Commit graph

1333 commits

Author SHA1 Message Date
Leonardo de Moura
ef6a27fe84 feat(util/script_state): add join method to Lua threads
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-12-04 09:15:09 -08:00
Leonardo de Moura
def186a9cd fix(util/stackinfo): try to fix incorrect main thread stack size on OSX
This fix tries to fix two failures on our unit tests.
     tests/kernel/normalizer
     tests/kernel/type_checker

Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-12-04 08:56:46 -08:00
Leonardo de Moura
d481cb251d chore(memcheck): add another suppression for LuaJIT
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-12-04 08:32:44 -08:00
Leonardo de Moura
fd9781d58d fix(util/stackinfo): compilation warning on mingw and cygwin
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-12-04 08:26:50 -08:00
Leonardo de Moura
ef069e39b0 chore(*): replace to_expr with to_nonnull_expr (when appropriate)
The goal is to make the Lua API more robust.

Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-12-03 12:53:53 -08:00
Leonardo de Moura
bcc8b67592 chore(*): consistent file name convention
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-12-03 12:40:52 -08:00
Leonardo de Moura
8e53643b61 feat(library/fo_unify): first order unification
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-12-03 12:21:21 -08:00
Leonardo de Moura
f80106a895 chore(*): use 'explicit operator bool' everywhere.
operator bool() may produce unwanted conversions.
For example, we had the following bug in the code base.

...
   object const & obj = find_object(const_name(n));
   if (obj && obj.is_builtin() && obj.get_name() == n)
...

obj.get_name() has type lean::name
n              has type lean::expr

Both have 'operator bool()', then the compiler uses the operator to
convert them to Boolean, and then compare the result.
Of course, this is not our intention.

After this commit, the compiler correctly signs the error.
The correct code is

...
   object const & obj = find_object(const_name(n));
   if (obj && obj.is_builtin() && obj.get_name() == const_name(n))
...

Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-12-02 23:02:45 -08:00
Leonardo de Moura
d79b2babd3 fix(*): typo
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-12-02 08:46:47 -08:00
Leonardo de Moura
dd62af1641 feat(frontends/parser): simplified theorem definition using tactical proof
When using tactics for proving theorems, a common pattern is

     Theorem T : <proposition> := _.
          apply <tactic>.
          ...
          done.

This commit allows the user to write the simplified form:

     Theorem T : <proposition>.
          apply <tactic>.
          ...
          done.

Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-12-02 08:20:18 -08:00
Leonardo de Moura
25978118df feat(library/tactic): add beta-reduction tactic
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-12-02 08:10:51 -08:00
Leonardo de Moura
39b99683a8 fix(util/stackinfo): handle error codes in the Linux version
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-12-02 07:56:53 -08:00
Soonho Kong
31b26f53ad fix(util/stackinfo): fix typo 2013-12-02 00:03:57 -05:00
Soonho Kong
7776f4b24b fix(util/stackinfo): fix preprocessor directive 2013-12-01 23:44:58 -05:00
Soonho Kong
4de3b772fd feat(util/stackinfo): implement get_stack_size (Mac OSX version) 2013-12-01 22:24:12 -05:00
Soonho Kong
a2d6918348 fix(library/rewriter): use Abst axiom in lambda_body RW 2013-12-01 22:24:12 -05:00
Soonho Kong
0553d29078 test(library/rewriter): add lambda_rewrite tests 2013-12-01 22:24:12 -05:00
Leonardo de Moura
75f8d56eb1 fix(util/stackinfo): memory leak at get_stack_size
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-12-01 19:17:28 -08:00
Leonardo de Moura
74dfdd02de feat(util): add primitives for checking the amount of available stack space
Recursive functions that may go very deep should invoke the function check_stack. It throws an exception if the amount of stack space is limited.

The function check_system() is syntax sugar for
    check_interrupted();
    check_stack();

Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-12-01 17:19:27 -08:00
Leonardo de Moura
1ec8f9d536 feat(kernel): add abstraction (aka function extensionality) axiom
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-12-01 13:57:14 -08:00
Leonardo de Moura
09f98ecddc feat(library/tactic): add unfold_tactic() that unfolds every non-hidden definition
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-12-01 10:41:05 -08:00
Leonardo de Moura
70e06f8e86 feat(library/hidden_defs): hidden definitions are just hints for tactics and solvers
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-12-01 10:27:27 -08:00
Leonardo de Moura
ca53a5a1cc feat(library/tactic): add unfold tactic
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-12-01 08:51:56 -08:00
Leonardo de Moura
737e634556 fix(util/list): bug in map template
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-12-01 08:51:24 -08:00
Leonardo de Moura
568931ccb1 refactor(library/tactic): remove duplicate code, add add_proofs
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-12-01 07:55:01 -08:00
Leonardo de Moura
f91c4901e8 feat(library/tactic): add absurd_tactic
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-12-01 07:55:00 -08:00
Leonardo de Moura
bf2adb20e7 feat(library/tactic): add disj_hyp_tactic
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-12-01 07:55:00 -08:00
Soonho Kong
6a6b69ddf4 test(library/rewriter): add test for depth RW 2013-12-01 01:59:21 -05:00
Soonho Kong
064e3fe20d refactor(library/rewriter): rename lc => ti 2013-12-01 01:59:20 -05:00
Soonho Kong
506cca0ac1 feat(library/rewriter): implement depth RW 2013-12-01 01:59:20 -05:00
Soonho Kong
ae0508128f refactor(library/rewriter): move apply_rewriter_fn into rewriter.h 2013-12-01 00:57:09 -05:00
Soonho Kong
6d5adb0429 chore(.gitignore): update .gitignore 2013-12-01 00:48:44 -05:00
Soonho Kong
d7ba5e3893 doc(library/rewriter): add doxygen annotations for rewrite_* funcs 2013-12-01 00:47:53 -05:00
Leonardo de Moura
1a221d8bbe feat(library/tactic): add focus tactical
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-11-30 11:28:38 -08:00
Leonardo de Moura
7ff791eb9f feat(util/name_set): add mk_unique (with respect to a name_set)
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-11-30 11:28:38 -08:00
Leonardo de Moura
6da13cc245 feat(util/list): map_append template
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-11-30 11:28:38 -08:00
Leonardo de Moura
fe79bbf2b7 feat(util/list): filter template
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-11-30 11:28:38 -08:00
Soonho Kong
f205dd0763 fix(library/rewriter): unused variable warnings 2013-11-30 07:05:18 -05:00
Leonardo de Moura
83aaf64318 fix(library/tactic): memory leaks
Proof/Cex builders and tactics implemented in Lua had a "strong reference" to script_state. If they are stored in the Lua state, then we get a cyclic reference.
That is, script_state points to these objects, and they point back to script_state.

To avoid this memory leak, this commit defines a weak reference for script_state objects. The Proof/Cex builders and tactics now store a weak reference to the Lua state.

Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-11-30 00:44:39 -08:00
Leonardo de Moura
2372567a6e fix(util/luaref): warnings produced by valgrind
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-11-30 00:34:38 -08:00
Leonardo de Moura
a7027a1d00 feat(library/tactic): polish tactic API, and add new example showing how to implement tactics using Lua
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-11-30 00:16:39 -08:00
Soonho Kong
aed8b1fc73 fix(tests/library/rewriter): app_rewriter1_tst
There was a bug in the app_rewriter1_tst. If we apply the ADD_COMM RW to
f(0), then the result should be f(0) since there is nothing to do for
ADD_COMM.

    f(0) = f(0)

The proof for this equality should be Refl(Nat, f(0)). But it was

    Refl(Nat -> Nat, f)

which is wrong. Somehow, the previous kernel didn't detect this type
mismatch and recent changes of the kernel found the problem.

I fixed the bug and re-enable the test as it was.
2013-11-30 02:25:30 -05:00
Soonho Kong
f5f7380fbe feat(library/rewriter): add apply_rewriter_fn which will be used in depth RW 2013-11-30 02:25:30 -05:00
Soonho Kong
1d76a6f71d feat(library/rewriter): add rewrite_* functions
rewrite_* functions take the rewriting results of the sub-components and
construct the rewriting result for the main component.

For instance, rewrite_app function takes env, ctx, and the value v s.t.

v = (e_0 e_1 ... e_n)

and the rewriting results for e_i's as a vector(buffer)

(e'_0, pf_0 -- proof of e_0 = e'_0)
(e'_1, pf_1 -- proof of e_1 = e'_1)
...
(e'_n, pf_n -- proof of e_n = e'_n).

Then rewrite_app function construct the new v'

v' = (e'_0 e'_1 ... e'_n)

and the proof of v = v' which is constructed with pf_i's.

These functions are used in the component rewriters such as app_RW and
let_type_RW, as well as more complicated rewriters such as depth
rewriter.
2013-11-30 02:25:29 -05:00
Leonardo de Moura
a9eb2a9307 feat(kernel/builtin): add is_* functions
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-11-29 11:35:58 -08:00
Leonardo de Moura
18eb9e427f fix(library/tactic): compilation warning
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-11-29 10:35:14 -08:00
Leonardo de Moura
98897b467d feat(frontends/lean/parser): add support for Lua expression code blocks
In expression code blocks, we do not have to write a "return".
After this commit, the argument of an apply command is a Lua expression instead of a Lua block of code. That is, we can now write

apply (** REPEAT(ORELSE(imp_tactic, conj_tactic, conj_hyp_tactic, assumption_tactic)) **)

instead of

apply (** return REPEAT(ORELSE(imp_tactic, conj_tactic, conj_hyp_tactic, assumption_tactic)) **)

Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-11-29 10:21:24 -08:00
Leonardo de Moura
f9874cd675 feat(library/tactic): add to_tactic_ext, it allows functions that return tactics to be used where a tactic is expected
For example, after this commit, we can write

simple_tac = REPEAT(ORELSE(imp_tactic, conj_tactic)) .. assumption_tactic

instead of

simple_tac = REPEAT(ORELSE(imp_tactic(), conj_tactic())) .. assumption_tactic()

Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-11-29 09:40:21 -08:00
Leonardo de Moura
20a36e98ec feat(library/elaborator): modify how elaborator handles constraints of the form ?M << P and P << ?M, where P is a proposition.
Before this commit, the elaborator would only assign ?M <- P, if P was normalized. This is bad since normalization may "destroy" the structure of P.

For example, consider the constraint
[a : Bool; b : Bool; c : Bool] ⊢ ?M::1 ≺ implies a (implies b (and a b))

Before this, ?M::1 will not be assigned to the "implies-term" because the "implies-term" is not normalized yet.
So, the elaborator would continue to process the constraint, and convert it into:

[a : Bool; b : Bool; c : Bool] ⊢ ?M::1 ≺ if Bool a (if Bool b (if Bool (if Bool a (if Bool b false true) true) false true) true) true

Now, ?M::1 is assigned to the term
     if Bool a (if Bool b (if Bool (if Bool a (if Bool b false true) true) false true) true) true

This is bad, since the original structure was lost.

This commit also contains an example that only works after the commit is applied.

Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-11-29 09:15:01 -08:00
Leonardo de Moura
066dacea31 feat(library/tactic): add show_tactic, and optional '.' in the end of tactic command
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-11-29 01:33:26 -08:00