import data.equiv.basic import group_theory.subgroup universes u v section open tactic interactive (parse loc.wildcard) interactive.types (location) lean.parser (many ident) run_cmd mk_simp_attr `with_zero_simp meta def tactic.with_zero_cases : list expr → tactic unit | (h::t) := seq (induction h [] (some `with_zero.cases_on) >> skip) $ tactic.with_zero_cases t | [] := do try (interactive.norm_cast loc.wildcard), try (tactic.interactive.simp_core {} assumption ff [] [`with_zero_simp] loc.wildcard), try (do exfalso, assumption) /-- Case bashing for with_zero. If `x₁, ... x_n` have type `with_zero α` then `with_zero cases x₁ ... x_n` will split according to whether each `x_i` is zero or coerced from `α` then run `norm_cast at *`, try to simplify using the simp rules `with_zero_simp`, and try to get a contradiction. -/ meta def tactic.interactive.with_zero_cases (l : parse $ many ident) := l.mmap tactic.get_local >>= tactic.with_zero_cases end namespace with_zero variables {α : Type u} {β : Type v} @[simp, with_zero_simp] lemma zero_le [preorder α] {x : with_zero α} : 0 ≤ x := by { intros y hy, cases hy } @[simp, with_zero_simp] lemma zero_lt_coe [preorder α] {a : α} : (0 : with_zero α) < a := ⟨a, rfl, λ y hy, by cases hy⟩ @[simp, with_zero_simp] lemma not_coe_eq_zero [preorder α] {x : α} : ¬ (x : with_zero α) = 0 := λ h, option.no_confusion h @[elim_cast] lemma coe_le_coe [preorder α] {x y : α} : (x : with_zero α) ≤ (y : with_zero α) ↔ x ≤ y := ⟨λ h, by rcases (h x rfl) with ⟨z, ⟨h2⟩, h3⟩; exact h3, λ _ _ h, ⟨y, rfl, by cases h ; assumption⟩⟩ @[elim_cast] lemma coe_lt_coe [preorder α] {x y : α} : (x : with_zero α) < (y : with_zero α) ↔ x < y := by repeat { rw [lt_iff_le_not_le, coe_le_coe] } -- TODO: replace `coe_one` in mathlib by this one, which seems to be stated as needed by norm_cast. -- Same remark applies to the next two lemmas. @[elim_cast] lemma coe_one' [has_one α] : (1 : with_zero α) = ((1 : α) : with_zero α) := rfl @[move_cast] lemma inv_coe' {α : Type*} [has_inv α] (a : α) : ((a⁻¹ : α) : with_zero α) = (a : with_zero α)⁻¹ := rfl @[move_cast] lemma mul_coe' {α : Type*} [has_mul α] (a b : α) : ((a * b : α) : with_zero α) = (a : with_zero α) * b := rfl attribute [elim_cast] coe_inj @[simp] lemma le_zero_iff_eq_zero [preorder α] {x : with_zero α} : x ≤ 0 ↔ x = 0 := begin with_zero_cases x, intro h, rcases h x rfl with ⟨_, h, _⟩, exact option.no_confusion h, end @[simp] lemma not_coe_le_zero [preorder α] (x : α) : ¬ (x : with_zero α) ≤ 0 := begin intro h, rw le_zero_iff_eq_zero at h, exact not_coe_eq_zero h, end @[simp] lemma not_lt_zero [preorder α] (x : with_zero α) : ¬ x < 0 := begin intro h, with_zero_cases x, exact lt_irrefl _ h, exact not_coe_le_zero x (le_of_lt h), end @[simp] lemma map_zero {f : α → β} : map f 0 = 0 := option.map_none' @[simp, elim_cast] lemma map_coe {f : α → β} {a : α} : map f (a : with_zero α) = f a := option.map_some' @[simp] lemma map_id {α : Type*} : map (id : α → α) = id := option.map_id lemma map_comp {α β γ : Type*} (f : α → β) (g : β → γ) (r : with_zero α) : map (g ∘ f) r = map g (map f r) := by cases r; refl @[simp] lemma map_eq_zero_iff {f : α → β} {a : with_zero α} : map f a = 0 ↔ a = 0 := ⟨λ h, by with_zero_cases a, λ h, by simp [h]⟩ lemma injective_map {f : α → β} (H : function.injective f) : function.injective (map f) := option.injective_map H lemma map_monotone [preorder α] [preorder β] {f : α → β} (H : monotone f) : monotone (map f) := λ x y, by { with_zero_cases x y, exact λ h, H h } lemma map_strict_mono [linear_order α] [partial_order β] {f : α → β} (H : ∀ a b, a < b → f a < f b) : strict_mono (map f) := λ x y, by { with_zero_cases x y, exact λ h, H _ _ h } lemma map_le [preorder α] [preorder β] {f : α → β} (H : ∀ a b : α, a ≤ b ↔ f a ≤ f b) (x y : with_zero α) : x ≤ y ↔ map f x ≤ map f y := by { with_zero_cases x y, exact H x y } @[move_cast] lemma coe_min (x y : α) [decidable_linear_order α] : ((min x y : α) : with_zero α) = min x y := begin by_cases h: x ≤ y, { simp [min_eq_left, h] }, { simp [min_eq_right, le_of_not_le h] } end section group variables [group α] lemma mul_left_cancel : ∀ {x : with_zero α} (h : x ≠ 0) {y z : with_zero α}, x * y = x * z → y = z | 0 h := false.elim $ h rfl | (a : α) h := λ y z h2, begin have h3 : (a⁻¹ : with_zero α) * (a * y) = a⁻¹ * (a * z) := by rw h2, rwa [←mul_assoc, ←mul_assoc, mul_left_inv _ h, one_mul, one_mul] at h3, end lemma mul_right_cancel : ∀ {x : with_zero α} (h : x ≠ 0) {y z : with_zero α}, y * x = z * x → y = z | 0 h := false.elim $ h rfl | (a : α) h := λ y z h2, begin have h3 : (y * a) * a⁻¹ = (z * a) * a⁻¹ := by rw h2, rwa [mul_assoc, mul_assoc, mul_right_inv _ h, mul_one, mul_one] at h3, end lemma mul_inv_eq_of_eq_mul : ∀ {x : with_zero α} (h : x ≠ 0) {y z : with_zero α}, y = z * x → y * x⁻¹ = z | 0 h := false.elim $ h rfl | (x : α) h := λ _ _ _, mul_right_cancel h (by rwa [mul_assoc, mul_left_inv _ h, mul_one]) lemma eq_mul_inv_of_mul_eq {x : with_zero α} (h : x ≠ 0) {y z : with_zero α} (h2 : z * x = y) : z = y * x⁻¹ := eq.symm $ mul_inv_eq_of_eq_mul h h2.symm end group end with_zero