michael/lean2
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions
4321 commits 3 branches 0 tags 28 MiB
Commit graph

4321 commits

This branch
This branch
All branches
Author SHA1 Message Date
Leonardo de Moura
4946f55290 refactor(frontends/lean): constant/axiom are top-level commands, parameter/variable/hypothesis/conjecture are section/context-level commands 
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
 2014-10-02 17:55:34 -07:00
Soonho Kong
91f789c3db feat(emacs/lean-cmd): add SYNC command 2014-10-02 17:30:03 -07:00
Soonho Kong
a25f38cd6b fix(bin/linja): avoid using realpath on msys2 2014-10-02 17:25:14 -07:00
Soonho Kong
229caec0a4 fix(bin/linja): fix typo 2014-10-02 17:25:13 -07:00
Soonho Kong
5e551db9be chore(CMakeLists.txt): use $PYTHON_EXECUTABLE instead of "python" 2014-10-02 17:25:13 -07:00
Soonho Kong
fd11b28d15 doc(make/msys2): update instructions 2014-10-02 17:25:12 -07:00
Leonardo de Moura
f78d831de3 refactor(frontends/lean): remove hardcoded Type', and define it using notation 2014-10-02 14:29:51 -07:00
Leonardo de Moura
76d21900a2 feat(library): add aliases for some sorts 2014-10-02 14:29:51 -07:00
Leonardo de Moura
4cb54ac825 feat(frontends/lean/elaborator): more strict test for bad universe solution 2014-10-02 14:29:51 -07:00
Leonardo de Moura
98e66586e9 feat(frontends/lean/elaborator): elaborator rejects 'Type' if the universe is explicit 2014-10-02 14:29:51 -07:00
Soonho Kong
fc19fbf26d chore(.travis.osx.yml): bump g++ version to 4.9.1 2014-10-02 10:44:37 -07:00
Leonardo de Moura
db9671d7c3 fix(frontends/lean/decl_cmds): remove assertion that does not hold anymore 2014-10-02 09:19:40 -07:00
Leonardo de Moura
d42fd657fe refactor(library): is_inhabited "theorems" should be "definitions", they are "data" 2014-10-02 09:00:34 -07:00
Leonardo de Moura
9a75298892 feat(emacs): highlight Type' 2014-10-02 08:04:00 -07:00
Leonardo de Moura
5bd8e9d141 fix(frontends/lean/decl_cmds): allow private transparent definitions 
Example:
section
  universe l
  private definition T := Type.{max 1 l}
  ...
end
 2014-10-02 07:56:01 -07:00
Leonardo de Moura
153e3927ac feat(frontends/lean/elaborator): modify '!' semantics: it stops consuming arguments as soon it finds an argument that does not occur in the rest of the type. 2014-10-01 18:50:17 -07:00
Leonardo de Moura
e64d5c4a4a refactor(library/data/nat): use new operator '!' 2014-10-01 18:39:47 -07:00
Leonardo de Moura
a978e30c81 refactor(library/data/nat): rename to_nat to of_num 2014-10-01 17:52:33 -07:00
Leonardo de Moura
716cd4d651 refactor(library): rename namespace eq_ops to eq.ops 2014-10-01 17:51:17 -07:00
Leonardo de Moura
ead827d6b7 feat(frontends/lean): add ! operator the "dual" of @, closes #220 2014-10-01 17:13:41 -07:00
Leonardo de Moura
bc6ebf34be feat(library/data/bool): do not use ! as notation for bnot, rename band/bor -> and/or 2014-10-01 17:00:03 -07:00
Soonho Kong
72beb438e6 fix(emacs/lean-server): merge server-trace-mode to debug-mode 2014-10-01 14:38:23 -07:00
Leonardo de Moura
c46ec6a548 fix(frontends/lean): missing type information for INFO, fixes #218 2014-10-01 14:29:07 -07:00
Leonardo de Moura
f863d82e69 fix(frontends/lean/pp): pp was not taking into account new namespace name resolution rules, fixes #216 2014-10-01 11:24:45 -07:00
Leonardo de Moura
3657d4c3ab feat(frontends/lean): coercion num -> int even when nat is not open, closes #219 
I also had to mark the coercions as reducible.
Otherwise, given the constraint

          ?M (int.of_num 0) =?= (int.of_nat (nat.to_nat 0))

the unifier will not generate the solution

          ?M := fun x, x
 2014-10-01 11:09:10 -07:00
Leonardo de Moura
966366e18e feat(kernel/inductive): relaxed rules for defining datatypes with explicit universes, closes #217 2014-10-01 10:56:05 -07:00
Leonardo de Moura
263b081e69 fix(frontends/lean/inductive_cmd): temporary aliases must take explicit universes into account, fixes #215 2014-10-01 10:24:44 -07:00
Leonardo de Moura
71077f5c89 feat(frontends/lean/parser): allow _ at level expressions 2014-10-01 10:24:44 -07:00
Soonho Kong
3775d07d5f fix(emacs/lean-company): use lean-grab-hname for findp auto-completion 
fix #214
 2014-10-01 09:53:48 -07:00
Leonardo de Moura
2730e5163e feat(frontends/lean): allow 'sorry' implicit argument to access the whole context, and avoid cryptic error message 
See new test for explanation.
 2014-09-30 18:04:04 -07:00
Soonho Kong
efdf7420b5 fix(bin/linja): disable "\r" trick for messages with flycheck 2014-09-30 17:36:15 -07:00
Soonho Kong
89a929e9cd feat(emacs/lean-syntax): add more word constituents to syntax table 2014-09-30 15:41:18 -07:00
Soonho Kong
9024cb23fd fix(emacs/lean-util): fix lean-grab-id 2014-09-30 15:41:16 -07:00
Soonho Kong
f4e118de76 fix(bin/linja): replace '\' with '/' for flycheck output on windows 2014-09-30 15:41:12 -07:00
Soonho Kong
5deb6e10db fix(bin/linja): use Popen.communicate instead of Popen.wait 
Popen.wait() can cause deadlock.

Reference:
https://docs.python.org/2/library/subprocess.html#subprocess.Popen.wait
 2014-09-30 15:41:08 -07:00
Soonho Kong
562cc2c8f3 feat(bin/linja): add --verbose option 2014-09-30 15:41:04 -07:00
Soonho Kong
d5e77c05cb fix(emacs/lean-flycheck): fix error/warning pattern to work on windows 2014-09-30 15:40:57 -07:00
Leonardo de Moura
3ca1264f61 refactor(library): mark 'decidable' theorems as definitions 
If we don't do that, then any 'if' term that uses one of these theorems
will get "stuck". That is, the kernel will not be able to reduce them
because theorems are always opaque
 2014-09-30 09:02:37 -07:00
Soonho Kong
41acc10d49 feat(emacs/lean-flycheck): detect and use flycheck error buffer width 
Detect the width of flycheck error list buffer and set pp.width when
calling flycheck. It uses the same heuristic that `split-window-sensibly`
function is using.

A user can also set `lean-flycheck-msg-width` custom variable to enforce
a certain width.

Close #205.
 2014-09-30 05:54:46 -07:00
Leonardo de Moura
486839881c feat(*): use environment fingerprint to detect when the cache cannot be used because the configuration changed, closes #75 
We are not taking into the account the options object yet.
 2014-09-29 18:30:00 -07:00
Leonardo de Moura
19dec32844 feat(library): add environment fingerprint 2014-09-29 18:30:00 -07:00
Leonardo de Moura
530997638a feat(library/definition_cache): store the hashcode of direct dependencies 
Actually we store the has code of their types.
 2014-09-29 18:30:00 -07:00
Soonho Kong
b538b3e0bf fix(emacs/lean-company): handle singleton candidate case 
If we have a singleton auto-complete candidate (e.g. proof_irrel) and it
does not matched with input prefix (e.g. foo), we add the input
prefix (foo) as an extra candidate to prevent the
candidate (proof_irrel) from being automatically selected and completed.
 2014-09-29 17:06:01 -07:00
Leonardo de Moura
113879a7dd feat(frontends/lean/server): sort exact matches by size in FINDP 2014-09-29 16:44:55 -07:00
Leonardo de Moura
334977b8bd fix(frontends/lean/info_manager): crash when accessing INFO in the end of the file 2014-09-29 16:44:55 -07:00
Leonardo de Moura
a0c37b231f feat(kernel/expr): add hash_bi function that takes binder information into account 2014-09-29 16:44:55 -07:00
Soonho Kong
7ab8b8acb7 fix(emacs/lean-company): respect lean syntax to extract findp prefix 
close #207
 2014-09-29 16:00:00 -07:00
Leonardo de Moura
0d6d746d98 feat(frontends/lean): check modification time of imported files 
Signed-off-by: Leonardo de Moura <leonardo@microsoft.com>
 2014-09-29 15:17:27 -07:00
Soonho Kong
6c2e576dff fix(emacs/lean-mode): trigger eri-indent only at the beginning of line 2014-09-29 12:53:22 -07:00
Leonardo de Moura
1a1e9a2340 feat(build): add 'CONSERVE_MEMORY' cmake option 
When CONSERVE_MEMORY=ON and gcc is being used, we provide additional
command line options that will minimize the amount of memory used by gcc
when compiling Lean.
This is an attempt to fix the "out-of-memory" failures when building
Lean at Travis.
 2014-09-29 12:44:29 -07:00