aboutsummaryrefslogtreecommitdiffhomepage
path: root/pretyping/evarutil.ml
Commit message (Collapse)AuthorAge
* Generic support for open terms in tacticsGravatar herbelin2009-12-21
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | We renounced to distribute evars to constr and bindings and to let tactics do the merge. There are now two disciplines: - the general case is that the holes in tactic arguments are pushed to the general sigma of the goal so that tactics have no such low-level tclEVARS, Evd.merge, or check_evars to do: - what takes tclEVARS and check_evars in charge is now a new tactical of name tclWITHHOLES (this tactical has a flag to support tactics in either the "e"- mode and the non "e"- mode); - the merge of goal evars and holes is now done generically at interpretation time (in tacinterp) and as a side-effect it also anticipates the possibility to refer to evars of the goal in the arguments; - with this approach, we don't need such constr/open_constr or bindings/ebindings variants and we can get rid of all ugly inj_open-style coercions; - some tactics however needs to have the exact subset of holes known; this is the case e.g. of "rewrite !c" which morally reevaluates c at each new rewriting step; this kind of tactics still receive a specific sigma around their arguments and they have to merge evars and call tclWITHHOLES by themselves. Changes so that each specific tactics can take benefit of this generic support remain to be done. git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@12603 85f007b7-540e-0410-9357-904b9bb8a0f7
* In "progress", extending the set of evars w/o solving an existing one isGravatar herbelin2009-12-21
| | | | | | | | | no longer considered a progress (this prepares generally having tactics with arguments that contains holes that are added to the goal sigma). Incidentally, made that "clear" now restricts evars only if the restriction is really needed. git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@12602 85f007b7-540e-0410-9357-904b9bb8a0f7
* Promote evar_defs to evar_map (in Evd)Gravatar glondu2009-11-11
| | | | git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@12502 85f007b7-540e-0410-9357-904b9bb8a0f7
* A bit of cleaning around name generation + creation of dedicated file namegen.mlGravatar herbelin2009-11-09
| | | | git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@12485 85f007b7-540e-0410-9357-904b9bb8a0f7
* Reverted an incorrect code simplification in function status_changedGravatar herbelin2009-11-02
| | | | | | | (rewrote it differently to avoid being trapped again). git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@12460 85f007b7-540e-0410-9357-904b9bb8a0f7
* Take constraints into account in the "instantiate" tacticGravatar herbelin2009-10-30
| | | | | | | | (building an example that needs it to do). git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@12449 85f007b7-540e-0410-9357-904b9bb8a0f7
* Integrate a few improvements on typeclasses and Program from the equations ↵Gravatar msozeau2009-10-28
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | branch and remove equations stuff which moves to a separate plugin. Classes: - Ability to define classes post-hoc from constants or inductive types. - Correctly rebuild the hint database associated to local hypotheses when they are changed by a [Hint Extern] in typeclass resolution. Tactics and proofs: - Change [revert] so that it keeps let-ins (but not [generalize]). - Various improvements to the [generalize_eqs] tactic to make it more robust and produce the smallest proof terms possible. Move [specialize_hypothesis] in tactics.ml as it goes hand in hand with [generalize_eqs]. - A few new general purpose tactics in Program.Tactics like [revert_until] - Make transitive closure well-foundedness proofs transparent. - More uniform testing for metas/evars in pretyping/unification.ml (might introduce a few changes in the contribs). Program: - Better sorting of dependencies in obligations. - Ability to start a Program definition from just a type and no obligations, automatically adding an obligation for this type. - In compilation of Program's well-founded definitions, make the functional a separate definition for easier reasoning. - Add a hint database for every Program populated by [Hint Unfold]s for every defined obligation constant. git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@12440 85f007b7-540e-0410-9357-904b9bb8a0f7
* Delete trailing whitespaces in all *.{v,ml*} filesGravatar glondu2009-09-17
| | | | git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@12337 85f007b7-540e-0410-9357-904b9bb8a0f7
* Made unification patch 12268 even more ad hoc (if evars remain in aGravatar herbelin2009-08-13
| | | | | | | term, its type is not the smallest one - actually, we would have to reduce the term too but it would be more costly). git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@12274 85f007b7-540e-0410-9357-904b9bb8a0f7
* Relatively ad hoc fix to an ill-typed instantiation bug in typeGravatar herbelin2009-08-11
| | | | | | | | | | | inference (see file failure/evar1.v) + fix of some CUMUL problems that were in the wrong direction. We assume for the fix that ill-typed unification problems come from subtyping where we don't know yet if a coercion has to be inserted or not, and hence are of the CUMUL form. More on suspending problems of the form ?n <= Type or Prop <= ?n has to be done yet. git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@12268 85f007b7-540e-0410-9357-904b9bb8a0f7
* Don't use recent ocaml tolerance for pattern "ProjectVar _" whenGravatar herbelin2009-07-08
| | | | | | | ProjectVar is a constant constructor (anyway, use of _ for constant constructor was here by mistake). git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@12232 85f007b7-540e-0410-9357-904b9bb8a0f7
* Add heuristic based on non-linearity of evars to determine whether anGravatar herbelin2009-07-08
| | | | | | | | evar is dependent or not (solve bug #2123). git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@12228 85f007b7-540e-0410-9357-904b9bb8a0f7
* Fixing bug 2129 (evar introduced to remember an ambiguous projectionGravatar herbelin2009-07-08
| | | | | | | | | had wrong type). At least two problems remain: - projection involving evar should check the type are compatible, - instances of filtered evars should not be shrinked as all values are needed to ensure the well-typedness of the instanciated restricted evars. git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@12226 85f007b7-540e-0410-9357-904b9bb8a0f7
* Jolification : tentative de supprimer les "( evd)" et associés quiGravatar aspiwack2009-07-07
| | | | | | | | | traînaient un peu partout dans le code depuis la fusion d'evar_map et evar_defs. Début du travail d'uniformisation des noms donnés aux evar_defs à travers le code. git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@12224 85f007b7-540e-0410-9357-904b9bb8a0f7
* documented a bug of pattern unification with metasGravatar barras2009-06-22
| | | | git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@12203 85f007b7-540e-0410-9357-904b9bb8a0f7
* A try at using sort variables during unification. Instead of refreshingGravatar msozeau2009-05-23
| | | | | | | | | | | | | universes as usual, we add the new universes to the sort constraints and do unification modulo those ([constr_unify_with_sorts]): this allows to instanciate Type i with Prop for example and keep track of it. The sort constraints are thrown away at the end of unification for the moment, but we can detect inconsistencies during unification. Make unification more symmetric as well w.r.t. substitution of defined metas. git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@12137 85f007b7-540e-0410-9357-904b9bb8a0f7
* - Fixing #2090 (occur check missing when trying to solve evar-evar equation).Gravatar herbelin2009-04-25
| | | | | | | | | | - Adding test file related to commit 12080 (bug #2091). - Cleaning old parsing stuff from 8.0. - Support for camlp5 in base_include. git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@12106 85f007b7-540e-0410-9357-904b9bb8a0f7
* Some dead code removal + cleanupsGravatar letouzey2009-04-08
| | | | | | | | | | | | | | This commit concerns about the first half of the useless code mentionned by Oug for coqtop (without plugins). For the moment, Oug is used in a mode where any elements mentionned in a .mli is considered to be precious. This already allows to detect and remove about 600 lines, and more is still to come. Among the interesting points, the type Entries.specification_entry and its constructors SPExxx were never used. Large parts of cases.ml (and hence subtac_cases.ml) were also useless. git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@12069 85f007b7-540e-0410-9357-904b9bb8a0f7
* - Fixing bug #2084 (unification not checking sort constraints), hopingGravatar herbelin2009-04-08
| | | | | | | | | | | | | it does not cause a time penalty. - Removing of get_type_of_with_meta made possible by the evar_defs/evar_map merge. - Adding unfolding of Meta in reductionops (this assumes that reduction does not move Metas across binders...) - Renaming newly created fold_map_rel_context into map_rel_context_in_env. git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@12061 85f007b7-540e-0410-9357-904b9bb8a0f7
* Fix obvious bug in evars_of_named_context.Gravatar msozeau2009-04-03
| | | | git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@12049 85f007b7-540e-0410-9357-904b9bb8a0f7
* Fixes to make Ynot compile with the trunk:Gravatar msozeau2009-03-20
| | | | | | | | | | | | | | | | - Choose one of the possible instances of an evar when considering remaining unification constraints: otherwise we just do nothing and some evars remain uninstantiated. - Normalise the goal w.r.t. evars before subst, to avoid a double vision problem: the substituted variable appears only in an instance of an evar and when we try the rewrite it has been substituted making the dependency disappear. - Hack to correcly handle let-in annotations which are internalized as casts: they're really typing constraints. Shouldn't we just change the AST to have the type at rawconstr let-in nodes? git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@11998 85f007b7-540e-0410-9357-904b9bb8a0f7
* Backtrack sur la mémoïsation de nf_evar.Gravatar aspiwack2009-03-04
| | | | | | | | | | | | | | | | L'expérience prouve que ce n'est pas franchement concluant. On peut se risquer à une explication : - nf_evar, version mémoïsée n'est pas tail recursive - On retarde la substitution des hypothèses de l'evar en échange de faire moins de substitutions d'evars. Intuitivement c'est intéressant seulement si il y a plus de substitutions d'evar dupliquées que d'hypothèses dupliquées. Ce qui ne doit pas être le cas (ne serait-ce que parce que dupliquer une evar duplique aussi ses variables libres). This reverts commit 066a564021788e995eb166ad6ed6e55611d6f593. git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@11958 85f007b7-540e-0410-9357-904b9bb8a0f7
* =?utf-8?q?Tentative=20d'optimisation=20(en=20temps)=20sur=20[nf=5Fevar]=20et ↵Gravatar aspiwack2009-02-27
| | | | | | | | | | | | | | | | | | | | =20[whd=5Fevar]=20:=20les =20evar=5Fdefs=20gardent=20un=20cache=20des=20appels=20pr=C3=A9c=C3=A9dents.=20Le=20d=C3=A9faut=20de=20la =20m=C3=A9thodologie=20est=20que=20=C3=A7a=20int=C3=A9ragit=20assez=20mal=20avec=20la=20substitution=20des =20hypoth=C3=A8ses=20de=20l'evar=20(qui=20n'est=20pas=20mise=20en=20cache).=20En=20particulier=20les =20deux=20fonctions=20ne=20sont=20plus=20r=C3=A9cursives=20terminales.=20De=20plus=20un=20appel=20=C3=A0 =20l'une=20des=20deux=20fera=20n=C3=A9cessairement=20un=20parcours=20du=20terme=20pour=20appliquer =20la=20substitution.?= MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 8bit D'un point de vue de l'effet observer, ça a un effet assez léger sur le trunk, je suis curieux de voir les effets sur les contribs. git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@11950 85f007b7-540e-0410-9357-904b9bb8a0f7
* On passe les last_mods (un des champs de Evd.evar_defs) de listGravatar aspiwack2009-02-20
| | | | | | | | | | d'entiers à ensemble d'entier. A première vue l'optimisation a un effet négligeable. Mais vu l'utilisation qui est fait des des last_mods (des mem à répétition). C'est plus raisonnable ainsi. git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@11942 85f007b7-540e-0410-9357-904b9bb8a0f7
* On remplace evar_map par evar_defs (seul evar_defs est désormais exporté Gravatar aspiwack2009-02-19
| | | | | | | | | | | | | | | | | | | | | | | | | | | | par Evd). Ça s'accompagne de quelques autres modifications de l'interface (certaines fonctions étaient des doublons, ou des conversions entre evar_map et evar_defs). J'ai modifié un peu la structure de evd.ml aussi, pour éviter des fonctions redéfinies deux fois (i.e. définies trois fois !), j'ai introduit des sous-modules pour les différentes couches. Il y a à l'heure actuelle une pénalité en performance assez sévère (due principalement à la nouvelle mouture de Evd.merge, si mon diagnostique est correct). Mais fera l'objet de plusieurs optimisations dans les commits à venir. Un peu plus ennuyeux, la test-suite du mode déclaratif ne passe plus. Un appel de Decl_proof_instr.mark_as_done visiblement, je suis pour l'instant incapable de comprendre ce qui cause cette erreur. J'espère qu'on pourra le déterminer rapidement. Ce commit est le tout premier commit dans le trunk en rapport avec les évolution futures de la machine de preuve, en vue en particulier d'obtenir un "vrai refine". git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@11939 85f007b7-540e-0410-9357-904b9bb8a0f7
* memoized is_ground_envGravatar barras2009-02-09
| | | | git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@11893 85f007b7-540e-0410-9357-904b9bb8a0f7
* pushed evar reduction in kernelGravatar barras2009-02-06
| | | | git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@11889 85f007b7-540e-0410-9357-904b9bb8a0f7
* From v8.2 to trunk:Gravatar herbelin2009-02-06
| | | | | | | | | - commit 11871 (Miller's pattern detection bug) + a corresponding test - commit 11883 (.ml4 to .cmxs in coq_makefile) git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@11884 85f007b7-540e-0410-9357-904b9bb8a0f7
* - Fixing bug 1891 (abusive instantiations of evar arguments inGravatar herbelin2009-01-20
| | | | | | | | | evar-evar problems). - Fixing target "make programs". git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@11817 85f007b7-540e-0410-9357-904b9bb8a0f7
* Backporting from v8.2 to trunk:Gravatar herbelin2009-01-18
| | | | | | | | | | - Filtering of doc compilation messages (11793,11795,11796). - Fixing bug #1925 and cleaning around bug #1894 (11796, 11801). - Adding some tests. git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@11802 85f007b7-540e-0410-9357-904b9bb8a0f7
* DISCLAIMERGravatar puech2009-01-17
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | ========== This big patch is commited here with a HUGE experimental tag on it. It is probably not a finished job. The aim of committing it now, as agreed with Hugo, is to get some feedback from potential users to identify more clearly the directions the implementation could take. So please feel free to mail me any remarks, bug reports or advices at <puech@cs.unibo.it>. Here are the changes induced by it : For the user ============ * Search tools have been reimplemented to be faster and more general. Affected are [SearchPattern], [SearchRewrite] and [Search] (not [SearchAbout] yet). Changes are: - All of them accept general constructions, and previous syntactical limitations are abolished. In particular, one can for example [SearchPattern (nat -> Prop)], which will find [isSucc], but also [le], [gt] etc. - Patterns are typed. This means that you cannot search mistyped expressions anymore. I'm not sure if it's a good or a bad thing though (especially regarding coercions)... * New tool to automatically infer (some) Record/Typeclasses instances. Usage : [Record/Class *Infer* X := ...] flags a record/class as subject to instance search. There is also an option to activate/deactivate the search [Set/Unset Autoinstance]. It works by finding combinations of definitions (actually all kinds of objects) which forms a record instance, possibly parameterized. It is activated at two moments: - A complete search is done when defining a new record, to find all possible instances that could have been formed with past definitions. Example: Require Import List. Record Infer Monoid A (op:A->A->A) e := { assoc : forall x y z, op x (op y z) = op (op x y) z; idl : forall x, x = op x e ; idr : forall x, x = op e x }. new instance Monoid_autoinstance_1 : (Monoid nat plus 0) [...] - At each new declaration (Definition, Axiom, Inductive), a search is made to find instances involving the new object. Example: Parameter app_nil_beg : forall A (l:list A), l = nil ++ l. new instance Build_Monoid_autoinstance_12 : (forall H : Type, Monoid (list H) app nil) := (fun H : Type => Build_Monoid (list H) app nil ass_app (app_nil_beg H) (app_nil_end H)) For the developper ================== * New yet-to-be-named datastructure in [lib/dnet.ml]. Should do efficient one-to-many or many-to-one non-linear first-order filtering, faster than traditional methods like discrimination nets (so yes, the name of the file should probably be changed). * Comes with its application to Coq's terms [pretyping/term_dnet.ml]. Terms are represented so that you can search for patterns under products as fast as you would do not under products, and facilities are provided to express other kind of searches (head of application, under equality, whatever you need that can be expressed as a pattern) * A global repository of all objects defined and imported is maintained [toplevel/libtypes.ml], with all search facilities described before. * A certain kind of proof search in [toplevel/autoinstance.ml]. For the moment it is specialized on finding instances, but it should be generalizable and reusable (more on this in a few months :-). The bad news ============ * Compile time should increase by 0 to 15% (depending on the size of the Requires done). This could be optimized greatly by not performing substitutions on modules which are not functors I think. There may also be some inefficiency sources left in my code though... * Vo's also gain a little bit of weight (20%). That's inevitable if I wanted to store the big datastructure of objects, but could also be optimized some more. git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@11794 85f007b7-540e-0410-9357-904b9bb8a0f7
* Completed 11745 (move of jprover to user contribs) and cleaned 11743Gravatar herbelin2009-01-05
| | | | | | | | (detection of Miller's pattern) git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@11748 85f007b7-540e-0410-9357-904b9bb8a0f7
* Fixed bugs #2001 (search_guard was overwriting the guard index givenGravatar herbelin2009-01-04
| | | | | | | | by user) and #2017 (unification pattern test too crude leading to regression wrt to 8.1). git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@11743 85f007b7-540e-0410-9357-904b9bb8a0f7
* fixed problem with r11612Gravatar barras2008-11-21
| | | | git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@11614 85f007b7-540e-0410-9357-904b9bb8a0f7
* - Fixed bug 1968 (inversion failing due to a Not_found bug introduced inGravatar herbelin2008-11-09
| | | | | | | | | | | | Evarutil.check_and_clear_in_constr in V8.2 revision 11309 and trunk revision 11300). - Improved various error messages related to inversion, evars and case analysis (including the removal of the obsolete dependent/non dependent distinction). git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@11561 85f007b7-540e-0410-9357-904b9bb8a0f7
* Fix universe problem appearing ConCaT using the existing infrastructure forGravatar msozeau2008-11-07
| | | | | | | | | | declaring additional conversion problems when unifying the type of an evar instance and the evar's declared type. Fix the corresponding conversion heuristic which failed due to (misplaced?) assertions when faced with general conversion problems. git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@11549 85f007b7-540e-0410-9357-904b9bb8a0f7
* Fix in the unification algorithm using evars: unify types of evarGravatar msozeau2008-11-05
| | | | | | | | | | | | instances and the corresponding evar's type if it contains existentials to avoid dangling evars. No noticeable performance impact (at least on the stdlib). Subsumes (and fixes) the (broken) fix in unification.ml that was previously patched by M. Puech. Improve error messages related to existential variables and type classes. git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@11543 85f007b7-540e-0410-9357-904b9bb8a0f7
* Partial fix for bug #1948: recompute order of dependencies betweenGravatar msozeau2008-09-25
| | | | | | | | | evars at the end of unification as later evars can refer to previous ones. This removes the assumption that evars are already ordered in eterm's code. git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@11419 85f007b7-540e-0410-9357-904b9bb8a0f7
* Évolutions diverses et variées.Gravatar herbelin2008-08-04
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | - Correction divers messages d'erreur - lorsque rien à réécrire dans une hyp, - lorsqu'une variable ltac n'est pas liée, - correction anomalie en présence de ?id dans le "as" de induction, - correction mauvais env dans message d'erreur de unify_0. - Diverses extensions et améliorations - "specialize" : - extension au cas (fun x1 ... xn => H u1 ... un), - renommage au même endroit. - "assert" et "pose proof" peuvent réutiliser la même hyp comme "specialize". - "induction" - intro des IH toujours au sommet même si induction sur var quantifiée, - ajout d'un hack pour la reconnaissance de schémas inductifs comme N_ind_double mais il reste du boulot pour reconnaître (et/ou réordonner) les composantes d'un schéma dont les hypothèses ne sont pas dans l'ordre standard, - vérification de longueur et éventuelle complétion des intropatterns dans le cas de sous-patterns destructifs dans induction (par exemple "destruct n as [|[|]]" sur "forall n, n=0" ne mettait pas le n dans le contexte), - localisation des erreurs d'intropattern, - ajout d'un pattern optionnel après "as" pour forcer une égalité et la nommer (*). - "apply" accepte plusieurs arguments séparés par des virgules (*). - Plus de robustesse pour clear en présence d'evars. - Amélioration affichage TacFun dans Print Ltac. - Vieux pb espace en trop en tête d'affichage des tactiques EXTEND résolu (incidemment, ça remodifie une nouvelle fois le test output Fixpoint.v !). - Fusion VTactic/VFun dans l'espoir. - Mise en place d'un système de trace de la pile des appels Ltac (tout en préservant certains aspects de la récursivité terminale - cf bug #468). - Tactiques primitives - ajout de "move before" dans les tactiques primitives et ajout des syntaxes move before et move dependent au niveau utilisateur (*), - internal_cut peuvent faire du remplacement de nom d'hypothèse existant, - suppression de Intro_replacing et du code sous-traitant - Nettoyage - Suppression cible et fichiers minicoq non portés depuis longtemps. (*) Extensions de syntaxe qu'il pourrait être opportun de discuter git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@11300 85f007b7-540e-0410-9357-904b9bb8a0f7
* Modification rapide du message d'erreur lorsqu'un _ ne peut êtreGravatar herbelin2008-07-18
| | | | | | | | effacé dans un intro-pattern (suggéré par ssreflect). git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@11235 85f007b7-540e-0410-9357-904b9bb8a0f7
* Uniformisation du format des messages d'erreur (commencent par uneGravatar herbelin2008-07-17
| | | | | | | | | | majuscule - si pas un ident ou un terme - et se terminent par un point). Restent quelques utilisations de "error" qui sont liées à des usages internes, ne faudrait-il pas utiliser des exceptions plus spécifiques à la place ? git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@11230 85f007b7-540e-0410-9357-904b9bb8a0f7
* Correction d'un bug dans l'analyse des contraintes non résoluesGravatar herbelin2008-06-29
| | | | | | | | | | | | (report de 11190 de v8.2 vers trunk). Le code fautif n'était en fait plus utilisé car les contraintes de la forme ?n[..,x,..,x,..] := x que ce code analysait sont finalement résolues par une heuristique de consider_remaining_univ_constraints consistant à reproduire le comportement de la 8.1. git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@11192 85f007b7-540e-0410-9357-904b9bb8a0f7
* Various improvements in handling of evars in general and typingGravatar msozeau2008-06-21
| | | | | | | | | | | | | | | | | | | | | | | constraints in Program: - normalize types and defs of local fixpoints before checking the guardness condition to avoid having to give type annotations, e.g: << Definition fold (A B : Set) (f : B -> A -> B) : B -> tree A -> B := fix aux acc t := match t with | Leaf x => f acc x | Node l => fold_left aux l acc end. >> - add new inh_coerce_to_prod to allow coercion of the typing constraint to a product before trying to split it. It's a noop in standard mode, and forgets subsets in Program. Allows to typecheck: << (λ x, x) : { f : nat -> nat | ... } >>. - Better handling of the "abstract" typing constraints in Program. - Correctly normalize w.r.t. evars. the tycon given by users in Program. git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@11156 85f007b7-540e-0410-9357-904b9bb8a0f7
* Correction bug 1878 (utilisation de extend_evar déplacée là où uneGravatar herbelin2008-06-14
| | | | | | | | | | restriction du contexte était attendue) + suppression warning + amélioration affichage en cas de clause "at" incorrecte + report commit 11121 (correction bug 1367) de la 8.2 vers le trunk. git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@11128 85f007b7-540e-0410-9357-904b9bb8a0f7
* - Correction bug highlighting "Module" dans CoqideGravatar herbelin2008-05-28
| | | | | | | | | - Divers code mort (evarutil.ml, Bvector.v) - MAJ perf-analysis git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@11004 85f007b7-540e-0410-9357-904b9bb8a0f7
* Correction d'un bug de l'unification pattern qui oubliait d'expanserGravatar herbelin2008-05-20
| | | | | | | | les alias avant de déclarer qu'une evar n'était appliquée qu'à des variables. git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@10956 85f007b7-540e-0410-9357-904b9bb8a0f7
* Mise en place d'un algorithme d'inversion des contraintes de type lorsGravatar herbelin2008-05-05
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | du filtrage. Cela permet de détecter les cas impossibles et de simuler les contraintes d'inversion exprimables sous la forme d'un assignement des arguments du constructeurs (cf le cas de Vtail dans Bvector.v). Si l'on filtre sur t:I u1 .. un, et que chaque ui a la forme vi(wi) avec vi composé uniquement de constructeurs, et que le résultat final est P(w1,...,wn) (qui est éventuellement lui-même une evar) alors on construit le prédicat Q:=fun x1 .. xn y => match x1 .. xn y with | v1(z) .. vn(z) t => P(z) | _ .. _ _ => ?evar-speciale-cas-impossible end qui vérifiera bien que Q u1 .. un = P(w1,..,wp). En raison de limitations de l'unification (on aurait besoin d'eta conversion pour résoudre des problèmes du genre "terme rigide == match x with _ => ?evar end", et besoin d'instanciation par constructeurs pour des cas comme "A(y) = match ?evar with C x => A(x) end"), je n'ai pas réussi à traiter le cas général. Aussi, on adopte une stratégie pragmatique consistant à tester plusieurs prédicats possibles : - si un type final est donné, on essaie d'abord l'algorithme de Matthieu et sinon le nouvel algorithme (permet par exemple de traiter certains cas d'élimination dépendante de Bvector.v), - s'il n'y a pas de type final, on essaie d'abord le nouvel algo et sinon, on essaie avec un prédicat sans dépendance (permet de traiter des cas compliqués comme celui de par cas sur I' dans le fichier Case13.v de la test-suite). Dans la pratique, il y a beaucoup de changement dans le code de compile_case. - Par exemple, la compilation est maintenant toujours appelé avec un prédicat (là où l'on pouvait avoir None, on a maintenant toujours au moins une evar). - En revanche, le membre droit des clauses est maintenant optionnel. Si c'est None, c'est qu'on se trouve dans le cas d'une branche impossible au moment du calcul du prédicat de retour. - Aussi, on renonce aux PrLetIn et PrProd dans l'expression du predicat de retour mais il faut savoir que c'est maintenant la liste des tomatchs qui spécifie le contexte exact dans lequel le prédicat de retour est bien typé. - Et d'autres... git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@10883 85f007b7-540e-0410-9357-904b9bb8a0f7
* Correction bug 1835 + correction bug occur-check résultant en unGravatar herbelin2008-04-18
| | | | | | | | | | | | "cannot define an evar twice" --Cette ligne, et les suivantes ci-dessous, seront ignorées-- M tacred.ml M evarutil.ml git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@10815 85f007b7-540e-0410-9357-904b9bb8a0f7
* Modifications diverses et variées :Gravatar herbelin2008-03-30
| | | | | | | | | | | | | - Nouvel essai de prise en compte unfold dans apply (unification.ml) - Correction bug commit précédent (constrintern.ml) - Correction bug d'explication des evars non résolues (evarutil.ml) - Fenêtre de query coqide plus large (command_windows.ml) - Orthographe tauto (tauto.ml4) - Crédits (ConstructiveEpsilon.v) git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@10731 85f007b7-540e-0410-9357-904b9bb8a0f7
* Une passe sur l'unification des evars (suite aux commits 10124, 10125, 10145)Gravatar herbelin2008-03-10
| | | | | | | | | - Correction bug des filtres dans define_evar_as_abstraction - Nettoyage, documentation et réorganisations diverses git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@10650 85f007b7-540e-0410-9357-904b9bb8a0f7
generated by cgit v1.2.3 (git 2.39.1) at 2024-07-25 23:14:50 +0000