Commit graph

4400 commits

Author SHA1 Message Date
Leonardo de Moura
69f50adb2e fix(frontends/lean/server): must save the starting environment/options when reprocessing file, fixes #209 2014-09-26 15:36:47 -07:00
Leonardo de Moura
a3e38dc8a0 feat(frontends/lean): allow users to define "numeral notation" 2014-09-26 14:55:23 -07:00
Leonardo de Moura
9928390605 fix(library/match): memory leak that only happens with compiling with clang++ 2014-09-26 10:24:38 -07:00
Leonardo de Moura
631ebc6c4b fix(tests/lean/uni_bug1): make sure test does not produce a 'used sorry' warning.
Thus, the output behavior is the same when we compile lean with/without IGNORE_SORRY=ON
2014-09-26 09:42:31 -07:00
Leonardo de Moura
6bf905aea8 fix(frontends/lean/pp): do not invoke type checker on expressions
containing free variables.

This could happened when the pretty printer was used from Lua to print
nested subterms
2014-09-26 09:38:36 -07:00
Leonardo de Moura
8143b51c7e feat(build): add 'CROSS_COMPILE' cmake option
When CROSS_COMPILE=ON, the Lean standard library will not be compiled.
2014-09-26 09:26:40 -07:00
Leonardo de Moura
f49a610995 fix(.travis.yml): (try to fix) random build failures at travis
At https://bitcointalk.org/index.php?topic=304389.0, people suggest
that the failure

    g++-4.8: internal compiler error: Killed (program cc1plus)

usually happens when the system does not have enough free memory.
A possible workaround is to disable parallel build, and compile file by
file. I added the option `-j1` to Ninja.
2014-09-26 09:05:30 -07:00
Leonardo de Moura
480bc639ea feat(build): add IGNORE_SORRY cmake option
It allows us to perform nightly builds and avoid distracting warning
messages on CDASH.
2014-09-26 08:55:54 -07:00
Leonardo de Moura
f05bb9daeb fix(util/memory): warning when compiling with clang++ 2014-09-26 08:42:47 -07:00
Soonho Kong
4b60499f7d fix(emacs/lean-company): take out '@' from prefix for findp
fix #207
2014-09-26 08:53:48 +02:00
Leonardo de Moura
d02ab15c88 fix(frontends/lean/proof_qed_elaborator): must also create
metavar_closure before solving nested proof_qed

The bug was exposed by the new policy for handling class-instance
resolution. In the new policy, we reject partial solutions.
The bug fixed in this commit was being masked by a partial solution that
was being "completed" later.
2014-09-25 20:07:51 -07:00
Leonardo de Moura
c775da16ec feat(frontends/lean/elaborator): discard partial solution during
class-instance resolution, use only tactic_hints associated with
classes, enforce is_strict
2014-09-25 19:46:08 -07:00
Leonardo de Moura
318fec43a4 feat(frontends/lean/elaborator): use tactic_hints for unsolved placeholders 2014-09-25 17:54:10 -07:00
Leonardo de Moura
bb1c6d44ac fix(frontends/lean/elaborator): missing register_meta 2014-09-25 17:52:36 -07:00
Leonardo de Moura
bcb30f6232 fix(frontends/lean/placeholder_elaborator): missing 'return' 2014-09-25 15:00:06 -07:00
Leonardo de Moura
03f71c73dc perf(kernel/instantiate): cache result of instantiate_type_univ_params
and instantiate_value_univ_params
2014-09-25 13:24:43 -07:00
Leonardo de Moura
23ddacad19 perf(frontends/lean/class): improve is_ext_class procedure 2014-09-25 12:26:09 -07:00
Leonardo de Moura
28796593e3 fix(kernel/type_checker): bug in new is_opaque method 2014-09-25 12:24:14 -07:00
Leonardo de Moura
1d92097781 refactor(kernel/declaration): return reference to type/value/name 2014-09-25 12:17:04 -07:00
Leonardo de Moura
d4236e40b4 feat(kernel/type_checker): expose is_opaque 2014-09-25 11:19:54 -07:00
Leonardo de Moura
8747f12118 refactor(frontends/lean/elaborator): remove unnecessary
set_local_context_for method
2014-09-25 10:21:31 -07:00
Leonardo de Moura
b8eb65aac2 perf(frontends/lean/placeholder_elaborator): reuse local_context, this
is possible now because local_context is a mainly "functional object"
2014-09-25 10:11:41 -07:00
Leonardo de Moura
2e2d2d21f1 refactor(local_context): local_context::scope auxiliary object is not
needed anymore
2014-09-25 09:59:27 -07:00
Leonardo de Moura
cd87539de5 fix(library/unifier): bug in the new next_delta_unfold_case_split 2014-09-25 09:56:32 -07:00
Leonardo de Moura
09162e5fea refactor(frontends/lean/local_context): remove name_generator from local_context 2014-09-25 09:44:34 -07:00
Leonardo de Moura
354c456639 refactor(frontends/lean/local_context): move mvar2meta mapping to elaborator 2014-09-25 09:31:03 -07:00
Leonardo de Moura
fce1113b80 refactor(frontends/lean/coercion_elaborator): simplify
coercion_elaborator interface
2014-09-25 08:48:31 -07:00
Leonardo de Moura
a61b95a87e refactor(frontends/lean/proof_qed_elaborator): simplify
proof_qed_elaborator interface
2014-09-25 08:38:02 -07:00
Leonardo de Moura
18cfce60b9 refactor(frontends/lean/local_context): simplify local_context representation 2014-09-25 08:13:27 -07:00
Leonardo de Moura
0488ee4ee2 perf(library/unifier): process delta expansion case split lazily 2014-09-24 19:16:12 -07:00
Leonardo de Moura
d4c0c01e0b fix(build): add missing file 2014-09-24 12:54:17 -07:00
Leonardo de Moura
516c0c73b9 refactor(*): remove dependency to thread_local C++11 keyword, the
current compilers have several bugs associated with it

We use the simpler __thread (gcc and clang) and
__declspec(thread) (visual studio).
2014-09-24 12:51:04 -07:00
Leonardo de Moura
ca1b8ca80f refactor(util/memory_pool): simplify memory_pool, it is not a template anymore 2014-09-24 10:48:32 -07:00
Leonardo de Moura
41433a4002 refactor(util/rb_tree): remove optimization that creates problems for
some compilers
2014-09-24 10:33:25 -07:00
Leonardo de Moura
9b61b18eaa perf(kernel/replace_visitor): use more expensive/precise cache
It does not use pointer equality, but structural equality
2014-09-24 10:12:29 -07:00
Leonardo de Moura
5489e46ce5 refactor(util/numerics): explicit initialization/finalization 2014-09-24 10:12:29 -07:00
Leonardo de Moura
4205368718 fix(util/sexpr): missing explicit initialization/finalization 2014-09-24 10:12:29 -07:00
Leonardo de Moura
7e84d5df3d refactor(util): explicit initialization/finalization 2014-09-24 10:12:29 -07:00
Leonardo de Moura
8466115665 fix(frontends/lean/class): missing explicit initialization/finalization 2014-09-24 10:12:29 -07:00
Leonardo de Moura
dbe1763b1a refactor(frontends/lean/server): explicit initialization/finalization 2014-09-24 10:12:29 -07:00
Leonardo de Moura
358074ae3d refactor(kernel/record): remove kernel extension for records, we will
implement it outside of the kernel on top of the inductive datatypes
2014-09-24 10:12:28 -07:00
Leonardo de Moura
da481c3274 refactor(kernel): explicit initialization/finalization
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
2014-09-24 10:12:28 -07:00
Leonardo de Moura
4dd7abb14e refactor(library): explicit initialization/finalization 2014-09-23 10:45:14 -07:00
Leonardo de Moura
27ff42d2e0 refactor(library): remove dead files 2014-09-23 10:28:20 -07:00
Leonardo de Moura
83b22823a6 refactor(library/tactic): explicit initialization/finalization 2014-09-23 10:06:15 -07:00
Leonardo de Moura
29d6bff785 refactor(frontends/lean): explicit initialization/finalization 2014-09-23 10:00:36 -07:00
Leonardo de Moura
79cfb32ec7 refactor(util): explicit initialization/finalization 2014-09-23 08:13:33 -07:00
Leonardo de Moura
4437a65d0b refactor(frontends/lean/builtin_cmds): explicit token initialization 2014-09-22 19:22:53 -07:00
Leonardo de Moura
531046626a refactor(*): explicit initialization/finalization for environment extensions 2014-09-22 17:30:29 -07:00
Leonardo de Moura
b6781711b1 refactor(*): explicit initialization/finalization for serialization
modules, expression annotations, and tactics
2014-09-22 15:26:41 -07:00