b181fd83ef
The `conv` tactic tries to close “trivial” goals after itself. As of now, it uses `try rfl`, which means it can close goals that are only trivial after reducing with default transparency. This is suboptimal * this can require a fair amount of unfolding, and possibly slow down the proof a lot. And the user cannot even prevent it. * it does not match what `rw` does, and a user might expect the two to behave the same. So this PR changes it to `with_reducible rfl`, matching `rw`’s behavior. I considered `with_reducible eq_refl` to only solve trivial goals that involve equality, but not other relations (e.g. `Perm xs xs`), but a discussion on mathlib pointed out that it’s expected and desirable to solve more general reflexive goals: https://leanprover.zulipchat.com/#narrow/stream/270676-lean4/topic/Closing.20after.20.60rw.60.2C.20.60conv.60.3A.20.60eq_refl.60.20instead.20of.20.60rfl.60/near/429851605
119 lines
3.6 KiB
Text
119 lines
3.6 KiB
Text
inductive ListSplit : List α → Type _
|
||
| split l₁ l₂ : ListSplit (l₁ ++ l₂)
|
||
|
||
def splitList : (l : List α) → ListSplit l
|
||
| [] => ListSplit.split [] []
|
||
| h :: t => ListSplit.split [h] t
|
||
|
||
@[simp] def ListSplit.left {as : List α} : ListSplit as → List α
|
||
| split a b => a
|
||
|
||
@[simp] def ListSplit.right {as : List α} : ListSplit as → List α
|
||
| split a b => b
|
||
|
||
/-- Helper theorem for justifying termination. -/
|
||
theorem splitList_length (as : List α) (h₁ : as.length > 1) (h₂ : as = bs) : (splitList as).left.length < bs.length ∧ (splitList as).right.length < bs.length := by
|
||
match as with
|
||
| [] => contradiction
|
||
| a :: as => simp_arith [← h₂, splitList]; simp_arith at h₁; assumption
|
||
|
||
def len : List α → Nat
|
||
| [] => 0
|
||
| a :: [] => 1
|
||
| l@h₁:(a :: b :: as) =>
|
||
-- Remark: we didn't use `_` because we currently don't have a way for getting a hypothesis stating that the previous two case were not taken here.
|
||
-- h₁ : l = a :: b :: as
|
||
match h₂ : splitList l with
|
||
| ListSplit.split fst snd =>
|
||
-- Remark: `match` refined `h₁`s type to `h₁ : fst ++ snd = a :: b :: as`
|
||
-- h₂ : HEq (splitList l) (ListSplit.split fst snd)
|
||
have := splitList_length (fst ++ snd) (by simp_arith [h₁]) h₁
|
||
-- The following two proofs ase used to justify the recursive applications `len fst` and `len snd`
|
||
have dec₁ : fst.length < as.length + 2 := by subst l; simp_arith [eq_of_heq h₂] at this |- ; simp [this]
|
||
have dec₂ : snd.length < as.length + 2 := by subst l; simp_arith [eq_of_heq h₂] at this |- ; simp [this]
|
||
len fst + len snd
|
||
termination_by xs => xs.length
|
||
|
||
theorem len_nil : len ([] : List α) = 0 := by
|
||
simp [len]
|
||
|
||
-- The `simp [len]` above generated the following equation theorems for len
|
||
#check @len.eq_1
|
||
#check @len.eq_2
|
||
#check @len.eq_3
|
||
|
||
theorem len_1 (a : α) : len [a] = 1 := by
|
||
simp [len]
|
||
|
||
theorem len_2 (a b : α) (bs : List α) : len (a::b::bs) = 1 + len (b::bs) := by
|
||
conv => lhs; unfold len
|
||
rfl
|
||
|
||
-- The `unfold` tactic above generated the following theorem
|
||
#check @len.eq_def
|
||
|
||
theorem len_cons (a : α) (as : List α) : len (a::as) = 1 + len as := by
|
||
cases as with
|
||
| nil => simp [len_1, len_nil]
|
||
| cons b bs => simp [len_2]
|
||
|
||
theorem listlen : ∀ l : List α, l.length = len l := by
|
||
intro l
|
||
induction l with
|
||
| nil => rfl
|
||
| cons h t ih =>
|
||
simp [List.length, len_cons, ih]
|
||
rw [Nat.add_comm]
|
||
|
||
namespace Ex2
|
||
|
||
/--
|
||
`len` example again but with the proofs at `decreasing_by`
|
||
-/
|
||
def len : List α → Nat
|
||
| [] => 0
|
||
| a :: [] => 1
|
||
| l@h₁:(a :: b :: as) =>
|
||
match h₂ : l, h₃ : splitList l with
|
||
| _, ListSplit.split fst snd =>
|
||
len fst + len snd
|
||
termination_by xs => xs.length
|
||
decreasing_by
|
||
all_goals
|
||
simp_wf
|
||
have := splitList_length (fst ++ snd) (by simp_arith [h₁]) h₁
|
||
subst h₂
|
||
simp_arith [eq_of_heq h₃] at this |- ; simp [this]
|
||
|
||
theorem len_nil : len ([] : List α) = 0 := by
|
||
simp [len]
|
||
|
||
-- The `simp [len]` above generated the following equation theorems for len
|
||
#check @len.eq_1
|
||
#check @len.eq_2
|
||
#check @len.eq_3
|
||
|
||
theorem len_1 (a : α) : len [a] = 1 := by
|
||
simp [len]
|
||
|
||
theorem len_2 (a b : α) (bs : List α) : len (a::b::bs) = 1 + len (b::bs) := by
|
||
conv => lhs; unfold len
|
||
rfl
|
||
|
||
-- The `unfold` tactic above generated the following theorem
|
||
#check @len.eq_def
|
||
|
||
theorem len_cons (a : α) (as : List α) : len (a::as) = 1 + len as := by
|
||
cases as with
|
||
| nil => simp [len_1, len_nil]
|
||
| cons b bs => simp [len_2]
|
||
|
||
theorem listlen : ∀ l : List α, l.length = len l := by
|
||
intro l
|
||
induction l with
|
||
| nil => rfl
|
||
| cons h t ih =>
|
||
simp [List.length, len_cons, ih]
|
||
rw [Nat.add_comm]
|
||
|
||
end Ex2
|