max/lean4-htt
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions 2
25230 commits 1 branch 0 tags 5.3 GiB
Commit graph

3378 commits

Author SHA1 Message Date
Sebastian Ullrich
2833c61a60 fix: respect preresolved names at resolveConst* 
This makes sure we can properly quote e.g. `deriving` clauses and avoids
a suspicious `eraseMacroScopes` call (though not at `Elab.Syntax`, since
categories do not have to be declaration names)
 2021-07-30 07:17:50 -07:00
Leonardo de Moura
b25bb78e2a feat: improve DiscrTree 
Try to improve the performance issue described at #587.

The issue is that Mathlib contains thousands of theorems where the
associated key for the discrimination tree is just
`Key.other`. The indexing is not effective for them. This happens because

1- Lambda expressions are indexed using `Key.other`. The
   discrimination tree mainly focus on the first-order structure.

2- It unfolds reducible constants when inserting and retrieving
   entries. The motivation is that users expect simp theorems to fire
   modulo reducible constants.

Then, we have many theorems such as
```lean
map ?g ∘ map ?f = map (?g ∘ ?f)
```
when we expand the function composition on the left-hand side, we get
```lean
fun (x : List ?α) => map ?g (map ?f x)
```
Which is indexed as `Key.other`.

We should not avoid the `Array`s in the discrimination tree nodes
If the index is working effectively, these arrays are all very small.

In this commit, we try to address the problem by using a different
approach. When processing the root of a pattern, we interrupt
reduction as soon as the we hit something that would be indexed
as `Key.other`. Note that, in Lean 3, the root of a pattern also
receives special treatment.
 2021-07-29 16:08:26 -07:00
Leonardo de Moura
49a87ceb4d feat: add basic isDefEq cache 2021-07-28 16:29:44 -07:00
Leonardo de Moura
51e03837f5 fix: exact and refine succeed if they produce no new metavariables 
closes #492
 2021-07-27 18:30:14 -07:00
Leonardo de Moura
6d05daf73b feat: add flag for allowing synthetic opaque mvars to be assigned at isDefEq 
See issue #492
TODO: add a mechanism for detecting new metavariables.
 2021-07-27 17:58:08 -07:00
Leonardo de Moura
8c12a264ee fix: offset support at isDefEq should not use HAdd.hAdd 
fixes #550
 2021-07-27 16:16:03 -07:00
Leonardo de Moura
3f22d5f624 feat: take auto params into account in the structure instance notation 
closes #461
 2021-07-27 15:49:23 -07:00
Leonardo de Moura
714cadfb31 fix: bug at structure instance notation 
It was exposed by the second example at #461.
 2021-07-27 11:56:33 -07:00
Wojciech Nawrocki
43190e0e63 feat: FromToJson for recursive inductives 2021-07-24 10:47:38 +02:00
Wojciech Nawrocki
d6893a3e1f fix: more robust LspEncoding 2021-07-24 10:45:28 +02:00
Wojciech Nawrocki
b3d9e90695 feat: IO.getRandomBytes 2021-07-24 10:45:28 +02:00
Wojciech Nawrocki
ef34cfd513 chore: more extensive LspEncoding test 2021-07-24 10:45:28 +02:00
Wojciech Nawrocki
9664fc88e0 chore: add test 2021-07-24 10:45:28 +02:00
Leonardo de Moura
cce6165d4e perf: refine tryHeuristic 2021-07-23 12:04:11 -07:00
Wojciech Nawrocki
1ba802418b fix: ToJson for single-field constructors 2021-07-08 09:01:06 +02:00
Leonardo de Moura
5e694d4b69 fix: fixes #536 2021-06-29 18:24:46 -07:00
Leonardo de Moura
90a79a0b06 chore: remove command universes 
Now, `universe` may declare many universes. The goal is to make it
consistent with the `variable` command
 2021-06-29 17:01:07 -07:00
Gabriel Ebner
3cff5ceb99 perf: make trace[...] ... notation lazy 2021-06-23 00:07:27 -07:00
Sebastian Ullrich
30a0954424 refactor: revert MonadRef changes 2021-06-21 10:17:26 -07:00
Sebastian Ullrich
da4c46370d feat: store elaborator declaration name in info tree 2021-06-21 10:17:26 -07:00
Daniel Selsam
ded51882a0
feat: pp motives and misc delab fixes 2021-06-13 00:06:27 +02:00
Leonardo de Moura
a435f3d641 feat: reduce s.1 =?= v to s =?= ⟨v⟩ if structure has a single field 
This feature was suggested by @dselsam at PR #521.
It closes #509
 2021-06-11 11:23:19 -07:00
Gabriel Ebner
94e299a730 fix: instantiate mvars in rewrite tactic 2021-06-11 10:27:05 -07:00
Reijo Jaakkola
32897440dc
fix: change IO.FS.Handle.read to return empty array at EOF 
Make EOF handling in IO.FS.Handle consistent with EOF handling in
IO.FS.Handle.getLine. Previously returned error at EOF which ended up
causing segmentation fault. Remove the declaration of g_io_error_eof,
since it becomes redundant.

Closes #349
 2021-06-08 13:17:53 +02:00
Daniel Fabian
9200de01ef refactor: fix code review comments. 2021-06-06 06:40:09 -07:00
Daniel Fabian
968ae18f20 fix: deal with params for inductive predicates. 2021-06-06 06:40:09 -07:00
Daniel Fabian
b7ecc1acc3 refactor: Make the non-below version of a premise in the below type for inductive predicates implicit. 
Since it is always fully implied by the below version thereof, it carries no real information and shouldn't be used in pattern matching.
 2021-06-06 06:40:09 -07:00
Daniel Fabian
822c551aa2 test: Add a bunch of test for structural recursion on predicates. 2021-06-06 06:40:09 -07:00
Daniel Fabian
ec6f7d9bd6 feat: Implement structural recursion for inductive predicates. 2021-06-06 06:40:09 -07:00
Daniel Fabian
4b7cb058d3 feat: Add support for inductive types to FromJson and ToJson handlers. 2021-06-05 13:53:10 +02:00
Daniel Fabian
06d1d3ae07 fix: Use UInt64 in deriving handler for Hashable. 2021-06-03 06:38:44 -07:00
Leonardo de Moura
4062dee864 fix: fixes #498 2021-05-31 15:42:13 -07:00
Leonardo de Moura
a247249880 feat: add configuration option for disabling proof irrelevance at MetaM 2021-05-27 13:37:26 -07:00
Leonardo de Moura
b91e22af2b fix: fixes #241 2021-05-22 19:10:07 -07:00
Daniel Fabian
c426a816a1 refactor: Make the non-below version of a premise in the below type for inductive predicates implicit. 
Since it is always fully implied by the below version thereof, it carries no real information and shouldn't be used in pattern matching.
 2021-05-22 18:09:32 -07:00
Leonardo de Moura
28b055463f fix: fixes #471 2021-05-22 15:42:52 -07:00
Leonardo de Moura
edb203ca54 fix: fixes #481 2021-05-21 20:40:26 -07:00
Leonardo de Moura
af4485f40e fix: fixes #482 2021-05-21 19:20:24 -07:00
Leonardo de Moura
8eceb07caf feat: new discriminant refinement procedure 2021-05-21 18:08:11 -07:00
Mac Malone
a6dc9e4ef3 feat: class abbrev now supports a type spec (+ test) 2021-05-20 15:23:30 -07:00
Mac Malone
6c07536b33 feat: simplified, improved class abbrev (+ tests) 2021-05-20 15:23:29 -07:00
Daniel Fabian
42bd44ab82 refactor: Capture environment modification in mkMatcher. 
Doing this allows us to add the declaration in the backtracking case of structural recursion.
 2021-05-20 15:20:16 -07:00
Sebastian Ullrich
a02c6fd3eb chore: adapt stdlib & tests 2021-05-20 15:17:36 -07:00
Daniel Fabian
ab0ef229ac feat: add getBelowIndices. 2021-05-19 07:28:14 -07:00
Daniel Fabian
91ecbb5b5c feat: Add withMkMatcherInput. 
This is the inverse function to `mkMatcher`, i.e. a way to turn a matcher into an input.
 2021-05-19 07:28:14 -07:00
Daniel Fabian
cf030a1634 refactor: Add MkMatcherInput. 
Since we are going to make `mkMatcher` reversible, it's quite useful to have the input be a `structure`. This way we make sure, that the inverse function always returns the same type as `mkMatcher` needs as input.
 2021-05-19 07:28:14 -07:00
Leonardo de Moura
8bbe6cac02 chore: fix test 2021-05-17 14:47:24 -07:00
Leonardo de Moura
53b2ceea51 fix: missing withoutModifyingState at elabSimpConfig 2021-05-16 13:07:13 -07:00
Leonardo de Moura
ac90052139 feat: add option for controlling how deep we go when trying to discharge simp theorem hypotheses 2021-05-16 12:32:05 -07:00
Leonardo de Moura
a498a64490 chore: disable injectivity theorems generation for big structure tests 
The test was producing a stack overflow in debug mode in CI.
 2021-05-15 21:30:40 -07:00