Simplified tokenizer

src/ast/AST.go

@@ -1,4 +1,7 @@
 package ast
 
+// Node is an interface used for all types of AST nodes.
 type Node any
+
+// AST is an abstract syntax tree which is simply a list of nodes.
 type AST []Node

src/token/Tokenize.go

@@ -27,246 +27,29 @@ func Tokenize(buffer []byte) List {
 		case '\n':
 			tokens = append(tokens, Token{Kind: NewLine, Position: i, Length: 1})
 		case '-':
-			if len(tokens) == 0 || tokens[len(tokens)-1].IsOperator() || tokens[len(tokens)-1].IsExpressionStart() || tokens[len(tokens)-1].IsKeyword() {
+			tokens, i = dash(tokens, buffer, i)
-				tokens = append(tokens, Token{Kind: Negate, Position: i, Length: 1})
-			} else {
-				if i+1 < Position(len(buffer)) {
-					switch buffer[i+1] {
-					case '=':
-						tokens = append(tokens, Token{Kind: SubAssign, Position: i, Length: 2})
-						i++
-					case '>':
-						tokens = append(tokens, Token{Kind: ReturnType, Position: i, Length: 2})
-						i++
-					default:
-						tokens = append(tokens, Token{Kind: Sub, Position: i, Length: 1})
-					}
-				} else {
-					tokens = append(tokens, Token{Kind: Sub, Position: i, Length: 1})
-				}
-			}
-
 		case '/':
-			if i+1 < Position(len(buffer)) && buffer[i+1] == '/' {
+			tokens, i = slash(tokens, buffer, i)
-				position := i
-
-				for i < Position(len(buffer)) && buffer[i] != '\n' {
-					i++
-				}
-
-				tokens = append(tokens, Token{Kind: Comment, Position: position, Length: Length(i - position)})
-			} else {
-				position := i
-				i++
-
-				for i < Position(len(buffer)) && isOperator(buffer[i]) {
-					i++
-				}
-
-				kind := Invalid
-
-				switch string(buffer[position:i]) {
-				case "/":
-					kind = Div
-				case "/=":
-					kind = DivAssign
-				}
-
-				tokens = append(tokens, Token{Kind: kind, Position: position, Length: Length(i - position)})
-			}
-
 			continue
-
 		case '"', '\'':
-			limiter := buffer[i]
+			tokens, i = quote(tokens, buffer, i)
-			start := i
-			end := Position(len(buffer))
-			i++
-
-			for i < Position(len(buffer)) {
-				if buffer[i] == limiter && (buffer[i-1] != '\\' || buffer[i-2] == '\\') {
-					end = i + 1
-					i++
-					break
-				}
-
-				i++
-			}
-
-			kind := String
-
-			if limiter == '\'' {
-				kind = Rune
-			}
-
-			tokens = append(tokens, Token{Kind: kind, Position: start, Length: Length(end - start)})
 			continue
-
 		case '0':
-			position := i
+			tokens, i = zero(tokens, buffer, i)
-			i++
-
-			if i >= Position(len(buffer)) {
-				tokens = append(tokens, Token{Kind: Number, Position: position, Length: 1})
-				break
-			}
-
-			filter := isDigit
-
-			switch buffer[i] {
-			case 'x':
-				i++
-				filter = isHexDigit
-
-			case 'b':
-				i++
-				filter = isBinaryDigit
-
-			case 'o':
-				i++
-				filter = isOctalDigit
-			}
-
-			for i < Position(len(buffer)) && filter(buffer[i]) {
-				i++
-			}
-
-			tokens = append(tokens, Token{Kind: Number, Position: position, Length: Length(i - position)})
 			continue
-
 		default:
 			if isIdentifierStart(buffer[i]) {
-				position := i
+				tokens, i = identifier(tokens, buffer, i)
-				i++
-
-				for i < Position(len(buffer)) && isIdentifier(buffer[i]) {
-					i++
-				}
-
-				identifier := buffer[position:i]
-				kind := Identifier
-
-				switch string(identifier) {
-				case "assert":
-					kind = Assert
-				case "if":
-					kind = If
-				case "else":
-					kind = Else
-				case "import":
-					kind = Import
-				case "loop":
-					kind = Loop
-				case "return":
-					kind = Return
-				case "switch":
-					kind = Switch
-				}
-
-				tokens = append(tokens, Token{Kind: kind, Position: position, Length: Length(len(identifier))})
 				continue
 			}
 
 			if isDigit(buffer[i]) {
-				position := i
+				tokens, i = digit(tokens, buffer, i)
-				i++
-
-				for i < Position(len(buffer)) && isDigit(buffer[i]) {
-					i++
-				}
-
-				last := len(tokens) - 1
-
-				if len(tokens) > 0 && tokens[last].Kind == Negate {
-					tokens[last].Kind = Number
-					tokens[last].Length = Length(i-position) + 1
-				} else {
-					tokens = append(tokens, Token{Kind: Number, Position: position, Length: Length(i - position)})
-				}
-
 				continue
 			}
 
 			if isOperator(buffer[i]) {
-				position := i
+				tokens, i = operator(tokens, buffer, i)
-				i++
-
-				for i < Position(len(buffer)) && isOperator(buffer[i]) {
-					i++
-				}
-
-				kind := Invalid
-
-				switch string(buffer[position:i]) {
-				case "!":
-					kind = Not
-				case "!=":
-					kind = NotEqual
-				case "%":
-					kind = Mod
-				case "%=":
-					kind = ModAssign
-				case "&":
-					kind = And
-				case "&&":
-					kind = LogicalAnd
-				case "&=":
-					kind = AndAssign
-				case "*":
-					kind = Mul
-				case "*=":
-					kind = MulAssign
-				case "+":
-					kind = Add
-				case "+=":
-					kind = AddAssign
-				// case "-":
-				// 	kind = Sub
-				// case "-=":
-				// 	kind = SubAssign
-				// case "->":
-				// 	kind = ReturnType
-				case ".":
-					kind = Period
-				// case "/":
-				// 	kind = Div
-				// case "/=":
-				// 	kind = DivAssign
-				case ":=":
-					kind = Define
-				case "<":
-					kind = Less
-				case "<<":
-					kind = Shl
-				case "<<=":
-					kind = ShlAssign
-				case "<=":
-					kind = LessEqual
-				case "=":
-					kind = Assign
-				case "==":
-					kind = Equal
-				case ">":
-					kind = Greater
-				case ">=":
-					kind = GreaterEqual
-				case ">>":
-					kind = Shr
-				case ">>=":
-					kind = ShrAssign
-				case "^":
-					kind = Xor
-				case "^=":
-					kind = XorAssign
-				case "|":
-					kind = Or
-				case "|=":
-					kind = OrAssign
-				case "||":
-					kind = LogicalOr
-				}
-
-				tokens = append(tokens, Token{Kind: kind, Position: position, Length: Length(i - position)})
 				continue
 			}
 
@@ -279,40 +62,3 @@ func Tokenize(buffer []byte) List {
 	tokens = append(tokens, Token{Kind: EOF, Position: i, Length: 0})
 	return tokens
 }
-
-func isIdentifier(c byte) bool {
-	return isLetter(c) || isDigit(c) || c == '_'
-}
-
-func isIdentifierStart(c byte) bool {
-	return isLetter(c) || c == '_'
-}
-
-func isLetter(c byte) bool {
-	return (c >= 'a' && c <= 'z') || (c >= 'A' && c <= 'Z')
-}
-
-func isDigit(c byte) bool {
-	return c >= '0' && c <= '9'
-}
-
-func isHexDigit(c byte) bool {
-	return (c >= '0' && c <= '9') || (c >= 'A' && c <= 'F')
-}
-
-func isBinaryDigit(c byte) bool {
-	return c == '0' || c == '1'
-}
-
-func isOctalDigit(c byte) bool {
-	return c >= '0' && c <= '7'
-}
-
-func isOperator(c byte) bool {
-	switch c {
-	case '=', ':', '.', '+', '-', '*', '/', '<', '>', '&', '|', '^', '%', '!':
-		return true
-	default:
-		return false
-	}
-}

src/token/dash.go

@@ -0,0 +1,25 @@
+package token
+
+// dash handles all tokens starting with '-'.
+func dash(tokens List, buffer []byte, i Position) (List, Position) {
+	if len(tokens) == 0 || tokens[len(tokens)-1].IsOperator() || tokens[len(tokens)-1].IsExpressionStart() || tokens[len(tokens)-1].IsKeyword() {
+		tokens = append(tokens, Token{Kind: Negate, Position: i, Length: 1})
+	} else {
+		if i+1 < Position(len(buffer)) {
+			switch buffer[i+1] {
+			case '=':
+				tokens = append(tokens, Token{Kind: SubAssign, Position: i, Length: 2})
+				i++
+			case '>':
+				tokens = append(tokens, Token{Kind: ReturnType, Position: i, Length: 2})
+				i++
+			default:
+				tokens = append(tokens, Token{Kind: Sub, Position: i, Length: 1})
+			}
+		} else {
+			tokens = append(tokens, Token{Kind: Sub, Position: i, Length: 1})
+		}
+	}
+
+	return tokens, i
+}

src/token/digit.go

@@ -0,0 +1,38 @@
+package token
+
+// digit handles all tokens that qualify as a digit.
+func digit(tokens List, buffer []byte, i Position) (List, Position) {
+	position := i
+	i++
+
+	for i < Position(len(buffer)) && isDigit(buffer[i]) {
+		i++
+	}
+
+	last := len(tokens) - 1
+
+	if len(tokens) > 0 && tokens[last].Kind == Negate {
+		tokens[last].Kind = Number
+		tokens[last].Length = Length(i-position) + 1
+	} else {
+		tokens = append(tokens, Token{Kind: Number, Position: position, Length: Length(i - position)})
+	}
+
+	return tokens, i
+}
+
+func isDigit(c byte) bool {
+	return c >= '0' && c <= '9'
+}
+
+func isHexDigit(c byte) bool {
+	return (c >= '0' && c <= '9') || (c >= 'A' && c <= 'F')
+}
+
+func isBinaryDigit(c byte) bool {
+	return c == '0' || c == '1'
+}
+
+func isOctalDigit(c byte) bool {
+	return c >= '0' && c <= '7'
+}

src/token/identifier.go

@@ -0,0 +1,46 @@
+package token
+
+// identifier handles all tokens that qualify as an identifier.
+func identifier(tokens List, buffer []byte, i Position) (List, Position) {
+	position := i
+	i++
+
+	for i < Position(len(buffer)) && isIdentifier(buffer[i]) {
+		i++
+	}
+
+	identifier := buffer[position:i]
+	kind := Identifier
+
+	switch string(identifier) {
+	case "assert":
+		kind = Assert
+	case "if":
+		kind = If
+	case "else":
+		kind = Else
+	case "import":
+		kind = Import
+	case "loop":
+		kind = Loop
+	case "return":
+		kind = Return
+	case "switch":
+		kind = Switch
+	}
+
+	tokens = append(tokens, Token{Kind: kind, Position: position, Length: Length(len(identifier))})
+	return tokens, i
+}
+
+func isIdentifier(c byte) bool {
+	return isLetter(c) || isDigit(c) || c == '_'
+}
+
+func isIdentifierStart(c byte) bool {
+	return isLetter(c) || c == '_'
+}
+
+func isLetter(c byte) bool {
+	return (c >= 'a' && c <= 'z') || (c >= 'A' && c <= 'Z')
+}

src/token/operator.go

@@ -0,0 +1,84 @@
+package token
+
+// operator handles all tokens that qualify as an operator.
+func operator(tokens List, buffer []byte, i Position) (List, Position) {
+	position := i
+	i++
+
+	for i < Position(len(buffer)) && isOperator(buffer[i]) {
+		i++
+	}
+
+	kind := Invalid
+
+	switch string(buffer[position:i]) {
+	case "!":
+		kind = Not
+	case "!=":
+		kind = NotEqual
+	case "%":
+		kind = Mod
+	case "%=":
+		kind = ModAssign
+	case "&":
+		kind = And
+	case "&&":
+		kind = LogicalAnd
+	case "&=":
+		kind = AndAssign
+	case "*":
+		kind = Mul
+	case "*=":
+		kind = MulAssign
+	case "+":
+		kind = Add
+	case "+=":
+		kind = AddAssign
+	case ".":
+		kind = Period
+	case ":=":
+		kind = Define
+	case "<":
+		kind = Less
+	case "<<":
+		kind = Shl
+	case "<<=":
+		kind = ShlAssign
+	case "<=":
+		kind = LessEqual
+	case "=":
+		kind = Assign
+	case "==":
+		kind = Equal
+	case ">":
+		kind = Greater
+	case ">=":
+		kind = GreaterEqual
+	case ">>":
+		kind = Shr
+	case ">>=":
+		kind = ShrAssign
+	case "^":
+		kind = Xor
+	case "^=":
+		kind = XorAssign
+	case "|":
+		kind = Or
+	case "|=":
+		kind = OrAssign
+	case "||":
+		kind = LogicalOr
+	}
+
+	tokens = append(tokens, Token{Kind: kind, Position: position, Length: Length(i - position)})
+	return tokens, i
+}
+
+func isOperator(c byte) bool {
+	switch c {
+	case '=', ':', '.', '+', '-', '*', '/', '<', '>', '&', '|', '^', '%', '!':
+		return true
+	default:
+		return false
+	}
+}

src/token/quote.go

@@ -0,0 +1,28 @@
+package token
+
+// quote handles all tokens starting with a single or double quote.
+func quote(tokens List, buffer []byte, i Position) (List, Position) {
+	limiter := buffer[i]
+	start := i
+	end := Position(len(buffer))
+	i++
+
+	for i < Position(len(buffer)) {
+		if buffer[i] == limiter && (buffer[i-1] != '\\' || buffer[i-2] == '\\') {
+			end = i + 1
+			i++
+			break
+		}
+
+		i++
+	}
+
+	kind := String
+
+	if limiter == '\'' {
+		kind = Rune
+	}
+
+	tokens = append(tokens, Token{Kind: kind, Position: start, Length: Length(end - start)})
+	return tokens, i
+}

src/token/slash.go

@@ -0,0 +1,34 @@
+package token
+
+// slash handles all tokens starting with '/'.
+func slash(tokens List, buffer []byte, i Position) (List, Position) {
+	if i+1 < Position(len(buffer)) && buffer[i+1] == '/' {
+		position := i
+
+		for i < Position(len(buffer)) && buffer[i] != '\n' {
+			i++
+		}
+
+		tokens = append(tokens, Token{Kind: Comment, Position: position, Length: Length(i - position)})
+	} else {
+		position := i
+		i++
+
+		for i < Position(len(buffer)) && isOperator(buffer[i]) {
+			i++
+		}
+
+		kind := Invalid
+
+		switch string(buffer[position:i]) {
+		case "/":
+			kind = Div
+		case "/=":
+			kind = DivAssign
+		}
+
+		tokens = append(tokens, Token{Kind: kind, Position: position, Length: Length(i - position)})
+	}
+
+	return tokens, i
+}

src/token/zero.go

@@ -0,0 +1,35 @@
+package token
+
+// zero handles all tokens starting with a '0'.
+func zero(tokens List, buffer []byte, i Position) (List, Position) {
+	position := i
+	i++
+
+	if i >= Position(len(buffer)) {
+		tokens = append(tokens, Token{Kind: Number, Position: position, Length: 1})
+		return tokens, i
+	}
+
+	filter := isDigit
+
+	switch buffer[i] {
+	case 'x':
+		i++
+		filter = isHexDigit
+
+	case 'b':
+		i++
+		filter = isBinaryDigit
+
+	case 'o':
+		i++
+		filter = isOctalDigit
+	}
+
+	for i < Position(len(buffer)) && filter(buffer[i]) {
+		i++
+	}
+
+	tokens = append(tokens, Token{Kind: Number, Position: position, Length: Length(i - position)})
+	return tokens, i
+}