golox/parser/parser.go

/* Description: This file contains the recursivde descent parser implementation.
 * The parser is responsible for parsing the tokens generated by the scanner.
 *
 * The grammar is as follows:
 * expression     → equality ;
 * equality       → comparison ( ( "!=" | "==" ) comparison )* ;
 * comparison     → term ( ( ">" | ">=" | "<" | "<=" ) term )* ;
 * term           → factor ( ( "-" | "+" ) factor )* ;
 * factor         → unary ( ( "/" | "*" ) unary )* ;
 * unary          → ( "!" | "-" ) unary
 *               | primary ;
 * primary        → NUMBER | STRING | "true" | "false" | "nil"
 *               | "(" expression ")" ;
 */
package parser

import (
	"golox/errors"
	"golox/token"
)

// Parser is a recursive descent parser.
type Parser struct {
	tokens    []token.Token
	current   int
	errLogger errors.Logger
}

// New creates a new Parser.
func New(tokens []token.Token, el errors.Logger) *Parser {
	return &Parser{tokens: tokens, current: 0, errLogger: el}
}

// Parse parses the tokens and returns the AST.
func (p *Parser) Parse() Expr {
	return p.expression()
}

// expression → equality ;
func (p *Parser) expression() Expr {
	return p.equality()
}

// equality → comparison ( ( "!=" | "==" ) comparison )* ;
func (p *Parser) equality() Expr {
	expr := p.comparison()

	for p.match(token.BANG_EQUAL, token.EQUAL_EQUAL) {
		operator := p.previous()
		right := p.comparison()
		expr = &BinaryExpr{Left: expr, Operator: operator, Right: right}
	}

	return expr
}

// comparison → term ( ( ">" | ">=" | "<" | "<=" ) term )* ;
func (p *Parser) comparison() Expr {
	expr := p.term()

	for p.match(token.GREATER, token.GREATER_EQUAL, token.LESS, token.LESS_EQUAL) {
		operator := p.previous()
		right := p.term()
		expr = &BinaryExpr{Left: expr, Operator: operator, Right: right}
	}

	return expr
}

// term → factor ( ( "-" | "+" ) factor )* ;
func (p *Parser) term() Expr {
	expr := p.factor()

	for p.match(token.MINUS, token.PLUS) {
		operator := p.previous()
		right := p.factor()
		expr = &BinaryExpr{Left: expr, Operator: operator, Right: right}
	}

	return expr
}

// factor → unary ( ( "/" | "*" ) unary )* ;
func (p *Parser) factor() Expr {
	expr := p.unary()

	for p.match(token.SLASH, token.STAR) {
		operator := p.previous()
		right := p.unary()
		expr = &BinaryExpr{Left: expr, Operator: operator, Right: right}
	}

	return expr
}

// unary → ( "!" | "-" ) unary | primary ;
func (p *Parser) unary() Expr {
	if p.match(token.BANG, token.MINUS) {
		operator := p.previous()
		right := p.unary()
		return &UnaryExpr{Operator: operator, Right: right}
	}

	return p.primary()
}

// primary → NUMBER | STRING | "true" | "false" | "nil" | "(" expression ")" ;
func (p *Parser) primary() Expr {
	switch {
	case p.match(token.FALSE):
		return &LiteralExpr{Value: false}
	case p.match(token.TRUE):
		return &LiteralExpr{Value: true}
	case p.match(token.NIL):
		return &LiteralExpr{Value: nil}
	case p.match(token.NUMBER, token.STRING):
		return &LiteralExpr{Value: p.previous().Literal}
	case p.match(token.LEFT_PAREN):
		expr := p.expression()
		err := p.consume(token.RIGHT_PAREN, "Expect ')' after expression.")
		if err != nil {
			return err
		}
		return &GroupingExpr{Expression: expr}
	}

	return p.newErrorExpr(p.peek(), "Expect expression.")
}

// match checks if the current token is any of the given types.
func (p *Parser) match(types ...token.TokenType) bool {
	for _, t := range types {
		if p.check(t) {
			p.advance()
			return true
		}
	}

	return false
}

// consume consumes the current token if it is of the given type.
func (p *Parser) consume(tt token.TokenType, message string) *ErrorExpr {
	if p.check(tt) {
		p.advance()
		return nil
	}

	return p.newErrorExpr(p.peek(), message)
}

// check checks if the current token is of the given type.
func (p *Parser) check(tt token.TokenType) bool {
	if p.isAtEnd() {
		return false
	}

	return p.peek().Type == tt
}

// advance advances the current token and returns the previous token.
func (p *Parser) advance() token.Token {
	if !p.isAtEnd() {
		p.current++
	}

	return p.previous()
}

// isAtEnd checks if the parser has reached the end of the tokens.
func (p *Parser) isAtEnd() bool {
	return p.peek().Type == token.EOF
}

// peek returns the current token.
func (p *Parser) peek() token.Token {
	return p.tokens[p.current]
}

// previous returns the previous token.
func (p *Parser) previous() token.Token {
	return p.tokens[p.current-1]
}

// newErrorExpr creates a new ErrorExpr and reports the error.
func (p *Parser) newErrorExpr(t token.Token, message string) *ErrorExpr {
	p.errLogger.ErrorAtToken(t, message)
	return &ErrorExpr{Value: message}
}

// synchronize synchronizes the parser after an error.
func (p *Parser) synchronize() {
	p.advance()

	for !p.isAtEnd() {
		if p.previous().Type == token.SEMICOLON {
			return
		}

		switch p.peek().Type {
		case token.CLASS, token.FUN, token.VAR, token.FOR, token.IF, token.WHILE, token.PRINT, token.RETURN:
			return
		}

		p.advance()
	}
}