Add repeating functions to DSL (#161)

These allow you to spell literal expressions like `a{3,}` or `a{2,5}`
while using the Regex DSL.

Author: natecook1000

Sources/VariadicsGenerator/VariadicsGenerator.swift

@@ -151,6 +151,8 @@ struct VariadicsGenerator: ParsableCommand {
         print("\(kind.rawValue) ", terminator: "", to: &standardError)
         emitQuantifier(kind: kind, arity: arity)
       }
+      print("repeating ", terminator: "", to: &standardError)
+      emitRepeating(arity: arity)
       print(to: &standardError)
     }
 
@@ -307,69 +309,140 @@ struct VariadicsGenerator: ParsableCommand {
       }
     }
   }
+  
+  struct QuantifierParameters {
+    var disfavored: String
+    var genericParams: String
+    var whereClause: String
+    var quantifiedCaptures: String
+    var matchType: String
+    
+    var repeatingWhereClause: String {
+      whereClause.isEmpty
+        ? "where R.Bound == Int"
+        : whereClause + ", R.Bound == Int"
+    }
+    
+    init(kind: QuantifierKind, arity: Int) {
+      self.disfavored = arity == 0 ? "@_disfavoredOverload\n" : ""
+      self.genericParams = {
+        var result = ""
+        if arity > 0 {
+          result += "W"
+          result += (0..<arity).map { ", C\($0)" }.joined()
+          result += ", "
+        }
+        result += "Component: \(regexProtocolName)"
+        return result
+      }()
+      
+      let captures = (0..<arity).map { "C\($0)" }
+      let capturesJoined = captures.joined(separator: ", ")
+      self.whereClause = arity == 0 ? "" :
+        "where Component.Match == (W, \(capturesJoined))"
+      self.quantifiedCaptures = {
+        switch kind {
+        case .zeroOrOne, .zeroOrMore:
+          return captures.map { "\($0)?" }.joined(separator: ", ")
+        case .oneOrMore:
+          return capturesJoined
+        }
+      }()
+      self.matchType = arity == 0
+        ? baseMatchTypeName
+        : "(\(baseMatchTypeName), \(quantifiedCaptures))"
+    }
+  }
 
   func emitQuantifier(kind: QuantifierKind, arity: Int) {
     assert(arity >= 0)
-    let genericParams: String = {
-      var result = ""
-      if arity > 0 {
-        result += "W"
-        result += (0..<arity).map { ", C\($0)" }.joined()
-        result += ", "
-      }
-      result += "Component: \(regexProtocolName)"
-      return result
-    }()
-    let captures = (0..<arity).map { "C\($0)" }
-    let capturesJoined = captures.joined(separator: ", ")
-    let whereClause: String = arity == 0 ? "" :
-      "where Component.Match == (W, \(capturesJoined))"
-    let quantifiedCaptures: String = {
-      switch kind {
-      case .zeroOrOne, .zeroOrMore:
-        return captures.map { "\($0)?" }.joined(separator: ", ")
-      case .oneOrMore:
-        return capturesJoined
-      }
-    }()
-    let matchType = arity == 0 ? baseMatchTypeName : "(\(baseMatchTypeName), \(quantifiedCaptures))"
+    let params = QuantifierParameters(kind: kind, arity: arity)
     output("""
-      \(arity == 0 ? "@_disfavoredOverload" : "")
-      public func \(kind.rawValue)<\(genericParams)>(
+      \(params.disfavored)\
+      public func \(kind.rawValue)<\(params.genericParams)>(
         _ component: Component,
         _ behavior: QuantificationBehavior = .eagerly
-      ) -> \(regexTypeName)<\(matchType)> \(whereClause) {
+      ) -> \(regexTypeName)<\(params.matchType)> \(params.whereClause) {
         .init(node: .quantification(.\(kind.astQuantifierAmount), behavior.astKind, component.regex.root))
       }
 
-      \(arity == 0 ? "@_disfavoredOverload" : "")
-      public func \(kind.rawValue)<\(genericParams)>(
+      \(params.disfavored)\
+      public func \(kind.rawValue)<\(params.genericParams)>(
         _ behavior: QuantificationBehavior = .eagerly,
         @RegexBuilder _ component: () -> Component
-      ) -> \(regexTypeName)<\(matchType)> \(whereClause) {
+      ) -> \(regexTypeName)<\(params.matchType)> \(params.whereClause) {
         .init(node: .quantification(.\(kind.astQuantifierAmount), behavior.astKind, component().regex.root))
       }
 
-      \(arity == 0 ? "@_disfavoredOverload" : "")
-      public postfix func \(kind.operatorName)<\(genericParams)>(
+      \(params.disfavored)\
+      public postfix func \(kind.operatorName)<\(params.genericParams)>(
         _ component: Component
-      ) -> \(regexTypeName)<\(matchType)> \(whereClause) {
+      ) -> \(regexTypeName)<\(params.matchType)> \(params.whereClause) {
         .init(node: .quantification(.\(kind.astQuantifierAmount), .eager, component.regex.root))
       }
 
       \(kind == .zeroOrOne ?
         """
         extension RegexBuilder {
-          public static func buildLimitedAvailability<\(genericParams)>(
+          public static func buildLimitedAvailability<\(params.genericParams)>(
             _ component: Component
-          ) -> \(regexTypeName)<\(matchType)> \(whereClause) {
+          ) -> \(regexTypeName)<\(params.matchType)> \(params.whereClause) {
             .init(node: .quantification(.\(kind.astQuantifierAmount), .eager, component.regex.root))
           }
         }
         """ : "")
 
       """)
   }
+  
+  func emitRepeating(arity: Int) {
+    assert(arity >= 0)
+    // `repeat(..<5)` has the same generic semantics as zeroOrMore
+    let params = QuantifierParameters(kind: .zeroOrMore, arity: arity)
+    // TODO: Could `repeat(count:)` have the same generic semantics as oneOrMore?
+    // We would need to prohibit `repeat(count: 0)`; can only happen at runtime
+    
+    output("""
+      \(params.disfavored)\
+      public func repeating<\(params.genericParams)>(
+        _ component: Component,
+        count: Int
+      ) -> \(regexTypeName)<\(params.matchType)> \(params.whereClause) {
+        assert(count > 0, "Must specify a positive count")
+        // TODO: Emit a warning about `repeatMatch(count: 0)` or `repeatMatch(count: 1)`
+        return Regex(node: .quantification(.exactly(.init(faking: count)), .eager, component.regex.root))
+      }
+
+      \(params.disfavored)\
+      public func repeating<\(params.genericParams)>(
+        count: Int,
+        @RegexBuilder _ component: () -> Component
+      ) -> \(regexTypeName)<\(params.matchType)> \(params.whereClause) {
+        assert(count > 0, "Must specify a positive count")
+        // TODO: Emit a warning about `repeatMatch(count: 0)` or `repeatMatch(count: 1)`
+        return Regex(node: .quantification(.exactly(.init(faking: count)), .eager, component().regex.root))
+      }
+      
+      \(params.disfavored)\
+      public func repeating<\(params.genericParams), R: RangeExpression>(
+        _ component: Component,
+        _ expression: R,
+        _ behavior: QuantificationBehavior = .eagerly
+      ) -> \(regexTypeName)<\(params.matchType)> \(params.repeatingWhereClause) {
+        .init(node: .repeating(expression.relative(to: 0..<Int.max), behavior, component.regex.root))
+      }
+
+      \(params.disfavored)\
+      public func repeating<\(params.genericParams), R: RangeExpression>(
+        _ expression: R,
+        _ behavior: QuantificationBehavior = .eagerly,
+        @RegexBuilder _ component: () -> Component
+      ) -> \(regexTypeName)<\(params.matchType)> \(params.repeatingWhereClause) {
+        .init(node: .repeating(expression.relative(to: 0..<Int.max), behavior, component().regex.root))
+      }
+      
+      """)
+  }
 
   func emitAlternation(leftArity: Int, rightArity: Int) {
     let leftGenParams: String = {

Sources/_StringProcessing/RegexDSL/DSLTree.swift

@@ -307,3 +307,33 @@ extension DSLTree.Node {
     return .alternation([self, newNode])
   }
 }
+
+extension DSLTree.Node {
+  /// Generates a DSLTree node for a repeated range of the given DSLTree node.
+  /// Individual public API functions are in the generated Variadics.swift file.
+  static func repeating(
+    _ range: Range<Int>,
+    _ behavior: QuantificationBehavior,
+    _ node: DSLTree.Node
+  ) -> DSLTree.Node {
+    // TODO: Throw these as errors
+    assert(range.lowerBound >= 0, "Cannot specify a negative lower bound")
+    assert(!range.isEmpty, "Cannot specify an empty range")
+    
+    switch (range.lowerBound, range.upperBound) {
+    case (0, Int.max): // 0...
+      return .quantification(.zeroOrMore, behavior.astKind, node)
+    case (1, Int.max): // 1...
+      return .quantification(.oneOrMore, behavior.astKind, node)
+    case _ where range.count == 1: // ..<1 or ...0 or any range with count == 1
+      // Note: `behavior` is ignored in this case
+      return .quantification(.exactly(.init(faking: range.lowerBound)), .eager, node)
+    case (0, _): // 0..<n or 0...n or ..<n or ...n
+      return .quantification(.upToN(.init(faking: range.upperBound)), behavior.astKind, node)
+    case (_, Int.max): // n...
+      return .quantification(.nOrMore(.init(faking: range.lowerBound)), behavior.astKind, node)
+    default: // any other range
+      return .quantification(.range(.init(faking: range.lowerBound), .init(faking: range.upperBound)), behavior.astKind, node)
+    }
+  }
+}