Merge pull request #23706 from milseman/5_1_gab_barg

[5.1][SE-0248] String Gaps

Author: milseman

stdlib/public/Darwin/Foundation/NSRange.swift

@@ -175,17 +175,33 @@ extension Range where Bound == Int {
 }
 
 extension Range where Bound == String.Index {
-  public init?(_ range: NSRange, in string: __shared String) {
+  private init?<S: StringProtocol>(
+    _ range: NSRange, _genericIn string: __shared S
+  ) {
+    // Corresponding stdlib version
+    guard #available(macOS 9999, iOS 9999, tvOS 9999, watchOS 9999, *) else {
+      fatalError()
+    }
     let u = string.utf16
     guard range.location != NSNotFound,
-      let start = u.index(u.startIndex, offsetBy: range.location, limitedBy: u.endIndex),
-      let end = u.index(u.startIndex, offsetBy: range.location + range.length, limitedBy: u.endIndex),
+      let start = u.index(
+        u.startIndex, offsetBy: range.location, limitedBy: u.endIndex),
+      let end = u.index(
+        start, offsetBy: range.length, limitedBy: u.endIndex),
       let lowerBound = String.Index(start, within: string),
       let upperBound = String.Index(end, within: string)
     else { return nil }
-    
+
     self = lowerBound..<upperBound
   }
+
+  public init?(_ range: NSRange, in string: __shared String) {
+    self.init(range, _genericIn: string)
+  }
+  @available(macOS 9999, iOS 9999, tvOS 9999, watchOS 9999, *)
+  public init?<S: StringProtocol>(_ range: NSRange, in string: __shared S) {
+    self.init(range, _genericIn: string)
+  }
 }
 
 extension NSRange : CustomReflectable {

stdlib/public/core/ASCII.swift

@@ -23,6 +23,10 @@ extension Unicode.ASCII : Unicode.Encoding {
     return EncodedScalar(0x1a) // U+001A SUBSTITUTE; best we can do for ASCII
   }
 
+  /// Returns whether the given code unit represents an ASCII scalar
+  @_alwaysEmitIntoClient
+  public static func isASCII(_ x: CodeUnit) -> Bool { return UTF8.isASCII(x) }
+
   @inline(__always)
   @inlinable
   public static func _isScalar(_ x: CodeUnit) -> Bool {

stdlib/public/core/CMakeLists.txt

@@ -90,6 +90,7 @@ set(SWIFTLIB_ESSENTIAL
   KeyValuePairs.swift
   LazyCollection.swift
   LazySequence.swift
+  LegacyABI.swift
   LifetimeManager.swift
   ManagedBuffer.swift
   Map.swift

stdlib/public/core/Character.swift

@@ -88,25 +88,26 @@ extension Character {
 }
 
 extension Character {
-  @usableFromInline
-  typealias UTF8View = String.UTF8View
+  /// A view of a character's contents as a collection of UTF-8 code units. See
+  /// String.UTF8View for more information
+  public typealias UTF8View = String.UTF8View
 
+  /// A UTF-8 encoding of `self`.
   @inlinable
-  internal var utf8: UTF8View {
-    return _str.utf8
-  }
-  @usableFromInline
-  typealias UTF16View = String.UTF16View
+  public var utf8: UTF8View { return _str.utf8 }
+
+  /// A view of a character's contents as a collection of UTF-16 code units. See
+  /// String.UTF16View for more information
+  public typealias UTF16View = String.UTF16View
 
+  /// A UTF-16 encoding of `self`.
   @inlinable
-  internal var utf16: UTF16View {
-    return _str.utf16
-  }
+  public var utf16: UTF16View { return _str.utf16 }
+
   public typealias UnicodeScalarView = String.UnicodeScalarView
+
   @inlinable
-  public var unicodeScalars: UnicodeScalarView {
-    return _str.unicodeScalars
-  }
+  public var unicodeScalars: UnicodeScalarView { return _str.unicodeScalars }
 }
 
 extension Character :

stdlib/public/core/GroupInfo.json

@@ -223,6 +223,7 @@
     "Equatable.swift",
     "Comparable.swift",
     "Codable.swift",
+    "LegacyABI.swift",
     "MigrationSupport.swift"
   ],
   "Result": [

stdlib/public/core/LegacyABI.swift

@@ -0,0 +1,41 @@
+//===----------------------------------------------------------------------===//
+//
+// This source file is part of the Swift.org open source project
+//
+// Copyright (c) 2014 - 2019 Apple Inc. and the Swift project authors
+// Licensed under Apache License v2.0 with Runtime Library Exception
+//
+// See https://swift.org/LICENSE.txt for license information
+// See https://swift.org/CONTRIBUTORS.txt for the list of Swift project authors
+//
+//===----------------------------------------------------------------------===//
+
+// This file contains non-API (or underscored) declarations that are needed to
+// be kept around for ABI compatibility
+
+extension Unicode.UTF16 {
+  @available(*, unavailable, renamed: "Unicode.UTF16.isASCII")
+  @inlinable
+  public static func _isASCII(_ x: CodeUnit) -> Bool  {
+    return Unicode.UTF16.isASCII(x)
+  }
+}
+
+@available(*, unavailable, renamed: "Unicode.UTF8.isASCII")
+@inlinable
+internal func _isASCII(_ x: UInt8) -> Bool {
+  return Unicode.UTF8.isASCII(x)
+}
+
+@available(*, unavailable, renamed: "Unicode.UTF8.isContinuation")
+@inlinable
+internal func _isContinuation(_ x: UInt8) -> Bool {
+  return UTF8.isContinuation(x)
+}
+
+extension Substring {
+@available(*, unavailable, renamed: "Substring.base")
+  @inlinable
+  internal var _wholeString: String { return base }
+}
+

stdlib/public/core/StringComparison.swift

@@ -239,7 +239,7 @@ private func _findBoundary(
   }
 
   // Back up to scalar boundary
-  while _isContinuation(utf8[_unchecked: idx]) {
+  while UTF8.isContinuation(utf8[_unchecked: idx]) {
     idx &-= 1
   }
 

stdlib/public/core/StringCreate.swift

@@ -193,8 +193,8 @@ extension String {
   internal static func _fromSubstring(
     _ substring: __shared Substring
   ) -> String {
-    if substring._offsetRange == substring._wholeString._offsetRange {
-      return substring._wholeString
+    if substring._offsetRange == substring.base._offsetRange {
+      return substring.base
     }
 
     return String._copying(substring)

stdlib/public/core/StringIndexConversions.swift

@@ -11,6 +11,60 @@
 //===----------------------------------------------------------------------===//
 
 extension String.Index {
+  private init?<S: StringProtocol>(
+    _ idx: String.Index, _genericWithin target: S
+  ) {
+    guard target._wholeGuts.isOnGraphemeClusterBoundary(idx),
+          idx >= target.startIndex && idx <= target.endIndex
+    else {
+      return nil
+    }
+
+    self = idx
+  }
+
+  /// Creates an index in the given string that corresponds exactly to the
+  /// specified position.
+  ///
+  /// If the index passed as `sourcePosition` represents the start of an
+  /// extended grapheme cluster---the element type of a string---then the
+  /// initializer succeeds.
+  ///
+  /// The following example converts the position of the Unicode scalar `"e"`
+  /// into its corresponding position in the string. The character at that
+  /// position is the composed `"é"` character.
+  ///
+  ///     let cafe = "Cafe\u{0301}"
+  ///     print(cafe)
+  ///     // Prints "Café"
+  ///
+  ///     let scalarsIndex = cafe.unicodeScalars.firstIndex(of: "e")!
+  ///     let stringIndex = String.Index(scalarsIndex, within: cafe)!
+  ///
+  ///     print(cafe[...stringIndex])
+  ///     // Prints "Café"
+  ///
+  /// If the index passed as `sourcePosition` doesn't have an exact
+  /// corresponding position in `target`, the result of the initializer is
+  /// `nil`. For example, an attempt to convert the position of the combining
+  /// acute accent (`"\u{0301}"`) fails. Combining Unicode scalars do not have
+  /// their own position in a string.
+  ///
+  ///     let nextScalarsIndex = cafe.unicodeScalars.index(after: scalarsIndex)
+  ///     let nextStringIndex = String.Index(nextScalarsIndex, within: cafe)
+  ///
+  ///     print(nextStringIndex)
+  ///     // Prints "nil"
+  ///
+  /// - Parameters:
+  ///   - sourcePosition: A position in a view of the `target` parameter.
+  ///     `sourcePosition` must be a valid index of at least one of the views
+  ///     of `target`.
+  ///   - target: The string referenced by the resulting index.
+  public init?(_ sourcePosition: String.Index, within target: String) {
+    self.init(sourcePosition, _genericWithin: target)
+  }
+
   /// Creates an index in the given string that corresponds exactly to the
   /// specified position.
   ///
@@ -49,14 +103,11 @@ extension String.Index {
   ///     `sourcePosition` must be a valid index of at least one of the views
   ///     of `target`.
   ///   - target: The string referenced by the resulting index.
-  public init?(
-    _ sourcePosition: String.Index,
-    within target: String
+  @available(macOS 9999, iOS 9999, tvOS 9999, watchOS 9999, *)
+  public init?<S: StringProtocol>(
+    _ sourcePosition: String.Index, within target: S
   ) {
-    guard target._guts.isOnGraphemeClusterBoundary(sourcePosition) else {
-      return nil
-    }
-    self = sourcePosition
+    self.init(sourcePosition, _genericWithin: target)
   }
 
   /// Returns the position in the given UTF-8 view that corresponds exactly to
@@ -81,7 +132,7 @@ extension String.Index {
   ///   position of a UTF-16 trailing surrogate returns `nil`.
   public func samePosition(
     in utf8: String.UTF8View
-    ) -> String.UTF8View.Index? {
+  ) -> String.UTF8View.Index? {
     return String.UTF8View.Index(self, within: utf8)
   }
 

stdlib/public/core/StringNormalization.swift

@@ -108,7 +108,7 @@ extension UnsafeBufferPointer where Element == UInt8 {
     if index == 0 || index == count {
       return true
     }
-    assert(!_isContinuation(self[_unchecked: index]))
+    assert(!UTF8.isContinuation(self[_unchecked: index]))
 
     // Sub-300 latiny fast-path
     if self[_unchecked: index] < 0xCC { return true }
@@ -165,7 +165,7 @@ extension UnsafeBufferPointer where Element == UInt8 {
       _internalInvariant(index == count)
       return true
     }
-    return !_isContinuation(self[index])
+    return !UTF8.isContinuation(self[index])
   }
 
 }

stdlib/public/core/StringUTF8Validation.swift

@@ -7,19 +7,15 @@ private func _isNotOverlong_F0(_ x: UInt8) -> Bool {
 }
 
 private func _isNotOverlong_F4(_ x: UInt8) -> Bool {
-  return _isContinuation(x) && x <= 0x8F
+  return UTF8.isContinuation(x) && x <= 0x8F
 }
 
 private func _isNotOverlong_E0(_ x: UInt8) -> Bool {
   return (0xA0...0xBF).contains(x)
 }
 
 private func _isNotOverlong_ED(_ x: UInt8) -> Bool {
-  return _isContinuation(x) && x <= 0x9F
-}
-
-private func _isASCII_cmp(_ x: UInt8) -> Bool {
-  return x <= 0x7F
+  return UTF8.isContinuation(x) && x <= 0x9F
 }
 
 internal struct UTF8ExtraInfo: Equatable {
@@ -48,7 +44,7 @@ internal func validateUTF8(_ buf: UnsafeBufferPointer<UInt8>) -> UTF8ValidationR
     guard f(cu) else { throw UTF8ValidationError() }
   }
   @inline(__always) func guaranteeContinuation() throws {
-    try guaranteeIn(_isContinuation)
+    try guaranteeIn(UTF8.isContinuation)
   }
 
   func _legacyInvalidLengthCalculation(_ _buffer: (_storage: UInt32, ())) -> Int {
@@ -94,7 +90,7 @@ internal func validateUTF8(_ buf: UnsafeBufferPointer<UInt8>) -> UTF8ValidationR
     var endIndex = buf.startIndex
     var iter = buf.makeIterator()
     _ = iter.next()
-    while let cu = iter.next(), !_isASCII(cu) && !_isUTF8MultiByteLeading(cu) {
+    while let cu = iter.next(), UTF8.isContinuation(cu) {
       endIndex += 1
     }
     let illegalRange = Range(buf.startIndex...endIndex)
@@ -107,7 +103,7 @@ internal func validateUTF8(_ buf: UnsafeBufferPointer<UInt8>) -> UTF8ValidationR
   do {
     var isASCII = true
     while let cu = iter.next() {
-      if _isASCII(cu) { lastValidIndex &+= 1; continue }
+      if UTF8.isASCII(cu) { lastValidIndex &+= 1; continue }
       isASCII = false
       if _slowPath(!_isUTF8MultiByteLeading(cu)) {
         throw UTF8ValidationError()

stdlib/public/core/StringUTF8View.swift

@@ -415,7 +415,7 @@ extension String.UTF8View {
 
     let (scalar, scalarLen) = _guts.foreignErrorCorrectedScalar(
       startingAt: i.strippingTranscoding)
-    let utf8Len = _numUTF8CodeUnits(scalar)
+    let utf8Len = UTF8.width(scalar)
 
     if utf8Len == 1 {
       _internalInvariant(i.transcodedOffset == 0)
@@ -442,7 +442,7 @@ extension String.UTF8View {
 
     let (scalar, scalarLen) = _guts.foreignErrorCorrectedScalar(
       endingAt: i)
-    let utf8Len = _numUTF8CodeUnits(scalar)
+    let utf8Len = UTF8.width(scalar)
     return i.encoded(offsetBy: -scalarLen).transcoded(withOffset: utf8Len &- 1)
   }
 

stdlib/public/core/StringUnicodeScalarView.swift

@@ -418,7 +418,7 @@ extension String.UnicodeScalarView {
   internal func _foreignIndex(after i: Index) -> Index {
     _internalInvariant(_guts.isForeign)
     let cu = _guts.foreignErrorCorrectedUTF16CodeUnit(at: i)
-    let len = _isLeadingSurrogate(cu) ? 2 : 1
+    let len = UTF16.isLeadSurrogate(cu) ? 2 : 1
 
     return i.encoded(offsetBy: len)
   }
@@ -429,7 +429,7 @@ extension String.UnicodeScalarView {
     _internalInvariant(_guts.isForeign)
     let priorIdx = i.priorEncoded
     let cu = _guts.foreignErrorCorrectedUTF16CodeUnit(at: priorIdx)
-    let len = _isTrailingSurrogate(cu) ? 2 : 1
+    let len = UTF16.isTrailSurrogate(cu) ? 2 : 1
 
     return i.encoded(offsetBy: -len)
   }