Parse Swift operators as link components for operators with symbol diacritics (#1125)

Author: d-ronnqvist

Sources/SwiftDocC/Infrastructure/Link Resolution/PathHierarchy+PathComponent.swift

@@ -248,7 +248,7 @@ private struct PathComponentScanner {
 
     mutating func _scanOperatorName() -> Substring? {
         // If the next component is a Swift operator, parse the full operator before splitting on "/" ("/" may appear in the operator name)
-        if remaining.unicodeScalars.prefix(3).allSatisfy(\.isValidSwiftOperatorHead) {
+        if remaining.unicodeScalars.prefix(3).isValidSwiftOperator() {
             return scanUntil(index: remaining.firstIndex(of: Self.swiftOperatorEnd)) + scan(length: 1)
         }
 
@@ -354,6 +354,20 @@ private extension StringProtocol {
     }
 }
 
+private extension Collection<Unicode.Scalar> {
+    /// Determines if this sequence of unicode scalars represent a valid Swift operator name
+    /// - Complexity: O(_n_), where _n_ is the length of the collection.
+    func isValidSwiftOperator() -> Bool {
+        // See https://docs.swift.org/swift-book/documentation/the-swift-programming-language/lexicalstructure#Operators
+        
+        // The first character of an operator supports fewer characters than the rest of the operator name
+        guard let first, first.isValidSwiftOperatorHead else {
+            return false
+        }
+        return dropFirst().allSatisfy { $0.isValidSwiftOperatorCharacter }
+    }
+}
+
 private extension Unicode.Scalar {
     /// Checks if this unicode scalar is a valid C99 Extended Identifier.
     var isValidC99ExtendedIdentifier: Bool {
@@ -547,10 +561,11 @@ private extension Unicode.Scalar {
         }
     }
 
+    /// A Boolean value that indicates if this scalar is a valid Swift operator head.
     var isValidSwiftOperatorHead: Bool {
         // See https://docs.swift.org/swift-book/documentation/the-swift-programming-language/lexicalstructure#Operators
         switch value {
-        case 
+        case
             // ! % & * + - . / < = > ? ^| ~
             0x21, 0x25, 0x26, 0x2A, 0x2B, 0x2D...0x2F, 0x3C, 0x3D...0x3F, 0x5E, 0x7C, 0x7E,
             // ¡ ¢ £ ¤ ¥ ¦ §
@@ -573,6 +588,12 @@ private extension Unicode.Scalar {
             0x2041 ... 0x2053,
             // ⁕ ⁖ ⁗ ⁘ ⁙ ⁚ ⁛ ⁜ ⁝ ⁞
             0x2055 ... 0x205E,
+            // Arrows
+            0x2190 ... 0x21FF,
+            // Mathematical Operators
+            0x2200 ... 0x22FF,
+            // Miscellaneous Technical
+            0x2300 ... 0x23FF,
             // Box Drawing
             0x2500 ... 0x257F,
             // Block Elements
@@ -610,4 +631,31 @@ private extension Unicode.Scalar {
             return false
         }
     }
+    
+    /// A Boolean value that indicates if this scalar is a valid Swift operator character (after the first character).
+    var isValidSwiftOperatorCharacter: Bool {
+        // See https://docs.swift.org/swift-book/documentation/the-swift-programming-language/lexicalstructure#Operators
+        if isValidSwiftOperatorHead {
+            return true
+        }
+        
+        switch value {
+        case
+            // Combining Diacritical Marks
+            0x0300 ... 0x036F,
+            // Combining Diacritical Marks Supplement
+            0x1DC0 ... 0x1DFF,
+            // Combining Diacritical Marks for Symbols
+            0x20D0 ... 0x20FF,
+            // Variation Selectors
+            0xFE00 ... 0xFE0F,
+            // Combining Half Marks
+            0xFE20 ... 0xFE2F,
+            // Variation Selectors Supplement
+            0xE0100 ... 0xE01EF:
+            return true
+        default:
+            return false
+        }
+    }
 }

Tests/SwiftDocCTests/Infrastructure/PathHierarchyTests.swift

@@ -3453,6 +3453,9 @@ class PathHierarchyTests: XCTestCase {
         assertParsedPathComponents("MyNumber//=(_:_:)", [("MyNumber", nil), ("/=(_:_:)", nil)])
         assertParsedPathComponents("MyNumber////=(_:_:)", [("MyNumber", nil), ("///=(_:_:)", nil)])
         assertParsedPathComponents("MyNumber/+/-(_:_:)", [("MyNumber", nil), ("+/-(_:_:)", nil)])
+        
+        // "☜⃩" is a symbol with a symbol diacritic mark.
+        assertParsedPathComponents("☜⃩/(_:_:)", [("☜⃩/(_:_:)", nil)])
 
         // Check parsing return values and parameter types
         assertParsedPathComponents("..<(_:_:)->Bool", [("..<(_:_:)", .typeSignature(parameterTypes: nil, returnTypes: ["Bool"]))])