aboutsummaryrefslogtreecommitdiff
path: root/dev
AgeCommit message (Collapse)Author
2017-05-25Merge PR#637: Short cleaning of the interpretation path for constr_with_bindingsMaxime Dénès
2017-05-25Merge PR#481: [option] Remove support for non-synchronous options.Maxime Dénès
2017-05-25Merge PR#406: coq makefile2Maxime Dénès
2017-05-25[location] [travis] Add overlays for located_switchEmilio Jesus Gallego Arias
2017-05-24Merge branch 'trunk' into located_switchEmilio Jesus Gallego Arias
2017-05-24[option] Remove support for non-synchronous options.Emilio Jesus Gallego Arias
Inspired by https://coq.inria.fr/bugs/show_bug.cgi?id=5229 , which this PR solves, I propose to remove support for non-synchronous options. It seems the few uses of `optsync = false` we legacy and shouldn't have any impact. Moreover, non synchronous options may create particularly tricky situations as for instance, they won't be propagated to workers.
2017-05-23Add parsers-examples target to fiat-parsers ciJason Gross
This tests a bit more of fiat-parsers, adding an extra ~3 minutes to the build.
2017-05-23overlay for UniMathEnrico Tassi
2017-05-23[vernac] Remove `Save thm id.` command.Emilio Jesus Gallego Arias
We'd like to cleanup the `proof_end` type so we can have a smaller path in proof save. Note that the construction: ``` Goal Type. ⋮ Save id. ``` has to be handled by the STM in the same path as Defined (but with an opaque flag), as `Save id` will alter the environment and cannot be processed in parallel. We thus try to simply such paths a bit, as complexity of `lemmas.ml` seems like an issue these days. The form `Save Theorem id` doesn't really seem used, and moreover we should really add a type of "Goal", and unify syntax. It is often the case that beginners try `Goal addnC n : n + 0 = n." etc...
2017-05-22Clarifying the interpretation path for the "constr_with_binding" argument.Hugo Herbelin
This fixes an inconsistency introduced in 554a6c806 (svn r12603) where both interp_constr_with_bindings and interp_open_constr_with_bindings were going through interp_open_constr (no type classes so as to not to commit too early on irreversible choices, accepting unresolved holes). We fix this by having interp_constr_with_bindings going to interp_constr (using type classes and failing on unresolved evars). The external impact is that any TACTIC EXTEND which refers to constr_with_binding has now to decide whether it intends it to use what the name suggest (using type classes and to fail if evars remain unresolved), thus keeping constr_with_binding, or the actual behavior which requires to use open_constr_with_bindings for strict compatibility.
2017-05-18Fix a typoJason Gross
2017-05-17Merge PR#633: An extension of EXTEND and notations to make standard parsing ↵Maxime Dénès
tricks available to users
2017-05-17Merge branch 'v8.6'Pierre-Marie Pédrot
2017-05-16Adding support for using grammar entries returning no value in EXTEND.Hugo Herbelin
2017-05-10Switch bedrock to mit-plv baseJason Gross
2017-05-08Fix warnings in top_printersGaetan Gilbert
Note that [@@@ocaml.warning "-32"] caused an error with Drop. It was put there because I didn't realise the warning was about a real issue.
2017-05-02Merge PR#582: Fix warningsMaxime Dénès
2017-05-01Add bmsherman/topology to the ciJason Gross
This development of @bmsherman tests universe polymorphism and setoid rewriting in type, and should build with v8.6 and trunk.
2017-05-01More consistent writing of de Bruijn.Théo Zimmermann
2017-05-01Fix for bug 5507. Mispelt de Bruijn.Théo Zimmermann
2017-04-27Remove unused [open] statementsGaetan Gilbert
2017-04-27Locally disable some warnings.Gaetan Gilbert
2017-04-27Merge PR#568: Remove tactic compatibility layerMaxime Dénès
2017-04-27Document the API changes.Pierre-Marie Pédrot
2017-04-27Merge branch 'v8.6'Pierre-Marie Pédrot
2017-04-25[located] [doc] Improve docs for `CAst.ast`.Emilio Jesus Gallego Arias
2017-04-25[travis] mathcomp and fiat overlay for #402Emilio Jesus Gallego Arias
2017-04-25[location] Document changes.Emilio Jesus Gallego Arias
2017-04-25[location] Remove `Loc.internal_ghost`Emilio Jesus Gallego Arias
`internal_ghost` was an artifact to ease porting of the ml4 rules. Now that the location is optional we can finally get rid of it.
2017-04-25[location] Remove Loc.ghost.Emilio Jesus Gallego Arias
Now it is a private field, locations are optional.
2017-04-24[location] Switch glob_constr to Loc.locatedEmilio Jesus Gallego Arias
2017-04-24Adding a dedicated travis overlay for fiat-parsers.Pierre-Marie Pédrot
2017-04-20Add bedrock targets src and facadeJason Gross
2017-04-19Documenting EConstr for developpers.Pierre-Marie Pédrot
2017-04-15Merge branch 'v8.6' into trunkMaxime Dénès
2017-04-14Fix EOL characters in xml protocol documentation.Maxime Dénès
2017-04-14Merge PR#563: add XML protocol doc for 8.6Maxime Dénès
2017-04-14[travis] Use the lite target for fiat-crypto.Maxime Dénès
2017-04-13update XML protocol doc to 8.6Paul Steckler
2017-04-13add XML protocol doc for 8.5Paul Steckler
2017-04-12Merge PR#441: Port Toplevel to the Stm APIMaxime Dénès
2017-04-12[stm] Improve error messages on add/parse.Emilio Jesus Gallego Arias
As suggested by @psteckler in #524 , we give more explicit information about what is wrong. We also provide some debug information for the possible dangerous case of having the tip go out of sync with the real installed state (which will make parsing fail if there was some changes to the parser). We also fix a couple of typos noticed by Paul, thanks Paul.
2017-04-12[stm] Port the toplevel to the STM.Emilio Jesus Gallego Arias
- We clean-up `Vernac` and make it use the STM API. - Now functions in `Vernac` for use in the toplevel and compiler take an starting `Stateid.t`. - Duplicated `Stm.interp` entry point is removed. - The XML protocol call `interp` is disabled.
2017-04-12[stm] Move main parsing entry point to the STM.Emilio Jesus Gallego Arias
Mainly due to notations, proof modes and plugins, parsing in Coq is stateful, so we expose a state-aware parsing API in the STM. This is a first move to unify all the parsing entry points in the Stm and the toplevel, and allows STM clients to control their input stream properly. This greatly helps for instance, with whole-document parsing. This commit supersedes PR#204.
2017-04-12Merge PR#422: Supporting all kinds of binders, including 'pat, in syntax of ↵Maxime Dénès
record fields.
2017-04-11Merge PR#379: Introducing evar-insensitive constrsMaxime Dénès
2017-04-07Merge PR#461: [camlpX] Remove camlp4 compat layer.Maxime Dénès
2017-04-07Fixes for Drop. to work (decl_mode removal and toplevel -> vernac)Matthieu Sozeau
2017-04-07Merge branch 'master' into econstrPierre-Marie Pédrot
2017-04-07[travis] Overlay for PR#461: Camlp4 removal.Emilio Jesus Gallego Arias