class Scanner:
def __init__(self, problem_file, tokens, separators):
self.__problem_file = problem_file
self.__tokens = tokens
self.__separators = separators
self.__pif = HashTable()
self.__st = SymbolTable()
def scan(self):
line_count = 0
for line in self.__problem_file.read().splitlines():
line_count += 1
for token in self.tokenize(line):
if token in self.__tokens:
self.__pif.add(token, (-1, 1))
elif isinstance(token, Identifier) or isinstance(
token, Constant):
token_id = self.__st.add(token)
self.__pif.add(token, token_id)
else:
print('Lexical error for token:', token, ', at line:',
str(line_count))
return
def tokenize(self, line):
return line.split(str(self.__separators))
def get_pif(self):
return self.__pif
def get_st(self):
return self.__st
def main():
symbol_table = SymbolTable()
v1 = Identifier('a', 5)
v2 = Identifier('b', 7)
v3 = Identifier('c', 12)
v4 = Identifier('a', 14)
c1 = Constant('d', 7)
symbol_table.add(v1)
symbol_table.add(v2)
symbol_table.add(v3)
symbol_table.add(c1)
symbol_table.add(v4)
v2.set_value(42)
symbol_table.add(v2)
c2 = Constant('c', 17)
symbol_table.add(c2)
symbol_table.print_symbol_table()
def syntatic(stack, input):
# Algoritmo descendente sem backtracking
table = SymbolTable()
X = stack.top()
a = input.top()[1]
lex = input.top()[0]
callfuncao = False
declarado = False
function = False
while X is not "$":
print("X = {}\ta = {}\t lexema = {}".format(X, a, lex))
# Se X é î (token de î = 17)
if X == 17:
##
# print('1')
##
stack.pop()
X = stack.top()
# Incluindo alterações no arquivo
file.write("--> X == 17 (Retirou elemento vazio da pilha)")
print_stacks(stack, input)
# Se X é terminal
elif X <= 50:
##
# print('2')
##
if X == a:
if a == 41 and declarado:
declarado = False
# callfuncao
if a == 27:
name = input.stack[-2][0]
callfuncao = True
if not table.search(name):
print(
"\nErro na linha {}: Função não declarada".format(
int(input.top()[2]) + 1))
exit()
# Se for uma chamada de funçao
if a == 7 and callfuncao:
function = True
params = []
if input.stack[-2][0] == '(':
i = -3
p = input.stack[i][0]
while p != ')':
if p != ',':
try:
p = table.table[p]['type']
except:
if input.stack[i][1] == 5:
p = 'integer'
if input.stack[i][1] == 6:
p = 'float'
if input.stack[i][1] == 8:
p = 'char'
params.append(p)
i -= 1
p = input.stack[i][0]
if not params:
params = None
if params != table.table[name]['params']:
print(
"\nErro na linha {}: Parametros incorretos".format(
int(input.top()[2]) + 1))
exit()
callfuncao = False
stack.pop()
input.pop()
X = stack.top()
a = input.top()[1]
lex = input.top()[0]
continue
# Atribuição
if a == 7 and input.stack[-2][0] == '=':
if not table.search(input.top()[0]):
print("\nErro na linha {}: Variável não declarada.".
format(int(input.top()[2]) + 1))
exit()
type = input.stack[-3][1]
if type == 5:
if not (table.table[input.stack[-1][0]]['type']
== 'integer'):
print("\nErro na linha {}: Atribuição incorreta".
format(int(input.top()[2]) + 1))
exit()
elif type == 6:
if not (table.table[input.stack[-1][0]]['type']
== 'float'):
print("\nErro na linha {}: Atribuição incorreta".
format(int(input.top()[2]) + 1))
exit()
elif type == 8:
if not (table.table[input.stack[-1][0]]['type']
== 'char'):
print("\nErro na linha {}: Atribuição incorreta".
format(int(input.top()[2]) + 1))
exit()
# Se for uma variavel
if a == 7 and not declarado and not function:
if not table.search(input.top()[0]):
# Nome de função sem parametos
if input.stack[-2][0] == '{':
# Add (name, cat, type, level, params=None)
table.add(input.top()[0], 'f', 0, 0)
# Nome de função com parametros
elif input.stack[-2][0] == '(':
params = []
i = -3
p = input.stack[i][0]
while p != ')':
if p != ';':
params.append(p)
i -= 1
p = input.stack[i][0]
# Add (name, cat, type, level, params=None)
table.add(input.top()[0], 'f', 0, 0, params=params)
# Declaração de uma variável
elif input.stack[-2][0] == ':':
type = input.stack[-3][0]
table.add(input.top()[0], 'v', type, 0)
# Declaração de mais de uma variável
elif input.stack[-2][0] == ',':
declarado = True
i = -3
variaveis = []
variaveis.append(input.top()[0])
p = input.stack[i][0]
while p != ':':
if p == ',':
i -= 1
else:
variaveis.append(p)
i -= 1
p = input.stack[i][0]
type = input.stack[i - 1][0]
for v in variaveis:
# Add (name, cat, type, level, params=None)
table.add(v, 'v', type, 0)
else:
print("\nErro na linha {}: Variável não declarada".
format(int(input.top()[2]) + 1))
exit()
else:
# Não está declarando variável
if input.stack[-2][0] != ',' or input.stack[-2][
0] == ':':
stack.pop()
input.pop()
X = stack.top()
a = input.top()[1]
lex = input.top()[0]
continue
# Erro
print("\nErro na linha {}: Redeclaração de variáveis".
format(int(input.top()[2]) + 1))
table.print()
exit()
stack.pop()
input.pop()
# Incluindo alterações no arquivo
file.write(
"--> X == a (Retirou os elementos da pilha e da entrada)")
print_stacks(stack, input)
if input.top() == '$' and stack.top() == '$':
break
X = stack.top()
a = input.top()[1]
lex = input.top()[0]
continue
else:
##
# print('4')
##
print("\nErro na linha {}".format(int(input.top()[2]) + 1))
exit()
# Se X é não teminal
elif X > 50:
##
# print('5')
##
try:
##
# print('6')
##
r = parser[X][a]
s = rules[r]
stack.pop()
for i in reversed(list(s)):
stack.push(i)
X = stack.top()
# Incluindo alterações no arquivo
file.write("--> X > 50 (Incluiu regra {} na pilha)".format(r))
print_stacks(stack, input)
continue
except:
##
# print('7')
##
#print("\nStack: {}".format(stack))
#print("input: {}".format(input))
print("\nErro na linha {}".format(int(input.top()[2]) + 1))
exit()
class AnalisadorLexico:
def __init__(self, fileName):
self.lexemes = {
'>': 'MAIOR',
'<': 'MENOR',
'>=': 'MAIOR_IGUAL',
'<=': 'MENOR_IGUAL',
'!=': 'DIFERENTE',
'==': 'IGUAL',
'=': 'ATRIBUICAO',
'+': 'SOMA',
'-': 'SUBTRACAO',
'*': 'MULTIPLICACAO',
'/': 'DIVISAO',
',': 'VIRGULA',
'.': 'PONTO',
';': 'PONTO_VIRGULA',
'(': 'ABRE_PARENTESE',
')': 'FECHA_PARENTESE',
'{': 'ABRE_CHAVE',
'}': 'FECHA_CHAVE',
'[': 'ABRE_COLCHETE',
']': 'FECHA_COLCHETE'
}
self.symbolTable = SymbolTable()
self.tokens = []
self.errors = []
self.indice = 0
self.reserved = [
"int", "char", "float", "struct", "if", "else", "while", "void",
"return"
]
self.operators = [
"=", "<=", "<", ">", ">=", "==", "!=", "+", "-", "*", "/"
]
self.separators = [",", ".", "[", "{", "(", ")", "}", "]", ";"]
self.skip = ['\t', ' ', '\n', '\r']
self.letters = "abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ"
self.digits = "0123456789"
self.reader = BufferedFileReader(fileName)
self.root = Noh(None, True, 'programa')
self.arvore = ArvoreSintatica(self.root)
def analisar(self):
try:
current_character = self.reader.getChar()
self.reader.goForward()
next_character = self.reader.getChar()
lexeme = ""
#Comentarios
if current_character == '/' and next_character == '*':
self.reader.goForward()
current_character = self.reader.getChar()
self.reader.goForward()
next_character = self.reader.getChar()
while not (next_character == '/' and current_character == '*'):
current_character = self.reader.getChar()
self.reader.goForward()
next_character = self.reader.getChar()
self.reader.goForward()
#Ids e reservadas
elif (current_character in self.letters):
location = [self.reader.getLine(), self.reader.getColumn()]
while (current_character in self.letters) or (current_character
in self.digits):
lexeme += current_character
current_character = self.reader.getChar()
self.reader.goForward()
self.reader.goBackwards()
if lexeme in self.reserved:
self.tokens.append(
Token("RESERVADO", lexeme, location, None))
else:
self.tokens.append(
Token("ID", None, location,
self.symbolTable.add(lexeme)))
#Operadores
elif current_character in self.operators or current_character == '!':
joint = ''
joint += current_character
current_character = self.reader.getChar()
self.reader.goForward()
if current_character in self.operators:
joint += current_character
if joint in self.operators:
self.tokens.append(
Token(self.lexemes[joint], joint, [
self.reader.getLine(),
self.reader.getColumn()
], None))
else:
#PanicMode
self.tokens.append(
Token(self.lexemes[joint[0]], joint[0], [
self.reader.getLine(),
self.reader.getColumn()
], None))
self.tokens.append(
Token(self.lexemes[joint[1]], joint[1], [
self.reader.getLine(),
self.reader.getColumn()
], None))
self.errors.append(
"Warning -> Dois operandos diferentes juntos que nao agem em conjunto. Linha: "
+ str(self.reader.getLine()) + " Coluna: " +
str(self.reader.getColumn()))
elif joint in self.operators:
self.tokens.append(
Token(self.lexemes[joint], joint,
[self.reader.getLine(),
self.reader.getColumn()], None))
self.reader.goBackwards()
else:
#PanicMode
self.errors.append(
"Warning -> exclamacao ignorada. Linha: " +
str(self.reader.getLine()) + " Coluna: " +
str(self.reader.getColumn() - 1))
#Constantes Numericas
elif current_character in self.digits:
floating = False
exponencial = False
notIncluded = True
number = ''
while current_character in self.digits or current_character == '.' or current_character == 'E':
if current_character in self.digits:
number += current_character
elif current_character == '.' and ((not floating) and
(not exponencial)):
floating = True
number += current_character
elif current_character == '.' and (floating
or exponencial):
self.errors(
"Error -> Constante numerica invalida. Linha: " +
str(self.reader.getLine()) + " Coluna: " +
str(self.reader.getColumn()))
elif current_character == 'E' and (not exponencial):
if number[len(number) - 1] == '.':
#PanicMode
number += '0' + current_character
exponencial = True
self.errors.append(
"Warning -> Constante numerica convergida. Linha: "
+ str(self.reader.getLine()) + " Coluna: " +
str(self.reader.getColumn()))
else:
number += current_character
exponencial = True
current_character = self.reader.getChar()
self.reader.goForward()
if current_character == '+' or current_character == '-':
self.tokens.append(
Token("CONSTANTE NUMERICA", None, [
self.reader.getLine(),
self.reader.getColumn()
], self.symbolTable.add(number)))
self.tokens.append(
Token(self.lexemes[current_character],
current_character, [
self.reader.getLine(),
self.reader.getColumn()
], None))
notIncluded = False
break
elif current_character in self.digits:
number += current_character
current_character = self.reader.getChar()
self.reader.goForward()
if current_character in self.letters and current_character != 'E':
#PanicMode
self.reader.goBackwards()
self.errors.append(
"Warning -> Digitos ignorados no inicio de um identificador. Linha: "
+ str(self.reader.getLine()) + " Coluna: " +
str(self.reader.getColumn()))
elif notIncluded:
self.reader.goBackwards()
self.tokens.append(
Token("CONSTANTE NUMERICA", None,
[self.reader.getLine(),
self.reader.getColumn()],
self.symbolTable.add(number)))
#Separadores
elif current_character in self.separators:
self.tokens.append(
Token(self.lexemes[current_character], current_character,
[self.reader.getLine(),
self.reader.getColumn()], None))
elif current_character not in self.skip:
self.errors.append(
"Error -> Caractere nao suportado. Linha: " +
str(self.reader.getLine()) + " Coluna: " +
str(self.reader.getColumn()))
self.analisar()
except EndOfBufferException as e:
self.errors.append(str(e))
except EndOfFileException as e:
pass
def imprimir_tokens(self):
print("\nTokens (<tipo, valor/id, linha, coluna>):")
for t in self.tokens:
print(t)
def peek_token(self):
if (self.indice >= len(self.tokens)):
return None
return self.tokens[self.indice]
def peek_second_token(self):
if ((self.indice + 1) >= len(self.tokens)):
return None
return self.tokens[self.indice + 1]
def pop_token(self):
if (self.indice == len(self.tokens)):
return None
else:
self.indice += 1
return self.tokens[self.indice - 1]
def analise_sintatica(self):
self.programa()
def programa(self):
if (self.peek_token() is not None):
noh1 = Noh(self.arvore.root, False, 'declaracao_lista')
self.arvore.root.addFilho(noh1)
self.declaracao_lista(noh1)
def declaracao_lista(self, noh):
if (self.peek_token() is not None):
noh1 = Noh(noh, False, 'declaracao')
noh.addFilho(noh1)
self.declaracao(noh1)
noh2 = Noh(noh, False, 'declaracao_lista\'')
noh.addFilho(noh2)
self.declaracao_lista1(noh2)
def declaracao_lista1(self, noh):
if (self.peek_token() is not None):
primeiros_declaracao = ['int', 'float', 'char', 'void', 'struct']
if (self.peek_token().lexeme in primeiros_declaracao):
noh1 = Noh(noh, False, 'declaracao')
noh.addFilho(noh1)
self.declaracao(noh1)
noh2 = Noh(noh, False, 'declaracao_lista\'')
noh.addFilho(noh2)
self.declaracao_lista1(noh2)
def declaracao(self, noh):
if (self.peek_token() is not None):
primeiros_tipo_especificador = ['int', 'float', 'char', 'void']
if (self.peek_token().lexeme in primeiros_tipo_especificador):
noh1 = Noh(noh, False, 'tipo_especificador')
noh.addFilho(noh1)
self.tipo_especificador(noh1)
self.match(noh, "ID")
noh2 = Noh(noh, False, 'declaracao\'')
noh.addFilho(noh2)
self.declaracao1(noh2)
else:
noh1 = Noh(noh, False, 'estrutura')
noh.addFilho(noh1)
self.estrutura(noh1)
def declaracao1(self, noh):
if (self.peek_token() is not None):
if (self.peek_token().lexeme == '('):
noh1 = Noh(noh, False, 'fun_declaracao')
noh.addFilho(noh1)
self.fun_declaracao(noh1)
else:
noh1 = Noh(noh, False, 'var_declaracao')
noh.addFilho(noh1)
self.var_declaracao(noh1)
def var_declaracao(self, noh):
if (self.peek_token() is not None):
noh1 = Noh(noh, False, 'var_declaracao\'')
noh.addFilho(noh1)
self.var_declaracao1(noh1)
def var_declaracao1(self, noh):
if (self.peek_token() is not None):
if (self.peek_token().lexemeType == self.lexemes['[']):
self.match(noh, '[')
self.match(noh, 'CONSTANTE NUMERICA')
self.match(noh, ']')
noh1 = Noh(noh, False, 'var_declaracao\'')
noh.addFilho(noh1)
self.var_declaracao1(noh1)
elif (self.peek_token().lexemeType == self.lexemes['=']):
self.match(noh, '=')
noh1 = Noh(noh, False, 'expressao')
noh.addFilho(noh1)
self.expressao(noh1)
noh2 = Noh(noh, False, 'var_declaracao\'')
noh.addFilho(noh2)
self.var_declaracao1(noh2)
else:
self.match(noh, ';')
def tipo_especificador(self, noh):
if (self.peek_token() is not None):
if (self.peek_token().lexeme == "int"):
self.match(noh, "int")
elif (self.peek_token().lexeme == "float"):
self.match(noh, "float")
elif (self.peek_token().lexeme == "char"):
self.match(noh, "char")
else:
self.match(noh, "void")
def estrutura(self, noh):
if (self.peek_token() is not None):
self.match(noh, 'struct')
self.match(noh, 'ID')
self.match(noh, '{')
noh1 = Noh(noh, False, 'atributos_declaracao')
noh.addFilho(noh1)
self.atributos_declaracao(noh1)
self.match(noh, '}')
def atributos_declaracao(self, noh):
if (self.peek_token() is not None):
primeiros_tipo_especificador = ['int', 'float', 'char', 'void']
if (self.peek_token().lexeme in primeiros_tipo_especificador):
noh1 = Noh(noh, False, 'tipo_especificador')
noh.addFilho(noh1)
self.tipo_especificador(noh1)
self.match(noh, 'ID')
noh2 = Noh(noh, False, 'var_declaracao')
noh.addFilho(noh2)
self.var_declaracao(noh2)
noh3 = Noh(noh, False, 'atributos_declaracao\'')
noh.addFilho(noh3)
self.atributos_declaracao1(noh3)
else:
noh1 = Noh(noh, False, 'estrutura')
noh.addFilho(noh1)
self.estrutura(noh1)
noh2 = Noh(noh, False, 'declaracao\'')
noh.addFilho(noh2)
self.atributos_declaracao1(noh2)
def atributos_declaracao1(self, noh):
if (self.peek_token() is not None):
primeiros_tipo_especificador = ['int', 'float', 'char', 'void']
if (self.peek_token().lexeme in primeiros_tipo_especificador):
noh1 = Noh(noh, False, 'tipo_especificador')
noh.addFilho(noh1)
self.tipo_especificador(noh1)
self.match(noh, 'ID')
noh2 = Noh(noh, False, 'var_declaracao')
noh.addFilho(noh2)
self.var_declaracao(noh2)
noh3 = Noh(noh, False, 'atributos_declaracao\'')
noh.addFilho(noh3)
self.atributos_declaracao1(noh3)
elif (self.peek_token().lexeme == 'struct'):
noh1 = Noh(noh, False, 'estrutura')
noh.addFilho(noh1)
self.estrutura(noh1)
noh2 = Noh(noh, False, 'atributos_declaracao\'')
noh.addFilho(noh2)
self.atributos_declaracao1(noh2)
def fun_declaracao(self, noh):
if (self.peek_token() is not None):
self.match(noh, '(')
noh1 = Noh(noh, False, 'params')
noh.addFilho(noh1)
self.params(noh1)
self.match(noh, ')')
noh2 = Noh(noh, False, 'composto_decl')
noh.addFilho(noh2)
self.composto_decl(noh2)
def params(self, noh):
if (self.peek_token() is not None):
if (self.peek_token().lexeme == 'void'):
self.match(noh, 'void')
else:
noh1 = Noh(noh, False, 'param_lista')
noh.addFilho(noh1)
self.param_lista(noh1)
def param_lista(self, noh):
if (self.peek_token() is not None):
primeiros_tipo_especificador = ['int', 'float', 'char', 'void']
if (self.peek_token().lexeme in primeiros_tipo_especificador):
noh1 = Noh(noh, False, 'param')
noh.addFilho(noh1)
self.param(noh1)
noh2 = Noh(noh, False, 'param_lista')
noh.addFilho(noh2)
self.param_lista(noh2)
elif (self.peek_token().lexeme == ','):
self.match(noh, ',')
noh1 = Noh(noh, False, 'param')
noh.addFilho(noh1)
self.param(noh1)
noh2 = Noh(noh, False, 'param_lista')
noh.addFilho(noh2)
self.param_lista(noh2)
def param(self, noh):
if (self.peek_token() is not None):
noh2 = Noh(noh, False, 'tipo_especificador')
noh.addFilho(noh2)
self.tipo_especificador(noh2)
self.match(noh, "ID")
noh1 = Noh(noh, False, 'param\'')
noh.addFilho(noh1)
self.param1(noh1)
def param1(self, noh):
if (self.peek_token() is not None):
if (self.peek_token().lexemeType == self.lexemes['[']):
self.match(noh, '[')
self.match(noh, ']')
noh1 = Noh(noh, False, 'param\'')
noh.addFilho(noh1)
self.param1(noh1)
def composto_decl(self, noh):
if (self.peek_token() is not None):
self.match(noh, "{")
noh1 = Noh(noh, False, 'instrucoes')
noh.addFilho(noh1)
self.instrucoes(noh1)
self.match(noh, "}")
def instrucoes(self, noh):
if (self.peek_token() is not None):
primeiros_comando = [
'{', 'if', 'while', 'return', '(', 'ID', 'CONSTANTE NUMERICA'
]
primeiros_local_declaracoes = ["int", "char", "float", "void"]
if (self.peek_token().lexeme in primeiros_comando
or self.peek_token().lexemeType in primeiros_comando):
noh2 = Noh(noh, False, 'comando_lista')
noh.addFilho(noh2)
self.comando_lista(noh2)
noh1 = Noh(noh, False, 'instrucoes')
noh.addFilho(noh1)
self.instrucoes(noh1)
elif (self.peek_token().lexeme in primeiros_local_declaracoes):
noh2 = Noh(noh, False, 'local_declaracoes')
noh.addFilho(noh2)
self.local_declaracoes(noh2)
noh1 = Noh(noh, False, 'instrucoes')
noh.addFilho(noh1)
self.instrucoes(noh1)
def local_declaracoes(self, noh):
if (self.peek_token() is not None):
primeiros_tipo_especificador = ["int", "char", "float", "void"]
if (self.peek_token().lexeme in primeiros_tipo_especificador):
noh1 = Noh(noh, False, 'tipo_especificador')
noh.addFilho(noh1)
self.tipo_especificador(noh1)
self.match(noh, 'ID')
noh3 = Noh(noh, False, 'var_declaracao')
noh.addFilho(noh3)
self.var_declaracao(noh3)
noh2 = Noh(noh, False, 'local_declaracoes')
noh.addFilho(noh2)
self.local_declaracoes(noh2)
elif (self.peek_token().lexeme == 'struct'):
noh1 = Noh(noh, False, 'estrutura')
noh.addFilho(noh1)
self.estrutura(noh1)
noh2 = Noh(noh, False, 'local_declaracoes')
noh.addFilho(noh2)
self.local_declaracoes(noh2)
def comando_lista(self, noh):
if (self.peek_token() is not None):
primeiros_comando = [
'{', 'if', 'while', 'return', '(', 'ID', 'CONSTANTE NUMERICA'
]
if (self.peek_token().lexeme in primeiros_comando
or self.peek_token().lexemeType in primeiros_comando):
noh1 = Noh(noh, False, 'comando')
noh.addFilho(noh1)
self.comando(noh1)
noh2 = Noh(noh, False, 'comando_lista')
noh.addFilho(noh2)
self.comando_lista(noh2)
def comando(self, noh):
if (self.peek_token() is not None):
if (self.peek_token().lexemeType == self.lexemes['{']):
noh1 = Noh(noh, False, 'composto_decl')
noh.addFilho(noh1)
self.composto_decl(noh1)
elif (self.peek_token().lexeme == "if"):
noh1 = Noh(noh, False, 'selecao_decl')
noh.addFilho(noh1)
self.selecao_decl(noh1)
elif (self.peek_token().lexeme == "while"):
noh1 = Noh(noh, False, 'iteracao_decl')
noh.addFilho(noh1)
self.iteracao_decl(noh1)
elif (self.peek_token().lexeme == "return"):
noh1 = Noh(noh, False, 'retorno_decl')
noh.addFilho(noh1)
self.retorno_decl(noh1)
else:
noh1 = Noh(noh, False, 'expressao_decl')
noh.addFilho(noh1)
self.expressao_decl(noh1)
def expressao_decl(self, noh):
if (self.peek_token() is not None):
if (self.peek_token().lexeme == ';'):
self.match(noh, ';')
else:
noh1 = Noh(noh, False, 'expressao')
noh.addFilho(noh1)
self.expressao(noh1)
self.match(noh, ";")
def selecao_decl(self, noh):
if (self.peek_token() is not None):
self.match(noh, "if")
self.match(noh, "(")
noh1 = Noh(noh, False, 'expressao')
noh.addFilho(noh1)
self.expressao(noh1)
self.match(noh, ")")
noh2 = Noh(noh, False, 'comando')
noh.addFilho(noh2)
self.comando(noh2)
noh3 = Noh(noh, False, 'else_decl')
noh.addFilho(noh3)
self.else_decl(noh3)
def else_decl(self, noh):
if (self.peek_token() is not None):
if (self.peek_token().lexeme == "else"):
self.match(noh, "else")
noh2 = Noh(noh, False, 'comando')
noh.addFilho(noh2)
self.comando(noh2)
def iteracao_decl(self, noh):
if (self.peek_token() is not None):
self.match(noh, "while")
self.match(noh, "(")
noh1 = Noh(noh, False, 'expressao')
noh.addFilho(noh1)
self.expressao(noh1)
self.match(noh, ")")
noh2 = Noh(noh, False, 'comando')
noh.addFilho(noh2)
self.comando(noh2)
def retorno_decl(self, noh):
if (self.peek_token() is not None):
self.match(noh, "return")
noh2 = Noh(noh, False, 'retorno_decl\'')
noh.addFilho(noh2)
self.retorno_decl1(noh2)
self.match(noh, ";")
def retorno_decl1(self, noh):
if (self.peek_token() is not None):
if (self.peek_token().lexemeType != self.lexemes[';']):
noh2 = Noh(noh, False, 'expressao')
noh.addFilho(noh2)
self.expressao(noh2)
def expressao(self, noh):
if (self.peek_token() is not None):
if (self.peek_token().lexemeType == self.lexemes['(']
or self.peek_token().lexemeType == 'CONSTANTE NUMERICA'):
noh2 = Noh(noh, False, 'expressao_simples')
noh.addFilho(noh2)
self.expressao_simples(noh2)
else:
if (self.peek_second_token().lexemeType == self.lexemes['(']):
noh2 = Noh(noh, False, 'ativacao')
noh.addFilho(noh2)
self.ativacao(noh2)
noh1 = Noh(noh, False, 'termo\'')
noh.addFilho(noh1)
self.termo1(noh1)
noh3 = Noh(noh, False, 'expressao_soma\'')
noh.addFilho(noh3)
self.expressao_soma1(noh3)
noh4 = Noh(noh, False, 'expressao_simples\'')
noh.addFilho(noh4)
self.expressao_simples1(noh4)
else:
noh1 = Noh(noh, False, 'var')
noh.addFilho(noh1)
self.var(noh1)
if (self.peek_token().lexemeType == self.lexemes['=']):
self.match(noh, '=')
noh2 = Noh(noh, False, 'expressao')
noh.addFilho(noh2)
self.expressao(noh2)
else:
noh2 = Noh(noh, False, 'termo\'')
noh.addFilho(noh2)
self.termo1(noh2)
noh3 = Noh(noh, False, 'expressao_soma\'')
noh.addFilho(noh3)
self.expressao_soma1(noh3)
noh4 = Noh(noh, False, 'expressao_simples\'')
noh.addFilho(noh4)
self.expressao_simples1(noh4)
def var(self, noh):
if (self.peek_token() is not None):
self.match(noh, 'ID')
noh1 = Noh(noh, False, 'var\'')
noh.addFilho(noh1)
self.var1(noh1)
def var1(self, noh):
if (self.peek_token() is not None):
if (self.peek_token().lexemeType == self.lexemes['[']):
self.match(noh, '[')
noh2 = Noh(noh, False, 'expressao')
noh.addFilho(noh2)
self.expressao(noh2)
self.match(noh, ']')
noh1 = Noh(noh, False, 'var\'')
noh.addFilho(noh1)
self.var1(noh1)
def expressao_simples(self, noh):
if (self.peek_token() is not None):
noh1 = Noh(noh, False, 'expressao_soma')
noh.addFilho(noh1)
self.expressao_soma(noh1)
noh2 = Noh(noh, False, 'expressao_simples\'')
noh.addFilho(noh2)
self.expressao_simples1(noh2)
def expressao_simples1(self, noh):
if (self.peek_token() is not None):
primeiros_relacional = ['<', '>', '==', '!=', '<=', '>=']
if (self.peek_token().lexeme in primeiros_relacional):
noh1 = Noh(noh, False, 'relacional')
noh.addFilho(noh1)
self.relacional(noh1)
noh2 = Noh(noh, False, 'expressao_soma')
noh.addFilho(noh2)
self.expressao_soma(noh2)
def expressao_soma(self, noh):
if (self.peek_token() is not None):
noh1 = Noh(noh, False, 'termo')
noh.addFilho(noh1)
self.termo(noh1)
noh2 = Noh(noh, False, 'expressao_soma\'')
noh.addFilho(noh2)
self.expressao_soma1(noh2)
def expressao_soma1(self, noh):
if (self.peek_token() is not None):
primeiros_soma = ['+', '-']
if (self.peek_token().lexeme in primeiros_soma):
noh1 = Noh(noh, False, 'soma')
noh.addFilho(noh1)
self.soma(noh1)
noh2 = Noh(noh, False, 'termo')
noh.addFilho(noh2)
self.termo(noh2)
noh3 = Noh(noh, False, 'expressao_soma\'')
noh.addFilho(noh3)
self.expressao_soma1(noh3)
def termo(self, noh):
if (self.peek_token() is not None):
noh1 = Noh(noh, False, 'fator')
noh.addFilho(noh1)
self.fator(noh1)
noh2 = Noh(noh, False, 'termo\'')
noh.addFilho(noh2)
self.termo1(noh2)
def termo1(self, noh):
if (self.peek_token() is not None):
primeiros_mult = ['*', '/']
if (self.peek_token().lexeme in primeiros_mult):
noh1 = Noh(noh, False, 'mult')
noh.addFilho(noh1)
self.mult(noh1)
noh2 = Noh(noh, False, 'fator')
noh.addFilho(noh2)
self.fator(noh2)
noh3 = Noh(noh, False, 'termo\'')
noh.addFilho(noh3)
self.termo1(noh3)
def fator(self, noh):
if (self.peek_token() is not None):
if (self.peek_token().lexemeType == self.lexemes['(']):
self.match(noh, '(')
noh1 = Noh(noh, False, 'expressao')
noh.addFilho(noh1)
self.expressao(noh1)
self.match(noh, ')')
elif (self.peek_token().lexemeType == 'CONSTANTE NUMERICA'):
self.match(noh, 'CONSTANTE NUMERICA')
elif (self.peek_second_token().lexemeType == self.lexemes['(']):
noh1 = Noh(noh, False, 'ativacao')
noh.addFilho(noh1)
self.ativacao(noh1)
else:
noh1 = Noh(noh, False, 'var')
noh.addFilho(noh1)
self.var(noh1)
def ativacao(self, noh):
if (self.peek_token() is not None):
self.match(noh, 'ID')
self.match(noh, '(')
noh1 = Noh(noh, False, 'args')
noh.addFilho(noh1)
self.args(noh1)
self.match(noh, ')')
def args(self, noh):
if (self.peek_token() is not None):
noh1 = Noh(noh, False, 'arg_lista')
noh.addFilho(noh1)
self.arg_lista(noh1)
def arg_lista(self, noh):
if (self.peek_token() is not None):
primeiros_expressao = [
'ID', 'CONSTANTE NUMERICA', self.lexemes['(']
]
if (self.peek_token().lexemeType in primeiros_expressao):
noh1 = Noh(noh, False, 'expressao')
noh.addFilho(noh1)
self.expressao(noh1)
noh2 = Noh(noh, False, 'arg_lista')
noh.addFilho(noh2)
self.arg_lista(noh2)
elif (self.peek_token().lexeme == ','):
self.match(noh, ',')
noh1 = Noh(noh, False, 'expressao')
noh.addFilho(noh1)
self.expressao(noh1)
noh2 = Noh(noh, False, 'arg_lista')
noh.addFilho(noh2)
self.arg_lista(noh2)
def relacional(self, noh):
if (self.peek_token() is not None):
if (self.peek_token().lexemeType == self.lexemes['>']):
self.match(noh, '>')
elif (self.peek_token().lexemeType == self.lexemes['<']):
self.match(noh, '<')
elif (self.peek_token().lexemeType == self.lexemes['>=']):
self.match(noh, '>=')
elif (self.peek_token().lexemeType == self.lexemes['<=']):
self.match(noh, '<=')
elif (self.peek_token().lexemeType == self.lexemes['!=']):
self.match(noh, '!=')
else:
self.match(noh, '==')
def soma(self, noh):
if (self.peek_token() is not None):
if (self.peek_token().lexemeType == self.lexemes['+']):
self.match(noh, '+')
else:
self.match(noh, '-')
def mult(self, noh):
if (self.peek_token() is not None):
if (self.peek_token().lexemeType == self.lexemes['*']):
self.match(noh, '*')
else:
self.match(noh, '/')
def match(self, noh, lexeme):
token = self.peek_token()
if (token is not None):
if (lexeme == 'ID' or lexeme == 'CONSTANTE NUMERICA'):
if (token.lexemeType != lexeme):
self.errors.append('Syntax Error -> Esperado \'' + lexeme +
'\' - encontrado \'' + str(token) +
'\'')
else:
noh1 = Noh(noh, False, token)
noh.addFilho(noh1)
else:
if (token.lexeme != lexeme):
#panic mode
if (lexeme == ';'):
self.errors.append('Syntax Error -> Esperado \'' +
lexeme + '\' - encontrado \'' +
str(token) +
'\'\nERRO FOI IGNORADO')
noh1 = Noh(noh, False, ';')
noh.addFilho(noh1)
return
elif (lexeme == 'struct'):
self.errors.append('Syntax Error -> Esperado \'' +
lexeme + '\' - encontrado \'' +
str(token) + '\'\n')
return
self.errors.append('Syntax Error -> Esperado \'' + lexeme +
'\' - encontrado \'' + str(token) +
'\'')
else:
noh1 = Noh(noh, False, token)
noh.addFilho(noh1)
else:
self.errors.append('Syntax Error -> Esperado \'' + lexeme +
'\' - encontrado \'nada\'')
return
self.pop_token()
#------------------------------------------------------------------------------------------------------------------------------------------
#------------------------------------------------------------------------------------------------------------------------------------------
#------------------------------------------------------------------------------------------------------------------------------------------
#------------------------------------------------------------------------------------------------------------------------------------------
def imprimir_tabela_simbolos(self):
print("\nTabela de Simbolos:")
print(self.symbolTable)
def imprimir_erros(self):
if not self.errors:
return
print("Errors:")
for e in self.errors:
print(e)
def imprimirArvore(self):
print("Arvore Sintatica:\n")
print(self.arvore)
class Interpreter:
def __init__(self, code_string=None):
self._KEYWORDS = ['read', 'write']
self._token = None
self._line = 0
self._tokenizer = Tokenizer(code_string, ['+','-','/','*','(',')',':='], ['\n',' '])
self._symboltable = SymbolTable()
def reset(self):
self._line = 0
self._token = None
self._tokenizer.clear()
def interpret(self, code_string=None):
if code_string is not None:
self._tokenizer.append(code_string)
self._consume()
self.program()
def _consume(self, _nomable=None):
if _nomable == '$$':
self.reset()
return True
if _nomable == 'id':
self._symboltable.add(self._token, self._line)
# TODO: add current token to AST
self._token = self._tokenizer.next()
def _is_token_id(self, _id=None):
if self._token is None:
raise ParseError(self._line, 'unexpected EOF')
if _id is None:
_id = self._token
if self._symboltable.has(_id):
return True
elif _id.isalpha() and _id not in self._KEYWORDS:
return True
else:
return False
def _is_token_num(self, _num=None):
if self._token is None:
raise ParseError(self._line, 'unexpected EOF')
if _num is None:
_num = self._token
if _num.isdigit():
return True
else:
return False
def _is_token_id_or_num(self, _token=None):
if _token is None:
_token = self._token
if self._is_token_id(_token) or self._is_token_num(_token):
return True
else:
return False
def _match(self, expected):
# TODO: might conflict with id's named 'id' or 'number'
if expected == self._token or expected in ['id', 'number']:
self._consume(self._token)
else:
raise TokenError(self._line, self._token, expected)
def _skip(self):
pass
def program(self):
if self._token in ['read', 'write', '$$'] or self._is_token_id():
self._stmt_list()
self._match('$$')
else:
raise ParseError(self._line, 'program')
def _stmt_list(self):
if self._token == '$$':
self._skip()
elif self._token in ['read', 'write'] or self._is_token_id():
self._line += 1
self._stmt()
self._stmt_list()
else:
raise ParseError(self._line, 'stmt_list')
def _stmt(self):
if self._token == 'read':
self._match('read')
self._match('id')
elif self._token == 'write':
self._match('write')
self._expr()
elif self._is_token_id():
self._match('id')
self._match(':=')
self._expr()
else:
raise ParseError(self._line, 'stmt')
def _expr(self):
if self._token == '(' or self._is_token_id_or_num():
self._term()
self._term_tail()
else:
raise ParseError(self._line, 'expr')
def _term_tail(self):
if self._token in ['+', '-']:
self._add_op()
self._term()
self._term_tail()
elif self._token in [')', 'read', 'write', '$$'] or self._is_token_id():
self._skip()
else:
raise ParseError(self._line, 'term_tail')
def _term(self):
if self._token == '(' or self._is_token_id_or_num():
self._factor()
self._factor_tail()
else:
raise ParseError(self._line, 'term')
def _factor_tail(self):
if self._token in ['*', '/']:
self._mult_op()
self._factor()
self._factor_tail()
elif self._token in ['+', '-', ')', 'read', 'write', '$$'] or self._is_token_id():
self._skip()
else:
raise ParseError(self._line, 'factor_tail')
def _factor(self):
if self._token == '(':
self._match('(')
self._expr()
self._match(')')
elif self._is_token_id():
self._match('id')
elif self._is_token_num():
self._match('number')
else:
raise ParseError(self._line, 'factor')
def _add_op(self):
if self._token == '+':
self._match('+')
elif self._token == '-':
self._match('-')
else:
raise ParseError(self._line, 'add_op')
def _mult_op(self):
if self._token == '*':
self._match('*')
elif self._token == '/':
self._match('/')
else:
raise ParseError(self._line, 'mult_op')
def run_ui(self):
flag = True
while flag:
self.print_menu()
cmd = int(input("Your comand : > "))
if cmd == 0:
flag = False
if cmd == 1:
inputSequnceStr = input(
"Give your input sequence (without space )")
input_sequence_list = []
for char in inputSequnceStr:
input_sequence_list.append(char)
g = Grammar.fromFile("grammar.txt")
lr0 = LZero(g)
print(lr0.parse(input_sequence_list))
print(lr0.get_derivation_string(input_sequence_list))
if cmd == 2:
scanner = Scanner()
identifierTable = SymbolTable()
constantsTable = SymbolTable()
pif = ProgramInternalForm()
fileName = 'program.txt'
with open(fileName, 'r') as file:
lineNo = 0
for line in file:
lineNo += 1
for token in scanner.tokenGenerator(
line.strip('\n'), separators):
if token == ' ':
continue
if token in separators + operators + reservedWords:
pif.add(codification[token], -1)
elif scanner.isIdentifier(token):
id = identifierTable.add(token)
pif.add(codification['identifier'], id)
elif scanner.isConstant(token):
id = constantsTable.add(token)
pif.add(codification['constant'], id)
else:
raise Exception('Unknown token ' + token +
' at line ' + str(lineNo))
print('Program Internal Form: \n', pif)
print('Identifier Table: \n', identifierTable)
print('Constants Table: \n', constantsTable)
#turning the PIF into an input stack
inverseCodification = dict([[codification[key], key]
for key in codification])
for code in pif.getCodes():
print(code, ' : ', inverseCodification[code])
inputStack = [str(code) for code in pif.getCodes()]
print(inputStack)
g = Grammar.fromFile("my_grammar.txt")
g.P = [('S1', ['.', g.S])] + g.P
g.N += ['S1']
lr0 = LZero(g)
print(g)
print(lr0.parse(inputStack))
#SEQUENCE OF DERIVATIONS IN CURSU 6
print(lr0.get_derivation_string_program(inputStack))
class ProgramBuilder():
""" manages the in memory IR for LLVM """
def __init__(self, input_fName, output_fName):
self.symbolTable = SymbolTable()
self.parser = Parser(input_fName)
self.output_fName = output_fName
self._has_errors = False
# llvm setup
binding.initialize()
binding.initialize_native_target()
binding.initialize_native_asmprinter()
def has_errors(self):
""" checks the flags for scanner, parser or codegen errors"""
return self._has_errors or self.parser._has_errors or self.parser.scanner._has_errors
def initialize_module(self, name):
# create module
self.module = ir.Module(name=name)
self.module.triple = binding.get_default_triple()
# load builtin I/O functions
self.load_builtins()
# create main function
func_type = ir.FunctionType(ir.VoidType(), [], False)
base_func = ir.Function(self.module, func_type, name="main")
block = base_func.append_basic_block(name="entry")
# create llvm ir builder and set at start of the main block
self.builder = ir.IRBuilder(block)
def load_builtins(self):
""" create builtin get and put functions (empty for now)"""
# name returnType Param
builtins = [
("putbool", ir.VoidType(), [ir.IntType(1)]),
("putstring", ir.VoidType(), [ir.PointerType(ir.IntType(8))]),
("putinteger", ir.VoidType(), [ir.IntType(32)]),
("putfloat", ir.VoidType(), [ir.FloatType()]),
("getbool", ir.IntType(1), []), ("getinteger", ir.IntType(32), []),
("getfloat", ir.FloatType(), []),
("main.StringEquals", ir.IntType(1),
[ir.PointerType(ir.IntType(8)),
ir.PointerType(ir.IntType(8))]),
("getstring", ir.PointerType(ir.IntType(8)), [])
]
for entry in builtins:
fType = ir.FunctionType(entry[1], entry[2])
func = ir.Function(self.module, fType, name=entry[0])
symbol = Symbol(entry[0], func, fType, id_type='function')
self.symbolTable.add(symbol)
def output_ir(self):
# signal return from main function
# self.builder.ret_void()
output = str(self.module)
with open(self.output_fName, "w") as fd:
fd.write(output)
def generate_module_code(self):
""" runs the parse loop and calls codegen functions for the input program"""
# parse top level module info
moduleName = self.parser.parse_program_header()
if not moduleName:
print("could not parse module header")
return None
self.initialize_module(moduleName)
print("generating module: {}".format(moduleName))
# parse and generate top level declarations (variable or function)
res = self.parser.parse_top_level_declaration()
while res != tkn.BEGIN:
if res != None: # and (not self.has_errors()):
ir = res.codegen(self.builder, self.symbolTable, self.module)
if ir == None:
self._has_errors = True
print("\n".join(res.getErrors()))
res = self.parser.parse_top_level_declaration()
# parse and generate top level statements
res = self.parser.parse_top_level_statement()
while res != tkn.EOF:
if res != None: #and (not self.has_errors()):
ir = res.codegen(self.builder, self.symbolTable, self.module)
if ir == None:
self._has_errors = True
print("\n".join(res.getErrors()))
res = self.parser.parse_top_level_statement()
if not self.has_errors():
print("writing IR to LLVM assembly\n")
self.output_ir()
else:
print("errors detected, compilation aborted")
class HackAssembler:
def __init__(self, asm_filename):
self.asm_filename = asm_filename
self.symbol_table = SymbolTable()
self.parser = Parser(self.symbol_table)
self.binary_translator = BinaryTraslator(self.parser)
self.next_open_memory_address = 16
self.labels_parsed = 0
self.output_string = ''
def compile(self):
# self.__remove_commented_and_empty_lines()
self.__scan_for_labels()
self.__scan_for_variables()
self.__translate_to_binary()
self.__write_to_out_file()
# def __remove_commented_and_empty_lines(self):
# self.__read_file_by_line(self.__remove_commented_or_empty_line)
# def __remove_commented_or_empty_line(self, line, cnt):
# line = line.split('//')[0].strip().replace(' ', '')
# if len(line) > 0:
# self.output_string += "{}\n".format(line)
def __scan_for_labels(self):
self.__read_file_by_line(self.__add_value_to_symbol_table_if_label)
def __scan_for_variables(self):
self.__read_file_by_line(self.__add_value_to_symbol_table_if_variable)
def __translate_to_binary(self):
self.__read_file_by_line(self.__translate_line_to_binary)
def __write_to_out_file(self):
base_filename = self.asm_filename.split('.asm')[0]
hack_filename = "{}_Brayden.hack".format(base_filename)
with open(hack_filename, "w") as text_file:
text_file.write(self.output_string)
def __read_file_by_line(self, block):
lines_processed = 0
with open(self.asm_filename) as fp:
for line in fp:
line = self.__preprocess_line(line)
if len(line) > 0:
block(line, lines_processed)
lines_processed += 1
def __preprocess_line(self, line):
return line.split('//')[0].strip().replace(' ', '')
def __add_value_to_symbol_table_if_label(self, line, cnt):
if self.parser.is_label(line):
self.symbol_table.add(self.parser.label_value(line),
cnt - self.labels_parsed)
self.labels_parsed += 1
def __add_value_to_symbol_table_if_variable(self, line, cnt):
if self.parser.is_variable(line):
if not self.symbol_table.contains(
self.parser.variable_value(line)):
self.symbol_table.add(self.parser.variable_value(line),
self.next_open_memory_address)
self.__increment_next_open_memory_address()
def __translate_line_to_binary(self, line, _):
binary_command = self.binary_translator.translate(line)
if binary_command:
self.output_string += '{}\n'.format(binary_command)
def __increment_next_open_memory_address(self):
self.next_open_memory_address += 1
