrong.
      
      source/store.ml
        Fix a bug where, if only the ~min or the ~max options were used,
        they were ignored ...
      
      source/parse_luc.ml:
        Make sure that ~min and ~max are fed with values, not complex expressions.
      
        pre-process lucky files with cpp.
      
      source/sim2chro.ml:
        put the locals after the outputs because it is the order into
        which their values appear.
      
      source/gen_stubs.ml:
        Do not change the oracle compiler if it is provided.
      
      Project-Description: Lurette
      1c8385d0
    • Erwan Jahier's avatar
      lurette 1.28 Wed, 21 Apr 2004 09:38:03 +0200 by jahier · 6de2c0e9
      Erwan Jahier authored
      Parent-Version:      1.27
      Version-Log:
      
      Add #locs pragmas so that sim2chro is able to print local var in green.
      
      Project-Description: Lurette
      6de2c0e9
    • Erwan Jahier's avatar
      lurette 1.24 Thu, 11 Mar 2004 15:19:19 +0100 by jahier · fa741106
      Erwan Jahier authored
      Parent-Version:      1.23
      Version-Log:
      
      source/lustreExp.ml:
         use a map indexed by the var name instead of a list of
         variables as the search in such a list is done very often
         during the type checking.
      
      Project-Description: Lurette
      fa741106
    • Erwan Jahier's avatar
      lurette 1.23 Mon, 08 Mar 2004 14:10:38 +0100 by jahier · 18e59b0d
      Erwan Jahier authored
      Parent-Version:      1.22
      Version-Log:
      
      source/*.ml*:
         perform a sed 's/[ ]*$//' on all files to remove
         trailing space.
      
      Project-Description: Lurette
      18e59b0d
    • Erwan Jahier's avatar
      lurette 1.22 Mon, 08 Mar 2004 13:46:15 +0100 by jahier · ceac869e
      Erwan Jahier authored
      Parent-Version:      1.21
      Version-Log:
      
      Add a couple of scade files in the distrib so that it is easier
      to test scade programs under linux using the generated c files.
      
      source/value.ml
         use T and F instead of t and f for booleans (more visible under xlurette).
      
      share/Makefile.lurette.in:
         Reorganize (clean-up) the file
      
      xlurette/xlurette_glade_main.ml:
      share/plot:
         Fix the gnuplot problems (+ <<set mouse>> in share/plot).
      
      source/util.ml:
      source/lurettetop.ml:
         Move the gnuplot stuff into lurettetop.
      
      Project-Description: Lurette
      ceac869e
    • Erwan Jahier's avatar
      lurette 1.20 Fri, 13 Feb 2004 16:03:37 +0100 by jahier · b3d4d2c7
      Erwan Jahier authored
      Parent-Version:      1.19
      Version-Log:
      
      source/store.ml:
         Fix a bug found by Bertrand where it was necessary to
         apply substitutions each time a constraint is added
         Maybe it is sometimes overkill, but it is safe,
         and performance does not seem to suffer that much.
      
      source/parse_luc.ml:
         Do not open the .luc file twice, and keep the whole file
         in a string to call the lustre expression parser.
         This was an attempt to make the debbuger work again
         with big .luc file, but it fails... The Stream lib really
         seems to suck AFA ocamldebug is concerned.
      
         Also shout when a wrong flag is used.
      
      source/lurette.ml:
         when an assertion is violated, put the corresponding vector
         in the .rif file.
      
      Project-Description: Lurette
      b3d4d2c7
    • Erwan Jahier's avatar
      lurette 1.19 Mon, 19 Jan 2004 14:36:43 +0100 by jahier · caa32faa
      Erwan Jahier authored
      Parent-Version:      1.18
      Version-Log:
      
      Add support for structured types in lurette.
      
      source/type.ml/.mli:
         New file containing lucky var type definition (moved from var.ml).
      
      source/gen_stubs_*.ml:
      source/parse_c*.ml:
      
      Project-Description: Lurette
      caa32faa
    • Erwan Jahier's avatar
      lurette 1.17 Thu, 08 Jan 2004 09:06:39 +0100 by jahier · 06abf72c
      Erwan Jahier authored
      Parent-Version:      1.16
      Version-Log:
      
      source/env_state.ml
         Fix several bugs in the way automata run in parallel were handled.
         (actually, i knew that this feature was broken).
      
         Also add support to handle cases where the same variable has
         several min (take the max), max (take the  min), alias, default,
         init (abort) option values.
         I am not sure if i should authorize that for min and max though.
      
      Project-Description: Lurette
      06abf72c
    • Erwan Jahier's avatar
      lurette 1.15 Mon, 05 Jan 2004 17:46:15 +0100 by jahier · e1350c78
      Erwan Jahier authored
      Parent-Version:      1.14
      Version-Log:
      
      source/util.ml:
         Redirect third party tools error msgs in the main windows rather than in the
      misc.
      
      source/lucky.ml:
      source/run_aut.ml:
         Indicate the removed paths in the verbose mode only.
      
      source/run_aut.ml:
         Fix a bug where the execution was looping when no more formula
         were satisfiable (when all formula mode was on).
      
      source/fair_bddd.ml:
         Raise No_numeric_solution if the polyhedron volume
         is smaller then epsilon.
      
      source/lurettetop.ml:
         Recompile the Lustre program if necessary.
      
      source/gen_fake_lutin.ml:
         Declare the variables in the same order as in the sut program.
      
      source/store.ml:
      source/polyhedron.ml:
         Fix a bug signaled by Claude where lurette was aborting
         when solving integer constraints. It was due to a bad
         implementation of is_point_in_poly which did not
         convert floats to ints when necessary (this bug was introduced
         rather recently when i provided my own version of
         is_point_in_poly... and of course i have non reg-test
         with integer constraints...).
      
      Project-Description: Lurette
      e1350c78
    • Erwan Jahier's avatar
      lurette 1.13 Fri, 12 Dec 2003 14:21:26 +0100 by jahier · 21390374
      Erwan Jahier authored
      Parent-Version:      1.12
      Version-Log:
      
      aesthetic changes in mli files so that
      the automatically generated doc looks ok.
      
      Project-Description: Lurette
      21390374
    • Erwan Jahier's avatar
      lurette 1.12 Tue, 09 Dec 2003 15:44:28 +0100 by jahier · c3ab83b6
      Erwan Jahier authored
      Parent-Version:      1.11
      Version-Log:
      
      Various fixes in the C scade parsing.
      
      Project-Description: Lurette
      c3ab83b6
    • Erwan Jahier's avatar
      lurette 1.11 Wed, 26 Nov 2003 08:52:30 +0100 by jahier · b3c5c602
      Erwan Jahier authored
      Parent-Version:      1.10
      Version-Log:
      
      No more side effects are done via the Env_state module.
      
      Project-Description: Lurette
      b3c5c602
    • Erwan Jahier's avatar
      lurette 1.9 Tue, 18 Nov 2003 11:17:12 +0100 by jahier · 67c17ef7
      Erwan Jahier authored
      Parent-Version:      1.8
      Version-Log:
      
      A lot (too much...) of changes, mainly renaming files, and types, and
      functions,
      change the interface, move from one module to another, etc.
      
      Bref, a lot of cleanning.
      
      Project-Description: Lurette
      67c17ef7
    • Erwan Jahier's avatar
      lurette 1.8 Wed, 22 Oct 2003 18:30:52 +0200 by jahier · a3e8a67b
      Erwan Jahier authored
      Parent-Version:      1.7
      Version-Log:
      
      Implement a new drawing policy in polyhedron.
      
      The idea is to base the base (which is not so expensive), and to draw
      in othogonal hypercube that is as small as possible until a point in
      the polyhedron is found.  However, it may happens in high dimension (>4)
      that the volume of the cube is much bigger. In that case, after a few tries,
      i call the old (unfair) method.
      
      polyDram.ml/mli:
         new module containing all the thing related to the drawing inside
         a convex polyhedron.
      
         Give up the idea of random walk (not clear what should be the default
         step and length in big dimension).
      
      Project-Description: Lurette
      a3e8a67b
    • Erwan Jahier's avatar
      lurette 1.7 Fri, 10 Oct 2003 14:01:43 +0200 by jahier · 79b18f3b
      Erwan Jahier authored
      Parent-Version:      1.6
      Version-Log:
      
      source/fair_bddd.ml: [new]
      source/fair_bddd.mli: [new]
         The same as bddd, but taking polyhedron volume into account.
      
      source/bddd.ml:
         It is useless to store both bdd and not bdd, because it is just
         so easy to recompute the sol nb of not bdd (by switchinf the
         pair of sol numbers!).
      
      source/lurettetop.ml:
      source/lurette.ml:
      source/luc_exe.ml:
         add a --compute-poly-volume option that makes lurette use
         fair_bddd instead of bddd.
      
      Project-Description: Lurette
      79b18f3b
    • Erwan Jahier's avatar
      lurette 1.4 Wed, 24 Sep 2003 15:48:02 +0200 by jahier · 5bf375a5
      Erwan Jahier authored
      Parent-Version:      1.3
      Version-Log:
      
      source/store.ml:
      source/solver.ml:
         use map instead as hashtbl for the range base store.
         It is much less error-prone (evil side effect!),
         leads to clearer code (no need to copy the store, which was particularly
         bad when done outside that module...), and is as efficient.
      
      Project-Description: Lurette
      5bf375a5
    • Erwan Jahier's avatar
      lurette 1.3 Tue, 23 Sep 2003 16:56:51 +0200 by jahier · d66521bd
      Erwan Jahier authored
      Parent-Version:      1.2
      Version-Log:
      
      source/store.ml:
      source/store.ml:
      source/solver.ml:
         Add a numeric thickness at the numeric level.
      
      Project-Description: Lurette
      d66521bd
    • Erwan Jahier's avatar
      lurette 1.1 Mon, 08 Sep 2003 16:02:09 +0200 by jahier · 4aed6dc3
      Erwan Jahier authored
      Parent-Version:      0.143
      Version-Log:
      
      Major change in the lucky syntax.
      
      Also add structured types (in lucky only, for the time being).
      
      Project-Description: Lurette
      4aed6dc3
    • Erwan Jahier's avatar
      lurette 0.132 Tue, 15 Apr 2003 17:38:13 +0200 by jahier · 557de1e9
      Erwan Jahier authored
      Parent-Version:      0.131
      Version-Log:
      
      Fix 3 bugs discovered by Pascal.
      
      one was due to a confusion with variable starting with _pre
      during formula_to_string.
      
      another one was in store.ml: i was not reinjecting the solutions
      when retriving values from store.substl
      
      a third one was due to the fact that I was not applying substitution
      rigth-to-left whereas I should.
      
      source/luc_exe.ml:
         Add 2 options --help and --step-number that let one set a bound
         on the number of steps to perform.
      
      source/env.ml:
      source/parse_env.ml:
         Provide a mean to give default values to pre variables.
      
      ihm/xlurette/xlurette_glade_main.ml
      ihm/xlurette/xlurette.glade:
        Add Combo boxes to browse lustre file nodes.
      
      source/
      Add the possibility to give default vales to pre variables.
      
      Project-Description: Lurette
      557de1e9
    • Erwan Jahier's avatar
      lurette 0.127 Mon, 24 Feb 2003 09:14:09 +0100 by jahier · 09d218f2
      Erwan Jahier authored
      Parent-Version:      0.126
      Version-Log:
      
      source/command_line_luc_exe.ml:
      source/command_line_luc_exe.mli:
      source/luc_exe.ml:
      source/store.ml:
      source/lurette.ml:
      source/lurettetop.ml:
         Put back the verteces draw. It was not such a great idea to remove
         after all...
      
         Also change verteces into vertices (oops...).
      
      Project-Description: Lurette
      09d218f2
    • Erwan Jahier's avatar
      lurette 0.126 Fri, 21 Feb 2003 18:49:20 +0100 by jahier · 5a165497
      Erwan Jahier authored
      Parent-Version:      0.125
      Version-Log:
      
      source/command_line_luc_exe.ml:
      source/command_line_luc_exe.mli:
      source/command_line.ml:
      source/command_line.mli:
      source/luc_exe.ml:
      source/lurette.ml:
      source/lurettetop.ml:
         Add the possibility to set the precision from lucky and lurette command
      lines.
      
         Remove the draw-vertices stuff.
      
      source/command_line_luc_exe.ml:
      source/command_line_luc_exe.mli:
      source/luc_exe.ml:
         Also add the --edges and the inside options for lucky.
      
      source/parse_env.ml:
        Allow to forget empty fields in the automata format.
      
      source/store.ml:
         Plug the new drawing heuristic
      
      Project-Description: Lurette
      5a165497
    • Erwan Jahier's avatar
      lurette 0.124 Fri, 21 Feb 2003 14:29:21 +0100 by jahier · 3ea22e95
      Erwan Jahier authored
      Parent-Version:      0.123
      Version-Log:
      
      Implement the polyhedron drawing primitives.
      
      source/util.ml.in:
         Change my_string_of_float so that it truncates floats 4 digits after the
      dot.
         The rational is that, for testing, we do not care the extra precision and
         too much precision may kill Polka. This precison ougth to be a parameter
         though.
      
      source/solver.ml:
      source/polyhedron.ml:
      source/store.ml:
          Implement the polyhedron drawing primitives.
      
      source/store.ml:
         Apply the substitutions in store.subtsl to the cstr to add.
      
      Project-Description: Lurette
      3ea22e95
    • Erwan Jahier's avatar
      lurette 0.123 Tue, 18 Feb 2003 09:49:47 +0100 by jahier · a13cc8e1
      Erwan Jahier authored
      Parent-Version:      0.122
      Version-Log:
      
      source/rnumsolver.mli:
      source/rnumsolver.ml:
      source/store.mli:
      source/store.ml:
         Rename rnumsolver into store.
      
      source/solver.ml:
      source/parse_env.ml:
         Do not explicitely handle dis-equalities but transform it using
         not and = instead. The rational for this change is to avoid code duplication
         (which was buggy !!!), but also to allow handle dis-equalities with
         the optimized algorithm I had for eqalities (for free) !
      
      Project-Description: Lurette
      a13cc8e1
    • Erwan Jahier's avatar
      lurette 0.122 Mon, 17 Feb 2003 14:50:50 +0100 by jahier · 78f40b65
      Erwan Jahier authored
      Parent-Version:      0.121
      Version-Log:
      
      When delayed constraints are remaining at bdd leaves, we were
      printing a message saying that one shoud use a polyhedron solver
      based instead.  Now, I really call such a polyhedron solver in order
      to solve those constraints.
      
      This means that I use to different solvers to handle interval and
      polyhedron of dimension > 1.
      
      It also means that the satisfiability of constraints over polyhedra
      is only checked at bdd leaves, which, in some circumstances, migth be
      inefficient. The point is that, if we chose to check the formula
      satisfiability during the bdd traversal, we take the risk that a very
      polyhedron is created whereas it was not necessary (because of
      forthcoming equalities). And creating polyhedron with big (>15)
      dimension simply runs forever, which is really bad.
      
      nb : drawing in poyhedron not yet plugged.
      
      source/polyhedron.ml:
      source/polyhedron.mli: (new files)
          misc functions over polyhedra. The rational for creating that file
          is that rnumsolver is already much too big (and should be splitted
          some more btw).
      
      source/solver.ml:
         One of the main change is that now, draw_in_bdd returns 2 stores.
         One is Range based, the other one is polyhedron-based.
      
      source/rnumsolver.mli:
      source/rnumsolver.ml:
         Add the function switch_to_polyhedron_representation which
         handle delayed constraint if necessary. It is meant to be called
         at bdd leaves.
      
      source/solver.ml:
      source/rnumsolver.ml:
         Also do not use split_store anymore but add_constraint. This allows
         a much more elegant and efficient implementation of draw_in_bdd and
         draw_in_bdd_eq (Now the store is buid when and only when a new branch
         should be tried because of backtracking). It is also more efficient.
      
      source/constraint.mli
      source/constraint.ml
      source/ne.mli
      source/ne.ml
         Implement get_vars that retreive the variables occuring in a contraint
         or an expression.
      
      Project-Description: Lurette
      78f40b65