Commit graph

125 commits

Author SHA1 Message Date
Leonardo de Moura
15f0899efb refactor(*): replace LEAN_THREAD_LOCAL with MK_THREAD_LOCAL_GET, the new macro uses the Boost thread_local_ptr instead of 'thread_local' directive
Motivation: clang++ on OSX does not support 'thread_local'.

Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2014-06-07 10:18:36 -07:00
Leonardo de Moura
91df9a5550 feat(util/thread): add LEAN_THREAD_PTR macro, it uses boost::thread_specific_ptr instead of thread_local keyword when we compile with Boost
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2014-06-07 08:16:20 -07:00
Leonardo de Moura
89a7898054 fix(*): static variable initialization problems
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2014-06-06 18:44:12 -07:00
Leonardo de Moura
4a43016735 fix(kernel/expr): expr_cache initialization
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2014-06-06 18:00:26 -07:00
Leonardo de Moura
bab430af43 chore(kernel/expr): cleanup
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2014-06-04 14:51:36 -07:00
Leonardo de Moura
4a25e7442a feat(kernel/expr): add optional expression caching (aka "partial" hash-consing)
We do not enforce full hash-consing because we would need to synchronize
the access to the hashtable/cache.

Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2014-06-03 15:35:47 -07:00
Leonardo de Moura
2be9bcef78 feat(library/coercion): add coercion management implementation
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2014-05-24 19:28:42 -07:00
Leonardo de Moura
2344fb9476 feat(kernel/expr): use the expression depth in the hash code
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2014-05-23 14:16:51 -07:00
Leonardo de Moura
edc8af7bb3 refactor(kernel): reduce code duplication
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2014-05-19 11:11:19 -07:00
Leonardo de Moura
5c7d3c79c4 feat(kernel/expr): improve get_app_args interface
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2014-05-17 19:19:10 -07:00
Leonardo de Moura
5ce134e24e chore(kernel): binder => binding where appropriate
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2014-05-17 11:37:27 -07:00
Leonardo de Moura
33ae79cd9e refactor(kernel): move shallow copy function to library
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2014-05-17 11:20:24 -07:00
Leonardo de Moura
d625c9a26c refactor(kernel): move max_sharing to library
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2014-05-17 11:15:08 -07:00
Leonardo de Moura
e6c52d17a6 fix(kernel/expr): memory leak introduced today
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2014-05-16 16:21:31 -07:00
Leonardo de Moura
862c5e354d feat(kernel/expr): attach auxiliary name (for pretty printing) into local constants
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2014-05-16 13:08:09 -07:00
Leonardo de Moura
40b3129e7b refactor(kernel): improve names
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2014-05-16 11:28:05 -07:00
Leonardo de Moura
d6d72ba80e refactor(kernel): add binder structure
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2014-05-16 10:51:54 -07:00
Leonardo de Moura
660b9299ad refactor(kernel): (de)serialization procedures don't need to be in the kernel
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2014-05-15 17:23:36 -07:00
Leonardo de Moura
606e6226c2 fix(kernel/type_checker): the type checker cache was not taking into account binder information
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2014-05-14 17:14:57 -07:00
Leonardo de Moura
5d61d23bf6 refactor(kernel): move mk_tele_pi to expr.h, and rename it mk_pi
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2014-05-14 16:24:55 -07:00
Leonardo de Moura
893e4d4ae2 fix(kernel/expr): bug in m_free_var_range let-expr field
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2014-05-14 14:28:47 -07:00
Leonardo de Moura
2593f65ce1 feat(kernel): add tag to kernel expressions
Frontends can used tags to associate expressions with line number information.
The update_* procedures automatically propagate tags.

Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2014-05-12 16:33:25 -07:00
Leonardo de Moura
65486b2dec refactor(kernel/expr): add macro_definition smart pointer
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2014-05-01 18:05:46 -07:00
Leonardo de Moura
a5ae4b6570 feat(kernel): add binder annotations (implicit, cast)
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2014-05-01 10:35:02 -07:00
Leonardo de Moura
b2518f873d feat(kernel/type_checker): handle ensure_pi when whnf is a meta application
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2014-04-25 18:17:32 -07:00
Leonardo de Moura
63b6564b78 refactor(kernel): define extension_context API for normalizer_extesions and macro_definitions
A particular macro definition may use the extra information to retrieve information necessary to expand/type a given macro application. Example: it may need to invoke whnf.

Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2014-04-25 15:43:17 -07:00
Leonardo de Moura
4842ae4fc7 refactor(kernel): store macro arguments in the macro_expr
Before this commit, we "stored" macro arguments using applications.
This representation had some issues. Suppose we use [m a] to denote a macro
application. In the old representation, ([m a] b) and [m a b] would have
the same representation. Another problem is that some procedures (e.g., type inference)
would not have a clean implementation.

Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2014-04-25 15:02:56 -07:00
Leonardo de Moura
0d7902f9eb feat(kernel/type_checker): finish is_convertible predicate
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2014-04-23 20:16:14 -07:00
Leonardo de Moura
9fda7e2529 fix(kernel): incorrect optimization in max_sharing_fn class
The resultant expression may failed to be fully shared.
Add an example that demonstrates the problem.

Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2014-04-22 15:38:22 -07:00
Leonardo de Moura
984ac03ac7 refactor(kernel): replace kernel object with definition, disable affected files
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2014-04-17 16:10:47 -07:00
Leonardo de Moura
cb479a75ae fix(kernel/expr): make sure we cannot create a free variable with index uint_max, reason: get_free_var_range would return an incorrect value
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2014-04-17 13:25:21 -07:00
Leonardo de Moura
dc864bf7b9 feat(kernel): store free variable range in composite expressions
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2014-04-17 11:31:48 -07:00
Leonardo de Moura
bc8379256a refactor(kernel): remove pairs from kernel
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2014-04-17 10:52:07 -07:00
Leonardo de Moura
9b161b825f refactor(kernel): instantiate_params
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2014-03-18 10:27:57 -07:00
Leonardo de Moura
3939b93c2d refactor(kernel): substitution
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2014-03-18 10:27:57 -07:00
Leonardo de Moura
916301bdfb refactor(kernel): parametric kernel objects
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2014-03-18 10:27:57 -07:00
Leonardo de Moura
3c6002e969 refactor(kernel): add mk_rev_app, update_rev_app, implement instantiate_metavars functions, modify instantiate (free vars) API
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2014-03-18 10:27:56 -07:00
Leonardo de Moura
501435f6fc feat(kernel): add has_local predicate
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2014-03-18 10:27:56 -07:00
Leonardo de Moura
eb046c11fb refactor(kernel): the type in let-exprs is not optional anymore, if the user does not provide it, we use a metavariable
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2014-03-18 10:27:55 -07:00
Leonardo de Moura
737fe6830f test(tests/kernel): adjust expr tests
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2014-03-18 10:27:55 -07:00
Leonardo de Moura
f986963a95 refactor(kernel): serializer and deserializer
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2014-03-18 10:27:55 -07:00
Leonardo de Moura
74f74d2f79 refactor(kernel): shallow copy procedure
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2014-03-18 10:27:55 -07:00
Leonardo de Moura
d17990ed78 refactor(kernel): add formatter and simplify contexts
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2014-03-18 10:27:55 -07:00
Leonardo de Moura
53ee205dc6 fix(kernel): memory corruption bugs
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2014-03-18 10:27:55 -07:00
Leonardo de Moura
72e1678ad9 refactor(kernel): cleanup instantiate and abstract procedures, implement update procedures
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2014-03-18 10:27:55 -07:00
Leonardo de Moura
23988f528c refactor(kernel/expr): add expr constructors, and expression equality test
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2014-03-18 10:27:54 -07:00
Leonardo de Moura
02413d7c44 refactor(kernel/expr): adding suport for universe polymorphism, and simplify metavariable representation
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2014-03-18 10:27:54 -07:00
Leonardo de Moura
6d7ec9d7b6 refactor(kernel): add heterogeneous equality back to expr
The main motivation is that we will be able to move equalities between universes.

For example, suppose we have
    A : (Type i)
    B : (Type i)
    H : @eq (Type j) A B
where j > i

We didn't find any trick for deducing (@eq (Type i) A B) from H.
Before this commit, heterogeneous equality as a constant with type

   heq : {A B : (Type U)} : A -> B -> Bool

So, from H, we would only be able to deduce

   (@heq (Type j) (Type j) A B)

Not being able to move the equality back to a smaller universe is
problematic in several cases. I list some instances in the end of the commit message.

With this commit, Heterogeneous equality is a special kind of expression.
It is not a constant anymore. From H, we can deduce

   H1 : A == B

That is, we are essentially "erasing" the universes when we move to heterogeneous equality.
Now, since A and B have (Type i), we can deduce (@eq (Type i) A B) from H1. The proof term is

  (to_eq (Type i) A B (to_heq (Type j) A B H))  :  (@eq (Type i) A B)

So, it remains to explain why we need this feature.

For example, suppose we want to state the Pi extensionality axiom.

axiom hpiext {A A' : (Type U)} {B : A → (Type U)} {B' : A' → (Type U)} :
      A = A' → (∀ x x', x == x' → B x == B' x') → (∀ x, B x) == (∀ x, B' x)

This axiom produces an "inflated" equality at (Type U) when we treat heterogeneous
equality as a constant. The conclusion

     (∀ x, B x) == (∀ x, B' x)

is syntax sugar for

   (@heq (Type U) (Type U) (∀ x : A, B x) (∀ x : A', B' x))

Even if A, A', B, B' live in a much smaller universe.

As I described above, it doesn't seem to be a way to move this equality back to a smaller universe.

So, if we wanted to keep the heterogeneous equality as a constant, it seems we would
have to support axiom schemas. That is, hpiext would be parametrized by the universes where
A, A', B and B'. Another possibility would be to have universe polymorphism like Agda.
None of the solutions seem attractive.

So, we decided to have heterogeneous equality as a special kind of expression.
And use the trick above to move equalities back to the right universe.

BTW, the parser is not creating the new heterogeneous equalities yet.
Moreover, kernel.lean still contains a constant name heq2 that is the heterogeneous
equality as a constant.

Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2014-02-07 10:28:10 -08:00
Leonardo de Moura
5e5ab1429d feat(frontends/lean): parse and pretty print sigma types
This commit also fixes some bugs in the implementation of Sigma types.

Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2014-02-03 18:16:00 -08:00
Leonardo de Moura
8eec289ce1 feat(kernel): add dependent pairs
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2014-02-03 16:52:49 -08:00