Limit CC ranges to single-scalar bounds

Character class ranges don't work well with multi-scalar inputs,
in either the range or the matched character. This change limits
range endpoints to single-scalar characters and matches only
characters that are themselves a single scalar.

Fixes issue swiftlang#407, which now displays this behavior:

```
try /[1-2]/.wholeMatch(in: "1️⃣")     // nil
try /[12]/.wholeMatch(in: "1️⃣")      // nil
try /(?U)[\d]/.wholeMatch(in: "1️⃣")  // nil
```

Author: natecook1000

Sources/_RegexParser/Regex/AST/Atom.swift

@@ -771,7 +771,7 @@ extension AST.Atom {
   /// range.
   public var isValidCharacterClassRangeBound: Bool {
     // If we have a literal character value for this, it can be used as a bound.
-    if literalCharacterValue != nil { return true }
+    if literalCharacterValue?.hasExactlyOneScalar == true { return true }
     switch kind {
     // \cx, \C-x, \M-x, \M-\C-x, \N{...}
     case .keyboardControl, .keyboardMeta, .keyboardMetaControl, .namedCharacter:

Sources/_RegexParser/Utility/Misc.swift

@@ -19,6 +19,13 @@ extension Substring {
   var string: String { String(self) }
 }
 
+extension Character {
+  /// Whether this character is made up of exactly one Unicode scalar value.
+  public var hasExactlyOneScalar: Bool {
+    unicodeScalars.index(after: unicodeScalars.startIndex) == unicodeScalars.endIndex
+  }
+}
+
 extension CustomStringConvertible {
   @_alwaysEmitIntoClient
   public var halfWidthCornerQuoted: String {

Sources/_StringProcessing/ConsumerInterface.swift

@@ -248,10 +248,10 @@ extension DSLTree.CustomCharacterClass.Member {
       return c
     case let .range(low, high):
       // TODO:
-      guard let lhs = low.literalCharacterValue else {
+      guard let lhs = low.literalCharacterValue, lhs.hasExactlyOneScalar else {
         throw Unsupported("\(low) in range")
       }
-      guard let rhs = high.literalCharacterValue else {
+      guard let rhs = high.literalCharacterValue, rhs.hasExactlyOneScalar else {
         throw Unsupported("\(high) in range")
       }
 
@@ -262,6 +262,7 @@ extension DSLTree.CustomCharacterClass.Member {
         return { input, bounds in
           // TODO: check for out of bounds?
           let curIdx = bounds.lowerBound
+          guard input[curIdx].hasExactlyOneScalar else { return nil }
           if (lhsLower...rhsLower).contains(input[curIdx].lowercased()) {
             // TODO: semantic level
             return input.index(after: curIdx)
@@ -273,6 +274,7 @@ extension DSLTree.CustomCharacterClass.Member {
         return { input, bounds in
           // TODO: check for out of bounds?
           let curIdx = bounds.lowerBound
+          guard input[curIdx].hasExactlyOneScalar else { return nil }
           if (lhs...rhs).contains(input[curIdx]) {
             // TODO: semantic level
             return input.index(after: curIdx)

Sources/_StringProcessing/Unicode/CharacterProps.swift

@@ -11,10 +11,3 @@
 
 
 // TODO
-
-extension Character {
-  /// Whether this character is made up of exactly one Unicode scalar value.
-  var hasExactlyOneScalar: Bool {
-    unicodeScalars.index(after: unicodeScalars.startIndex) == unicodeScalars.endIndex
-  }
-}

Sources/_StringProcessing/_CharacterClassModel.swift

@@ -103,6 +103,11 @@ public struct _CharacterClassModel: Hashable {
           return c == character
         }
       case .range(let range):
+        // Ranges can be formed with single-scalar characters, and can only
+        // match as such.
+        // FIXME: Convert to canonical composed version before testing?
+        guard character.hasExactlyOneScalar else { return false }
+        
         if options.isCaseInsensitive {
           let newLower = range.lowerBound.lowercased()
           let newUpper = range.upperBound.lowercased()

Tests/RegexTests/ParseTests.swift

@@ -2534,6 +2534,8 @@ extension RegexTests {
     diagnosticTest("[[:=:]]", .emptyProperty)
 
     diagnosticTest(#"|([\d-c])?"#, .invalidCharacterClassRangeOperand)
+    diagnosticTest(#"|([🇦🇫-🇿🇼])?"#, .invalidCharacterClassRangeOperand)
+    diagnosticTest(#"|([👨‍👩‍👦-👩‍👩‍👧‍👧])?"#, .invalidCharacterClassRangeOperand)
 
     diagnosticTest(#"[_-A]"#, .invalidCharacterRange(from: "_", to: "A"))
     diagnosticTest(#"(?i)[_-A]"#, .invalidCharacterRange(from: "_", to: "A"))