Commit graph

1484 commits

Author SHA1 Message Date
Kim Morrison
2eb478787f
chore: split Int.DivModLemmas into Bootstrap and Lemmas (#7162)
This PR splits `Int.DivModLemmas` into a `Bootstrap` and `Lemmas` file,
where it is possible to use `omega` in `Lemmas`.

I'm going to add more theory, particularly about `fdiv` and `tdiv` to
the `Lemmas` file, and would prefer to have access to `omega`.
2025-02-20 12:05:09 +00:00
Kim Morrison
2960b384af
feat: strengthen Int.tdiv_eq_ediv (#7158)
This PR strengthens `Int.tdiv_eq_ediv`, by dropping an unnecessary
hypothesis, in preparation for further work on `ediv`/`tdiv`/`fdiv`
lemmas.
2025-02-20 05:57:47 +00:00
Leonardo de Moura
2c2b3641f1
feat: helper theorem for cutsat (#7156)
This PR adds a helper theorem that will be used in divisibility
constraint conflict resolution during model construction.
2025-02-20 05:39:49 +00:00
Kim Morrison
6a4225bf04
chore: complete variable name linting for Vector (#7154) 2025-02-20 02:42:50 +00:00
Leonardo de Moura
c86073830f
feat: infrastructure for inequalities constraints in cutsat (#7152)
This PR implements the infrastructure for supporting integer inequality
constraints in the cutsat procedure.
2025-02-19 23:09:12 +00:00
Joachim Breitner
8672186a4e
refactor: use more fun_induction in Init.Data.Int.Linear (#7142)
This uses the improved functional induction theorems from #7140.
2025-02-19 22:56:23 +00:00
Joachim Breitner
5bee3288ac refactor: update proofs after stage0 update for #7140 2025-02-19 20:59:01 +01:00
Kim Morrison
8a2e21cfc4
chore: linting variable names in List/Array (#7146) 2025-02-19 12:45:02 +00:00
Kim Morrison
1d9b19189a chore: deprecate Array.get
fix test
2025-02-19 08:48:33 +11:00
Kim Morrison
d3c36bd7cf chore: use as[i] instead of as.get i 2025-02-19 08:48:33 +11:00
Kyle Miller
2d4c0017b8
chore: review uses of generalize (#7126)
This PR looks at some uses of the `generalize` tactic, especially when
used in conjunction with `induction`.
2025-02-18 14:07:40 +00:00
Markus Himmel
3a22035dad
feat: IntX.abs (#7131)
This PR adds `IntX.abs` functions. These are specified by `BitVec.abs`,
so they map `IntX.minValue` to `IntX.minValue`, similar to Rust's
`i8::abs`. In the future we might also have versions which take values
in `UIntX` and/or `Nat`.
2025-02-18 13:16:30 +00:00
Markus Himmel
a26c937650
feat: Repr and Hashable for IntX (#7128)
This PR adds `Repr` and `Hashable` instances for `IntX`.
2025-02-18 11:03:53 +00:00
Kyle Miller
219f36f499
chore: cleanup after stage0 update for #7103 (#7125)
This PR cleans up the bootstrapping workarounds introduced in #7103
(`induction` target generalization equation names).
2025-02-18 05:55:49 +00:00
Leonardo de Moura
b5bf7d4b87
feat: cutsat divisibility constraint solver proof by reflection theorems (#7124)
This PR adds the helper theorems for justifying the divisibility
constraint solver in the cutsat procedure used by the `grind` tactic.
2025-02-18 04:30:09 +00:00
Kyle Miller
6a59926592
feat: add generalization hypotheses to induction tactic (#7103)
This PR gives the `induction` tactic the ability to name hypotheses to
use when generalizing targets, just like in `cases`. For example,
`induction h : xs.length` leads to goals with hypotheses `h : xs.length
= 0` and `h : xs.length = n + 1`. Target handling is also slightly
modified for multi-target induction principles: it used to be that if
any target was not a free variable, all of the targets would be
generalized (thus causing free variables to lose their connection to the
local hypotheses they appear in); now only the non-free-variable targets
are generalized.

This gives `induction` the last basic feature of the mathlib
`induction'` tactic, which has been long-requested. Recent Zulip
discussion:
https://leanprover.zulipchat.com/#narrow/channel/270676-lean4/topic/To.20replace.20.60induction'.20h.20.3A.20f.20x.60/near/499482173
2025-02-18 03:46:23 +00:00
Kim Morrison
ca253ae4cf
chore: repair defeqs for List GetElem instances (#7121)
This PR repairs some defeq breakages from #7059.
2025-02-18 02:19:08 +00:00
Kim Morrison
4b307914fc
chore: cleanup duplicate theorems (#7113) 2025-02-18 01:46:12 +00:00
Markus Himmel
5a8b4459c8
feat: conversions between Float and finite integers (#7083)
This PR adds (value-based, not bitfield-based) conversion functions
between `Float`/`Float32` and `IntX`/`UIntX`.
2025-02-17 15:42:10 +00:00
Kim Morrison
f07e72fad7
chore: linting variable names for List (#7107) 2025-02-17 14:50:43 +00:00
Kim Morrison
88664e4a99
feat: complete aligning List/Array/Vector.finRange (#7106)
This PR completes the alignment of `List/Array/Vector.finRange` lemmas.
2025-02-17 06:11:43 +00:00
Kim Morrison
9d1fb9f4fa
feat: align Array/Vector.extract lemmas with List (#7105)
This PR completes aligning `Array/Vector.extract` lemmas with the lemmas
for `List.take` and `List.drop`.
2025-02-17 04:56:04 +00:00
Luisa Cicolini
6a17e62523
feat: add BitVec.[(getMsbD, msb)_extractLsb', (getLsbD, getMsbD, msb)_extractLsb] , add and_eq_decide, or_eq_decide, decide_eq_true_iff to bool_to_prop (#6792)
This PR adds theorems `BitVec.(getMsbD, msb)_(extractLsb', extractLsb),
getMsbD_extractLsb'_eq_getLsbD`.

---------

Co-authored-by: Siddharth <siddu.druid@gmail.com>
Co-authored-by: Alex Keizer <alex@keizer.dev>
Co-authored-by: Kim Morrison <kim@tqft.net>
Co-authored-by: Tobias Grosser <tobias@grosser.es>
Co-authored-by: Tobias Grosser <github@grosser.es>
2025-02-17 03:02:37 +00:00
Kim Morrison
1ce7047bf5
feat: cleanup of get and back functions on List/Array (#7059)
This PR moves away from using `List.get` / `List.get?` / `List.get!` and
`Array.get!`, in favour of using the `GetElem` mediated getters. In
particular it deprecates `List.get?`, `List.get!` and `Array.get?`. Also
adds `Array.back`, taking a proof, matching `List.getLast`.
2025-02-17 01:43:45 +00:00
Kitamado
6f5bb3e896
fix: allow trailing comma in array syntax (#7055)
This PR improves array and vector literal syntax by allowing trailing
commas. For example, `#[1, 2, 3,]`.

see: [Why Are Trailing Commas Not Allowed in Array
Literals?](https://leanprover.zulipchat.com/#narrow/channel/270676-lean4/topic/Why.20Are.20Trailing.20Commas.20Not.20Allowed.20in.20Array.20Literals.3F)

Note: we need to preserve the current name for the array syntax
(`«term#[_,]»`) to avoid a bootstrapping issue. The `FromJson`/`ToJson`
deriving handlers use array syntax in macros, and the stage0 version is
used in most of the prelude.
2025-02-16 19:26:23 +00:00
Leonardo de Moura
dd3652ecdc
feat: cutsat preparations (#7097)
This PR implements several modifications for the cutsat procedure in
`grind`.
- The maximal variable is now at the beginning of linear polynomials. 
- The old `LinearArith.Solver` was deleted, and the normalizer was moved
to `Simp`.
- cutsat first files were created, and basic infrastructure for
representing divisibility constraints was added.
2025-02-16 02:52:14 +00:00
Tobias Grosser
a9efbf04f4
feat: make BitVec.getElem the simp normal form and use it in ext (#5498)
This PR makes `BitVec.getElem` the simp normal form in case a proof is
available and changes `ext` to return `x[i]` + a hypothesis that proves
that we are in-bounds. This aligns `BitVec` further with the API
conventions of the Lean standard datatypes.

We move our proofs to this new normal form, which results in slightly
smaller proofs. With the exception of `getElem_ofFin`, no new API
surface is added as the `getElem` API has already been completed over
the previous months. We also move `getElem_shiftConcat_*` a bit higher
as they are needed in earlier proofs. To keep the changeset small, we do
not update the API of `BVDecide` but insert `←
BitVec.getLsbD_eq_getElem` at the few locations where it is needed.
Finally, we add a simproc for getElem, mirroring the existing ones for
getLsbD/getMsdD.

---------

Co-authored-by: Alex Keizer <alex@keizer.dev>
2025-02-16 00:04:56 +00:00
Leonardo de Moura
3a76ac5620
chore: cleanup and missing grind normalization rules (#7095)
This PR adds missing `grind` normalization rules, and removes dead
theorems.
2025-02-15 23:45:35 +00:00
Leonardo de Moura
747ea91c3a
refactor: add denote' functions to Int/Linear.lean (#7094)
This PR adds the functions `Poly.denote'`, `RelCnstr.denote'`, and
`DvdCnstr.denote'`. These functions are useful for representing the
denotation of normalized results in `simp +arith` and the `grind`
preprocessor. This PR also adjusts all auxiliary normalization theorems
to use them to represent the normalized constraints. Previously, we were
converting `RelCnstr` and `DvdCnstr` back into raw constraints. While
this overhead was reasonable for `simp +arith`, it is not for the cutsat
procedure, which has no need for raw constraints. All constraints have
already been normalized by the time they reach cutsat.
2025-02-15 22:10:23 +00:00
Leonardo de Moura
ecdc2d57f2
refactor: Int.Linear module (#7093)
This PR cleans up the `Int.Linear` module by normalizing function and
type names and adding documentation strings. We will use it to implement
cutsat in the `grind` tactic.
2025-02-15 19:20:18 +00:00
Leonardo de Moura
f4afcfc923
feat: divisibility constraint normalizer (#7092)
This PR implements divisibility constraint normalization in `simp
+arith`.
2025-02-15 04:20:40 +00:00
Leonardo de Moura
57aadf8af9
feat: add helper theorems for normalizing divisibility constraints (#7091)
This PR adds helper theorems for normalizing divisibility constraints.
They are going to be used to implement the cutsat procedure in the
`grind` tactic.
2025-02-15 02:44:49 +00:00
Markus Himmel
dd1a4188a0
feat: Fin.toNat (#7079)
This PR introduces `Fin.toNat` as an alias for `Fin.val`. We add this
function for discoverability and consistency reasons. The normal form
for proofs remains `Fin.val`, and there is a `simp` lemma rewriting
`Fin.toNat` to `Fin.val`.
2025-02-14 11:59:44 +00:00
Markus Himmel
ed42d068d4
feat: UIntX.ofNatTruncate (#7080)
This PR adds the functions `UIntX.ofNatTruncate` (the version for
`UInt32` already exists).
2025-02-14 11:59:41 +00:00
Markus Himmel
784444c7a9
feat: IntX.minValue, IntX.maxValue, IntX.ofIntLE, IntX.ofIntTruncate (#7081)
This PR adds functions `IntX.ofIntLE`, `IntX.ofIntTruncate`, which are
analogous to the unsigned counterparts `UIntX.ofNatLT` and
`UInt.ofNatTruncate`.
2025-02-14 11:59:37 +00:00
Markus Himmel
47548aa171
chore: rename UIntX.ofNatCore, UIntX.ofNat' -> UIntX.ofNatLT (#7071)
This PR unifies the existing functions `UIntX.ofNatCore` and
`UIntX.ofNat'` under a new name, `UIntX.ofNatLT`.
2025-02-14 06:58:15 +00:00
Leonardo de Moura
b26b781992
feat: simprocs for Int and Nat divides predicates (#7078)
This PR implements simprocs for `Int` and `Nat` divides predicates.
2025-02-14 05:43:38 +00:00
Leonardo de Moura
2c2a3a65b2
feat: support theorems for cutsat Div-Solve rule (#7077)
This PR proves the helper theorems for justifying the "Div-Solve" rule
in the cutsat procedure.
2025-02-14 04:55:58 +00:00
Markus Himmel
b38da34db2
chore: rename BitVec.ofNatLt -> BitVec.ofNatLT (#7064)
This PR renames `BitVec.ofNatLt` to `BitVec.ofNatLT` and sets up
deprecations for the old name.
2025-02-13 12:52:31 +00:00
Markus Himmel
4a900cc65c
chore: rename IntX.toNat -> IntX.toNatClampNeg (#7066)
This PR renames `IntX.toNat` to `IntX.toNatClampNeg` (to reduce
surprises) and sets up a deprecation.
2025-02-13 12:14:28 +00:00
Markus Himmel
a3fd2eb0fe
chore: make IntX constructor private, provide UIntX.toIntX (#7062)
This PR introduces the functions `UIntX.toIntX` as the public API to
obtain the `IntX` that is 2's complement equivalent to a given `UIntX`.
2025-02-13 11:29:31 +00:00
Markus Himmel
04fe72fee0
feat: missing conversion functions for ISize (#7063)
This PR adds `ISize.toInt8`, `ISize.toInt16`, `Int8.toISize`,
`Int16.toISize`.
2025-02-13 11:02:00 +00:00
Markus Himmel
7c9454edd2
feat: UIntX.ofFin (#7056)
This PR adds the `UIntX.ofFin` conversion functions.
2025-02-13 08:45:01 +00:00
Markus Himmel
1ecb4a43ae
chore: rename UIntX.val -> UIntX.toFin (#7050)
This PR renames the functions `UIntX.val` to `UIntX.toFin`.
2025-02-13 07:50:47 +00:00
Kim Morrison
ae9d12aeaa
chore: upstream an Int lemma (#7060) 2025-02-13 03:19:02 +00:00
Markus Himmel
9ff4d53d0b
chore: rename UIntX.mk -> UIntX.ofBitVec (#7046)
This PR renames `UIntX.mk` to `UIntX.ofBitVec` and adds deprecations.
2025-02-12 16:08:03 +00:00
Markus Himmel
b08fc5dfda
feat: IntX.ofBitVec (#7048)
This PR adds the functions `IntX.ofBitVec`.
2025-02-12 14:49:31 +00:00
Joachim Breitner
bdf4b792a8
feat: wf_preprocess for {List,Array}.Monadic functions (#7034)
This PR adds `wf_preprocess` theorems for
`{List,Array}.{foldlM,foldrM,mapM,filterMapM,flatMapM}`
2025-02-12 09:06:12 +00:00
Kim Morrison
3cf6fb2405
chore: fix linter.listVariables naming (#7044) 2025-02-12 05:17:39 +00:00
Leonardo de Moura
2a67a49f31
chore: simp_arith has been deprecated (#7043)
This PR deprecates the tactics `simp_arith`, `simp_arith!`,
`simp_all_arith` and `simp_all_arith!`. Users can just use the `+arith`
option.
2025-02-12 03:55:45 +00:00