From 79659457fb0377da019c274eb58e5de2ff95d9d0 Mon Sep 17 00:00:00 2001
From: Julia Markus Himmel <2065352+TwoFX@users.noreply.github.com>
Date: Thu, 7 May 2026 16:27:06 +0200
Subject: [PATCH] chore: reduce usage of `String.length` (#13681)

This PR reduces the usage of `String.length` in our codebase.

This is just the first step of many towards eliminating `String.length`.
---
 src/Init/Data/Nat/ToString.lean               |  1 +
 src/Init/Data/String.lean                     |  1 +
 src/Init/Data/String/Basic.lean               | 55 +--------------
 src/Init/Data/String/Extra.lean               |  1 +
 src/Init/Data/String/Lemmas.lean              |  1 +
 src/Init/Data/String/Lemmas/Intercalate.lean  |  5 --
 src/Init/Data/String/Lemmas/IsEmpty.lean      |  4 --
 src/Init/Data/String/Lemmas/Length.lean       | 31 +++++++++
 src/Init/Data/String/Lemmas/Modify.lean       |  4 --
 .../Data/String/Lemmas/Pattern/Basic.lean     |  8 +--
 src/Init/Data/String/Lemmas/Pattern/Char.lean | 12 ++--
 src/Init/Data/String/Lemmas/Pattern/Pred.lean |  4 +-
 .../String/Lemmas/Pattern/String/Basic.lean   |  8 +--
 .../String/Lemmas/Pattern/TakeDrop/Char.lean  |  4 +-
 src/Init/Data/String/Lemmas/Splits.lean       | 12 ++--
 src/Init/Data/String/Lemmas/TakeDrop.lean     |  8 +--
 src/Init/Data/String/Length.lean              | 68 +++++++++++++++++++
 src/Init/Data/String/TakeDrop.lean            |  8 +--
 src/Init/System/FilePath.lean                 |  1 +
 src/Init/System/IO.lean                       |  4 +-
 src/Init/System/Uri.lean                      |  1 +
 src/Lean/Data/EditDistance.lean               |  1 +
 src/Lean/Data/Name.lean                       |  1 +
 src/Lean/DocString/Links.lean                 | 47 ++++++-------
 src/Lean/DocString/Markdown.lean              |  1 +
 src/Lean/Message.lean                         |  5 +-
 src/Lean/Meta/MethodSpecs.lean                |  2 +-
 src/Lean/Util/Path.lean                       |  1 +
 src/Lean/Widget/TaggedText.lean               |  1 +
 src/Std/Http/Data/URI/Basic.lean              |  1 +
 src/Std/Internal/Parsec/String.lean           |  1 +
 src/Std/Time/Duration.lean                    |  1 +
 src/Std/Time/Format/Basic.lean                |  8 +--
 src/lake/Lake/Util/Cli.lean                   |  1 +
 src/lake/Lake/Util/String.lean                |  1 +
 src/lake/Lake/Util/Version.lean               |  1 +
 36 files changed, 180 insertions(+), 134 deletions(-)
 create mode 100644 src/Init/Data/String/Lemmas/Length.lean
 create mode 100644 src/Init/Data/String/Length.lean

diff --git a/src/Init/Data/Nat/ToString.lean b/src/Init/Data/Nat/ToString.lean
index 00dc083226..c8264d0a2a 100644
--- a/src/Init/Data/Nat/ToString.lean
+++ b/src/Init/Data/Nat/ToString.lean
@@ -10,6 +10,7 @@ public import Init.Data.Repr
 public import Init.Data.Char.Basic
 public import Init.Data.ToString.Basic
 public import Init.Data.String.Basic
+public import Init.Data.String.Length
 import Init.NotationExtra
 import all Init.Data.Repr
 import Init.Omega
diff --git a/src/Init/Data/String.lean b/src/Init/Data/String.lean
index d859797e13..70263d9230 100644
--- a/src/Init/Data/String.lean
+++ b/src/Init/Data/String.lean
@@ -31,3 +31,4 @@ public import Init.Data.String.Subslice
 public import Init.Data.String.Iter
 public import Init.Data.String.Iterate
 public import Init.Data.String.Hashable
+public import Init.Data.String.Length
diff --git a/src/Init/Data/String/Basic.lean b/src/Init/Data/String/Basic.lean
index d57fc02395..7ff696d6c2 100644
--- a/src/Init/Data/String/Basic.lean
+++ b/src/Init/Data/String/Basic.lean
@@ -264,28 +264,6 @@ theorem String.toList_empty : "".toList = [] := by
 theorem String.data_empty : "".toList = [] :=
   toList_empty
 
-/--
-Returns the length of a string in Unicode code points.
-
-Examples:
-* `"".length = 0`
-* `"abc".length = 3`
-* `"L∃∀N".length = 4`
--/
-@[extern "lean_string_length", expose, tagged_return]
-def String.length (b : @& String) : Nat :=
-  b.toList.length
-
-@[simp]
-theorem String.Internal.size_toArray {b : String} : (String.Internal.toArray b).size = b.length :=
-  (rfl)
-
-@[simp]
-theorem String.length_toList {s : String} : s.toList.length = s.length := (rfl)
-
-@[deprecated String.length_toList (since := "2025-10-30")]
-theorem String.length_data {b : String} : b.toList.length = b.length := (rfl)
-
 private theorem ByteArray.utf8Decode?go_eq_utf8Decode?go_extract {b : ByteArray} {hi : i ≤ b.size} {acc : Array Char} :
     utf8Decode?.go b i acc hi = (utf8Decode?.go (b.extract i b.size) 0 #[] (by simp)).map (acc ++ ·) := by
   fun_cases utf8Decode?.go b i acc hi with
@@ -436,14 +414,6 @@ theorem String.ofList_eq_empty_iff {l : List Char} : String.ofList l = "" ↔ l
 theorem List.asString_eq_empty_iff {l : List Char} : String.ofList l = "" ↔ l = [] :=
   String.ofList_eq_empty_iff
 
-@[simp]
-theorem String.length_ofList {l : List Char} : (String.ofList l).length = l.length := by
-  rw [← String.length_toList, String.toList_ofList]
-
-@[deprecated String.length_ofList (since := "2025-10-30")]
-theorem List.length_asString {l : List Char} : (String.ofList l).length = l.length :=
-  String.length_ofList
-
 end
 
 namespace String
@@ -531,7 +501,7 @@ theorem Pos.Raw.isValid_ofList {l : List Char} {p : Pos.Raw} :
   rw [isValid_iff_exists_append]
   refine ⟨?_, ?_⟩
   · rintro ⟨t₁, t₂, ht, rfl⟩
-    refine ⟨t₁.length, ?_⟩
+    refine ⟨t₁.toList.length, ?_⟩
     have := congrArg String.toList ht
     simp only [String.toList_ofList, String.toList_append] at this
     simp [this]
@@ -3127,8 +3097,6 @@ namespace String
 theorem ext {s₁ s₂ : String} (h : s₁.toList = s₂.toList) : s₁ = s₂ :=
   toList_injective h
 
-@[simp] theorem length_empty : "".length = 0 := by simp [← length_toList, toList_empty]
-
 @[deprecated singleton_eq_ofList (since := "2025-10-30")]
 theorem singleton_eq {c : Char} : String.singleton c = ofList [c] :=
   singleton_eq_ofList
@@ -3140,10 +3108,6 @@ theorem singleton_eq {c : Char} : String.singleton c = ofList [c] :=
 theorem data_singleton (c : Char) : (String.singleton c).toList = [c] :=
   toList_singleton c
 
-@[simp]
-theorem length_singleton {c : Char} : (String.singleton c).length = 1 := by
-  simp [← length_toList]
-
 @[simp]
 theorem toList_push (c : Char) : (String.push s c).toList = s.toList ++ [c] := by
   simp [← append_singleton]
@@ -3152,15 +3116,6 @@ theorem toList_push (c : Char) : (String.push s c).toList = s.toList ++ [c] := b
 theorem data_push (c : Char) : (String.push s c).toList = s.toList ++ [c] :=
   toList_push c
 
-@[simp] theorem length_push (c : Char) : (String.push s c).length = s.length + 1 := by
-  simp [← length_toList]
-
-@[simp] theorem length_pushn (c : Char) (n : Nat) : (pushn s c n).length = s.length + n := by
-  rw [pushn_eq_repeat_push]; induction n <;> simp [Nat.repeat, Nat.add_assoc, *]
-
-@[simp] theorem length_append (s t : String) : (s ++ t).length = s.length + t.length := by
-  simp [← length_toList]
-
 theorem lt_iff {s t : String} : s < t ↔ s.toList < t.toList := .rfl
 
 namespace Pos.Raw
@@ -3211,11 +3166,3 @@ theorem bytes_inj {s t : String} : s.toByteArray = t.toByteArray ↔ s = t :=
   toByteArray_inj
 
 end String
-
-namespace Char
-
-@[deprecated String.length_singleton (since := "2026-02-12")]
-theorem length_toString (c : Char) : c.toString.length = 1 := by
-  simp
-
-end Char
diff --git a/src/Init/Data/String/Extra.lean b/src/Init/Data/String/Extra.lean
index efd475a871..2489996230 100644
--- a/src/Init/Data/String/Extra.lean
+++ b/src/Init/Data/String/Extra.lean
@@ -11,6 +11,7 @@ public import Init.Data.String.Basic
 import all Init.Data.String.Basic
 import Init.Data.String.Search
 import Init.Data.String.Termination
+import Init.Data.String.Length
 
 public section
 
diff --git a/src/Init/Data/String/Lemmas.lean b/src/Init/Data/String/Lemmas.lean
index 93bf092505..7f82405d66 100644
--- a/src/Init/Data/String/Lemmas.lean
+++ b/src/Init/Data/String/Lemmas.lean
@@ -21,3 +21,4 @@ public import Init.Data.String.Lemmas.Iter
 public import Init.Data.String.Lemmas.Hashable
 public import Init.Data.String.Lemmas.TakeDrop
 public import Init.Data.String.Lemmas.StringOrder
+public import Init.Data.String.Lemmas.Length
diff --git a/src/Init/Data/String/Lemmas/Intercalate.lean b/src/Init/Data/String/Lemmas/Intercalate.lean
index 44e6ed80fb..a3f7796990 100644
--- a/src/Init/Data/String/Lemmas/Intercalate.lean
+++ b/src/Init/Data/String/Lemmas/Intercalate.lean
@@ -81,11 +81,6 @@ theorem toList_join {l : List String} : (String.join l).toList = l.flatMap Strin
 theorem join_append {l m : List String} : String.join (l ++ m) = String.join l ++ String.join m := by
   simp [← toList_inj]
 
-@[simp]
-theorem length_join {l : List String} : (String.join l).length = (l.map String.length).sum := by
-  simp only [← length_toList, toList_join, List.length_flatMap]
-  simp
-
 namespace Slice
 
 @[simp]
diff --git a/src/Init/Data/String/Lemmas/IsEmpty.lean b/src/Init/Data/String/Lemmas/IsEmpty.lean
index 1c92a169d0..4fc4909fa4 100644
--- a/src/Init/Data/String/Lemmas/IsEmpty.lean
+++ b/src/Init/Data/String/Lemmas/IsEmpty.lean
@@ -162,10 +162,6 @@ theorem Slice.isEmpty_slice_eq_false_iff {s : Slice} {p₁ p₂ h} :
     (s.slice p₁ p₂ h).isEmpty = false ↔ p₁ ≠ p₂ := by
   rw [ne_eq, ← isEmpty_slice (h := h), Bool.not_eq_true]
 
-@[simp]
-theorem length_eq_zero_iff {s : String} : s.length = 0 ↔ s = "" := by
-  simp [← length_toList]
-
 @[simp]
 theorem toByteArray_eq_empty_iff {s : String} :
     s.toByteArray = ByteArray.empty ↔ s = "" := by
diff --git a/src/Init/Data/String/Lemmas/Length.lean b/src/Init/Data/String/Lemmas/Length.lean
new file mode 100644
index 0000000000..81649fbb6c
--- /dev/null
+++ b/src/Init/Data/String/Lemmas/Length.lean
@@ -0,0 +1,31 @@
+/-
+Copyright (c) 2026 Lean FRO, LLC. All rights reserved.
+Released under Apache 2.0 license as described in the file LICENSE.
+Authors: Markus Himmel
+-/
+module
+
+prelude
+public import Init.Data.String.Length
+public import Init.Data.String.Modify
+import Init.Data.String.Lemmas.Modify
+import Init.Data.String.Lemmas.Intercalate
+
+public section
+
+namespace String
+
+@[simp]
+theorem length_eq_zero_iff {s : String} : s.length = 0 ↔ s = "" := by
+  simp [← length_toList]
+
+@[simp]
+theorem length_map {f : Char → Char} {s : String} : (s.map f).length = s.length := by
+  simp [← length_toList]
+
+@[simp]
+theorem length_join {l : List String} : (String.join l).length = (l.map String.length).sum := by
+  simp only [← length_toList, toList_join, List.length_flatMap]
+  simp [length_toList]
+
+end String
diff --git a/src/Init/Data/String/Lemmas/Modify.lean b/src/Init/Data/String/Lemmas/Modify.lean
index 5818a191d4..be3926c275 100644
--- a/src/Init/Data/String/Lemmas/Modify.lean
+++ b/src/Init/Data/String/Lemmas/Modify.lean
@@ -57,10 +57,6 @@ theorem map_eq_internal {f : Char → Char} {s : String} : s.map f = .ofList (s.
   apply String.toList_injective
   simp only [toList_map, toList_ofList]
 
-@[simp]
-theorem length_map {f : Char → Char} {s : String} : (s.map f).length = s.length := by
-  simp [← length_toList]
-
 @[simp]
 theorem map_eq_empty {f : Char → Char} {s : String} : s.map f = "" ↔ s = "" := by
   simp [← toList_eq_nil_iff]
diff --git a/src/Init/Data/String/Lemmas/Pattern/Basic.lean b/src/Init/Data/String/Lemmas/Pattern/Basic.lean
index 956269e15f..738b9eb2c9 100644
--- a/src/Init/Data/String/Lemmas/Pattern/Basic.lean
+++ b/src/Init/Data/String/Lemmas/Pattern/Basic.lean
@@ -301,8 +301,8 @@ This implies that the notion of match and longest match coincide.
 class NoPrefixPatternModel {ρ : Type} (pat : ρ) [PatternModel pat] : Prop where
   eq_empty (s t) : PatternModel.Matches pat s → PatternModel.Matches pat (s ++ t) → t = ""
 
-theorem NoPrefixPatternModel.of_length_eq {ρ : Type} {pat : ρ} [PatternModel pat]
-    (h : ∀ s t, PatternModel.Matches pat s → PatternModel.Matches pat t → s.length = t.length) :
+theorem NoPrefixPatternModel.of_length_toList_eq {ρ : Type} {pat : ρ} [PatternModel pat]
+    (h : ∀ s t, PatternModel.Matches pat s → PatternModel.Matches pat t → s.toList.length = t.toList.length) :
     NoPrefixPatternModel pat where
   eq_empty s t hs ht := by simpa using h s _ hs ht
 
@@ -322,8 +322,8 @@ This implies that the notion of reverse match and longest reverse match coincide
 class NoSuffixPatternModel {ρ : Type} (pat : ρ) [PatternModel pat] : Prop where
   eq_empty (s t) : PatternModel.Matches pat t → PatternModel.Matches pat (s ++ t) → s = ""
 
-theorem NoSuffixPatternModel.of_length_eq {ρ : Type} {pat : ρ} [PatternModel pat]
-    (h : ∀ s t, PatternModel.Matches pat s → PatternModel.Matches pat t → s.length = t.length) :
+theorem NoSuffixPatternModel.of_length_toList_eq {ρ : Type} {pat : ρ} [PatternModel pat]
+    (h : ∀ s t, PatternModel.Matches pat s → PatternModel.Matches pat t → s.toList.length = t.toList.length) :
     NoSuffixPatternModel pat where
   eq_empty s t hs ht := by simpa using h t _ hs ht
 
diff --git a/src/Init/Data/String/Lemmas/Pattern/Char.lean b/src/Init/Data/String/Lemmas/Pattern/Char.lean
index 5edab67ce5..c556173ec0 100644
--- a/src/Init/Data/String/Lemmas/Pattern/Char.lean
+++ b/src/Init/Data/String/Lemmas/Pattern/Char.lean
@@ -33,10 +33,10 @@ instance {c : Char} : StrictPatternModel c where
   not_matches_empty := by simp [PatternModel.Matches]
 
 instance {c : Char} : NoPrefixPatternModel c :=
-  .of_length_eq (by simp +contextual [PatternModel.Matches])
+  .of_length_toList_eq (by simp +contextual [PatternModel.Matches])
 
 instance {c : Char} : NoSuffixPatternModel c :=
-  .of_length_eq (by simp +contextual [PatternModel.Matches])
+  .of_length_toList_eq (by simp +contextual [PatternModel.Matches])
 
 theorem isMatch_iff {c : Char} {s : Slice} {pos : s.Pos} :
     IsMatch c pos ↔
@@ -186,12 +186,12 @@ theorem isLongestMatchAtChain_iff {c : Char} {s : Slice} {pos pos' : s.Pos} :
 
 theorem isLongestMatchAtChain_iff_toList {c : Char} {s : Slice} {pos pos' : s.Pos} :
     IsLongestMatchAtChain c pos pos' ↔
-      ∃ (h : pos ≤ pos'), (s.slice pos pos' h).copy.toList = List.replicate (s.slice pos pos' h).copy.length c := by
+      ∃ (h : pos ≤ pos'), (s.slice pos pos' h).copy.toList = List.replicate (s.slice pos pos' h).copy.toList.length c := by
   simp [isLongestMatchAtChain_iff_isLongestMatchAtChain_beq, CharPred.isLongestMatchAtChain_iff_toList,
     List.eq_replicate_iff]
 
 theorem isLongestMatchAtChain_startPos_endPos_iff_toList {c : Char} {s : Slice} :
-    IsLongestMatchAtChain c s.startPos s.endPos ↔ s.copy.toList = List.replicate s.copy.length c := by
+    IsLongestMatchAtChain c s.startPos s.endPos ↔ s.copy.toList = List.replicate s.copy.toList.length c := by
   simp [isLongestMatchAtChain_iff_isLongestMatchAtChain_beq,
     CharPred.isLongestMatchAtChain_startPos_endPos_iff_toList, List.eq_replicate_iff]
 
@@ -216,12 +216,12 @@ theorem isLongestRevMatchAtChain_iff {c : Char} {s : Slice} {pos pos' : s.Pos} :
 
 theorem isLongestRevMatchAtChain_iff_toList {c : Char} {s : Slice} {pos pos' : s.Pos} :
     IsLongestRevMatchAtChain c pos pos' ↔
-      ∃ (h : pos ≤ pos'), (s.slice pos pos' h).copy.toList = List.replicate (s.slice pos pos' h).copy.length c := by
+      ∃ (h : pos ≤ pos'), (s.slice pos pos' h).copy.toList = List.replicate (s.slice pos pos' h).copy.toList.length c := by
   simp [isLongestRevMatchAtChain_iff_isLongestRevMatchAtChain_beq, CharPred.isLongestRevMatchAtChain_iff_toList,
     List.eq_replicate_iff]
 
 theorem isLongestRevMatchAtChain_startPos_endPos_iff_toList {c : Char} {s : Slice} :
-    IsLongestRevMatchAtChain c s.startPos s.endPos ↔ s.copy.toList = List.replicate s.copy.length c := by
+    IsLongestRevMatchAtChain c s.startPos s.endPos ↔ s.copy.toList = List.replicate s.copy.toList.length c := by
   simp [isLongestRevMatchAtChain_iff_isLongestRevMatchAtChain_beq,
     CharPred.isLongestRevMatchAtChain_startPos_endPos_iff_toList, List.eq_replicate_iff]
 
diff --git a/src/Init/Data/String/Lemmas/Pattern/Pred.lean b/src/Init/Data/String/Lemmas/Pattern/Pred.lean
index d36ae40079..949c364d30 100644
--- a/src/Init/Data/String/Lemmas/Pattern/Pred.lean
+++ b/src/Init/Data/String/Lemmas/Pattern/Pred.lean
@@ -33,10 +33,10 @@ instance {p : Char → Bool} : StrictPatternModel p where
   not_matches_empty := by simp [PatternModel.Matches]
 
 instance {p : Char → Bool} : NoPrefixPatternModel p :=
-  .of_length_eq (by simp +contextual [PatternModel.Matches])
+  .of_length_toList_eq (by simp +contextual [PatternModel.Matches])
 
 instance {p : Char → Bool} : NoSuffixPatternModel p :=
-  .of_length_eq (by simp +contextual [PatternModel.Matches])
+  .of_length_toList_eq (by simp +contextual [PatternModel.Matches])
 
 theorem isMatch_iff {p : Char → Bool} {s : Slice} {pos : s.Pos} :
     IsMatch p pos ↔
diff --git a/src/Init/Data/String/Lemmas/Pattern/String/Basic.lean b/src/Init/Data/String/Lemmas/Pattern/String/Basic.lean
index 7cad6d2d3e..1861c24406 100644
--- a/src/Init/Data/String/Lemmas/Pattern/String/Basic.lean
+++ b/src/Init/Data/String/Lemmas/Pattern/String/Basic.lean
@@ -29,10 +29,10 @@ theorem strictPatternModel {pat : Slice} (hpat : pat.isEmpty = false) : StrictPa
   not_matches_empty := by simpa [PatternModel.Matches]
 
 instance {pat : Slice} : NoPrefixPatternModel pat :=
-  .of_length_eq (by simp +contextual [PatternModel.Matches])
+  .of_length_toList_eq (by simp +contextual [PatternModel.Matches])
 
 instance {pat : Slice} : NoSuffixPatternModel pat :=
-  .of_length_eq (by simp +contextual [PatternModel.Matches])
+  .of_length_toList_eq (by simp +contextual [PatternModel.Matches])
 
 theorem isMatch_iff {pat s : Slice} {pos : s.Pos} :
     IsMatch pat pos ↔ (s.sliceTo pos).copy = pat.copy := by
@@ -287,10 +287,10 @@ theorem strictPatternModel {pat : String} (h : pat ≠ "") : StrictPatternModel
   not_matches_empty := by simpa [PatternModel.Matches]
 
 instance {pat : String} : NoPrefixPatternModel pat :=
-  .of_length_eq (by simp +contextual [PatternModel.Matches])
+  .of_length_toList_eq (by simp +contextual [PatternModel.Matches])
 
 instance {pat : String} : NoSuffixPatternModel pat :=
-  .of_length_eq (by simp +contextual [PatternModel.Matches])
+  .of_length_toList_eq (by simp +contextual [PatternModel.Matches])
 
 theorem isMatch_iff_slice {pat : String} {s : Slice} {pos : s.Pos} :
     IsMatch (ρ := String) pat pos ↔ IsMatch (ρ := Slice) pat.toSlice pos := by
diff --git a/src/Init/Data/String/Lemmas/Pattern/TakeDrop/Char.lean b/src/Init/Data/String/Lemmas/Pattern/TakeDrop/Char.lean
index 9b8163fd8c..dda11b480c 100644
--- a/src/Init/Data/String/Lemmas/Pattern/TakeDrop/Char.lean
+++ b/src/Init/Data/String/Lemmas/Pattern/TakeDrop/Char.lean
@@ -87,7 +87,7 @@ theorem get_eq_of_lt_skipPrefixWhile_char {c : Char} {s : Slice} {pos : s.Pos} (
   Pos.get_eq_of_lt_skipWhile_char (Pos.startPos_le _) (by rwa [skipPrefixWhile_eq_skipWhile_startPos] at h)
 
 @[simp]
-theorem all_char_iff {c : Char} {s : Slice} : s.all c ↔ s.copy.toList = List.replicate s.copy.length c := by
+theorem all_char_iff {c : Char} {s : Slice} : s.all c ↔ s.copy.toList = List.replicate s.copy.toList.length c := by
   rw [Bool.eq_iff_iff]
   simp [Pattern.Model.all_eq_true_iff, Char.isLongestMatchAtChain_startPos_endPos_iff_toList]
 
@@ -123,7 +123,7 @@ theorem get_eq_of_skipSuffixWhile_le_char {c : Char} {s : Slice} {pos : s.Pos}
   Pos.get_eq_of_revSkipWhile_le_char h' (by rwa [skipSuffixWhile_eq_revSkipWhile_endPos] at h)
 
 @[simp]
-theorem revAll_char_iff {c : Char} {s : Slice} : s.revAll c ↔ s.copy.toList = List.replicate s.copy.length c := by
+theorem revAll_char_iff {c : Char} {s : Slice} : s.revAll c ↔ s.copy.toList = List.replicate s.copy.toList.length c := by
   rw [Bool.eq_iff_iff]
   simp [Pattern.Model.revAll_eq_true_iff, Char.isLongestRevMatchAtChain_startPos_endPos_iff_toList]
 
diff --git a/src/Init/Data/String/Lemmas/Splits.lean b/src/Init/Data/String/Lemmas/Splits.lean
index eda457a235..c9ef153a9a 100644
--- a/src/Init/Data/String/Lemmas/Splits.lean
+++ b/src/Init/Data/String/Lemmas/Splits.lean
@@ -760,9 +760,9 @@ theorem splits_nextn_startPos (s : String) (n : Nat) :
   simpa using s.splits_startPos.nextn n
 
 theorem Slice.Pos.Splits.prevn {s : Slice} {t₁ t₂ : String} {p : s.Pos} (h : p.Splits t₁ t₂) (n : Nat) :
-    (p.prevn n).Splits (String.ofList (t₁.toList.take (t₁.length - n))) (String.ofList (t₁.toList.drop (t₁.length - n)) ++ t₂) := by
+    (p.prevn n).Splits (String.ofList (t₁.toList.take (t₁.toList.length - n))) (String.ofList (t₁.toList.drop (t₁.toList.length - n)) ++ t₂) := by
   induction n generalizing p t₁ t₂ with
-  | zero => simpa [← String.length_toList]
+  | zero => simpa
   | succ n ih =>
     rw [Pos.prevn_add_one]
     split
@@ -771,16 +771,16 @@ theorem Slice.Pos.Splits.prevn {s : Slice} {t₁ t₂ : String} {p : s.Pos} (h :
       simpa [Nat.add_sub_add_right, List.take_append, List.drop_append, ← append_assoc] using ih h.prev
 
 theorem Slice.splits_prevn_endPos (s : Slice) (n : Nat) :
-    (s.endPos.prevn n).Splits (String.ofList (s.copy.toList.take (s.copy.length - n)))
-      (String.ofList (s.copy.toList.drop (s.copy.length - n))) := by
+    (s.endPos.prevn n).Splits (String.ofList (s.copy.toList.take (s.copy.toList.length - n)))
+      (String.ofList (s.copy.toList.drop (s.copy.toList.length - n))) := by
   simpa using s.splits_endPos.prevn n
 
 theorem Pos.Splits.prevn {s t₁ t₂ : String} {p : s.Pos} (h : p.Splits t₁ t₂) (n : Nat) :
-    (p.prevn n).Splits (String.ofList (t₁.toList.take (t₁.length - n))) (String.ofList (t₁.toList.drop (t₁.length - n)) ++ t₂) := by
+    (p.prevn n).Splits (String.ofList (t₁.toList.take (t₁.toList.length - n))) (String.ofList (t₁.toList.drop (t₁.toList.length - n)) ++ t₂) := by
   simpa [← splits_toSlice_iff, toSlice_prevn] using h.toSlice.prevn n
 
 theorem splits_prevn_endPos (s : String) (n : Nat) :
-    (s.endPos.prevn n).Splits (String.ofList (s.toList.take (s.length - n))) (String.ofList (s.toList.drop (s.length - n))) := by
+    (s.endPos.prevn n).Splits (String.ofList (s.toList.take (s.toList.length - n))) (String.ofList (s.toList.drop (s.toList.length - n))) := by
   simpa using s.splits_endPos.prevn n
 
 @[simp]
diff --git a/src/Init/Data/String/Lemmas/TakeDrop.lean b/src/Init/Data/String/Lemmas/TakeDrop.lean
index 83735c6d6c..3d6c988300 100644
--- a/src/Init/Data/String/Lemmas/TakeDrop.lean
+++ b/src/Init/Data/String/Lemmas/TakeDrop.lean
@@ -29,7 +29,7 @@ theorem dropEnd_eq_sliceTo {s : Slice} {n : Nat} : s.dropEnd n = s.sliceTo (s.en
 
 @[simp]
 theorem toList_copy_dropEnd {s : Slice} {n : Nat} :
-    (s.dropEnd n).copy.toList = s.copy.toList.take (s.copy.length - n) := by
+    (s.dropEnd n).copy.toList = s.copy.toList.take (s.copy.toList.length - n) := by
   simp [dropEnd_eq_sliceTo, (s.splits_prevn_endPos n).copy_sliceTo_eq]
 
 theorem take_eq_sliceTo {s : Slice} {n : Nat} : s.take n = s.sliceTo (s.startPos.nextn n) :=
@@ -44,7 +44,7 @@ theorem takeEnd_eq_sliceFrom {s : Slice} {n : Nat} : s.takeEnd n = s.sliceFrom (
 
 @[simp]
 theorem toList_copy_takeEnd {s : Slice} {n : Nat} :
-    (s.takeEnd n).copy.toList = s.copy.toList.drop (s.copy.length - n) := by
+    (s.takeEnd n).copy.toList = s.copy.toList.drop (s.copy.toList.length - n) := by
   simp [takeEnd_eq_sliceFrom, (s.splits_prevn_endPos n).copy_sliceFrom_eq]
 
 end Slice
@@ -63,7 +63,7 @@ theorem dropEnd_toSlice {s : String} {n : Nat} : s.toSlice.dropEnd n = s.dropEnd
 
 @[simp]
 theorem toList_copy_dropEnd {s : String} {n : Nat} :
-    (s.dropEnd n).copy.toList = s.toList.take (s.length - n) := by
+    (s.dropEnd n).copy.toList = s.toList.take (s.toList.length - n) := by
   simp [← dropEnd_toSlice]
 
 @[simp]
@@ -80,7 +80,7 @@ theorem takeEnd_toSlice {s : String} {n : Nat} : s.toSlice.takeEnd n = s.takeEnd
 
 @[simp]
 theorem toList_copy_takeEnd {s : String} {n : Nat} :
-    (s.takeEnd n).copy.toList = s.toList.drop (s.length - n) := by
+    (s.takeEnd n).copy.toList = s.toList.drop (s.toList.length - n) := by
   simp [← takeEnd_toSlice]
 
 end String
diff --git a/src/Init/Data/String/Length.lean b/src/Init/Data/String/Length.lean
new file mode 100644
index 0000000000..d5f601f24c
--- /dev/null
+++ b/src/Init/Data/String/Length.lean
@@ -0,0 +1,68 @@
+/-
+Copyright (c) 2026 Lean FRO, LLC. All rights reserved.
+Released under Apache 2.0 license as described in the file LICENSE.
+Authors: Julia Markus Himmel
+-/
+module
+
+prelude
+public import Init.Data.String.Basic
+import Init.Data.Char.Lemmas
+
+public section
+
+namespace String
+
+/--
+Returns the length of a string in Unicode code points.
+
+Examples:
+* `"".length = 0`
+* `"abc".length = 3`
+* `"L∃∀N".length = 4`
+-/
+@[extern "lean_string_length", expose, tagged_return]
+def length (b : @& String) : Nat :=
+  b.toList.length
+
+@[simp]
+theorem Internal.size_toArray {b : String} : (String.Internal.toArray b).size = b.length :=
+  (rfl)
+
+theorem length_toList {s : String} : s.toList.length = s.length := (rfl)
+
+@[deprecated String.length_toList (since := "2025-10-30")]
+theorem length_data {b : String} : b.toList.length = b.length := (rfl)
+
+@[simp]
+theorem length_ofList {l : List Char} : (String.ofList l).length = l.length := by
+  rw [← String.length_toList, String.toList_ofList]
+
+@[deprecated String.length_ofList (since := "2025-10-30")]
+theorem _root_.List.length_asString {l : List Char} : (String.ofList l).length = l.length :=
+  String.length_ofList
+
+@[simp] theorem length_empty : "".length = 0 := by simp [← length_toList, toList_empty]
+
+@[simp]
+theorem length_singleton {c : Char} : (String.singleton c).length = 1 := by
+  simp [← length_toList]
+
+@[simp] theorem length_push (c : Char) : (String.push s c).length = s.length + 1 := by
+  simp [← length_toList]
+
+@[simp] theorem length_pushn (c : Char) (n : Nat) : (pushn s c n).length = s.length + n := by
+  rw [pushn_eq_repeat_push]; induction n <;> simp [Nat.repeat, Nat.add_assoc, *]
+
+@[simp] theorem length_append (s t : String) : (s ++ t).length = s.length + t.length := by
+  simp [← length_toList]
+
+end String
+
+namespace Char
+
+@[deprecated String.length_singleton (since := "2026-02-12")]
+theorem length_toString (c : Char) : c.toString.length = 1 := by
+  simp
+
+end Char
diff --git a/src/Init/Data/String/TakeDrop.lean b/src/Init/Data/String/TakeDrop.lean
index f5f5ca62cc..b55ab71e52 100644
--- a/src/Init/Data/String/TakeDrop.lean
+++ b/src/Init/Data/String/TakeDrop.lean
@@ -28,7 +28,7 @@ open Slice.Pattern
 Returns a {name}`String.Slice` obtained by removing the specified number of characters (Unicode code
 points) from the start of the string.
 
-If {name}`n` is greater than {lean}`s.length`, returns an empty slice.
+If {name}`n` is greater than {lean}`s.toList.length`, returns an empty slice.
 
 This is a cheap operation because it does not allocate a new string to hold the result.
 To convert the result into a string, use {name}`String.Slice.copy`.
@@ -50,7 +50,7 @@ def Internal.dropImpl (s : String) (n : Nat) : String :=
 Returns a {name}`String.Slice` obtained by removing the specified number of characters (Unicode code
 points) from the end of the string.
 
-If {name}`n` is greater than {lean}`s.length`, returns an empty slice.
+If {name}`n` is greater than {lean}`s.toList.length`, returns an empty slice.
 
 This is a cheap operation because it does not allocate a new string to hold the result.
 To convert the result into a string, use {name}`String.Slice.copy`.
@@ -80,7 +80,7 @@ def Internal.dropRightImpl (s : String) (n : Nat) : String :=
 Returns a {name}`String.Slice` that contains the first {name}`n` characters (Unicode code points) of
 {name}`s`.
 
-If {name}`n` is greater than {lean}`s.length`, returns {lean}`s.toSlice`.
+If {name}`n` is greater than {lean}`s.toList.length`, returns {lean}`s.toSlice`.
 
 This is a cheap operation because it does not allocate a new string to hold the result.
 To convert the result into a string, use {name}`String.Slice.copy`.
@@ -99,7 +99,7 @@ Examples:
 Returns a {name}`String.Slice` that contains the last {name}`n` characters (Unicode code points) of
 {name}`s`.
 
-If {name}`n` is greater than {lean}`s.length`, returns {lean}`s.toSlice`.
+If {name}`n` is greater than {lean}`s.toList.length`, returns {lean}`s.toSlice`.
 
 This is a cheap operation because it does not allocate a new string to hold the result.
 To convert the result into a string, use {name}`String.Slice.copy`.
diff --git a/src/Init/System/FilePath.lean b/src/Init/System/FilePath.lean
index b826958e25..bd42b49b7b 100644
--- a/src/Init/System/FilePath.lean
+++ b/src/Init/System/FilePath.lean
@@ -11,6 +11,7 @@ import Init.Data.String.Search
 public import Init.Data.ToString.Basic
 import Init.Data.Iterators.Consumers.Collect
 import Init.System.Platform
+import Init.Data.String.Length
 
 public section
 
diff --git a/src/Init/System/IO.lean b/src/Init/System/IO.lean
index 4f529d57cd..64faef771c 100644
--- a/src/Init/System/IO.lean
+++ b/src/Init/System/IO.lean
@@ -1026,7 +1026,7 @@ and/or return data.
 partial def Handle.lines (h : Handle) : IO (Array String) := do
   let rec read (lines : Array String) := do
     let line ← h.getLine
-    if line.length == 0 then
+    if line.isEmpty then
       pure lines
     else if line.back == '\n' then
       let line := line.dropEnd 1 |>.copy
@@ -1780,7 +1780,7 @@ and/or return data.
 partial def lines (s : Stream) : IO (Array String) := do
   let rec read (lines : Array String) := do
     let line ← s.getLine
-    if line.length == 0 then
+    if line.isEmpty then
       pure lines
     else if line.back == '\n' then
       let line := line.dropEnd 1 |>.copy
diff --git a/src/Init/System/Uri.lean b/src/Init/System/Uri.lean
index 95420a8c90..033252f293 100644
--- a/src/Init/System/Uri.lean
+++ b/src/Init/System/Uri.lean
@@ -13,6 +13,7 @@ import Init.Data.String.Search
 import Init.Omega
 import Init.System.Platform
 import Init.While
+import Init.Data.String.Length
 
 public section
 
diff --git a/src/Lean/Data/EditDistance.lean b/src/Lean/Data/EditDistance.lean
index dff74584fd..6905dc5d3d 100644
--- a/src/Lean/Data/EditDistance.lean
+++ b/src/Lean/Data/EditDistance.lean
@@ -12,6 +12,7 @@ import Init.Data.Nat.Order
 import Init.Data.Order.Lemmas
 import Init.Data.Range
 import Init.While
+import Init.Data.String.Length
 
 set_option linter.missingDocs true
 
diff --git a/src/Lean/Data/Name.lean b/src/Lean/Data/Name.lean
index 00b39c322f..1392ad175a 100644
--- a/src/Lean/Data/Name.lean
+++ b/src/Lean/Data/Name.lean
@@ -11,6 +11,7 @@ import Init.Data.String.TakeDrop
 import Init.Data.Ord.String
 import Init.Data.Ord.UInt
 import Init.Data.String.Search
+import Init.Data.String.Length
 
 public section
 namespace Lean
diff --git a/src/Lean/DocString/Links.lean b/src/Lean/DocString/Links.lean
index d85813e5f9..0b57475bcb 100644
--- a/src/Lean/DocString/Links.lean
+++ b/src/Lean/DocString/Links.lean
@@ -12,6 +12,7 @@ import Init.Data.String.TakeDrop
 import Init.Data.String.Search
 import Init.Data.ToString.Macro
 import Init.While
+import Init.Data.String.Length
 
 public section
 
@@ -114,28 +115,28 @@ def rewriteManualLinksCore (s : String) : Id (Array (Lean.Syntax.Range × String
     let pre := iter
     iter := iter.next h
 
-    if !lookingAt scheme pre then
+    match pre.skip? scheme with
+    | none =>
       out := out.push c
       continue
-
-    let start := pre.nextn scheme.length
-    let mut iter' := start
-    while h' : ¬iter'.IsAtEnd do
-      let c' := iter'.get h'
-      let pre' := iter'
-      iter' := iter'.next h'
-      if urlChar c' && ¬iter'.IsAtEnd then
-        continue
-      match rw (s.extract start pre') with
-      | .error err =>
-        errors := errors.push (⟨pre.offset, pre'.offset⟩, err)
-        out := out.push c
-        break
-      | .ok path =>
-        out := out ++ manualRoot ++ path
-        out := out.push c'
-        iter := iter'
-        break
+    | some start =>
+      let mut iter' := start
+      while h' : ¬iter'.IsAtEnd do
+        let c' := iter'.get h'
+        let pre' := iter'
+        iter' := iter'.next h'
+        if urlChar c' && ¬iter'.IsAtEnd then
+          continue
+        match rw (s.extract start pre') with
+        | .error err =>
+          errors := errors.push (⟨pre.offset, pre'.offset⟩, err)
+          out := out.push c
+          break
+        | .ok path =>
+          out := out ++ manualRoot ++ path
+          out := out.push c'
+          iter := iter'
+          break
 
   pure (errors, out)
 
@@ -154,12 +155,6 @@ where
     c == '!' || c == '$' || c == '&' || c == '\'' || /- c == '(' || c == ')' || -/ c == '*' ||
     c == '+' || c == ',' || c == ';' || c == '='
 
-  /--
-  Returns `true` if `goal` is a prefix of the string at the position pointed to by `iter`.
-  -/
-  lookingAt (goal : String) {s : String} (iter : s.Pos) : Bool :=
-    String.Pos.Raw.substrEq s iter.offset goal 0 goal.rawEndPos.byteIdx
-
 /--
 Rewrites Lean reference manual links in `docstring` to point at the reference manual.
 
diff --git a/src/Lean/DocString/Markdown.lean b/src/Lean/DocString/Markdown.lean
index 13c3aaa865..6abbe77d74 100644
--- a/src/Lean/DocString/Markdown.lean
+++ b/src/Lean/DocString/Markdown.lean
@@ -11,6 +11,7 @@ prelude
 public import Lean.DocString.Types
 public import Init.Data.String.TakeDrop
 public import Init.Data.String.Search
+public import Init.Data.String.Length
 import Init.Data.ToString.Macro
 import Init.While
 
diff --git a/src/Lean/Message.lean b/src/Lean/Message.lean
index c82e6a18bb..d5552f455c 100644
--- a/src/Lean/Message.lean
+++ b/src/Lean/Message.lean
@@ -14,6 +14,7 @@ public import Lean.Util.Sorry
 import Init.Data.String.Search
 import Init.Data.Format.Macro
 import Init.Data.Iterators.Consumers.Collect
+import Init.Data.String.Length
 
 public section
 
@@ -688,7 +689,7 @@ def inlineExpr (e : Expr) (maxInlineLength := 30) : MessageData :=
     (fun ctx => do
       let msg := MessageData.ofExpr e
       let render ← msg.formatExpensively ctx
-      if render.length > maxInlineLength || render.any (· == '\n') then
+      if render.positions.length > maxInlineLength || render.any (· == '\n') then
         return indentD msg ++ "\n"
       else
         return " `" ++ msg ++ "` ")
@@ -704,7 +705,7 @@ def inlineExprTrailing (e : Expr) (maxInlineLength := 30) : MessageData :=
     (fun ctx => do
       let msg := MessageData.ofExpr e
       let render ← msg.formatExpensively ctx
-      if render.length > maxInlineLength || render.any (· == '\n') then
+      if render.positions.length > maxInlineLength || render.any (· == '\n') then
         return indentD msg
       else
         return " `" ++ msg ++ "`")
diff --git a/src/Lean/Meta/MethodSpecs.lean b/src/Lean/Meta/MethodSpecs.lean
index 458332e874..fcbb9ef031 100644
--- a/src/Lean/Meta/MethodSpecs.lean
+++ b/src/Lean/Meta/MethodSpecs.lean
@@ -127,7 +127,7 @@ def mkSpecTheoremName (env : Environment) (instName : Name) (privateSpecs : Bool
   if privateSpecs then mkPrivateName env thmName else thmName
 
 def startsWithFollowedByNumber (s p : String) : Bool :=
-  s.startsWith p && (s.drop p.length).isNat
+  (s.dropPrefix? p).any (·.isNat)
 
 def isSpecThmLikeSuffix (fieldName : Name) (s : String) : Bool :=
   s == s!"{fieldName}_spec" || startsWithFollowedByNumber s s!"{fieldName}_spec_"
diff --git a/src/Lean/Util/Path.lean b/src/Lean/Util/Path.lean
index b2af1878ec..09c0253b70 100644
--- a/src/Lean/Util/Path.lean
+++ b/src/Lean/Util/Path.lean
@@ -18,6 +18,7 @@ import Init.Data.String.TakeDrop
 import Init.Data.List.Monadic
 import Init.Data.Option.BasicAux
 import Init.Data.ToString.Macro
+import Init.Data.String.Length
 
 public section
 
diff --git a/src/Lean/Widget/TaggedText.lean b/src/Lean/Widget/TaggedText.lean
index 48895a5a4e..8b558bf91a 100644
--- a/src/Lean/Widget/TaggedText.lean
+++ b/src/Lean/Widget/TaggedText.lean
@@ -9,6 +9,7 @@ module
 prelude
 public import Lean.Server.Rpc.Basic
 import Init.Data.Array.GetLit
+import Init.Data.String.Length
 
 public section
 
diff --git a/src/Std/Http/Data/URI/Basic.lean b/src/Std/Http/Data/URI/Basic.lean
index e4bb328344..4adddfc594 100644
--- a/src/Std/Http/Data/URI/Basic.lean
+++ b/src/Std/Http/Data/URI/Basic.lean
@@ -11,6 +11,7 @@ public import Std.Net
 public import Std.Http.Internal
 public import Std.Http.Data.URI.Encoding
 public import Init.Data.String.Search
+public import Init.Data.String.Length
 
 public section
 
diff --git a/src/Std/Internal/Parsec/String.lean b/src/Std/Internal/Parsec/String.lean
index 851761c55b..746da76bc5 100644
--- a/src/Std/Internal/Parsec/String.lean
+++ b/src/Std/Internal/Parsec/String.lean
@@ -9,6 +9,7 @@ prelude
 public import Std.Internal.Parsec.Basic
 public import Init.Data.String.Slice
 public import Init.Data.String.Termination
+import Init.Data.String.Length
 
 public section
 
diff --git a/src/Std/Time/Duration.lean b/src/Std/Time/Duration.lean
index 35ff4ca3d8..bfce6f135d 100644
--- a/src/Std/Time/Duration.lean
+++ b/src/Std/Time/Duration.lean
@@ -8,6 +8,7 @@ module
 prelude
 public import Std.Time.Date
 public import Init.Data.String.Basic
+public import Init.Data.String.Length
 
 public section
 
diff --git a/src/Std/Time/Format/Basic.lean b/src/Std/Time/Format/Basic.lean
index 4f0f6a113d..a0da7d3293 100644
--- a/src/Std/Time/Format/Basic.lean
+++ b/src/Std/Time/Format/Basic.lean
@@ -442,7 +442,7 @@ inductive Modifier
 It takes a constructor function to build the `Modifier` and a classify function that maps the pattern length to a specific type.
 -/
 private def parseMod (constructor : α → Modifier) (classify : Nat → Option α) (p : String) : Parser Modifier :=
-  let len := p.length
+  let len := p.positions.length
   match classify len with
   | some res => pure (constructor res)
   | none => fail s!"invalid quantity of characters for '{p.front}'"
@@ -617,16 +617,16 @@ private def specParse (s : String) : Except String FormatString :=
 -- Pretty printer
 
 private def leftPad (n : Nat) (a : Char) (s : String) : String :=
-  "".pushn a (n -  s.length) ++ s
+  "".pushn a (n -  s.positions.length) ++ s
 
 private def rightPad (n : Nat) (a : Char) (s : String) : String :=
-  s ++ "".pushn a (n - s.length)
+  s ++ "".pushn a (n - s.positions.length)
 
 private def pad (size : Nat)  (n : Int) (cut : Bool := false) : String :=
   let (sign, n) := if n < 0 then ("-", -n) else ("", n)
 
   let numStr := toString n
-  if numStr.length > size then
+  if numStr.positions.length > size then
     sign ++ if cut then numStr.drop (numStr.length - size) |>.copy else numStr
   else
     sign ++ leftPad size '0' numStr
diff --git a/src/lake/Lake/Util/Cli.lean b/src/lake/Lake/Util/Cli.lean
index 22be41acb3..756ce27ce4 100644
--- a/src/lake/Lake/Util/Cli.lean
+++ b/src/lake/Lake/Util/Cli.lean
@@ -8,6 +8,7 @@ module
 prelude
 public import Init.Data.String.TakeDrop
 public import Init.Data.String.Search
+public import Init.Data.String.Length
 
 namespace Lake
 
diff --git a/src/lake/Lake/Util/String.lean b/src/lake/Lake/Util/String.lean
index 5fc925216f..a2758022fd 100644
--- a/src/lake/Lake/Util/String.lean
+++ b/src/lake/Lake/Util/String.lean
@@ -10,6 +10,7 @@ public import Init.Data.ToString.Basic
 import Init.Data.UInt.Lemmas
 import Init.Data.String.Basic
 import Init.Data.Nat.Fold
+import Init.Data.String.Length
 
 namespace Lake
 
diff --git a/src/lake/Lake/Util/Version.lean b/src/lake/Lake/Util/Version.lean
index bc897e7d13..fc4eb9fc09 100644
--- a/src/lake/Lake/Util/Version.lean
+++ b/src/lake/Lake/Util/Version.lean
@@ -13,6 +13,7 @@ import Init.Data.String.TakeDrop
 import Lean.Data.Trie
 import Init.Data.String.Search
 import Init.Omega
+import Init.Data.String.Length
 
 /-! # Version