Chapter 2

Task 4.6

main.go

@@ -7,6 +7,8 @@ import (
 	"strings"
 )
 
+var hasError = false
+
 func main() {
 	if len(os.Args) > 2 {
 		fmt.Println("Usage: lox [script]")
@@ -25,13 +27,15 @@ func parseFile(path string) {
 		os.Exit(1)
 	}
 
-	scanner := NewScanner()
+	run(string(file))
-	scanner.Parse(string(file))
+
+	if hasError {
+		os.Exit(1)
+	}
 }
 
 func parsePrompt() {
 	reader := bufio.NewReader(os.Stdin)
-	scanner := NewScanner()
 
 	for {
 		fmt.Print("> ")
@@ -42,6 +46,27 @@ func parsePrompt() {
 			return
 		}
 
-		scanner.Parse(text)
+		run(text)
+
+		hasError = false
 	}
 }
+
+func run(source string) {
+	scanner := NewScanner(source)
+	tokens := scanner.ScanTokens()
+
+	for _, token := range tokens {
+		fmt.Println(token.String())
+	}
+}
+
+// fuck is just the error function, but collides with a builtin so whatever
+func fuck(line int, msg string) {
+	report(line, "", msg)
+}
+
+func report(line int, where string, msg string) {
+	fmt.Printf("[line %d] Error %s: %s\n", line, where, msg)
+	hasError = true
+}

scanner.go

@@ -1,21 +1,157 @@
 package main
 
-import (
-	"fmt"
-)
-
 type Scanner struct {
-	tokens []Token
+	source  string
+	tokens  []Token
+	start   int
+	current int
+	line    int
 }
 
-func (s *Scanner) GetTokens() []Token {
+func (s *Scanner) ScanTokens() []Token {
+	for !s.isAtEnd() {
+		s.start = s.current
+		s.scanToken()
+	}
+
+	s.tokens = append(s.tokens, Token{
+		Type:    EOF,
+		Lexeme:  "",
+		Literal: "",
+		Line:    s.line,
+	})
 	return s.tokens
 }
 
-func (s *Scanner) Parse(source string) {
+func (s *Scanner) scanToken() {
-	fmt.Println(source)
+	c := s.advance()
+	switch c {
+	case "(":
+		s.addToken(LeftBrace)
+		break
+	case ")":
+		s.addToken(RightBrace)
+		break
+	case "{":
+		s.addToken(LeftBrace)
+		break
+	case "}":
+		s.addToken(RightBrace)
+		break
+	case ",":
+		s.addToken(Comma)
+		break
+	case ".":
+		s.addToken(Dot)
+		break
+	case "-":
+		s.addToken(Minus)
+		break
+	case "+":
+		s.addToken(Plus)
+		break
+	case ";":
+		s.addToken(Semicolon)
+		break
+	case "*":
+		s.addToken(Star)
+		break
+	case "!":
+		if s.match("=") {
+			s.addToken(BangEqual)
+		} else {
+			s.addToken(Bang)
+		}
+		break
+	case "=":
+		if s.match("=") {
+			s.addToken(EqualEqual)
+		} else {
+			s.addToken(Equal)
+		}
+		break
+	case "<":
+		if s.match("=") {
+			s.addToken(LessEqual)
+		} else {
+			s.addToken(Less)
+		}
+		break
+	case ">":
+		if s.match("=") {
+			s.addToken(GreaterEqual)
+		} else {
+			s.addToken(Greater)
+		}
+		break
+	case "/":
+		if s.match("/") {
+			for s.peek() != "\n" && !s.isAtEnd() {
+				s.advance()
+			}
+		} else {
+			s.addToken(Slash)
+		}
+		break
+	case " ":
+	case "\t":
+	case "\r":
+		// ignore whitespace
+		break
+	case "\n":
+		s.line += 1
+		break
+	default:
+		fuck(s.line, "Unexpected Char.")
+		break
+	}
 }
 
-func NewScanner() *Scanner {
+func (s *Scanner) advance() string {
-	return &Scanner{}
+	c := string(s.source[s.current])
+	s.current += 1
+	return c
+}
+
+func (s *Scanner) peek() string {
+	if s.isAtEnd() {
+		return ""
+	}
+	return string(s.source[s.current])
+}
+
+func (s *Scanner) match(expected string) bool {
+	if s.isAtEnd() {
+		return false
+	}
+	if string(s.source[s.current]) != expected {
+		return false
+	}
+	s.current += 1
+	return true
+}
+
+func (s *Scanner) addToken(token int) {
+	s.tokens = append(s.tokens, Token{
+		Type:    token,
+		Lexeme:  s.source[s.start:s.current],
+		Literal: "",
+		Line:    s.line,
+	})
+	//fmt.Printf("[%d %d %d] ", s.line, s.current, s.start)
+	//fmt.Println(s.tokens[len(s.tokens)-1])
+}
+
+func (s *Scanner) isAtEnd() bool {
+	return s.current >= len(s.source)
+}
+
+func NewScanner(source string) *Scanner {
+	return &Scanner{
+		source:  source,
+		tokens:  []Token{},
+		start:   0,
+		current: 0,
+		line:    0,
+	}
 }

scripts/example.lox

@@ -1 +1,3 @@
-print "Hello, World!";
+// this is a comment
+(( )){} // grouping stuff
+!*+-/ =<> <= == // operators

scripts/helloworld.lox

@@ -0,0 +1 @@
+print "Hello, World!";

tokens.go

@@ -1,5 +1,66 @@
 package main
 
+import (
+	"fmt"
+)
+
+const (
+	// Single Char tokens
+	LeftParen = iota
+	RightParen
+	LeftBrace
+	RightBrace
+	Comma
+	Dot
+	Minus
+	Plus
+	Semicolon
+	Slash
+	Star
+
+	// One or Two Char tokens
+	Bang
+	BangEqual
+	Equal
+	EqualEqual
+	Greater
+	GreaterEqual
+	Less
+	LessEqual
+
+	// Literals
+	Identifier
+	String
+	Number
+
+	// Keywords
+	And
+	Class
+	Else
+	False
+	Fun
+	For
+	If
+	Nah
+	Or
+	Print
+	Return
+	Super
+	This
+	True
+	Var
+	While
+
+	EOF
+)
+
 type Token struct {
-	token string
+	Type    int
+	Lexeme  string
+	Literal string
+	Line    int
+}
+
+func (t Token) String() string {
+	return fmt.Sprintf("%d %s %s", t.Type, t.Lexeme, t.Literal)
 }