Commit graph

116 commits

Author SHA1 Message Date
Kim Morrison
dcba6dfa7e
chore: failing grind test cases for linarith on ordered fields (#9756) 2025-08-06 09:31:09 +00:00
Kim Morrison
ed1ca47199
chore: add failing grind cutsat tests (#9751)
Further `grind` cutsat failures relative to `omega`, found using Anne's
tactic analysis tool in Mathlib.
2025-08-06 04:15:34 +00:00
Kim Morrison
6e06978961
chore: remove >6 month old deprecations (#9640) 2025-08-05 02:29:15 +00:00
Leonardo de Moura
7f22c0883b
perf: Expr.toPoly in grind (#9714)
This PR adds a version of `CommRing.Expr.toPoly` optimized for kernel
reduction. We use this function not only to implement `grind ring`, but
also to interface the ring module with `grind cutsat`.
2025-08-04 15:30:10 +00:00
Leonardo de Moura
2713c846f1
chore: update tests/lean/grind todo folder (#9683)
Remove examples that we have already moved to `tests/lean/run`, and add
notes for possible fixes.
2025-08-02 13:10:34 +00:00
Kim Morrison
062ac89c34
chore: failing test cases for grind regressions vs omega (#9656) 2025-08-01 02:19:16 +00:00
Kyle Miller
4575799f8e
chore: library style cleanup (#9654)
This PR cleans up the style of the library in anticipation of a future
PR that requires strict indentation for tactic sequences.
2025-07-31 21:28:59 +00:00
Kim Morrison
edade0cea8
chore: add failing grind test about exponents (#9611) 2025-07-29 02:53:43 +00:00
Kim Morrison
1726a61e88
chore: add failing test for grind and BitVec (#9608)
This PR adds a failing test for `grind`, reported by @eric-wieser.

```
example {x : BitVec 2} : x - 2 • x + x = 0 := by
  grind -- fails
```

There are several independent problems here!

1. Cutsat doesn't evaluate `2 ^ 2`:
```
-- [cutsat] Assignment satisfying linear constraints
-- [assign] 「2 ^ 2」 := 0
```

2. We don't normalize `3 * 2 • x` to `6 * x` in the ring solver:
```
-- [ring] Rings ▼
--   [] Ring `BitVec 2` ▼
--     [diseqs] Disequalities ▼
--       [_] ¬2 * x + 3 * 2 • x = 0
```
This should then give a contradiction because the characteristic of
`BitVec 2` is 4.

3. In `Int`, we're not normalizing `*` and `•`:
```
-- [ring] Rings ▼
--   [] Ring `Int` ▼
--     [basis] Basis ▼
--       [_] 2 * ↑x + -1 * ↑(2 • x) + -4 * ((2 * ↑x + -1 * ↑(2 • x)) / 4) + -1 * ((2 * ↑x + -1 * ↑(2 • x)) % 4) = 0
```
2025-07-29 02:03:30 +00:00
Kim Morrison
9399b2ee36
chore: add failing grind test (#9607)
This PR adds a failing grind test.
2025-07-29 01:36:53 +00:00
Kim Morrison
68adcdb475
chore: maintain failing grind tests about Nat as a semiring (#9501)
These tests (still failing until we embed a NatModule in its IntModule
envelope) had further broken because of name changes.
2025-07-24 06:44:39 +00:00
Kim Morrison
3cde12567f
feat: add HPow Int field to Field (#9500)
This PR adds a `HPow \a Int \a` field to `Lean.Grind.Field`, and
sufficient axioms to connect it to the operations, so that in future we
can reason about exponents in `grind`. To avoid collisions, we also move
the `HPow \a Nat \a` field in `Semiring` from the extends clause to a
field. Finally, we add some failing tests about normalizing exponents.
2025-07-24 06:00:11 +00:00
Leonardo de Moura
e9a55bfff7
chore: move test to run (#9183) 2025-07-04 00:33:39 +00:00
Kim Morrison
c06af84d9f
fix: refactor grind's module/ring design to avoid a diamond (#9168)
This PR resolves a defeq diamond, which caused a problem in Mathlib:
```
import Mathlib

example (R : Type) [I : Ring R] :
  @AddCommGroup.toGrindIntModule R (@Ring.toAddCommGroup R I) =
    @Lean.Grind.Ring.instIntModule R (@Ring.toGrindRing R I) := rfl -- fails
```
2025-07-03 06:50:46 +00:00
Leonardo de Moura
5f818826d1
feat: add [grind symbol <prio>] attribute (#9169)
This PR adds the attribute `[grind symbol <prio>]`. This is just the
first part of the PR.
2025-07-03 06:06:55 +00:00
Leonardo de Moura
535ce0b8fd
feat: pow_add normalization in grind (#9133)
This PR adds support for `a^(m+n)` in the `grind` normalizer.
2025-07-01 17:52:16 +00:00
Kim Morrison
6ab8ed34d0
chore: grind test case for matching Nat arguments (#9123) 2025-07-01 07:21:36 +00:00
Leonardo de Moura
b95b0069e7
feat: use comm ring module to normalize nonlinear polynomials in grind cutsat (#9074)
This PR uses the commutative ring module to normalize nonlinear
polynomials in `grind cutsat`. Examples:
```lean
example (a b : Nat) (h₁ : a + 1 ≠ a * b * a) (h₂ : a * a * b ≤ a + 1) : b * a^2 < a + 1 := by 
  grind

example (a b c : Int) (h₁ : a + 1 + c = b * a) (h₂ : c + 2*b*a = 0) : 6 * a * b - 2 * a ≤ 2 := by 
  grind
```
2025-06-29 11:09:29 +00:00
Leonardo de Moura
5ca6eadd50
feat: equations <num> = 0 in grind ring (#9062)
This PR implements support for equations `<num> = 0` in rings and fields
of unknown characteristic. Examples:
```lean
example [Field α] (a : α) : (2 * a)⁻¹ = a⁻¹ / 2 := by grind

example [Field α] (a : α) : (2 : α) ≠ 0 → 1 / a + 1 / (2 * a) = 3 / (2 * a) := by grind

example [CommRing α] (a b : α) (h₁ : a + 2 = a) (h₂ : 2*b + a = 0) : a = 0 := by
  grind

example [CommRing α] (a b : α) (h₁ : a + 6 = a) (h₂ : b + 9 = b) (h₂ : 3*b + a = 0) : a = 0 := by
  grind

example [CommRing α] (a b : α) (h₁ : a + 6 = a) (h₂ : b + 9 = b) (h₂ : 3*b + a = 0) : a = 0 := by
  grind

example [CommRing α] (a b : α) (h₁ : a + 2 = a) (h₂ : b = 0) : 4*a + b = 0 := by
  grind

example [CommRing α] (a b c : α) (h₁ : a + 6 = a) (h₂ : c = c + 9) (h : b + 3*c = 0) : 27*a + b = 0 := by
  grind

```
2025-06-28 14:28:42 +00:00
Kim Morrison
af22926d53
chore: updates to (failing) grind algebra tests (#8987) 2025-06-25 02:44:59 +00:00
Leonardo de Moura
f1021e4537
fix: congruence proof for over-applied terms (#8983)
This PR fixes a bug in congruence proof generation in `grind` for
over-applied functions.
2025-06-24 22:04:23 +00:00
Kim Morrison
cc493e688b
feat: embed a NatModule in its IntModule completion (#8963)
This PR embeds a NatModule into its IntModule completion, which is
injective when we have AddLeftCancel, and monotone when the modules are
ordered. Also adds some (failing) grind test cases that can be verified
once `grind` uses this embedding.
2025-06-24 05:30:43 +00:00
Kim Morrison
449bc31832
chore: adds (failing) grind algebra tests (#8961) 2025-06-24 03:51:39 +00:00
Leonardo de Moura
0941d53f6a
feat: semiring normalizer in grind (#8953)
This PR implements support for normalization for commutative semirings
that do not implement `AddRightCancel`. Examples:
```lean
variable (R : Type u) [CommSemiring R]

example (a b c : R) : a * (b + c) = a * c + b * a := by grind
example (a b : R) : (a + b)^2 = a^2 + 2 * a * b + b^2 := by grind
example (a b : R) : (a + 2 * b)^2 = a^2 + 4 * a * b + 4 * b^2 := by grind
example (a b : R) : (a + 2 * b)^2 = 4 * b^2 + b * 4 * a + a^2 := by grind
```
2025-06-24 01:09:22 +00:00
Leonardo de Moura
ba07e46368
refactor: simplify semiring normalization helper theorems (#8946)
This PR simplifies the semiring normalization theorem that will be used
by `grind`.
2025-06-23 23:20:20 +00:00
Kim Morrison
8f4b2909de
chore: cleanup of grind's order typeclasses (#8913)
This PR cleans up `grind`'s internal order typeclasses, removing
unnecessary duplication.
2025-06-22 23:36:48 +00:00
Leonardo de Moura
7531d16112
feat: (commutative) semiring support in grind (#8921)
This PR implements support for (commutative) semirings in `grind`. It
uses the Grothendieck completion to construct a (commutative) ring
`Lean.Grind.Ring.OfSemiring.Q α` from a (commutative) semiring `α`. This
construction is mostly useful for semirings that implement
`AddRightCancel α`. Otherwise, the function `toQ` is not injective.
Examples:
```lean
example (x y : Nat) : x^2*y = 1 → x*y^2 = y → y*x = 1 := by
  grind 

example [CommSemiring α] [AddRightCancel α] (x y : α) : x^2*y = 1 → x*y^2 = y → y*x = 1 := by
  grind

example (a b : Nat) : 3 * a * b = a * b * 3 := by grind

example (k z : Nat) : k * (z * 2 * (z * 2 + 1)) = z * (k * (2 * (z * 2 + 1))) := by grind

example [CommSemiring α] [AddRightCancel α] [IsCharP α 0] (x y : α) 
    : x^2*y = 1 → x*y^2 = y → x + y = 1 → False := by
  grind
```
2025-06-21 23:00:16 +00:00
Leonardo de Moura
588df4612a
fix: missing isEqFalse (#8893)
This PR fixes a bug in the `dvd` propagation function in cutsat.
2025-06-20 08:16:08 +00:00
Kim Morrison
743c60224a
chore: minimize grind panic (#8889) 2025-06-20 01:07:14 +00:00
Kim Morrison
d0c1053903
chore: add test case for grind panic (#8861)
This PR adds a (failing) test case for a panic caused by grind.
2025-06-18 08:00:17 +00:00
Markus Himmel
c16204615d
chore: add a failing grind test (#8858) 2025-06-18 07:14:56 +00:00
Kim Morrison
16e67dc738
feat: grind annotations for Nat.Bitwise (#8852)
This PR adds grind annotations for `Nat.testBit` and bitwise operations
on `Nat`.

(Also includes some in-progress tests for `BitVec`.)
2025-06-18 02:42:43 +00:00
Leonardo de Moura
4e96a4ff45
feat: eliminate equations in grind linarith (#8810)
This PR implements equality elimination in `grind linarith`. The current
implementation supports only `IntModule` and `IntModule` +
`NoNatZeroDivisors`
2025-06-16 09:31:13 +00:00
Kim Morrison
ddff851294
chore: cleanup of grind tests (#8806) 2025-06-16 02:47:46 +00:00
Leonardo de Moura
f86560d134
fix: grind bogus warning and missing normalization rule (#8797)
This PR adds small fixes
2025-06-15 03:44:53 +00:00
Leonardo de Moura
cc3dafe67a
fix: grind linarith internalization and HSMul support (#8796)
This PR fixes `grind linarith` term internalization and support for
`HSMul`.
2025-06-15 02:34:42 +00:00
Leonardo de Moura
1d971c8735
feat: Rabinowitsch transformation in grind (#8789)
This PR implements the Rabinowitsch transformation for `Field`
disequalities in `grind`. For example, this transformation is necessary
for solving:
```lean
example [Field α] (a : α) : a^2 = 0 → a = 0 := by
  grind
```
2025-06-14 22:22:40 +00:00
Leonardo de Moura
ff6eb56f5c
fix: natCast in grind cutsat (#8776)
This PR ensures that user provided `natCast` application are properly
internalized in the grind cutsat module.
2025-06-13 17:56:00 +00:00
Leonardo de Moura
4b7ea26d91
fix: add grind normalization theorem for Int.negSucc (#8775)
This PR adds a `grind` normalization theorem for `Int.negSucc`. Example:

```lean
example (p : Int) (n : Nat) (hmp : Int.negSucc (n + 1) + 1 = p)
    (hnm : Int.negSucc (n + 1 + 1) + 1 = Int.negSucc (n + 1)) : p = Int.negSucc n := by
  grind
```
2025-06-13 16:53:42 +00:00
Kim Morrison
b280b83c98
chore: add test case with bad grind pattern (#8757) 2025-06-13 01:06:02 +00:00
Kim Morrison
faffe86334
chore: add failing grind tests from Mathlib (#8737) 2025-06-12 05:57:32 +00:00
Kim Morrison
abfc49d0f7
chore: cleanup of grind tests (#8735) 2025-06-12 04:42:25 +00:00
Kim Morrison
eccc472e8d
chore: remove set_option grind.warning false (#8714)
This PR removes the now unnecessary `set_option grind.warning false`
statements, now that the warning is disabled by default.
2025-06-11 05:09:19 +00:00
Kim Morrison
b1980ef871
chore: cleanup notes about grind in LRAT (#8623)
This PR cleans up some notes about `grind` failures in the LRAT checker,
now that some `grind` bugs have been fixed.
2025-06-10 03:47:28 +00:00
Leonardo de Moura
8165ecc1db
fix: bug in the equality resolution procedure in grind (#8621)
This PR fixes a bug in the equality-resolution procedure used by
`grind`.
The procedure now performs a topological sort so that every simplified
theorem declaration is emitted **before** any place where it is
referenced.
Previously, applying equality resolution to
```lean
h : ∀ x, p x a → ∀ y, p y b → x ≠ y
```
in the example
```lean
example
  (p : Nat → Nat → Prop)
  (a b c : Nat)
  (h  : ∀ x, p x a → ∀ y, p y b → x ≠ y)
  (h₁ : p c a)
  (h₂ : p c b) :
  False := by
  grind
```
caused `grind` to produce the incorrect term
```lean
p ?y a → ∀ y, p y b → False
```
The patch eliminates this error, and the following correct simplified
theorem is generated
```lean
∀ y, p y a → p y b → False
```
2025-06-04 00:34:47 +00:00
Leonardo de Moura
344b52f999
fix: term internalization issue in grind (#8619)
This PR fixes an internalization (aka preprocessing) issue in `grind`
when applying injectivity theorems.
2025-06-04 00:13:51 +00:00
Kim Morrison
f7b6e155d4
chore: add failing grind test (#8608) 2025-06-03 07:45:38 +00:00
Kim Morrison
f4e86e310c
chore: add failing grind test (unknown metavariable) (#8607) 2025-06-03 07:00:56 +00:00
Kim Morrison
6c17ad8954
chore: add failing grind test (#8599)
`@[grind local]` currently doesn't work as expected on theorems in
namespaces.
2025-06-03 01:49:36 +00:00
Leonardo de Moura
f6c83f3dce
chore: adjust test (#8567)
It is working now
2025-06-01 00:21:23 +00:00