[6.0] Add a string-specific search algorithm

Sources/_StringProcessing/Algorithms/Algorithms/FirstRange.swift

@@ -21,8 +21,23 @@ extension Collection {
 }
 
 // MARK: Fixed pattern algorithms
+extension Substring {
+  func _firstRangeSubstring(
+    of other: Substring
+  ) -> Range<String.Index>? {
+    var searcher = SubstringSearcher(text: self, pattern: other)
+    return searcher.next()
+  }
+}
 
 extension Collection where Element: Equatable {
+  func _firstRangeGeneric<C: Collection>(
+    of other: C
+  ) -> Range<Index>? where C.Element == Element {
+    let searcher = ZSearcher<SubSequence>(pattern: Array(other), by: ==)
+    return searcher.search(self[...], in: startIndex..<endIndex)
+  }
+
   /// Finds and returns the range of the first occurrence of a given collection
   /// within this collection.
   ///
@@ -33,9 +48,19 @@ extension Collection where Element: Equatable {
   public func firstRange<C: Collection>(
     of other: C
   ) -> Range<Index>? where C.Element == Element {
-    // TODO: Use a more efficient search algorithm
-    let searcher = ZSearcher<SubSequence>(pattern: Array(other), by: ==)
-    return searcher.search(self[...], in: startIndex..<endIndex)
+    switch (self, other) {
+    case (let str as String, let other as String):
+      return str[...]._firstRangeSubstring(of: other[...]) as! Range<Index>?
+    case (let str as Substring, let other as String):
+      return str._firstRangeSubstring(of: other[...]) as! Range<Index>?
+    case (let str as String, let other as Substring):
+      return str[...]._firstRangeSubstring(of: other) as! Range<Index>?
+    case (let str as Substring, let other as Substring):
+      return str._firstRangeSubstring(of: other) as! Range<Index>?
+
+    default:
+      return _firstRangeGeneric(of: other)
+    }
   }
 }
 
@@ -50,8 +75,19 @@ extension BidirectionalCollection where Element: Comparable {
   public func firstRange<C: Collection>(
     of other: C
   ) -> Range<Index>? where C.Element == Element {
-    let searcher = ZSearcher<SubSequence>(pattern: Array(other), by: ==)
-    return searcher.search(self[...], in: startIndex..<endIndex)
+    switch (self, other) {
+    case (let str as String, let other as String):
+      return str[...]._firstRangeSubstring(of: other[...]) as! Range<Index>?
+    case (let str as Substring, let other as String):
+      return str._firstRangeSubstring(of: other[...]) as! Range<Index>?
+    case (let str as String, let other as Substring):
+      return str[...]._firstRangeSubstring(of: other) as! Range<Index>?
+    case (let str as Substring, let other as Substring):
+      return str._firstRangeSubstring(of: other) as! Range<Index>?
+
+    default:
+      return _firstRangeGeneric(of: other)
+    }
   }
 }
 

Sources/_StringProcessing/Algorithms/Algorithms/Ranges.swift

@@ -135,7 +135,7 @@ extension Collection where Element: Equatable {
   ) -> RangesCollection<ZSearcher<Self>> where C.Element == Element {
     _ranges(of: ZSearcher(pattern: Array(other), by: ==))
   }
-
+  
   // FIXME: Return `some Collection<Range<Index>>` for SE-0346
   /// Finds and returns the ranges of the all occurrences of a given sequence
   /// within the collection.
@@ -146,7 +146,19 @@ extension Collection where Element: Equatable {
   public func ranges<C: Collection>(
     of other: C
   ) -> [Range<Index>] where C.Element == Element {
-    Array(_ranges(of: other))
+    switch (self, other) {
+    case (let str as String, let other as String):
+      return Array(SubstringSearcher(text: str[...], pattern: other[...])) as! [Range<Index>]
+    case (let str as Substring, let other as String):
+      return Array(SubstringSearcher(text: str, pattern: other[...])) as! [Range<Index>]
+    case (let str as String, let other as Substring):
+      return Array(SubstringSearcher(text: str[...], pattern: other)) as! [Range<Index>]
+    case (let str as Substring, let other as Substring):
+      return Array(SubstringSearcher(text: str, pattern: other)) as! [Range<Index>]
+
+    default:
+      return Array(_ranges(of: other))
+    }
   }
 }
 

Sources/_StringProcessing/Algorithms/Algorithms/Replace.swift

@@ -11,6 +11,32 @@
 
 // MARK: `CollectionSearcher` algorithms
 
+extension Substring {
+  func _replacingSubstring(
+    _ other: Substring,
+    with replacement: Substring,
+    maxReplacements: Int = .max
+  ) -> String {
+    precondition(maxReplacements >= 0)
+
+    var result = String()
+    var index = startIndex
+
+    var rangeIterator = SubstringSearcher(text: self, pattern: other)
+    var replacementCount = 0
+    while replacementCount < maxReplacements, let range = rangeIterator.next() {
+      result.append(contentsOf: self[index..<range.lowerBound])
+      result.append(contentsOf: replacement)
+      index = range.upperBound
+
+      replacementCount += 1
+    }
+
+    result.append(contentsOf: self[index...])
+    return result
+  }
+}
+
 extension RangeReplaceableCollection {
   func _replacing<Ranges: Collection, Replacement: Collection>(
     _ ranges: Ranges,
@@ -35,19 +61,6 @@ extension RangeReplaceableCollection {
     result.append(contentsOf: self[index...])
     return result
   }
-
-  mutating func _replace<
-    Ranges: Collection, Replacement: Collection
-  >(
-    _ ranges: Ranges,
-    with replacement: Replacement,
-    maxReplacements: Int = .max
-  ) where Ranges.Element == Range<Index>, Replacement.Element == Element {
-    self = _replacing(
-      ranges,
-      with: replacement,
-      maxReplacements: maxReplacements)
-  }
 }
 
 // MARK: Fixed pattern algorithms
@@ -70,10 +83,40 @@ extension RangeReplaceableCollection where Element: Equatable {
     subrange: Range<Index>,
     maxReplacements: Int = .max
   ) -> Self where C.Element == Element, Replacement.Element == Element {
-    _replacing(
-      self[subrange]._ranges(of: other),
-      with: replacement,
-      maxReplacements: maxReplacements)
+    switch (self, other, replacement) {
+    case (let str as String, let other as String, let repl as String):
+      return str[...]._replacingSubstring(other[...], with: repl[...],
+                                maxReplacements: maxReplacements) as! Self
+    case (let str as Substring, let other as Substring, let repl as Substring):
+      return str._replacingSubstring(other, with: repl,
+                                maxReplacements: maxReplacements)[...] as! Self
+
+    case (let str as Substring, let other as String, let repl as String):
+      return str[...]._replacingSubstring(other[...], with: repl[...],
+                                maxReplacements: maxReplacements)[...] as! Self
+    case (let str as String, let other as Substring, let repl as String):
+      return str[...]._replacingSubstring(other[...], with: repl[...],
+                                maxReplacements: maxReplacements) as! Self
+    case (let str as String, let other as String, let repl as Substring):
+      return str[...]._replacingSubstring(other[...], with: repl[...],
+                                maxReplacements: maxReplacements) as! Self
+
+    case (let str as String, let other as Substring, let repl as Substring):
+      return str[...]._replacingSubstring(other[...], with: repl[...],
+                                maxReplacements: maxReplacements) as! Self
+    case (let str as Substring, let other as String, let repl as Substring):
+      return str[...]._replacingSubstring(other[...], with: repl[...],
+                                maxReplacements: maxReplacements)[...] as! Self
+    case (let str as Substring, let other as Substring, let repl as String):
+      return str[...]._replacingSubstring(other[...], with: repl[...],
+                                maxReplacements: maxReplacements)[...] as! Self
+
+    default:
+      return _replacing(
+        self[subrange]._ranges(of: other),
+        with: replacement,
+        maxReplacements: maxReplacements)
+    }
   }
 
   /// Returns a new collection in which all occurrences of a target sequence

Sources/_StringProcessing/Algorithms/Algorithms/Split.swift

@@ -148,10 +148,22 @@ extension Collection where Element: Equatable {
     maxSplits: Int = .max,
     omittingEmptySubsequences: Bool = true
   ) -> [SubSequence] where C.Element == Element {
-    Array(split(
-      by: ZSearcher(pattern: Array(separator), by: ==),
-      maxSplits: maxSplits,
-      omittingEmptySubsequences: omittingEmptySubsequences))
+    switch (self, separator) {
+    case (let str as String, let sep as String):
+      return str[...]._split(separator: sep, maxSplits: maxSplits, omittingEmptySubsequences: omittingEmptySubsequences) as! [SubSequence]
+    case (let str as String, let sep as Substring):
+      return str[...]._split(separator: sep, maxSplits: maxSplits, omittingEmptySubsequences: omittingEmptySubsequences) as! [SubSequence]
+    case (let str as Substring, let sep as String):
+      return str._split(separator: sep, maxSplits: maxSplits, omittingEmptySubsequences: omittingEmptySubsequences) as! [SubSequence]
+    case (let str as Substring, let sep as Substring):
+      return str._split(separator: sep, maxSplits: maxSplits, omittingEmptySubsequences: omittingEmptySubsequences) as! [SubSequence]
+
+    default:
+      return Array(split(
+        by: ZSearcher(pattern: Array(separator), by: ==),
+        maxSplits: maxSplits,
+        omittingEmptySubsequences: omittingEmptySubsequences))
+    }
   }
 }
 
@@ -174,8 +186,8 @@ extension StringProtocol where SubSequence == Substring {
     maxSplits: Int = .max,
     omittingEmptySubsequences: Bool = true
   ) -> [Substring] {
-    Array(split(
-      by: ZSearcher(pattern: Array(separator), by: ==),
+    Array(self[...].split(
+      by: SubstringSearcher(text: "" as Substring, pattern: separator[...]),
       maxSplits: maxSplits,
       omittingEmptySubsequences: omittingEmptySubsequences))
   }
@@ -187,8 +199,8 @@ extension StringProtocol where SubSequence == Substring {
     maxSplits: Int = .max,
     omittingEmptySubsequences: Bool = true
   ) -> [Substring] {
-    Array(split(
-      by: ZSearcher(pattern: Array(separator), by: ==),
+    Array(self[...].split(
+      by: SubstringSearcher(text: "" as Substring, pattern: separator[...]),
       maxSplits: maxSplits,
       omittingEmptySubsequences: omittingEmptySubsequences))
   }