def compile_subroutine_call(self):
# eat identifier
if not lexical_elements.is_identifier(self.tokenizer.current_token):
raise CompilationError(
"Expected valid identifier in subroutine call but found " +
self.tokenizer.current_token)
self.eat(self.tokenizer.current_token)
if self.tokenizer.current_token == "(":
self.eat("(")
self.compile_expression_list()
self.eat(")")
elif self.tokenizer.current_token == ".":
self.eat(".")
# eat subroutine name
if not lexical_elements.is_identifier(
self.tokenizer.current_token):
raise CompilationError(
"Expected valid identifier in subroutine call but found " +
self.tokenizer.current_token)
self.eat(self.tokenizer.current_token)
self.eat("(")
self.compile_expression_list()
self.eat(")")
else:
raise CompilationError(
"Expected '(' or '.' in subroutine call but found " +
self.tokenizer.current_token)
def compile_subroutine_dec(self):
while self.tokenizer.current_token in [CONSTRUCTOR, FUNCTION, METHOD]:
self.openNonTerminal(SUBROUTINE_DEC)
# eat subroutine type: method, function or constructor
self.eat(self.tokenizer.current_token)
# eat subroutine return type
if self.tokenizer.current_token != VOID and not self.is_valid_type(
self.tokenizer.current_token):
raise CompilationError(
"Expected valid subroutine return type or void but was " +
self.tokenizer.current_token)
self.eat(self.tokenizer.current_token)
# eat subroutine name
if not lexical_elements.is_identifier(
self.tokenizer.current_token):
raise CompilationError(
"Expected valid subroutine name but was " +
self.tokenizer.current_token)
self.eat(self.tokenizer.current_token)
# compile subroutine parameter list (possibly empty)
self.compile_parameter_list()
# compile subroutine body
self.compile_subroutine_body()
self.closeNonTerminal(SUBROUTINE_DEC)
def compile_term(self):
self.openNonTerminal(TERM)
if lexical_elements.is_int_constant(self.tokenizer.current_token):
self.eat(self.tokenizer.current_token)
elif lexical_elements.is_string_constant(self.tokenizer.current_token):
self.eat(self.tokenizer.current_token)
elif self.is_keyword_constant(self.tokenizer.current_token):
self.eat(self.tokenizer.current_token)
elif self.tokenizer.current_token == "(":
self.eat("(")
self.compile_expression()
self.eat(")")
elif self.tokenizer.current_token == "-" or self.tokenizer.current_token == "~":
self.eat(self.tokenizer.current_token)
self.compile_term()
elif lexical_elements.is_identifier(self.tokenizer.current_token):
if self.tokenizer.peek() == "[":
self.eat(self.tokenizer.current_token)
self.eat("[")
self.compile_expression()
self.eat("]")
elif self.tokenizer.peek() == "(" or self.tokenizer.peek() == ".":
self.compile_subroutine_call()
else:
self.eat(self.tokenizer.current_token)
self.closeNonTerminal(TERM)
def compile_class(self):
self.tokenizer.advance()
self.openNonTerminal(CLASS)
# eat 'class' keyword
self.eat(CLASS)
# eat class name
if not lexical_elements.is_identifier(self.tokenizer.current_token):
raise CompilationError(
"Class name must be a valid identifier but was " +
self.tokenizer.current_token)
self.eat(self.tokenizer.current_token)
# eat opening brace
self.eat("{")
# compile class variable declarations
self.compile_class_var_dec()
# compile class subroutines
self.compile_subroutine_dec()
# eat closing brace
self.eat("}")
self.closeNonTerminal(CLASS)
def eat_var_sequence(self):
self.eat_var_dec()
while self.tokenizer.current_token == ",":
self.eat(",")
if not lexical_elements.is_identifier(
self.tokenizer.current_token):
raise CompilationError(
"Expected valid variable name but was " +
self.tokenizer.current_token)
self.eat(self.tokenizer.current_token) # eat varName
def token_type(self, token):
if lexical_elements.is_symbol(token):
return SYMBOL
if lexical_elements.is_string_constant(token):
return STRING_CONSTANT
if lexical_elements.is_keyword(token):
return KEYWORD
if lexical_elements.is_int_constant(token):
return INTEGER_CONSTANT
if lexical_elements.is_identifier(token):
return IDENTIFIER
def eat_var_dec(self):
# eat var type
if not self.is_valid_type(self.tokenizer.current_token):
raise CompilationError("Expected valid var type but found " +
self.tokenizer.current_token)
self.eat(self.tokenizer.current_token)
# eat var name
if not lexical_elements.is_identifier(self.tokenizer.current_token):
raise CompilationError("Expected valid var name but found " +
self.tokenizer.current_token)
self.eat(self.tokenizer.current_token)
def compile_let_statement(self):
self.openNonTerminal(LET_STATEMENT)
# eat 'let' keyword
self.eat(LET)
# eat variable name
if not lexical_elements.is_identifier(self.tokenizer.current_token):
raise CompilationError("Expected valid variable name but was " +
self.tokenizer.current_token)
self.eat(self.tokenizer.current_token)
# eat potential array indexing expression
if self.tokenizer.current_token == "[":
self.eat("[")
self.compile_expression()
self.eat("]")
self.eat("=")
self.compile_expression()
self.eat(";")
self.closeNonTerminal(LET_STATEMENT)
def compile_term(self):
if lexical_elements.is_int_constant(self.tokenizer.current_token):
self.compile_int_constant()
elif lexical_elements.is_string_constant(self.tokenizer.current_token):
self.compile_string_constant()
elif self.tokenizer.is_keyword_constant(self.tokenizer.current_token):
self.compile_keyword_constant()
elif self.tokenizer.current_token == "(":
self.eat("(")
self.compile_expression()
self.eat(")")
elif self.tokenizer.current_token == "-" or self.tokenizer.current_token == "~":
self.compile_unary()
elif lexical_elements.is_identifier(self.tokenizer.current_token):
if self.tokenizer.peek() == "[":
self.compile_array_expression()
elif self.tokenizer.peek() == "(" or self.tokenizer.peek() == ".":
self.compile_subroutine_call()
else:
var_name = self.tokenizer.current_token
self.eat(var_name)
self.vm_writer.write_push(
segment_map[self.st.kind_of(var_name)],
self.st.index_of(var_name))
def is_valid_term(self, token):
return (lexical_elements.is_identifier(token)
or lexical_elements.is_int_constant(token)
or lexical_elements.is_string_constant(token)
or self.is_keyword_constant(token) or token == "("
or token == "~" or token == "-")
def is_valid_type(self, token):
return token in [INT, CHAR, BOOLEAN
] or lexical_elements.is_identifier(token)
def test_identifier(self):
self.assertTrue(lexical_elements.is_identifier("A"))
self.assertTrue(lexical_elements.is_identifier("AAA"))
self.assertTrue(lexical_elements.is_identifier("a"))
self.assertTrue(lexical_elements.is_identifier("abc"))
self.assertTrue(lexical_elements.is_identifier("_AbC"))
self.assertTrue(lexical_elements.is_identifier("A_Bc"))
self.assertTrue(lexical_elements.is_identifier("A123"))
self.assertTrue(lexical_elements.is_identifier("A_1_2_B"))
self.assertTrue(lexical_elements.is_identifier("_1"))
self.assertTrue(lexical_elements.is_identifier("_1wroi"))
self.assertTrue(lexical_elements.is_identifier("_123"))
self.assertTrue(lexical_elements.is_identifier("ABab_12345"))
self.assertFalse(lexical_elements.is_identifier("1A"))
self.assertFalse(lexical_elements.is_identifier("1 A"))