]>
matita.cs.unibo.it Git - helm.git/log
Ferruccio Guidi [Wed, 28 Oct 2009 19:27:29 +0000 (19:27 +0000)]
- lambda-delta: some fixes: now the grundlagen type-checkes through dual_rg
- matita/Makefile: we removed LAMBDA-TYPES from the nightly tests (too slow ???)
Claudio Sacerdoti Coen [Wed, 28 Oct 2009 16:52:01 +0000 (16:52 +0000)]
Works again
Claudio Sacerdoti Coen [Wed, 28 Oct 2009 16:29:46 +0000 (16:29 +0000)]
Ad-hoc management of ? vs out_scope in instantiate. The library passes again,
but there seems to be still more work to do in instantiate.
Claudio Sacerdoti Coen [Wed, 28 Oct 2009 16:28:50 +0000 (16:28 +0000)]
Bug fixed: the `IsTerm attribute is now added by mk_meta and it was not
preserved. The current code should be more stable.
Claudio Sacerdoti Coen [Wed, 28 Oct 2009 15:38:20 +0000 (15:38 +0000)]
1) new-style debugging/profiling code for old reduction
2) aliases are now dumped to stderr
Claudio Sacerdoti Coen [Wed, 28 Oct 2009 15:36:45 +0000 (15:36 +0000)]
Commented out code to optimize the case t1 vs t2 when t1 and t2 are both
meta closed. Why is it necessary, anyway?
Claudio Sacerdoti Coen [Wed, 28 Oct 2009 15:26:49 +0000 (15:26 +0000)]
One-shot aliases were no longer generated because of a bug (i.e. all aliases
retrieved from the universe used to have instance=0). Fixed by parameterizing
once again all the functions with a ~fix_instance function to fix the instance.
Claudio Sacerdoti Coen [Wed, 28 Oct 2009 15:00:14 +0000 (15:00 +0000)]
Different aliases, better equality inferred.
Claudio Sacerdoti Coen [Wed, 28 Oct 2009 14:51:38 +0000 (14:51 +0000)]
instances
Claudio Sacerdoti Coen [Wed, 28 Oct 2009 14:46:19 +0000 (14:46 +0000)]
instance fixed
Enrico Tassi [Wed, 28 Oct 2009 13:38:21 +0000 (13:38 +0000)]
better indentation
Enrico Tassi [Wed, 28 Oct 2009 13:37:07 +0000 (13:37 +0000)]
better indentation
Enrico Tassi [Wed, 28 Oct 2009 13:35:47 +0000 (13:35 +0000)]
better comments and indentation
Enrico Tassi [Wed, 28 Oct 2009 13:30:38 +0000 (13:30 +0000)]
use prop_only to filter instead of repeting the same function body
Enrico Tassi [Wed, 28 Oct 2009 13:27:48 +0000 (13:27 +0000)]
better logging
Enrico Tassi [Wed, 28 Oct 2009 13:24:55 +0000 (13:24 +0000)]
better logging and immediate pruning of new goals when
- length(goals) > maxwidth
- depth = maxdepth and goals <> []
Enrico Tassi [Wed, 28 Oct 2009 12:32:41 +0000 (12:32 +0000)]
auto navigates a real tree, not a flattened one
Enrico Tassi [Wed, 28 Oct 2009 12:22:10 +0000 (12:22 +0000)]
labels in group_by_tac
Enrico Tassi [Wed, 28 Oct 2009 12:20:50 +0000 (12:20 +0000)]
new data structures for auto
Enrico Tassi [Wed, 28 Oct 2009 12:19:04 +0000 (12:19 +0000)]
do not put " around node name, otherwise names like foo:f1 are not
accepted
Enrico Tassi [Wed, 28 Oct 2009 10:52:46 +0000 (10:52 +0000)]
export group_by_tac
Ferruccio Guidi [Wed, 28 Oct 2009 00:21:36 +0000 (00:21 +0000)]
drg->brg translation contibued (still bugged though)
Claudio Sacerdoti Coen [Mon, 26 Oct 2009 14:15:34 +0000 (14:15 +0000)]
qed => nqed.
Claudio Sacerdoti Coen [Mon, 26 Oct 2009 13:03:28 +0000 (13:03 +0000)]
Now the time required to eval a command is printed.
Enrico Tassi [Fri, 23 Oct 2009 14:04:42 +0000 (14:04 +0000)]
added code to print the tree
Enrico Tassi [Fri, 23 Oct 2009 13:45:01 +0000 (13:45 +0000)]
CSC proof made by paramod
Enrico Tassi [Fri, 23 Oct 2009 11:44:13 +0000 (11:44 +0000)]
more functions
Claudio Sacerdoti Coen [Fri, 23 Oct 2009 07:20:59 +0000 (07:20 +0000)]
Alias required now ??
Enrico Tassi [Thu, 22 Oct 2009 22:11:03 +0000 (22:11 +0000)]
new instantiate, only known bug is w.r.t. in/out scope and file matita/contribs/ng_assembly/compiler/environment.ma
Enrico Tassi [Thu, 22 Oct 2009 21:38:41 +0000 (21:38 +0000)]
the trie indexes terms up to 10 nested applications and skips applications with more then 50 arguments
Enrico Tassi [Thu, 22 Oct 2009 08:44:35 +0000 (08:44 +0000)]
more auto
Claudio Sacerdoti Coen [Wed, 21 Oct 2009 20:11:22 +0000 (20:11 +0000)]
Non general recursion implemented via recursion over unary (?) inductive
generated formal topologies.
Enrico Tassi [Wed, 21 Oct 2009 15:31:55 +0000 (15:31 +0000)]
first bits for the zipper
Enrico Tassi [Wed, 21 Oct 2009 14:52:00 +0000 (14:52 +0000)]
...
Enrico Tassi [Wed, 21 Oct 2009 14:49:00 +0000 (14:49 +0000)]
...
Enrico Tassi [Wed, 21 Oct 2009 14:27:38 +0000 (14:27 +0000)]
more printings
Enrico Tassi [Wed, 21 Oct 2009 13:39:28 +0000 (13:39 +0000)]
auto is smarter :-)
Enrico Tassi [Wed, 21 Oct 2009 13:37:34 +0000 (13:37 +0000)]
nauto:
- fixed indexing (now always dome in the same way)
- depth=x means: tray at depth=2 ... try at depth=x
- run auto on set of goals that are linked by occurring in each others types,
simply distribute if there is no dependency
Enrico Tassi [Wed, 21 Oct 2009 09:30:37 +0000 (09:30 +0000)]
...
Enrico Tassi [Wed, 21 Oct 2009 09:30:11 +0000 (09:30 +0000)]
fixed pictures
Enrico Tassi [Wed, 21 Oct 2009 09:12:01 +0000 (09:12 +0000)]
preserve sharing if map_term_fold_a
Enrico Tassi [Wed, 21 Oct 2009 09:11:23 +0000 (09:11 +0000)]
add XXX where I found a catch all statement
Enrico Tassi [Wed, 21 Oct 2009 09:09:47 +0000 (09:09 +0000)]
new sharing-preserving map with accumulator
Enrico Tassi [Wed, 21 Oct 2009 09:09:14 +0000 (09:09 +0000)]
apply the subst to the metasenv and to p
Claudio Sacerdoti Coen [Tue, 20 Oct 2009 14:59:41 +0000 (14:59 +0000)]
- Bug fixed: some assert failure were just failures (when processing terms
that do not satisfy the IRS condition).
- New test ng_bove.ma added to test the Bove-Capretta method (in CProp)
Wilmer Ricciotti [Tue, 20 Oct 2009 12:53:09 +0000 (12:53 +0000)]
...
Claudio Sacerdoti Coen [Mon, 19 Oct 2009 09:23:57 +0000 (09:23 +0000)]
Smarter implementation of instantiate to avoid re-checking twice the same term.
Enrico Tassi [Sun, 18 Oct 2009 18:46:36 +0000 (18:46 +0000)]
...
Enrico Tassi [Sat, 17 Oct 2009 17:39:59 +0000 (17:39 +0000)]
batch pdf generation
Enrico Tassi [Sat, 17 Oct 2009 09:53:04 +0000 (09:53 +0000)]
...
Enrico Tassi [Sat, 17 Oct 2009 07:03:03 +0000 (07:03 +0000)]
some more work
Enrico Tassi [Fri, 16 Oct 2009 14:02:54 +0000 (14:02 +0000)]
...
Enrico Tassi [Fri, 16 Oct 2009 14:01:18 +0000 (14:01 +0000)]
...
Enrico Tassi [Fri, 16 Oct 2009 12:41:41 +0000 (12:41 +0000)]
debug idem for auto added
Enrico Tassi [Fri, 16 Oct 2009 12:41:22 +0000 (12:41 +0000)]
new lambda instros and better logging
Enrico Tassi [Fri, 16 Oct 2009 12:40:36 +0000 (12:40 +0000)]
better indexing for auto
Enrico Tassi [Fri, 16 Oct 2009 12:40:16 +0000 (12:40 +0000)]
some work for auto
Enrico Tassi [Fri, 16 Oct 2009 12:40:05 +0000 (12:40 +0000)]
some work for auto
Enrico Tassi [Fri, 16 Oct 2009 12:39:27 +0000 (12:39 +0000)]
removed optimization potentially unsound
Enrico Tassi [Fri, 16 Oct 2009 12:36:12 +0000 (12:36 +0000)]
...
Claudio Sacerdoti Coen [Thu, 15 Oct 2009 13:59:31 +0000 (13:59 +0000)]
Profiling code integrated.
Cosimo Oliboni [Thu, 15 Oct 2009 08:47:14 +0000 (08:47 +0000)]
freescale porting, work in progress
Claudio Sacerdoti Coen [Wed, 14 Oct 2009 21:03:49 +0000 (21:03 +0000)]
Debugging improved.
Claudio Sacerdoti Coen [Wed, 14 Oct 2009 21:01:12 +0000 (21:01 +0000)]
Benchmarking integrated in folding/unfolding.
Cosimo Oliboni [Wed, 14 Oct 2009 20:52:07 +0000 (20:52 +0000)]
freescale porting, work in progress
Enrico Tassi [Wed, 14 Oct 2009 15:13:27 +0000 (15:13 +0000)]
cantor...
Enrico Tassi [Wed, 14 Oct 2009 15:05:15 +0000 (15:05 +0000)]
...
Enrico Tassi [Wed, 14 Oct 2009 14:38:02 +0000 (14:38 +0000)]
hints were not used by reduction machines on heads
Enrico Tassi [Wed, 14 Oct 2009 13:45:33 +0000 (13:45 +0000)]
...
Enrico Tassi [Wed, 14 Oct 2009 13:35:24 +0000 (13:35 +0000)]
ncheck is not erased
Enrico Tassi [Wed, 14 Oct 2009 13:12:32 +0000 (13:12 +0000)]
...
Enrico Tassi [Wed, 14 Oct 2009 13:10:22 +0000 (13:10 +0000)]
...
Enrico Tassi [Wed, 14 Oct 2009 13:09:04 +0000 (13:09 +0000)]
...
Enrico Tassi [Wed, 14 Oct 2009 13:08:11 +0000 (13:08 +0000)]
...
Enrico Tassi [Wed, 14 Oct 2009 12:52:06 +0000 (12:52 +0000)]
minor fix
Enrico Tassi [Wed, 14 Oct 2009 12:49:08 +0000 (12:49 +0000)]
tons of typo fixed
Claudio Sacerdoti Coen [Wed, 14 Oct 2009 11:19:03 +0000 (11:19 +0000)]
Error message fixed (dereferencing must be done eagerly, not when the error is
actually printed!)
Claudio Sacerdoti Coen [Wed, 14 Oct 2009 09:37:50 +0000 (09:37 +0000)]
Serious bug fixed: fix_sorts used to allow inference of Type[0]:Type[0]!
Enrico Tassi [Wed, 14 Oct 2009 08:51:13 +0000 (08:51 +0000)]
CProp uri fixed
Enrico Tassi [Tue, 13 Oct 2009 16:59:55 +0000 (16:59 +0000)]
...
Enrico Tassi [Tue, 13 Oct 2009 16:15:03 +0000 (16:15 +0000)]
...
Enrico Tassi [Tue, 13 Oct 2009 15:53:50 +0000 (15:53 +0000)]
...
Enrico Tassi [Tue, 13 Oct 2009 15:50:11 +0000 (15:50 +0000)]
...
Enrico Tassi [Tue, 13 Oct 2009 15:47:03 +0000 (15:47 +0000)]
...
Enrico Tassi [Tue, 13 Oct 2009 15:13:09 +0000 (15:13 +0000)]
...
Wilmer Ricciotti [Tue, 13 Oct 2009 14:57:30 +0000 (14:57 +0000)]
Experimental scripts for nth-order rewriting principles.
Enrico Tassi [Tue, 13 Oct 2009 13:36:32 +0000 (13:36 +0000)]
relocate is hopefully fixed once and for-all!
Enrico Tassi [Tue, 13 Oct 2009 11:27:03 +0000 (11:27 +0000)]
better screenshot
Cosimo Oliboni [Tue, 13 Oct 2009 11:18:49 +0000 (11:18 +0000)]
freescale porting, work in progress
Enrico Tassi [Tue, 13 Oct 2009 11:13:46 +0000 (11:13 +0000)]
relocate fixed
Enrico Tassi [Tue, 13 Oct 2009 10:14:50 +0000 (10:14 +0000)]
better ppcontext
Enrico Tassi [Tue, 13 Oct 2009 10:14:36 +0000 (10:14 +0000)]
better ppcontext
Enrico Tassi [Tue, 13 Oct 2009 10:14:07 +0000 (10:14 +0000)]
debug + relocate uses Prop instead of (Prop Prop)... we should think a better solution
Enrico Tassi [Tue, 13 Oct 2009 09:43:02 +0000 (09:43 +0000)]
one mode debug item
Enrico Tassi [Tue, 13 Oct 2009 09:36:39 +0000 (09:36 +0000)]
...
Enrico Tassi [Tue, 13 Oct 2009 09:33:40 +0000 (09:33 +0000)]
no mode middle age debug menu
Enrico Tassi [Tue, 13 Oct 2009 09:21:04 +0000 (09:21 +0000)]
more comments
Enrico Tassi [Tue, 13 Oct 2009 07:56:48 +0000 (07:56 +0000)]
...
Enrico Tassi [Mon, 12 Oct 2009 20:08:36 +0000 (20:08 +0000)]
...
Claudio Sacerdoti Coen [Mon, 12 Oct 2009 17:38:50 +0000 (17:38 +0000)]
1) Bug fixed: the case Meta(i) vs Meta(i) was handled in a particular way,
but the case Meta(i) vs (Appl (Meta(j),...)) that reduces to Meta(i) was
not. As a result, a tempeted self assegniment yielded strange errors.
Fixed by more aggressively unwinding the subst during fo_unif.
2) Major re-organization of the code to gain some speed in Oliboni's stuff.
The idea is that of introducing a new internal exception KeepReducing
used to signal that, after a fo_unif, it still makes sense to fall back
to machines. Only if it does not it makes sense to distinguish between
Failures and Uncertain and the latter test can now be implemented more
lazily w.r.t. the old version that used to call metas_of_term on the
unwinded machines (that are potentially HUGE).
With this modification, all Oliboni's tests terminate, even if they are
still very slow compared to the height=0 strategy. Moreover, the tests
show that unification on closed terms can still be 4x slower than conversion,
which is partially unexpected.