Commit graph

3180 commits

Author SHA1 Message Date
Leonardo de Moura
f335623530 feat(library/constructions/brec_on): add below and ibelow for new inductive datatype module 2018-09-05 14:46:03 -07:00
Leonardo de Moura
afb9584a63 feat(kernel): store at inductive_val whether the type is reflexive or not 2018-09-05 14:46:03 -07:00
Leonardo de Moura
6114e00cea chore(frontends/lean/decl_cmds): dead var 2018-09-05 10:22:07 -07:00
Sebastian Ullrich
3a4a052167 fix(frontends/lean/inductive_cmds): expand inductive fix to mutual inductives 2018-09-05 09:58:49 -07:00
Sebastian Ullrich
df725c1f41 perf(frontends/lean/builtin_exprs): fix regression from position refactoring
Even atomic do patterns were compiled into match expressions
2018-09-05 09:58:49 -07:00
Leonardo de Moura
4773a3be5f feat(library/constructions/no_confusion): add no_confusion_type for new inductive datatype module 2018-09-05 09:55:13 -07:00
Leonardo de Moura
78f4edaf57 chore(frontends/lean): remove info_manager and interactive modules 2018-09-04 17:22:16 -07:00
Leonardo de Moura
d8e7941116 feat(library/constructions/rec_on): add rec_on for new inductive datatype module 2018-09-04 17:22:16 -07:00
Leonardo de Moura
8ed89c6ac3 chore(library): remove normalize.cpp
The command `#reduce` was also temporarily removed.
2018-09-04 10:51:14 -07:00
Leonardo de Moura
1dfcea58f9 chore(*): remove some references to old inductive datatype module 2018-09-04 10:45:17 -07:00
Leonardo de Moura
10a7eccecd feat(library/constructions/cases_on): add cases_on for new inductive datatype module 2018-09-04 09:26:16 -07:00
Leonardo de Moura
af8f3c7bb4 chore(library/constructions): remove injective
It was mainly used to perform dependent elimination with nested
inductive datatypes produced by the inductive compiler.
2018-09-03 17:38:13 -07:00
Leonardo de Moura
cc748c0b17 chore(library/constructions): remove drec (dead code) 2018-09-03 17:32:16 -07:00
Leonardo de Moura
dd03747d22 chore(kernel): univ_param vs lparam, level_param_names ==> names, and other inconsistencies 2018-09-03 13:05:42 -07:00
Leonardo de Moura
7928dbb239 chore(kernel): get_constructor ==> get_cnstr 2018-09-03 12:30:49 -07:00
Leonardo de Moura
5e6d3511f4 chore(frontends/lean/print_cmd): display rule info 2018-09-03 12:27:25 -07:00
Sebastian Ullrich
76741ae695 feat(frontends/lean/elaborator): fall back to approximate position in errors 2018-09-02 18:34:57 -07:00
Sebastian Ullrich
ae70158c09 fix(frontends/lean/elaborator): position information for callees 2018-09-02 18:08:41 -07:00
Sebastian Ullrich
39cdae50ee feat(library,frontends/lean): use mdata instead of hacky cache for position information in preterms 2018-09-02 18:08:41 -07:00
Leonardo de Moura
517923d362 feat(kernel/inductive): generate recursors in the new inductive datatype module 2018-08-31 17:47:22 -07:00
Sebastian Ullrich
93d13ffea3 chore(library/init/lean): lean.parser.reader ~> lean.parser 2018-08-30 16:34:47 -07:00
Sebastian Ullrich
2528aee72b refactor(library/init/lean/parser/reader): use different monad stacks for different parts of the reader 2018-08-29 16:42:24 -07:00
Leonardo de Moura
b79794e601 feat(frontends/lean/print_cmd): add basic support for testing new inductive datatype module 2018-08-29 09:36:47 -07:00
Leonardo de Moura
101886ffae feat(kernel): proper constant_info and declaration objects for quot type 2018-08-28 13:46:31 -07:00
Sebastian Ullrich
46f734b1b1 refactor(library/init/lean/parser/reader): replace macro with syntax_node_kind
Also make sure that the name inside a node kind is the full name of the
declaration. This way, we cannot have accidentally conflicting node kind names.
2018-08-28 13:13:14 -07:00
Leonardo de Moura
776c977742 refactor(kernel): continue constant_info/declaration refactoring 2018-08-27 17:23:26 -07:00
Leonardo de Moura
ae18cee0ea chore(library/module): remove pos_info tracking
We will use a completely different approach in Lean4
2018-08-27 15:55:57 -07:00
Leonardo de Moura
27ef29a50f refactor(kernel/declaration): continue constant_info/declaration refactoring 2018-08-27 13:22:10 -07:00
Leonardo de Moura
7bf032d6fe chore(frontends/lean): remove calc-expression 2018-08-27 12:20:54 -07:00
Leonardo de Moura
d334bb1fa7 chore(*): remove more stuff 2018-08-23 15:56:31 -07:00
Leonardo de Moura
17ae59b5b0 chore(*): remove more stuff 2018-08-23 15:27:12 -07:00
Leonardo de Moura
05a6d8a791 chore(library/tactic): remove destruct_tactic, generalize_tactic and fun_info_tactics 2018-08-23 14:47:51 -07:00
Leonardo de Moura
88c8c560a9 chore(library/equations_compiler): do not generate equation lemmas 2018-08-23 14:04:37 -07:00
Leonardo de Moura
3abcd0c9e8 chore(frontends/lean/elaborator): remove invoke_tactic 2018-08-23 13:49:47 -07:00
Leonardo de Moura
e5c3f04937 chore(frontends/lean): remove tactic notation 2018-08-23 13:44:52 -07:00
Leonardo de Moura
d27ef3d3b0 chore(frontends/lean/builtin_cmds): remove tactic framework dependency 2018-08-23 13:34:38 -07:00
Leonardo de Moura
22ba0a1155 chore(library): remove inverse.cpp
We used this module to implement inductive_compiler pack/unpack functions
2018-08-23 13:16:27 -07:00
Leonardo de Moura
a10e81e6a6 chore(frontends/lean/elaborator): remove unfold tactic dependency 2018-08-23 11:57:38 -07:00
Sebastian Ullrich
61f8ebf4ef refactor(library/init/lean/parser/reader): destructure reader into monad and has_tokens typeclass 2018-08-23 10:38:59 -07:00
Leonardo de Moura
82095cc018 refactor(kernel): split declaration into declaration and constant_info
This is just another step towards the design described at commit 16598391a07d4a
2018-08-22 17:53:11 -07:00
Sebastian Ullrich
14a5246695 fix(frontends/lean/elaborator): nested parser should reset imports 2018-08-22 14:32:03 -07:00
Sebastian Ullrich
6384d118a8 fix(frontends/lean/builtin_cmds,library/vm/vm): fix debug build 2018-08-22 14:32:03 -07:00
Sebastian Ullrich
c69cc61292 feat(frontends/lean,library/init/lean/parser/reader/module): add node_choice! macro
Creates an inductive type from an ordered choice reader
2018-08-22 14:32:03 -07:00
Sebastian Ullrich
1b8e74123c feat(frontends/lean/{builtin_exprs,elaborator}): special-case try inside of node! 2018-08-22 14:32:03 -07:00
Sebastian Ullrich
26d696a937 feat(frontends/lean,library/init/lean/parser/reader/module): basic node! macro 2018-08-22 14:32:03 -07:00
Leonardo de Moura
f3e99286bb chore(kernel): remove certified_declaration 2018-08-22 12:11:34 -07:00
Leonardo de Moura
261dc999d0 refactor(frontends/lean/elaborator): mark thunk as opaque, and thunk A to A is now a coercion
@kha I was working in the new declaration type and using tasks there.
Since we don't have tasks yet in Lean, I decided to start refactoring
the `thunk` type. I defined it as:

```
-- TODO(Leo): mark as opaque, it is implemented by the new runtime
structure thunk (α : Type u) : Type u :=
(fn : unit → α)

def thunk.pure {α : Type u} (a : α) : thunk α :=
⟨λ _, a⟩

def thunk.get {α : Type u} (t : thunk α) : α :=
t.fn ()
```

The idea is to use the runtime primitives to implement them.
Then, I realized the support for `thunk`s in the elaborator are quite
hacky. Given `f x`, if `f`'s domain has type `thunk A`, we elaborate
`f x` as `f (fun _, x)` even if `x` has type `thunk A`.
This is quite bad, for example, suppose we have
```
def f (x : thunk A) := ...
```
Then, the following definition is type incorrect.
```
def g (x : thunk A) := f x
```
and we are forced to write
```
def g (x : thunk A) := f (x ())
```
The term `f (x ())` will be elaborated as `f (fun _, x ())` and an
unnecessary closure is created at runtime.

This mechanism inherited from Lean 3 is also incompatible with the
new thunk definition. Given `x : thunk A`, I want to write `x.get`
to retrieve the value instead of `x ()` as in Lean 3.
However, `x.get` expands into the nonsensical `(fun _, x).get`.

So, I decided to view the mapping `A` to `thunk A` as a "coercion".
I used double quotes, because it is a macro instead of a function.
If it were a coercion, then we would be using `thunk.pure` to coerce
values but this is not we want most of the time.
For example, given `f : thunk A -> B` and a term `t : A`, when we write
`f t`, we want it to be converted into `f (fun _, t)` instead of
`f (thunk.pure t)` which would eagerly compute `t`. The transformation
`t` into `fun _, t` is syntactic.
We cannot implement it using type classes. I implemented it as
a hard-coded extra case like the one from `Prop` to `bool`.
We can also add a coercion from `thunk A` to `A` to avoid the `.get`.

That being said, I had a few breakages in the code base since we only
use coercions when the given and expected type do not contain
metavariables.
2018-08-21 15:27:51 -07:00
Leonardo de Moura
6473bd1294 chore(frontends/lean/builtin_cmds): fix warning 2018-08-21 10:09:39 -07:00
Sebastian Ullrich
0936d4d81e feat(library/init/io): introduce has_eval class to customize #eval output 2018-08-21 08:43:09 -07:00
Leonardo de Moura
fd9bc9e15b feat(frontends/lean/builtin_cmds): write/read compacted region file 2018-08-20 16:35:37 -07:00