def repl():
line = input(">> ")
lexer = Lexer(line)
while (tok :=
lexer.next_token()).type not in {TokenType.ILLEGAL, TokenType.EOF}:
print(tok)
def test_next_token_on_source_code():
input = """
let five = 5;
let ten = 10;
let add = fn(x, y){
x + y;
};
let result = add(five, ten);
"""
expected = [
(TokenType.LET, "let"),
(TokenType.IDENTIFIER, "five"),
(TokenType.ASSIGN, "="),
(TokenType.INT, "5"),
(TokenType.SEMICOLON, ";"),
(TokenType.LET, "let"),
(TokenType.IDENTIFIER, "ten"),
(TokenType.ASSIGN, "="),
(TokenType.INT, "10"),
(TokenType.SEMICOLON, ";"),
(TokenType.LET, "let"),
(TokenType.IDENTIFIER, "add"),
(TokenType.ASSIGN, "="),
(TokenType.FUNCTION, "fn"),
(TokenType.LPAREN, "("),
(TokenType.IDENTIFIER, "x"),
(TokenType.COMMA, ","),
(TokenType.IDENTIFIER, "y"),
(TokenType.RPAREN, ")"),
(TokenType.LBRACE, "{"),
(TokenType.IDENTIFIER, "x"),
(TokenType.PLUS, "+"),
(TokenType.IDENTIFIER, "y"),
(TokenType.SEMICOLON, ";"),
(TokenType.RBRACE, "}"),
(TokenType.SEMICOLON, ";"),
(TokenType.LET, "let"),
(TokenType.IDENTIFIER, "result"),
(TokenType.ASSIGN, "="),
(TokenType.IDENTIFIER, "add"),
(TokenType.LPAREN, "("),
(TokenType.IDENTIFIER, "five"),
(TokenType.COMMA, ","),
(TokenType.IDENTIFIER, "ten"),
(TokenType.RPAREN, ")"),
(TokenType.SEMICOLON, ";"),
(TokenType.EOF, ""),
]
lexer = Lexer(input)
for expected_token, expected_literal in expected:
token = lexer.next_token()
assert token.type == expected_token
assert token.literal == expected_literal
def testReturnStatement(self):
inputs = '''return 5;
return 10;
return 838383; :'''
lex = Lexer(inputs)
curtoken = lex.nextToken()
peektoken = lex.nextToken()
pas = Parser(lex, curtoken, peektoken)
program = pas.parseProgram()
self.assertEquals(3, len(program.statements))
def testErrors(self):
inputs = '''let five 5;
let = 10;
let 838383; :'''
lex = Lexer(inputs)
curtoken = lex.nextToken()
peektoken = lex.nextToken()
pas = Parser(lex, curtoken, peektoken)
program = pas.parseProgram()
self.assertEquals(3, len(pas.errors))
print 'Number of error is {}'.format(len(pas.errors))
def testLetStatement(self):
inputs = '''let five = 5;
let ten = 10;
let foobar = 838383; :'''
lex = Lexer(inputs)
curtoken = lex.nextToken()
peektoken = lex.nextToken()
pas = Parser(lex, curtoken, peektoken)
program = pas.parseProgram()
self.assertNotEqual(program, None)
self.assertEquals(3, len(program.statements))
for i in range(len(program.statements) - 1):
self.assertEquals(program.statements[i].tokenLiteral(), 'let')
def test_if_expression():
source = "if (x < y) { x }"
l = Lexer(source)
p = Parser(l)
program = p.parse_program()
assert p.errors == []
assert len(program.statements) == 1
stmt = program.statements[0]
assert type(stmt) == ast.ExpressionStatement
assert type(stmt.expression) == ast.IfExpression
assert stmt.expression.condition.left.value == "x"
assert stmt.expression.condition.operator == "<"
assert stmt.expression.condition.right.value == "y"
# TODO Finish defining me please
assert len(stmt.expression.consequence.statements) == 1
def test_nextToken(self):
inputs = '''let five = 5;
let ten = 10;
let add = fn(x, y) {
x + y;
};
let result = add(five, ten);
!-/*5;
5 < 10 > 5;
10 == 10;
10 != 9;'''
l = Lexer(inputs)
while l.readPosition <= len(inputs):
tk = l.nextToken()
print 'Literal is ' + tk.literal
self.assertEquals(tk.tokenType, testToken[tk.literal])
def test_is_whitespace(self):
assert Lexer.is_whitespace(' ')
assert Lexer.is_whitespace('\t')
assert Lexer.is_whitespace('\n')
assert Lexer.is_whitespace('\r')
assert not Lexer.is_whitespace('4')
assert not Lexer.is_whitespace('f')
assert not Lexer.is_whitespace('$')
assert not Lexer.is_whitespace('Z')
def test_parsing_let_statements():
input = """
let x = 5;
let y = 10;
let foobar = 838383;
"""
lexer = Lexer(input)
parser = Parser(lexer)
def start():
env = object.Environment()
while True:
line = input(prompt)
l = Lexer(line)
p = Parser(l)
program = p.parse_program()
val = evaluator.eval(program, env)
print(val)
def test_string_literal_expression():
source = '"hello world";'
l = Lexer(source)
p = Parser(l)
program = p.parse_program()
assert p.errors == []
stmt = program.statements[0]
assert isinstance(stmt, ast.ExpressionStatement)
assert isinstance(stmt.expression, ast.StringLiteral)
assert stmt.expression.value == "hello world"
def test_function_parameter_parsing(source, expected):
l = Lexer(source)
p = Parser(l)
program = p.parse_program()
assert p.errors == []
stmt = program.statements[0]
assert type(stmt) == ast.ExpressionStatement
assert type(stmt.expression) == ast.FunctionLiteral
func: ast.FunctionLiteral = stmt.expression
assert len(func.params) == len(expected)
def test_parsing_empty_hash_literal():
source = "{}"
l = Lexer(source)
p = Parser(l)
program = p.parse_program()
assert p.errors == []
stmt = program.statements[0]
assert isinstance(stmt, ast.ExpressionStatement)
hash_literal = stmt.expression
assert isinstance(hash_literal, ast.HashLiteral)
assert len(hash_literal.pairs) == 0
def test_is_digit(self):
assert Lexer.is_digit('9')
assert not Lexer.is_digit(' ')
assert not Lexer.is_digit('f')
assert not Lexer.is_digit('U')
assert not Lexer.is_digit('$')
assert not Lexer.is_digit('_')
def test_call_expression_parsing():
source = "add(1, 2 * 3, 4 + 5);"
l = Lexer(source)
p = Parser(l)
program = p.parse_program()
assert p.errors == []
assert len(program.statements) == 1
stmt = program.statements[0]
assert isinstance(stmt, ast.ExpressionStatement)
assert isinstance(stmt.expression, ast.CallExpression)
assert stmt.expression.function.value == "add"
assert len(stmt.expression.arguments) == 3
assert stmt.expression == ast.CallExpression(
token=Token(tok_type="(", literal="("),
function=ast.Identifier(token=Token(tok_type="IDENT", literal="add"),
value="add"),
arguments=[
ast.IntegerLiteral(token=Token(tok_type="INT", literal="1"),
value=1),
ast.InfixExpression(
token=Token(tok_type="*", literal="*"),
left=ast.IntegerLiteral(token=Token(tok_type="INT",
literal="2"),
value=2),
operator="*",
right=ast.IntegerLiteral(token=Token(tok_type="INT",
literal="3"),
value=3),
),
ast.InfixExpression(
token=Token(tok_type="+", literal="+"),
left=ast.IntegerLiteral(token=Token(tok_type="INT",
literal="4"),
value=4),
operator="+",
right=ast.IntegerLiteral(token=Token(tok_type="INT",
literal="5"),
value=5),
),
],
)
assert str(stmt.expression) == "add(1, (2 * 3), (4 + 5))"
def test_next_token():
input = "=+(){},;"
expected = [
(TokenType.ASSIGN, "="),
(TokenType.PLUS, "+"),
(TokenType.LPAREN, "("),
(TokenType.RPAREN, ")"),
(TokenType.LBRACE, "{"),
(TokenType.RBRACE, "}"),
(TokenType.COMMA, ","),
(TokenType.SEMICOLON, ";"),
(TokenType.EOF, ""),
]
lexer = Lexer(input)
for expected_token, expected_literal in expected:
token = lexer.next_token()
assert token.type == expected_token
assert token.literal == expected_literal
def test_parsing_array_literal():
source = "[1, 2 * 2, 3 + 3];"
l = Lexer(source)
p = Parser(l)
program = p.parse_program()
assert p.errors == []
stmt = program.statements[0]
assert isinstance(stmt, ast.ExpressionStatement)
array = stmt.expression
assert isinstance(array, ast.ArrayLiteral)
assert len(array.elements) == 3
literal = array.elements[0]
assert literal.value == 1
assert literal.token_literal() == "1"
def test_parsing_index_expressions():
source = "myArray[1 + 1]"
l = Lexer(source)
p = Parser(l)
program = p.parse_program()
assert p.errors == []
stmt = program.statements[0]
assert isinstance(stmt, ast.ExpressionStatement)
expression = stmt.expression
assert isinstance(expression, ast.IndexExpression)
left = expression.left
assert isinstance(left, ast.Identifier)
assert left.value == "myArray"
assert left.token_literal() == "myArray"
check_infix_expression(expression.index, 1, "+", "1")
def test_let_statements(source, expected_identifier, expected_value):
l = Lexer(source)
p = Parser(l)
program = p.parse_program()
assert p.errors == []
assert program is not None
assert len(program.statements) == 1
stmt = program.statements[0]
check_let_statement(stmt, expected_identifier)
assert stmt.value.value == expected_value
def test_prefix_expressions(source, operator, expected):
l = Lexer(source)
p = Parser(l)
program = p.parse_program()
assert p.errors == []
assert len(program.statements) == 1
stmt = program.statements[0]
assert type(stmt) == ast.ExpressionStatement
assert type(stmt.expression) == ast.PrefixExpression
assert stmt.expression.operator == operator
assert stmt.expression.right.value == expected
def test_parsing_hash_literal():
source = '{"one": 1, "two": 2, "three": 3}'
l = Lexer(source)
p = Parser(l)
program = p.parse_program()
assert p.errors == []
stmt = program.statements[0]
assert isinstance(stmt, ast.ExpressionStatement)
hash_literal = stmt.expression
assert isinstance(hash_literal, ast.HashLiteral)
assert len(hash_literal.pairs) == 3
expected = {"one": 1, "two": 2, "three": 3}
for key, value in hash_literal.pairs.items():
assert isinstance(key, ast.StringLiteral)
check_integer_literal(value, expected[str(key)])
def test_function_literal_parsing():
source = "fn(x, y) { x + y; }"
l = Lexer(source)
p = Parser(l)
program = p.parse_program()
assert p.errors == []
assert len(program.statements) == 1
stmt = program.statements[0]
assert type(stmt) == ast.ExpressionStatement
assert type(stmt.expression) == ast.FunctionLiteral
assert [str(s) for s in stmt.expression.params] == ["x", "y"]
def test_return_statements():
source = """
return 5;
return 10;
return 993322;
"""
l = Lexer(source)
p = Parser(l)
program = p.parse_program()
assert p.errors == []
assert program is not None
assert len(program.statements) == 3
for stmt in program.statements:
assert type(stmt) == ast.ReturnStatement
assert stmt.token_literal() == "return"
def test_boolean_expression():
source = "true;"
l = Lexer(source)
p = Parser(l)
program = p.parse_program()
assert p.errors == []
assert len(program.statements) == 1
stmt = program.statements[0]
assert type(stmt) == ast.ExpressionStatement
assert type(stmt.expression) == ast.Boolean
ident = stmt.expression
assert ident.value == True
assert ident.token_literal() == "true"
def test_integer_literals():
source = "5;"
l = Lexer(source)
p = Parser(l)
program = p.parse_program()
assert p.errors == []
assert len(program.statements) == 1
stmt = program.statements[0]
assert type(stmt) == ast.ExpressionStatement
assert type(stmt.expression) == ast.IntegerLiteral
literal = stmt.expression
assert literal.value == 5
assert literal.token_literal() == "5"
def test_identifier_expression():
source = "foobar;"
l = Lexer(source)
p = Parser(l)
program = p.parse_program()
assert p.errors == []
assert len(program.statements) == 1
stmt = program.statements[0]
assert type(stmt) == ast.ExpressionStatement
assert type(stmt.expression) == ast.Identifier
ident = stmt.expression
assert ident.value == "foobar"
assert ident.token_literal() == "foobar"
def test_parsing_hash_literal_with_expressions():
source = '{"one": 0 + 1, "two": 10 - 8, "three": 15 / 5}'
l = Lexer(source)
p = Parser(l)
program = p.parse_program()
assert p.errors == []
stmt = program.statements[0]
assert isinstance(stmt, ast.ExpressionStatement)
hash_literal = stmt.expression
assert isinstance(hash_literal, ast.HashLiteral)
assert len(hash_literal.pairs) == 3
tests = {
"one": lambda e: check_infix_expression(e, 0, "+", 1),
"two": lambda e: check_infix_expression(e, 10, "-", 8),
"three": lambda e: check_infix_expression(e, 15, "/", 5),
}
for key, value in hash_literal.pairs.items():
assert isinstance(key, ast.StringLiteral)
test_fn = tests[str(key)]
test_fn(value)
def main():
hist_file = Path("~/.monkeyhist").expanduser()
# Create history file if it doesn't exist already
if not hist_file.exists():
hist_file.touch()
session = PromptSession(history=FileHistory(hist_file),
completer=monkey_completer,
style=style)
env = {}
while True:
try:
scanned = session.prompt(">> ",
lexer=PygmentsLexer(JavascriptLexer))
except KeyboardInterrupt:
continue
except EOFError:
break
else:
lexer = Lexer(scanned)
parser = Parser(lexer)
program = parser.parse_program()
if parser.errors:
for error in parser.errors:
print(error)
evaluated = Eval(program, env)
session.completer = WordCompleter(
list(set(token.keywords) | set(env)))
if evaluated is not None:
print(evaluated.inspect())
print("Farewell!")
def test_next_token_on_source_code_2():
input = """
!-/*5;
5 < 10 > 5;
if (5 < 10) {
return true;
} else {
return false;
}
10 == 10;
10 != 9;
"""
expected = [
(TokenType.BANG, "!"),
(TokenType.MINUS, "-"),
(TokenType.SLASH, "/"),
(TokenType.ASTERIK, "*"),
(TokenType.INT, "5"),
(TokenType.SEMICOLON, ";"),
(TokenType.INT, "5"),
(TokenType.LT, "<"),
(TokenType.INT, "10"),
(TokenType.GT, ">"),
(TokenType.INT, "5"),
(TokenType.SEMICOLON, ";"),
(TokenType.IF, "if"),
(TokenType.LPAREN, "("),
(TokenType.INT, "5"),
(TokenType.LT, "<"),
(TokenType.INT, "10"),
(TokenType.RPAREN, ")"),
(TokenType.LBRACE, "{"),
(TokenType.RETURN, "return"),
(TokenType.TRUE, "true"),
(TokenType.SEMICOLON, ";"),
(TokenType.RBRACE, "}"),
(TokenType.ELSE, "else"),
(TokenType.LBRACE, "{"),
(TokenType.RETURN, "return"),
(TokenType.FALSE, "false"),
(TokenType.SEMICOLON, ";"),
(TokenType.RBRACE, "}"),
(TokenType.INT, "10"),
(TokenType.EQ, "=="),
(TokenType.INT, "10"),
(TokenType.SEMICOLON, ";"),
(TokenType.INT, "10"),
(TokenType.NOT_EQ, "!="),
(TokenType.INT, "9"),
(TokenType.SEMICOLON, ";"),
(TokenType.EOF, ""),
]
lexer = Lexer(input)
for expected_token, expected_literal in expected:
token = lexer.next_token()
assert token.type == expected_token
assert token.literal == expected_literal
def test_next_token_more(self):
input = '''
let five = 5;
let ten = 10;
let add = fn(x, y) {
x + y;
};
let result = add(five, ten);
!-/*5
5 < 10 > 5
if (5 < 10) {
return true;
} else {
return false;
}
10 == 10;
10 != 9;
'''
tests = [
ExpectedToken(Token.LET, "let"),
ExpectedToken(Token.IDENTIFIER, "five"),
ExpectedToken(Token.ASSIGN, "="),
ExpectedToken(Token.INT, "5"),
ExpectedToken(Token.SEMICOLON, ";"),
ExpectedToken(Token.LET, "let"),
ExpectedToken(Token.IDENTIFIER, "ten"),
ExpectedToken(Token.ASSIGN, "="),
ExpectedToken(Token.INT, "10"),
ExpectedToken(Token.SEMICOLON, ";"),
ExpectedToken(Token.LET, "let"),
ExpectedToken(Token.IDENTIFIER, "add"),
ExpectedToken(Token.ASSIGN, "="),
ExpectedToken(Token.FUNCTION, "fn"),
ExpectedToken(Token.LPAREN, "("),
ExpectedToken(Token.IDENTIFIER, "x"),
ExpectedToken(Token.COMMA, ","),
ExpectedToken(Token.IDENTIFIER, "y"),
ExpectedToken(Token.RPAREN, ")"),
ExpectedToken(Token.LBRACE, "{"),
ExpectedToken(Token.IDENTIFIER, "x"),
ExpectedToken(Token.PLUS, "+"),
ExpectedToken(Token.IDENTIFIER, "y"),
ExpectedToken(Token.SEMICOLON, ";"),
ExpectedToken(Token.RBRACE, "}"),
ExpectedToken(Token.SEMICOLON, ";"),
ExpectedToken(Token.LET, "let"),
ExpectedToken(Token.IDENTIFIER, "result"),
ExpectedToken(Token.ASSIGN, "="),
ExpectedToken(Token.IDENTIFIER, "add"),
ExpectedToken(Token.LPAREN, "("),
ExpectedToken(Token.IDENTIFIER, "five"),
ExpectedToken(Token.COMMA, ","),
ExpectedToken(Token.IDENTIFIER, "ten"),
ExpectedToken(Token.RPAREN, ")"),
ExpectedToken(Token.SEMICOLON, ";"),
ExpectedToken(Token.BANG, "!"),
ExpectedToken(Token.MINUS, "-"),
ExpectedToken(Token.SLASH, "/"),
ExpectedToken(Token.ASTERISK, "*"),
ExpectedToken(Token.INT, "5"),
ExpectedToken(Token.INT, "5"),
ExpectedToken(Token.LT, "<"),
ExpectedToken(Token.INT, "10"),
ExpectedToken(Token.GT, ">"),
ExpectedToken(Token.INT, "5"),
ExpectedToken(Token.IF, "if"),
ExpectedToken(Token.LPAREN, "("),
ExpectedToken(Token.INT, "5"),
ExpectedToken(Token.LT, "<"),
ExpectedToken(Token.INT, "10"),
ExpectedToken(Token.RPAREN, ")"),
ExpectedToken(Token.LBRACE, "{"),
ExpectedToken(Token.RETURN, "return"),
ExpectedToken(Token.TRUE, "true"),
ExpectedToken(Token.SEMICOLON, ";"),
ExpectedToken(Token.RBRACE, "}"),
ExpectedToken(Token.ELSE, "else"),
ExpectedToken(Token.LBRACE, "{"),
ExpectedToken(Token.RETURN, "return"),
ExpectedToken(Token.FALSE, "false"),
ExpectedToken(Token.SEMICOLON, ";"),
ExpectedToken(Token.RBRACE, "}"),
ExpectedToken(Token.INT, "10"),
ExpectedToken(Token.EQ, "=="),
ExpectedToken(Token.INT, "10"),
ExpectedToken(Token.SEMICOLON, ";"),
ExpectedToken(Token.INT, "10"),
ExpectedToken(Token.NOT_EQ, "!="),
ExpectedToken(Token.INT, "9"),
ExpectedToken(Token.SEMICOLON, ";"),
ExpectedToken(Token.EOF, "")
]
lexer = Lexer(input)
for test in tests:
token = lexer.next_token()
assert token.type == test.expected_type
assert token.literal == test.expected_literal
