Ferruccio Guidi [Sun, 14 Dec 2014 19:17:51 +0000 (19:17 +0000)]
- we are moving from old (patched) management of sort inclusion
to new (proper) one via \lambda\delta 3
- some refactoring
- omega.out: committed for reference
- Makefiles: some bugs fixed
Ferruccio Guidi [Sat, 6 Dec 2014 19:17:17 +0000 (19:17 +0000)]
- helena: the improved attribute system allows to export the sorts of Pi's
Pi's of sort Prop are rendered as Forall's in the generated Matita script
- we updated the dtd accordingly
Ferruccio Guidi [Sun, 30 Nov 2014 17:58:24 +0000 (17:58 +0000)]
level disambiguation cmpleted! the Grafite file is succesfully generated.
Matite cannot compiled it yet since it seems to freeze on notion 2760.
We divided the Grafite file in 7 parts for convenience.
Ferruccio Guidi [Fri, 28 Nov 2014 19:28:08 +0000 (19:28 +0000)]
- bug fix in the static disambiguation of unified binders
allows to disambiguate the whole grundlagen
- 20 incompatibilities are detected (AutQE vs PTS)
- matita validates notion 115
- some refactoring
Ferruccio Guidi [Fri, 28 Nov 2014 15:23:48 +0000 (15:23 +0000)]
- the disambiguation of unified binders continues
- exportation to grafite is set up
- grundlagen_2: we disambiguate up to notion 6377, matita validates up
to notion 114
Ferruccio Guidi [Mon, 24 Nov 2014 23:07:39 +0000 (23:07 +0000)]
- static disambiguation of Automath unified binders
by position heuristics + degree heuristics fixed
[ grundlagen_2: now 1217 binders out of 47115 remain ambiguous ]
- brgReduction: we did not check the sort-inclusion flag
- new constraints system continues ...
Ferruccio Guidi [Sun, 23 Nov 2014 19:36:06 +0000 (19:36 +0000)]
- new attributes system
- new constraints system (starts)
- static disambiguation of Automath unified binders
by degree heuristics (C.E. Brown)
[ grundlagen_2: just 1960 binders out of 47115 remain ambiguous ]
- Makefile: XNLDIR is now relative
Ferruccio Guidi [Tue, 11 Nov 2014 12:23:46 +0000 (12:23 +0000)]
- commit completed :)
typechecking is approx. fast as before,
validation is slightly faster than typechecking.
for now the new RTM is slightly slower than the old one
- we commit three versions of the grundlagen:
0 (original)
1 (no eta-conversion)
2 (no eta-conversion, no @-typing)
Ferruccio Guidi [Mon, 10 Nov 2014 16:56:55 +0000 (16:56 +0000)]
we begin the commit of the validation procedure
(this includes some changes in the xml format of exported entities)
the procedure seems to introduce a delay, so this commit is partial ...
Ferruccio Guidi [Sun, 19 Oct 2014 21:56:33 +0000 (21:56 +0000)]
- grafiteParser: we added the comand "defined" as a presentational
alternative to "qed" for definitions.
- lambdadelta: we added an example about extended validity vs.
restricted validity + some minor corrections
Ferruccio Guidi [Sat, 4 Oct 2014 20:42:40 +0000 (20:42 +0000)]
- nnAuto: we catch TypeCheckerFailure generated at the end of
smart_apply (rel out of context) during equational reasoning,
and we treat it as an Error. TO BE UNDERSTTOD
(this TypeCheckerFailure dramatically interferes with automatic
proof search in \lambda\delta)
- lambdadelta: new definition of fpbg without fpbc + some refactoring
Ferruccio Guidi [Thu, 7 Aug 2014 17:34:13 +0000 (17:34 +0000)]
- bugfixed "aacr" allows to remove historical eta-conversions
in the apparatus for strong normalization
- minor bugfixes in annotations
- xhtbl tables updated for the bugfixed stylesheets