Commit graph

1526 commits

Author SHA1 Message Date
Leonardo de Moura
4ebb3c572a feat(kernel/environment): make the environment throw an exception when weak-ref has expired
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-11-16 18:35:17 -08:00
Leonardo de Moura
590b14570f feat(lua): improve error handling in Lua API
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-11-16 18:21:42 -08:00
Leonardo de Moura
390f7eaec8 test(kernel/typechecker): type checker
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-11-16 15:01:39 -08:00
Leonardo de Moura
516c5c8fea feat(lua): add metavar_env objects to Lua API
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-11-16 14:44:33 -08:00
Leonardo de Moura
c8fff45319 feat(lua): add justification objects to Lua API
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-11-16 11:12:58 -08:00
Leonardo de Moura
9398ea5a59 feat(util/shared_mutex): add support for recursive lock at shared_mutex
We need support for recursive locks. The main user of this class is
the environment object. This commit adds a test that demonstrates that
the shared_lock of the environment object may be recursively requested.
Before this fix, the Lean was deadlocking in this example.

Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-11-15 22:01:11 -08:00
Leonardo de Moura
209a2d10f7 fix(lua): replace std::mutex with std::recursive_mutex, add test that demonstrates the problem that is being fixed
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-11-15 21:26:16 -08:00
Leonardo de Moura
8525e8534b feat(lua): expose parse_expr and parse_commands from frontends/lean in the Lua API
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-11-15 16:11:26 -08:00
Leonardo de Moura
d9d9c05e4f refactor(lua/options): remove unnecessary function
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-11-15 12:53:27 -08:00
Leonardo de Moura
19533c811b feat(library/script_evaluator): add abstract class that exposes only the API needed by frontend objects
The main motivation is to break the remove the dependency frontends/lean <-- bindings/lua.
This dependency is undesirable because we want to expose the frontends/lean parser and pretty printer objects at bindings/lua.

Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-11-15 12:13:09 -08:00
Leonardo de Moura
1cb0262ec5 chore(memcheck): make the suppression entries for LuaJIT more general
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-11-15 11:34:50 -08:00
Leonardo de Moura
e7552d35e0 chore(memcheck): hide another tcmalloc bug
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-11-15 11:33:34 -08:00
Leonardo de Moura
a74412963a chore(build): only execute lua multi-threading tests when on cygwin or linux, and using g++
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-11-15 09:27:58 -08:00
Leonardo de Moura
5e170ce5fe chore(tests/lua): move examples that use threads to tests/lua/threads
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-11-15 09:11:58 -08:00
Leonardo de Moura
782335581b fix(lua/thread): uninitialized memory
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-11-15 09:07:18 -08:00
Leonardo de Moura
b5dcb93550 feat(lua): communication channels for threads
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-11-14 21:10:46 -08:00
Leonardo de Moura
bd1e9c7548 feat(lua): throw an exception if the user tries to create a thread and Lean was compiled without multi-threading support
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-11-14 19:06:36 -08:00
Leonardo de Moura
3a924a5fb1 perf(lua/name): improve to_name_ext performance
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-11-14 18:06:09 -08:00
Leonardo de Moura
cd6bd79d63 refactor(lua): cleanup Lua API, improve performance
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-11-14 17:33:46 -08:00
Leonardo de Moura
45858d54ae test(frontends/lean): add missing expected output
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-11-14 15:51:47 -08:00
Leonardo de Moura
6575566f9f fix(lua): rename lua_ref to lref, lua_ref is a macro in LuaJIT
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-11-14 15:50:56 -08:00
Leonardo de Moura
6b30ebab5e test(lua): use simplified Const creation
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-11-14 15:17:00 -08:00
Leonardo de Moura
09bed4786c feat(lua): add semantic attachments for builtin arithmetical values to Lua API, improve mk_constant
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-11-14 15:15:04 -08:00
Leonardo de Moura
05f254f605 refactor(lua): move lua_ref to separate file
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-11-14 14:41:54 -08:00
Leonardo de Moura
bdea42b2a9 style(lua): ignore bogus warning
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-11-14 14:13:06 -08:00
Leonardo de Moura
abe93dfec0 fix(lua/splay_tree): for_each method was crashing if the map was updated during for_each
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-11-14 13:48:23 -08:00
Leonardo de Moura
64cce595a5 feat(lua): add splay_maps to the Lua API
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-11-14 13:35:36 -08:00
Leonardo de Moura
8e56726116 fix(library/expr_lt): fix bug when using hash codes
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-11-14 13:18:33 -08:00
Leonardo de Moura
b0322787ff feat(lua): add has_metavar method
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-11-14 10:28:50 -08:00
Leonardo de Moura
9e445d1917 feat(lua): interrupt and sleep Lua APIs
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-11-14 10:04:32 -08:00
Leonardo de Moura
691893258d feat(kernel/expr): add hash code based on allocation time
The new hash code has the property that given expr_cell * c1 and expr_cell * c2,
if c1 != c2 then there is a high propbability that c1->hash_alloc() != c2->hash_alloc().

The structural hash code hash() does not have this property because we may have
c1 != c2, but c1 and c2 are structurally equal.

The new hash code is only compatible with pointer equality.
By compatible we mean, if c1 == c2, then c1->hash_alloc() == c2->hash_alloc().
This property is obvious because hash_alloc() does not have side-effects.

The test tests/lua/big.lua exposes the problem fixed by this commit.

Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-11-14 02:43:11 -08:00
Leonardo de Moura
ae7ea99b56 fix(lua): copy_values and add tests
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-11-13 21:31:27 -08:00
Leonardo de Moura
c759fc93f7 test(lua): object Lua API
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-11-13 20:59:28 -08:00
Leonardo de Moura
6e0fc0ca9b chore(build): include dl.so when compiling on Linux
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-11-13 18:44:28 -08:00
Leonardo de Moura
6964e374c0 fix(lua): luajit incompatibility
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-11-13 18:08:00 -08:00
Leonardo de Moura
cbc4254ff5 feat(lua): add occurs method to expr Lua API
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-11-13 17:10:09 -08:00
Leonardo de Moura
ad1180c5b4 fix(kernel/occurs): typos
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-11-13 17:04:56 -08:00
Leonardo de Moura
351ef867d2 feat(lua): add abstract, instantiate, has_free_vars, lift/lower free_vars to Lua API
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-11-13 17:02:49 -08:00
Leonardo de Moura
ed3cf8152b feat(lua): add for_each to expr Lua API
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-11-13 16:30:59 -08:00
Leonardo de Moura
eacd60de9c feat(frontends/lean): return the operator associated with constant expressions that are names of builtin values
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-11-13 16:08:21 -08:00
Leonardo de Moura
3dea7ae0d6 test(frontends/lean): example mixing Lean and Lua
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-11-13 16:07:15 -08:00
Leonardo de Moura
1bf6051866 test(lua): expr LUA API test
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-11-13 15:29:31 -08:00
Leonardo de Moura
9a22702383 feat(lua): make objects() and localobjects() methods return iterators in the environment LUA API
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-11-13 14:26:01 -08:00
Leonardo de Moura
8c52d47692 chore(lua): rename env() to get_env()
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-11-13 13:58:51 -08:00
Leonardo de Moura
ba0889265e refactor(lua): cleanup Lua bindings
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-11-13 13:55:05 -08:00
Leonardo de Moura
b5f0185729 fix(lua): replace lua_pushfstring with lua_pushstring
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-11-13 12:14:55 -08:00
Leonardo de Moura
a80adae1c3 feat(lua): add fields method to sexpr Lua API
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-11-13 12:10:24 -08:00
Leonardo de Moura
e2efce6b62 style(util/sexpr): name convetion for enumeration types
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-11-13 11:50:12 -08:00
Leonardo de Moura
450128e28b refactor(lua): cleanup Lua bindings, and add accessor/tester to expr Lua API
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-11-13 11:46:09 -08:00
Leonardo de Moura
c4c548dc5d feat(*): simplify interrupt propagation
Instead of having m_interrupted flags in several components. We use a thread_local global variable.
The new approach is much simpler to get right since there is no risk of "forgetting" to propagate
the set_interrupt method to sub-components.

The plan is to support set_interrupt methods and m_interrupted flags only in tactic objects.
We need to support them in tactics and tacticals because we want to implement combinators/tacticals such as (try_for T M) that fails if tactic T does not finish in M ms.
For example, consider the tactic:

    try-for (T1 ORELSE T2) 5

It tries the tactic (T1 ORELSE T2) for 5ms.
Thus, if T1 does not finish after 5ms an interrupt request is sent, and T1 is interrupted.
Now, if you do not have a m_interrupted flag marking each tactic, the ORELSE combinator will try T2.
The set_interrupt method for ORELSE tactical should turn on the m_interrupted flag.

Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2013-11-12 21:45:48 -08:00