]>
matita.cs.unibo.it Git - helm.git/log
Claudio Sacerdoti Coen [Mon, 4 Jul 2005 12:10:00 +0000 (12:10 +0000)]
New function (only partially implemented) to pretty print an Ast at the
semantics level.
Stefano Zacchiroli [Mon, 4 Jul 2005 12:08:55 +0000 (12:08 +0000)]
added some utility functions on filename suffixes
Claudio Sacerdoti Coen [Mon, 4 Jul 2005 11:59:55 +0000 (11:59 +0000)]
Use eval_from_stream in place of eval_from_stream_ref to avoid printing the
prompt.
Claudio Sacerdoti Coen [Mon, 4 Jul 2005 11:55:48 +0000 (11:55 +0000)]
include must not add aliases in the script!
Claudio Sacerdoti Coen [Mon, 4 Jul 2005 10:45:52 +0000 (10:45 +0000)]
"include" command implemented.
Claudio Sacerdoti Coen [Mon, 4 Jul 2005 10:38:32 +0000 (10:38 +0000)]
"include" command implemented.
Enrico Tassi [Mon, 4 Jul 2005 09:59:57 +0000 (09:59 +0000)]
matitaclean splitted in matitacleanLib and matitaclean.
Claudio Sacerdoti Coen [Mon, 4 Jul 2005 08:57:07 +0000 (08:57 +0000)]
alias declarations are now put in the .moo file.
Enrico Tassi [Mon, 4 Jul 2005 08:49:24 +0000 (08:49 +0000)]
matitaclean now call getter.ls using a buri with a trailing /
Claudio Sacerdoti Coen [Mon, 4 Jul 2005 07:40:28 +0000 (07:40 +0000)]
Improvements in matitatop: instead of doing exit -1 it enters the toploop
after a ^C.
Claudio Sacerdoti Coen [Sun, 3 Jul 2005 14:47:59 +0000 (14:47 +0000)]
Equality tactics are now used in the first half of the script.
Note: it seems that "elim H ?" is more powerful than "rewrite < H ?" since
it is able to discover the value of the implicit argument ?. To be improved.
Claudio Sacerdoti Coen [Sat, 2 Jul 2005 14:50:11 +0000 (14:50 +0000)]
The equality tactics are now exploited.
Claudio Sacerdoti Coen [Sat, 2 Jul 2005 14:42:46 +0000 (14:42 +0000)]
All the equalityTactics have now been ported to use both the equality of
Coq and that of Matita.
Note: the only problematic tactic is replace that must pick an equality without
any hint. Right now it always picks the one of matita.
Claudio Sacerdoti Coen [Sat, 2 Jul 2005 13:59:16 +0000 (13:59 +0000)]
New theorem: eq_ind_r.
Claudio Sacerdoti Coen [Sat, 2 Jul 2005 13:49:22 +0000 (13:49 +0000)]
cosmetic change: a space removed (to make tests/Makefile and library/Makefile
the same)
Alberto Griggio [Fri, 1 Jul 2005 19:08:05 +0000 (19:08 +0000)]
fixed bug in proof generation, new weight function to sort equalities, which
gives a huge speedup
Enrico Tassi [Fri, 1 Jul 2005 18:57:48 +0000 (18:57 +0000)]
may fix the nigtly build
Enrico Tassi [Fri, 1 Jul 2005 17:59:48 +0000 (17:59 +0000)]
more makefile work
Claudio Sacerdoti Coen [Fri, 1 Jul 2005 17:39:20 +0000 (17:39 +0000)]
make tests and make tests.opt now call their counterparts in library and tests
Claudio Sacerdoti Coen [Fri, 1 Jul 2005 17:38:35 +0000 (17:38 +0000)]
The tests are now in alphabetical order.
Claudio Sacerdoti Coen [Fri, 1 Jul 2005 17:31:33 +0000 (17:31 +0000)]
match.ma removed
Claudio Sacerdoti Coen [Fri, 1 Jul 2005 17:21:13 +0000 (17:21 +0000)]
* match.ma removed (it is now splitted in several files in library/*.ma)
* Makefile.tests removed (redundant with tests/Makefile)
* make tests and make tests.opt now call the same targets in library and tests
Claudio Sacerdoti Coen [Fri, 1 Jul 2005 17:16:00 +0000 (17:16 +0000)]
...
Claudio Sacerdoti Coen [Fri, 1 Jul 2005 17:14:39 +0000 (17:14 +0000)]
conflicting baseuri
Claudio Sacerdoti Coen [Fri, 1 Jul 2005 17:05:23 +0000 (17:05 +0000)]
matitaclean is necessary!
However, we need a fast ls method for this. Waiting for Zack's new getter...
Claudio Sacerdoti Coen [Fri, 1 Jul 2005 17:00:00 +0000 (17:00 +0000)]
...
Claudio Sacerdoti Coen [Fri, 1 Jul 2005 16:47:33 +0000 (16:47 +0000)]
* makefile improved
* new target-schema: *.opt
Claudio Sacerdoti Coen [Fri, 1 Jul 2005 16:40:54 +0000 (16:40 +0000)]
new targets:
opt
verbose.opt
Claudio Sacerdoti Coen [Fri, 1 Jul 2005 16:29:19 +0000 (16:29 +0000)]
...
Claudio Sacerdoti Coen [Fri, 1 Jul 2005 16:27:37 +0000 (16:27 +0000)]
New argument: the cleaner.
Claudio Sacerdoti Coen [Fri, 1 Jul 2005 16:24:15 +0000 (16:24 +0000)]
make all is now nicer
Claudio Sacerdoti Coen [Fri, 1 Jul 2005 16:22:42 +0000 (16:22 +0000)]
Two targets now:
make all
make verbose
Claudio Sacerdoti Coen [Fri, 1 Jul 2005 16:16:15 +0000 (16:16 +0000)]
make MATITAC="../scripts/do_tests.sh ../matitac /dev/null"
can now be used to benchmark the output
Claudio Sacerdoti Coen [Fri, 1 Jul 2005 16:01:12 +0000 (16:01 +0000)]
...
Claudio Sacerdoti Coen [Fri, 1 Jul 2005 16:00:53 +0000 (16:00 +0000)]
Makefile improved
Claudio Sacerdoti Coen [Fri, 1 Jul 2005 15:43:43 +0000 (15:43 +0000)]
The library of matita is borned! Long life to the library of matita.
Enrico Tassi [Fri, 1 Jul 2005 15:01:38 +0000 (15:01 +0000)]
ma -> moo in dependencies
Stefano Zacchiroli [Fri, 1 Jul 2005 14:57:18 +0000 (14:57 +0000)]
more logging information on received request
Stefano Zacchiroli [Fri, 1 Jul 2005 14:56:59 +0000 (14:56 +0000)]
uses relative OCAMLPATH
Enrico Tassi [Fri, 1 Jul 2005 14:55:19 +0000 (14:55 +0000)]
now maketests uses matitaclean (but not in the right way)
tests are made in the stupid order, not in the needed one
Enrico Tassi [Fri, 1 Jul 2005 14:40:19 +0000 (14:40 +0000)]
first snapshot of separate compilation
Enrico Tassi [Fri, 1 Jul 2005 14:38:10 +0000 (14:38 +0000)]
on is now a keyword (needed in let rec)
Enrico Tassi [Fri, 1 Jul 2005 14:37:39 +0000 (14:37 +0000)]
added uri_is_ind
Claudio Sacerdoti Coen [Fri, 1 Jul 2005 14:33:39 +0000 (14:33 +0000)]
matitaclean.opt, matitadep.opt
Alberto Griggio [Fri, 1 Jul 2005 14:28:02 +0000 (14:28 +0000)]
removed first Cic.term from type equality, added an int (weight of the equality)
Claudio Sacerdoti Coen [Fri, 1 Jul 2005 14:27:50 +0000 (14:27 +0000)]
...
Claudio Sacerdoti Coen [Fri, 1 Jul 2005 14:20:45 +0000 (14:20 +0000)]
TacticAst2Box no longer used.
Claudio Sacerdoti Coen [Fri, 1 Jul 2005 14:18:51 +0000 (14:18 +0000)]
Module TacticAst2Box unused!
Claudio Sacerdoti Coen [Fri, 1 Jul 2005 13:53:46 +0000 (13:53 +0000)]
count_pattern implemented
Claudio Sacerdoti Coen [Fri, 1 Jul 2005 13:49:56 +0000 (13:49 +0000)]
More cases implemented in tactic_count.
Claudio Sacerdoti Coen [Fri, 1 Jul 2005 13:10:27 +0000 (13:10 +0000)]
An interesting test for replace.
Claudio Sacerdoti Coen [Fri, 1 Jul 2005 13:06:43 +0000 (13:06 +0000)]
replace tactic reimplemented.
It is now able to replace several terms at once.
NOTE: it is still unable to replace terms in the hypotheses.
Claudio Sacerdoti Coen [Fri, 1 Jul 2005 13:05:51 +0000 (13:05 +0000)]
pattern_of function reimplemented. Now it takes a term and a list of subterms
and it returns the _minimal_ pattern for the subterms in the term.
Claudio Sacerdoti Coen [Fri, 1 Jul 2005 11:18:50 +0000 (11:18 +0000)]
prerr_endline -> debug_print
Claudio Sacerdoti Coen [Fri, 1 Jul 2005 11:18:42 +0000 (11:18 +0000)]
Signature of fwdSimpl changed to get rid of a warning.
Claudio Sacerdoti Coen [Fri, 1 Jul 2005 11:13:34 +0000 (11:13 +0000)]
The replace tactic is now working again. It can now replace simultaneously
n terms.
Claudio Sacerdoti Coen [Fri, 1 Jul 2005 11:13:10 +0000 (11:13 +0000)]
Signature of fwdSimpl changed to get rid of a warning.
Enrico Tassi [Fri, 1 Jul 2005 10:03:11 +0000 (10:03 +0000)]
uri_of_string now checks the uri is well formed
Claudio Sacerdoti Coen [Fri, 1 Jul 2005 09:39:24 +0000 (09:39 +0000)]
matitaclean
Claudio Sacerdoti Coen [Fri, 1 Jul 2005 09:38:55 +0000 (09:38 +0000)]
New syntax for patterns.
Claudio Sacerdoti Coen [Fri, 1 Jul 2005 09:34:07 +0000 (09:34 +0000)]
1. change_tac moved from PrimitiveTactics to ReductionTactics
2. fold_tac reimplemented using change_tac
Claudio Sacerdoti Coen [Fri, 1 Jul 2005 09:33:55 +0000 (09:33 +0000)]
New test.
Claudio Sacerdoti Coen [Fri, 1 Jul 2005 09:33:16 +0000 (09:33 +0000)]
\vdash added
Claudio Sacerdoti Coen [Fri, 1 Jul 2005 09:25:51 +0000 (09:25 +0000)]
Syntax of patterns changed again to make it non-ambiguous:
[in match t] [in sequent_pattern]
Enrico Tassi [Fri, 1 Jul 2005 08:06:46 +0000 (08:06 +0000)]
in case of performace improvement there is no performace loss message
Enrico Tassi [Thu, 30 Jun 2005 16:19:00 +0000 (16:19 +0000)]
matitaclean anapshot
Alberto Griggio [Thu, 30 Jun 2005 16:05:34 +0000 (16:05 +0000)]
proofs are now built lazily at the end of the computation
Claudio Sacerdoti Coen [Thu, 30 Jun 2005 14:56:12 +0000 (14:56 +0000)]
...
Claudio Sacerdoti Coen [Thu, 30 Jun 2005 14:52:17 +0000 (14:52 +0000)]
Signature and concrete syntax of fold fixed.
Claudio Sacerdoti Coen [Thu, 30 Jun 2005 14:48:12 +0000 (14:48 +0000)]
The pattern of a fold cannot have the "wanted" part (for the same reasons as
rewrite)
Claudio Sacerdoti Coen [Thu, 30 Jun 2005 14:46:31 +0000 (14:46 +0000)]
Signature and concrete syntax of fold fixed.
Ferruccio Guidi [Thu, 30 Jun 2005 14:27:19 +0000 (14:27 +0000)]
lapply and fwd improved
Claudio Sacerdoti Coen [Thu, 30 Jun 2005 14:25:17 +0000 (14:25 +0000)]
Text fixed (due to stricter semantics for naming the binders in the patterns).
Claudio Sacerdoti Coen [Thu, 30 Jun 2005 14:20:20 +0000 (14:20 +0000)]
The rewrite_* set of tactics is now working again. However, as before,
they do not work yet when the user asks to rewrite somethign also in the
hypotheses.
Claudio Sacerdoti Coen [Thu, 30 Jun 2005 13:17:30 +0000 (13:17 +0000)]
Bug fixed: a wrong lift made select unuseful when wanted was an open term.
Claudio Sacerdoti Coen [Thu, 30 Jun 2005 13:05:27 +0000 (13:05 +0000)]
Pretty printing of Cic.Implict (Some `Hole) is now "%"
Claudio Sacerdoti Coen [Thu, 30 Jun 2005 12:31:24 +0000 (12:31 +0000)]
1. rewrite_* and rewrite_back_* merged into one function
2. signature of rewrite_* improved
3. concrete syntax for the rewrite direction fixed to ">" and "<"
NOTE: rewrite_* is still no longer working.
Claudio Sacerdoti Coen [Thu, 30 Jun 2005 12:25:19 +0000 (12:25 +0000)]
1. rewrite_* and rewrite_back_* merged into one function
2. signature of rewrite_* improved
3. concrete syntax for the rewrite direction fixed to ">" and "<"
NOTE: rewrite_* is still no longer working.
Claudio Sacerdoti Coen [Thu, 30 Jun 2005 11:44:24 +0000 (11:44 +0000)]
Oooops. I forgot the convertibility test that makes the change tactic correct!
Claudio Sacerdoti Coen [Thu, 30 Jun 2005 11:44:06 +0000 (11:44 +0000)]
baseuri put back
Claudio Sacerdoti Coen [Thu, 30 Jun 2005 11:36:49 +0000 (11:36 +0000)]
The tactic change is now working again. Moreover, it now receives a pattern
in input; moreover it handles correctly the with_what argument when it must
be put in a context different from the one it was parsed in.
Claudio Sacerdoti Coen [Thu, 30 Jun 2005 11:35:30 +0000 (11:35 +0000)]
A bit of renaming in the code to make it more clear.
Claudio Sacerdoti Coen [Thu, 30 Jun 2005 10:34:59 +0000 (10:34 +0000)]
...
Claudio Sacerdoti Coen [Thu, 30 Jun 2005 10:24:36 +0000 (10:24 +0000)]
A simpler implementation of inversion that does not generate the dummy
hypothesis Hcut: 0=0.
Enrico Tassi [Thu, 30 Jun 2005 09:39:28 +0000 (09:39 +0000)]
buri_of_uri is now #xpointer aware.
Stefano Zacchiroli [Thu, 30 Jun 2005 09:36:02 +0000 (09:36 +0000)]
added license stuff
Stefano Zacchiroli [Thu, 30 Jun 2005 09:32:45 +0000 (09:32 +0000)]
better handling of script names
Stefano Zacchiroli [Thu, 30 Jun 2005 09:32:28 +0000 (09:32 +0000)]
added a debugging helper
Claudio Sacerdoti Coen [Thu, 30 Jun 2005 09:07:35 +0000 (09:07 +0000)]
Obsolete comment removed.
Claudio Sacerdoti Coen [Thu, 30 Jun 2005 08:48:22 +0000 (08:48 +0000)]
Stupid bug fixed (a completely erroneous assert false removed).
Claudio Sacerdoti Coen [Thu, 30 Jun 2005 08:44:39 +0000 (08:44 +0000)]
This commit makes ProofEngineHelpers.select reach its full potentials.
Its arguments are now a conjecture and a pattern and its output is
a conjecture-like structure made of physical pointers to matching subterms
(together with their contexts).
Enrico Tassi [Thu, 30 Jun 2005 07:54:34 +0000 (07:54 +0000)]
first matitadep snapshot
Claudio Sacerdoti Coen [Wed, 29 Jun 2005 16:24:08 +0000 (16:24 +0000)]
1. new syntax for patterns:
[t] [in p]
where t is the term to be matched and p is the pattern on the hypotheses
and on the conclusion (whose syntax is not changed).
All the tactics now use the new syntax.
2. several tactics have been changed to use the new kind of patterns
(that also have the optional term in them)
3. the signature of the select function has changed to require the context
(and return a context and not a "context patch"); moreover it performs
now both the search for roots and the search for subterms of the roots
that are alpha-convertible with the optional given term (if any)
WARNING!!!
The following tactics have been commented out for a while:
replace
rewrite
change
fold
Claudio Sacerdoti Coen [Wed, 29 Jun 2005 16:18:05 +0000 (16:18 +0000)]
1. new syntax for patterns:
[t] [in p]
where t is the term to be matched and p is the pattern on the hypotheses
and on the conclusion (whose syntax is not changed).
All the tactics now use the new syntax.
2. several tactics have been changed to use the new kind of patterns
(that also have the optional term in them)
3. the signature of the select function has changed to require the context
(and return a context and not a "context patch"); moreover it performs
now both the search for roots and the search for subterms of the roots
that are alpha-convertible with the optional given term (if any)
WARNING!!!
The following tactics have been commented out for a while:
replace
rewrite
change
fold
Enrico Tassi [Wed, 29 Jun 2005 15:41:37 +0000 (15:41 +0000)]
removed profiling function (now a stub is used instead)
added a comment in urimanager
Claudio Sacerdoti Coen [Wed, 29 Jun 2005 15:17:19 +0000 (15:17 +0000)]
Incredible bug of simpl fixed: the stack (in the terminology used for the
Krivine's machine) was processed in the wrong context. As a result List.nth
(to get a Rel in the context) used to raise Failure in several cases. The
Failure, however, was catched somewhere in the code of matita and the
failure of simpl was hidden in most of the cases.
Alberto Griggio [Wed, 29 Jun 2005 15:08:20 +0000 (15:08 +0000)]
various updates, removed proofs for now because they are the real bottleneck!!
Ferruccio Guidi [Wed, 29 Jun 2005 15:00:26 +0000 (15:00 +0000)]
lapply reimplemented using letin_tac
Claudio Sacerdoti Coen [Wed, 29 Jun 2005 14:34:20 +0000 (14:34 +0000)]
A few bug fixes. In particular parsing errors in matitatop when a file is
given to process used to result in entering the ocaml mode. Now the user is
left in the matita mode.
Enrico Tassi [Wed, 29 Jun 2005 14:26:38 +0000 (14:26 +0000)]
fix