aboutsummaryrefslogtreecommitdiff
path: root/toplevel/mltop.ml4
AgeCommit message (Collapse)Author
2009-10-21This big commit addresses two problems:soubiran
1- Management of the name-space in a modular development / sharing of non-logical objects. 2- Performance of atomic module operations (adding a module to the environment, subtyping ...). 1- There are 3 module constructions which derive equalities on fields from a module to another: Let P be a module path and foo a field of P Module M := P. Module M. Include P. ... End M. Declare Module K : S with Module M := P. In this 3 cases we don't want to be bothered by the duplication of names. Of course, M.foo delta reduce to P.foo but many non-logical features of coq do not work modulo conversion (they use eq_constr or constr_pat object). To engender a transparent name-space (ie using P.foo or M.foo is the same thing) we quotient the name-space by the equivalence relation on names induced by the 3 constructions above. To implement this, the types constant and mutual_inductive are now couples of kernel_names. The first projection correspond to the name used by the user and the second projection to the canonical name, for example the internal name of M.foo is (M.foo,P.foo). So: ************************************************************************************* * Use the eq_(con,mind,constructor,gr,egr...) function and not = on names values * ************************************************************************************* Map and Set indexed on names are ordered on user name for the kernel side and on canonical name outside. Thus we have sharing of notation, hints... for free (also for a posteriori declaration of them, ex: a notation on M.foo will be avaible on P.foo). If you want to use this, use the appropriate compare function defined in name.ml or libnames.ml. 2- No more time explosion (i hoppe) when using modules i have re-implemented atomic module operations so that they are all linear in the size of the module. We also have no more unique identifier (internal module names) for modules, it is now based on a section_path like mechanism => we have less substitutions to perform at require, module closing and subtyping but we pre-compute more information hence if we instanciate several functors then we have bigger vo. Last thing, the checker will not work well on vo(s) that contains one of the 3 constructions above, i will work on it soon... git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@12406 85f007b7-540e-0410-9357-904b9bb8a0f7
2009-09-29Add support for Local Declare ML Moduleglondu
Instead of failing with some obscure error message *after* loading the module, accept Local Declare ML Module with the usual semantics. git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@12366 85f007b7-540e-0410-9357-904b9bb8a0f7
2009-09-17Replace unprotected call to where_in_path by find_file_in_pathglondu
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@12340 85f007b7-540e-0410-9357-904b9bb8a0f7
2009-09-17Remove useless Liboject.export_function fieldglondu
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@12338 85f007b7-540e-0410-9357-904b9bb8a0f7
2009-09-17Delete trailing whitespaces in all *.{v,ml*} filesglondu
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@12337 85f007b7-540e-0410-9357-904b9bb8a0f7
2009-08-13Death of "survive_module" and "survive_section" (the first one washerbelin
only used to allow a module to be ended before the summaries were restored what can be solved by moving upwards the place where the summaries are restored). git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@12275 85f007b7-540e-0410-9357-904b9bb8a0f7
2009-08-06- Cleaning phase of the interfaces of libnames.ml and nametab.mlherbelin
(uniformisation of function names, classification). One of the most visible change is the renaming of section_path into full_path (the use of name section was obsolete due to the module system, but I don't know if the new name is the best chosen one - especially it remains some "sp" here and there). - Simplification of the interface of classify_object (first argument dropped). - Simplification of the code for vernac keyword "End". - Other small cleaning or dead code removal. git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@12265 85f007b7-540e-0410-9357-904b9bb8a0f7
2009-03-17- configure: affiche si le natdynlink est positionnebarras
- coq_makefile: utilise Coq_config pour avoir la liste des contribs - mltop: normalisation des noms de modules ML (majuscule) - Makefiles: introduction de fichiers %-mod.ml qui se chargent de faire les declarations de modules ML d'un plugin git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@11987 85f007b7-540e-0410-9357-904b9bb8a0f7
2009-03-14Better mechanism for loading initial pluginsletouzey
Instead of dirty hacks in toplevel/coqtop.ml, we simply add some Declare ML Module in Prelude.v. Gain: now that coqdep is clever enough, dependencies are automatic, and we can simplify the Makefile quite a lot: no more references to INITPLUGINSBEST and the like. Besides, mltop.ml4 can also be simplified a lot: by giving $(CONTRIBSTATIC) to coqmktop instead of contrib.cma, now coqtop is aware that it already contain the static plugins (or not), and subsequent ML Module are ignored correctly without us having to do anything :-) git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@11979 85f007b7-540e-0410-9357-904b9bb8a0f7
2009-02-03Allow to turn contrib/subtac into a (nat)dynlink'able pluginletouzey
Main issue was declare_summary being triggered too late in subtac_obligations, hence the associated init_function was _not_ being done by Lib.init(). Fixed for the moment by an ad-hoc launch of this init_function in subtac_obligations. In other plugins, this issue doesn't appear, since init_function is mostly putting back some empty set into a reference that was initially empty. No need to really run init_function in this case. For future plugins, we will nonetheless have to be careful about that. Of course, the (ref Obj.magic) was not exactly helpful in debugging this matter, see http://caml.inria.fr/mantis/view.php?id=4707 As said by Xavier, naughty naughty boys... git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@11877 85f007b7-540e-0410-9357-904b9bb8a0f7
2009-01-18Backporting from v8.2 to trunk:herbelin
- 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
2009-01-05Completed 11745 (move of jprover to user contribs) and cleaned 11743herbelin
(detection of Miller's pattern) git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@11748 85f007b7-540e-0410-9357-904b9bb8a0f7
2008-12-24- coq_makefile: target install now respects the original tree structureherbelin
of the archive to install in coq user-contrib installation directory. - Relaxed the validity check on identifiers from an error to a warning. - Added a filtering option to Print LoadPath. - Support for empty root in option -R. - Better handling of redundant paths in ml loadpath. - Makefile's: Added target initplugins and added initplugins to coqbinaries. git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@11713 85f007b7-540e-0410-9357-904b9bb8a0f7
2008-12-17Sequel of 11697: repair coqtop.byte when contribs are statically linked ↵letouzey
(+minor improvements) git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@11692 85f007b7-540e-0410-9357-904b9bb8a0f7
2008-12-16Take advantage of natdynlink when available: almost all contribs become ↵letouzey
loadable plugins - Any contrib foo leads to contrib/foo/foo_plugin.cmxs (and .cma for bytecode). - Features that were available without any Require are now loaded systematically when launching coqtop (see Coqtop.load_initial_plugins): extraction, jprover, cc, ground, dp, recdef, xml - The other plugins are loaded when a corresponding Require is done: quote, ring, field, setoid_ring, omega, romega, micromega, fourier - I experienced a crash (segfault) while turning subtac into a plugin, so this one stays statically linked into coqtop for now - When the ocaml version doesn't support natdynlink, or if "-natdynlink no" is explicitely given to configure, coqtop is statically linked with all of the above code as usual. Some messages [Ignore ML file Foo_plugin] may appear. - How should coqdep handle a "Declare ML Module "foo"" if foo is an archive and not a ml file ? For now, we suppose that the foo.{cmxs,cma} are at the same location as the .v during the build, but can be moved later in any place of the ml loadpath. This is clearly an experimentation. Feedback most welcome... git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@11687 85f007b7-540e-0410-9357-904b9bb8a0f7
2008-11-23- Synchronized subst_object with load_object (load_and_subst_objects)herbelin
and set Declare ML Module as a regular substitutive object so that Declare ML Module is treated at the right place in the order of appearance of substitutive declarations of a required module. - Note: The full load/import mechanism for modules is not so clear: the Require part of a Require Import inside a module is set outside the module at module closing but the Import part remains inside (why not to put the "special" objects in the module too?); moreover the "substitute" and "keep" objects of a module are desynchronised at module closing (is that really harmless/necessary?). - Treatment of .cmxs targets in coq_makefile and in coqdep. - Better make clean in coq_makefile generated makefiles. git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@11623 85f007b7-540e-0410-9357-904b9bb8a0f7
2008-11-22- Fixed minor bug #1994 in the tactic chapter of the manual [doc]herbelin
- Improved warning when found several path to the same file in path [mltop.ml4, system.ml] - Add support for "rewrite" on specific equality to true (i.e. eq_true) [Datatypes.v, tactics] PS: compilation test made over 11611 to shunt the archive-breaking 11612 and 11614 git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@11617 85f007b7-540e-0410-9357-904b9bb8a0f7
2008-10-29Remove calls to Dynlink.add_{interfaces,available_units} altogetherglondu
According to OCaml's Changes, all modules loaded so far (and statically linked) are available to dynamically loaded modules (since 3.07). Therefore, these calls seem irrelevant now. git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@11522 85f007b7-540e-0410-9357-904b9bb8a0f7
2008-10-28Native "Declare ML Module" when possibleglondu
Dynlink.add_{interfaces,available_units} are deprecated and not implemented natively. Currently, native "Declare ML Module" doesn't work because of this. Dynlink-related should be switched to the API introduced in OCaml 3.07. git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@11518 85f007b7-540e-0410-9357-904b9bb8a0f7
2008-07-17Uniformisation du format des messages d'erreur (commencent par uneherbelin
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
2008-06-29Lissage de la gestion des chemins de chargement de fichiers :herbelin
- Option -R fait maintenant des Import à tous les niveaux de la hiérarchie de répertoires. Par exemple, Require "Init.Wf" marche. - Option -I rend maintenant possible l'accès aux sous-répertoires via les noms qualifiés. Ainsi -R est exactement comme -I sauf qu'il rend récursivement visibles les noms non qualifiés. - Ajout option -I dir -as coqdir, et par symétrie, -R dir -as coqdir. - Ajout option -exclude-dir pour exclure certains sous-répertoires de la descente récursive de -R. - Amélioration message de localisation pour fichiers venant d'un "state". - Adaptation du checker (et ajout du test check_coq_overwriting qui semblait involontairement oublié dans l'option -R). git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@11188 85f007b7-540e-0410-9357-904b9bb8a0f7
2007-12-06Plus de combinateurs sont passés de Util à Option. Le module Options aspiwack
devient Flags. git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@10348 85f007b7-540e-0410-9357-904b9bb8a0f7
2007-09-15* Adding compability with ocaml 3.10 + camlp5 (rework of letouzey
the patch by S. Mimram) * for detecting architecture, also look for /bin/uname * restore the compatibility of kernel/byterun/coq_interp.c with ocaml 3.07 (caml_modify vs. modify). There is still an issue with this 3.07 and 64-bits architecture (see coqdev and a future bug report). git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@10122 85f007b7-540e-0410-9357-904b9bb8a0f7
2007-07-13New bootstrapping, improved, Makefile systemcorbinea
Documented in dev/doc/build-system.txt . git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@9992 85f007b7-540e-0410-9357-904b9bb8a0f7
2006-11-21Nettoyage de l'utilisation de l'expansion des macros ~ et $ dans les noms deherbelin
chemin physique : expansion uniquement pour Load, Add LoadPath, Declare ML Module, Cd, ... mais pas pour les options -I, -boot, -R, -load-vernac-file, ... git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@9398 85f007b7-540e-0410-9357-904b9bb8a0f7
2005-02-06Nettoyage et documentation de Libraryherbelin
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@6692 85f007b7-540e-0410-9357-904b9bb8a0f7
2004-07-17Backtrack sur l'utilisation de pa_macro car n'existait pas en 3.06herbelin
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@5941 85f007b7-540e-0410-9357-904b9bb8a0f7
2004-07-16Branchement sur pa_macro, pa_ifdef devenant obsolete en 3.08herbelin
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@5931 85f007b7-540e-0410-9357-904b9bb8a0f7
2004-07-16Nouvelle en-têteherbelin
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@5920 85f007b7-540e-0410-9357-904b9bb8a0f7
2003-10-07Correction du bug 335 et Export/Require Export dans un modulecoq
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@4534 85f007b7-540e-0410-9357-904b9bb8a0f7
2003-10-04Reparation plus juste de l'inefficacite avec loaded_modules (respecte l'ordre)herbelin
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@4527 85f007b7-540e-0410-9357-904b9bb8a0f7
2003-10-03Correction bug explosion de la taille de la liste loaded_modulesherbelin
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@4521 85f007b7-540e-0410-9357-904b9bb8a0f7
2003-03-12*** empty log message ***barras
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@3761 85f007b7-540e-0410-9357-904b9bb8a0f7
2003-02-13Chargement dynamique de .cmadelahaye
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@3676 85f007b7-540e-0410-9357-904b9bb8a0f7
2002-08-02Modules dans COQ\!\!\!\!coq
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@2957 85f007b7-540e-0410-9357-904b9bb8a0f7
2002-03-17Meilleure gestion de la reduction dans Fielddelahaye
Field term (nouveau) Injections dans l'interpreteur de tactiques Exportation de quelques entrees de grammaires Exportation de quelques fonctionnalites des definitions git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@2538 85f007b7-540e-0410-9357-904b9bb8a0f7
2001-12-13compat ocaml 3.03filliatr
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@2291 85f007b7-540e-0410-9357-904b9bb8a0f7
2001-11-05Oopsbarras
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@2159 85f007b7-540e-0410-9357-904b9bb8a0f7
2001-11-05GROS COMMIT:barras
- reduction du noyau (variables existentielles, fonctions auxiliaires pour inventer des noms, etc. deplacees hors de kernel/) - changement de noms de constructeurs des constr (suppression de "Is" et "Mut") git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@2158 85f007b7-540e-0410-9357-904b9bb8a0f7
2001-10-17Abstraction de l'immplementation de dirpath et implementation dans l'autre ↵herbelin
sens pour plus de partage entre chemins git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@2126 85f007b7-540e-0410-9357-904b9bb8a0f7
2001-10-17Amélioration mise en page Print ML Module et Print ML Moduleherbelin
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@2124 85f007b7-540e-0410-9357-904b9bb8a0f7
2001-09-20On ignore les répertoires qui ne correspondent pas à des identsherbelin
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@2016 85f007b7-540e-0410-9357-904b9bb8a0f7
2001-09-19Protection hd d'une liste videherbelin
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@2002 85f007b7-540e-0410-9357-904b9bb8a0f7
2001-08-10Parsingherbelin
- Typage renforcé dans les grammaires (distinction des vars et des metavars) - Disparition de SLAM au profit de ABSTRACT - Paths primitifs dans les quotations (syntaxe concrète à base de .) - Mise en place de identifier dès le type ast - Protection de identifier contre les effets de bord via un String.copy - Utilisation de module_ident (= identifier) dans les dir_path (au lieu de string) Table des noms qualifiés - Remplacement de la table de visibilité par une table qui ne cache plus les noms de modules et sections mais seulement les noms des constantes (e.g. Require A. ne cachera plus le contenu d'un éventuel module A déjà existant : seuls les noms de constructions de l'ancien A qui existent aussi dans le nouveau A seront cachés) - Renoncement à la possibilité d'accéder les formes non déchargées des constantes définies à l'intérieur de sections et simplification connexes (suppression de END-SECTION, une seule table de noms qui ne survit pas au discharge) - Utilisation de noms longs pour les modules, de noms qualifiés pour Require and co, tests de cohérence; pour être cohérent avec la non survie des tables de noms à la sortie des section, les require à l'intérieur d'une section eux aussi sont refaits à la fermeture de la section git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@1889 85f007b7-540e-0410-9357-904b9bb8a0f7
2001-05-23amelioration des messages d'erreurs vis a vis des evarsbarras
ajout automatique des chemins vers les sources au moment du Drop git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@1761 85f007b7-540e-0410-9357-904b9bb8a0f7
2001-04-03utilisation de Options.if_verbosefilliatr
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@1519 85f007b7-540e-0410-9357-904b9bb8a0f7
2001-03-15entetesfilliatr
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@1469 85f007b7-540e-0410-9357-904b9bb8a0f7
2001-02-07Centralisation des add_path dans Mltop a cause de la dependance en add_ml_dirherbelin
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@1344 85f007b7-540e-0410-9357-904b9bb8a0f7
2000-11-29load_path_entry structure simplified; field relative_subdir renamed to ↵sacerdot
coq_dirpath git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@1022 85f007b7-540e-0410-9357-904b9bb8a0f7
2000-11-29Ajout d'un alias à add_path, rec_add_path et all_subdirs pour associer un ↵herbelin
chemin Unix à un chemin Coq git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@1013 85f007b7-540e-0410-9357-904b9bb8a0f7