def main():
try:
opts, args = getopt.getopt(
sys.argv[1:], "ho:d:si:S:",
["help", "output=", "display=", "simple", "input=", "style="])
except getopt.GetoptError as err:
# print help information and exit:
print str(err)
usage()
sys.exit(1)
output_file = None
input_file = None
display_file = None
simple = False
fsm_style = "nfa"
for o, a in opts:
if o == "-s":
simple = True
elif o in ("-h", "--help"):
usage()
sys.exit()
elif o in ("-o", "--output"):
output_file = a
elif o in ("-i", "--input"):
input_file = a
elif o in ("-d", "--display"):
display_file = a
elif o in ("-S", "--style"):
fsm_style = a.lower()
else:
assert False, "unhandled option " + o
if (input_file == None or output_file == None):
print "Input file with PCRE patterns and output file for FSM must be specified!"
sys.exit(1)
fsm = b_automaton.b_Automaton()
p = parser.parser("pcre_parser")
p.load_file(input_file)
fsm.create_by_parser(p)
if fsm_style == "nfa":
fsm.remove_epsilons()
elif fsm_style == "pa":
fsm.thompson2glushkov()
elif fsm_style == "dpa":
fsm.thompson2reverse_glushkov()
else:
print "Unsupported automaton type: " + fsm_style
print "Supported automata styles are: nfa, pa, dpa"
print "See " + sys.argv[0] + " -h for details"
sys.exit(1)
if simple == True:
fsm.remove_char_classes()
fsm.save_to_timbuk(output_file, simple)
if display_file != None:
fsm.show(display_file, " ")
def _replace_length_restriction_with_a_closure(self, NFA):
"""
The first step in this construction replaces the length
restriction with a closure, and constructs the H-FA, with
the closure represented by a flag in the history buffer.
:param NFA: NFA
:type NFA: nfa_data
:returns: NFA without counting constraint
:rtype: nfa_data
"""
# identify counting transitions with exactly X counting
cnt_transitions = list()
for t in NFA.transitions:
if NFA.alphabet[t[1]].ctype == io_mapper["b_Sym_cnt_constr"]:
if NFA.alphabet[t[1]].m == NFA.alphabet[t[1]].n:
cnt_transitions.append(t)
# remove founded counting transtions
# and replace them with loop transitions
# and add epsilon tran. to next state
for t in cnt_transitions:
NFA.transitions.remove(t)
cnt_symbol = NFA.alphabet[t[1]]
self.flags_cnt[t[0]] = str(cnt_symbol.m)
NFA.transitions.add((t[0], t[1], t[0]))
NFA.transitions.add((t[0], -1, t[2]))
# replace cnt symbol at char or char class
# and add epsilon symbol into alphabet if does not exist
for t in cnt_transitions:
symbolID = t[1]
cnt_symbol = copy.deepcopy(NFA.alphabet[symbolID])
if cnt_symbol.ctype == io_mapper["b_Sym_cnt_constr"]:
if isinstance(cnt_symbol.symbol, str):
NFA.alphabet[symbolID] = b_Sym_char_class(
new_text=cnt_symbol._text,
charClass=set([cnt_symbol.symbol]),
new_id=cnt_symbol._id)
else:
NFA.alphabet[symbolID] = b_Sym_char_class(
new_text=cnt_symbol._text,
charClass=cnt_symbol.symbol,
new_id=cnt_symbol._id)
epsilonID = -1
if not epsilonID in NFA.alphabet:
NFA.alphabet[epsilonID] = b_Sym_char("Epsilon", "", -1)
# remove epsilons
aut = b_Automaton()
aut._automaton = NFA
aut.remove_epsilons()
return aut._automaton
def main():
try:
opts, args = getopt.getopt(sys.argv[1:], "ho:d:si:S:", ["help", "output=", "display=", "simple", "input=", "style="])
except getopt.GetoptError as err:
# print help information and exit:
print str(err)
usage()
sys.exit(1)
output_file = None
input_file = None
display_file = None
simple = False
fsm_style = "nfa"
for o, a in opts:
if o == "-s":
simple = True
elif o in ("-h", "--help"):
usage()
sys.exit()
elif o in ("-o", "--output"):
output_file = a
elif o in ("-i", "--input"):
input_file = a
elif o in ("-d", "--display"):
display_file = a
elif o in ("-S", "--style"):
fsm_style = a.lower()
else:
assert False, "unhandled option " + o
if (input_file == None or output_file == None):
print "Input file with PCRE patterns and output file for FSM must be specified!"
sys.exit(1)
fsm = b_automaton.b_Automaton()
p = parser.parser("pcre_parser")
p.load_file(input_file)
fsm.create_by_parser(p)
if fsm_style == "nfa":
fsm.remove_epsilons()
elif fsm_style == "pa":
fsm.thompson2glushkov()
elif fsm_style == "dpa":
fsm.thompson2reverse_glushkov()
else:
print "Unsupported automaton type: " + fsm_style
print "Supported automata styles are: nfa, pa, dpa"
print "See " + sys.argv[0] + " -h for details"
sys.exit(1)
if simple == True:
fsm.remove_char_classes()
fsm.save_to_timbuk(output_file, simple)
if display_file != None:
fsm.show(display_file, " ")
def main():
try:
opts, args = getopt.getopt(
sys.argv[1:], "ho:d:si:S:",
["help", "output=", "display=", "simple", "input=", "style="])
except getopt.GetoptError as err:
# print help information and exit:
print str(err)
usage()
sys.exit(1)
output_file = None
input_file = None
display_file = None
simple = False
fsm_style = "nfa"
for o, a in opts:
if o == "-s":
simple = True
elif o in ("-h", "--help"):
usage()
sys.exit()
elif o in ("-o", "--output"):
output_file = a
elif o in ("-i", "--input"):
input_file = a
elif o in ("-d", "--display"):
display_file = a
else:
assert False, "unhandled option " + o
if (input_file == None or output_file == None):
print "Input file with PCRE patterns and output file for FSM must be specified!"
sys.exit(1)
fsm = b_automaton.b_Automaton()
p = parser.parser("pcre_parser")
p.load_file(input_file)
fsm.create_by_parser(p)
fsm.remove_epsilons()
print "State num: ", fsm.get_state_num()
if simple == True:
fsm.remove_char_classes()
fsm.save_to_FA_format(output_file)
if display_file != None:
fsm.show(display_file, " ")
def main():
try:
opts, args = getopt.getopt(sys.argv[1:], "ho:d:si:S:", ["help", "output=", "display=", "simple", "input=", "style="])
except getopt.GetoptError as err:
# print help information and exit:
print str(err)
usage()
sys.exit(1)
output_file = None
input_file = None
display_file = None
simple = False
fsm_style = "nfa"
for o, a in opts:
if o == "-s":
simple = True
elif o in ("-h", "--help"):
usage()
sys.exit()
elif o in ("-o", "--output"):
output_file = a
elif o in ("-i", "--input"):
input_file = a
elif o in ("-d", "--display"):
display_file = a
else:
assert False, "unhandled option " + o
if (input_file == None or output_file == None):
print "Input file with PCRE patterns and output file for FSM must be specified!"
sys.exit(1)
fsm = b_automaton.b_Automaton()
p = parser.parser("pcre_parser")
p.load_file(input_file)
fsm.create_by_parser(p)
fsm.remove_epsilons()
print "State num: ", fsm.get_state_num()
if simple == True:
fsm.remove_char_classes()
fsm.save_to_FA_format(output_file)
if display_file != None:
fsm.show(display_file, " ")
def ProcessRule(parser):
f = open("out.out","a")
N_Automaton = b_Automaton()
N_Automaton.create_from_nfa_data(parser.get_nfa())
N_Automaton.get_automaton().Show("test_NFA.dot")
print("Automata joined")
D_Automaton = PHF_DFA()
D_Automaton.create_from_nfa_data(N_Automaton.get_automaton())
print("Determinising...")
D_Automaton.resolve_alphabet()
# D_Automaton.removeCharClasses()
# D_Automaton.reduce_alphabet()
D_Automaton.determinise(states_limit = 10000)
print("Minimising...")
D_Automaton.minimise() #Co to vlastne vypisuje?
D_Automaton.get_automaton().Show("test_min_dfa.dot")
stride0 = str(D_Automaton.get_state_num())+"/"+str(D_Automaton.get_trans_num())+"|"
f.write(stride0)
D_Automaton.stride_2()
stride2 = str(D_Automaton.get_state_num())+"/"+str(D_Automaton.get_trans_num())+"|"
f.write(stride2)
D_Automaton.stride_2()
stride4 = str(D_Automaton.get_state_num())+"/"+str(D_Automaton.get_trans_num())+"|"
f.write(stride4)
#D_Automaton.stride_2()
#stride8 = str(D_Automaton.get_state_num())+"/"+str(D_Automaton.get_trans_num())+"|"
#f.write(stride8)
f.write("\n")
f.close()
# D_Automaton.set_table_parameters((5,13))
# if(D_Automaton.generate_PHF_table() == False):
# print("Failed")
# exit()
# D_Automaton.search("ahojdfgffskf sdfgsdfgs dfg jkgsdgcjgfsdfg admin_root=php")
# D_Automaton.search(Data)
# except Exception:
# print("GENERATED EXCEPTION")
#D_Automaton.printConfFile("/home/galloth/configuration");
print("")
from netbench.pattern_match import sym_char, sym_char_class
from netbench.pattern_match import pcre_parser
from netbench.pattern_match.b_automaton import b_Automaton
from netbench.pattern_match.nfa_data import nfa_data
#from netbench.pattern_match.b_dfa import b_dfa
from phf_dfa import PHF_DFA
#FileName= "../../rules/Moduly/web-client.rules.pcre"
FileName= "../../rules/Moduly/web-php.rules.pcre"
parser = pcre_parser.pcre_parser()
parser.load_file(FileName)
print("Joining...")
N_Automaton = b_Automaton()
N_Automaton.create_from_nfa_data(parser.get_nfa())
#while (parser.next_line()):
# N_Automaton.join(parser.get_nfa())
N_Automaton.get_automaton().Show("test_NFA.dot")
print("Automata joined")
D_Automaton = PHF_DFA()
D_Automaton.create_from_nfa_data(N_Automaton.get_automaton())
print("Determinising...")
D_Automaton.determinise(states_limit = 10000)
print("Minimising...")
D_Automaton.minimise() #Co to vlastne vypisuje?
D_Automaton.get_automaton().Show("test_min_dfa.dot")
D_Automaton.get_automaton().SaveToFile("temp_automaton")
def test_get_nfa(self):
"""get_nfa()"""
# If attribute _position < 0, check returning None.
parser = pcre_parser()
self.assertTrue(parser._position < 0)
self.assertTrue(parser.get_nfa() == None)
# Try method on a few regular expressions.
# The results obtained compare with the manually completed machines.
# (Recommend to compare after the elimination of epsilon transition)
# 1) concatenation
parser = pcre_parser()
parser.set_text("/first/")
automat = b_Automaton()
automat._automaton = parser.get_nfa()
automat.remove_epsilons()
cp = automat.get_automaton()
result = nfa_data().load_from_file(
"test_data/(1)pcre_get_nfa.nfa_data")
self.assertTrue(
sorted(cp.states.keys()) == sorted(result.states.keys()))
self.assertTrue(cp.alphabet == result.alphabet)
self.assertTrue(cp.start == result.start)
self.assertTrue(cp.final == result.final)
self.assertTrue(cp.transitions == result.transitions)
self.assertTrue(cp.Flags == result.Flags)
# 2) branch (automat create char class), iteration *
parser = pcre_parser()
parser.set_text("/[ab]cd*/")
automat = b_Automaton()
automat._automaton = parser.get_nfa()
automat.remove_epsilons()
cp = automat.get_automaton()
result = nfa_data().load_from_file(
"test_data/(2)pcre_get_nfa.nfa_data")
self.assertTrue(
sorted(cp.states.keys()) == sorted(result.states.keys()))
self.assertTrue(cp.alphabet == result.alphabet)
self.assertTrue(cp.start == result.start)
self.assertTrue(cp.final == result.final)
self.assertTrue(cp.transitions == result.transitions)
self.assertTrue(cp.Flags == result.Flags)
# 3) try second RE (move to next line)
parser = pcre_parser()
parser.set_text("/abc/\n/ABC/\n")
parser.next_line()
automat = b_Automaton()
automat._automaton = parser.get_nfa()
automat.remove_epsilons()
cp = automat.get_automaton()
result = nfa_data().load_from_file(
"test_data/(3)pcre_get_nfa.nfa_data")
self.assertTrue(
sorted(cp.states.keys()) == sorted(result.states.keys()))
self.assertTrue(cp.alphabet == result.alphabet)
self.assertTrue(cp.start == result.start)
self.assertTrue(cp.final == result.final)
self.assertTrue(cp.transitions == result.transitions)
self.assertTrue(cp.Flags == result.Flags)
# 4) basic counting constratin
parser = pcre_parser()
parser.set_text("/ab{5}c/")
automat = b_Automaton()
automat._automaton = parser.get_nfa()
automat.remove_epsilons()
cp = automat.get_automaton()
result = nfa_data().load_from_file(
"test_data/(4)pcre_get_nfa.nfa_data")
self.assertTrue(
sorted(cp.states.keys()) == sorted(result.states.keys()))
self.assertTrue(cp.alphabet == result.alphabet)
self.assertTrue(cp.start == result.start)
self.assertTrue(cp.final == result.final)
self.assertTrue(cp.transitions == result.transitions)
self.assertTrue(cp.Flags == result.Flags)
# 5) branch, iteration +, harder counting constraint
parser = pcre_parser()
parser.set_text("/a[bc]+d{2,3}/")
automat = b_Automaton()
automat._automaton = parser.get_nfa()
automat.remove_epsilons()
cp = automat.get_automaton()
result = nfa_data().load_from_file(
"test_data/(5)pcre_get_nfa.nfa_data")
self.assertTrue(
sorted(cp.states.keys()) == sorted(result.states.keys()))
self.assertTrue(cp.alphabet == result.alphabet)
self.assertTrue(cp.start == result.start)
self.assertTrue(cp.final == result.final)
self.assertTrue(cp.transitions == result.transitions)
self.assertTrue(cp.Flags == result.Flags)
# 6) basic counting constratin, use param create_cnt_constr = True
parser = pcre_parser(create_cnt_constr=True)
parser.set_text("/ab{5}c/")
automat = b_Automaton()
automat._automaton = parser.get_nfa()
automat.remove_epsilons()
cp = automat.get_automaton()
result = nfa_data().load_from_file(
"test_data/(6)pcre_get_nfa.nfa_data")
self.assertTrue(
sorted(cp.states.keys()) == sorted(result.states.keys()))
self.assertTrue(cp.alphabet == result.alphabet)
self.assertTrue(cp.start == result.start)
self.assertTrue(cp.final == result.final)
self.assertTrue(cp.transitions == result.transitions)
self.assertTrue(cp.Flags == result.Flags)
# 7) branch, iteration +, harder counting constraint,
# use param create_cnt_constr = True
parser = pcre_parser(create_cnt_constr=True)
parser.set_text("/a[bc]+d{2,3}/")
automat = b_Automaton()
automat._automaton = parser.get_nfa()
automat.remove_epsilons()
cp = automat.get_automaton()
result = nfa_data().load_from_file(
"test_data/(7)pcre_get_nfa.nfa_data")
self.assertTrue(
sorted(cp.states.keys()) == sorted(result.states.keys()))
self.assertTrue(cp.alphabet == result.alphabet)
self.assertTrue(cp.start == result.start)
self.assertTrue(cp.final == result.final)
self.assertTrue(cp.transitions == result.transitions)
self.assertTrue(cp.Flags == result.Flags)
# 8) concatenation, with create_eof_symbols = True, no $
parser = pcre_parser(create_eof_symbols=True)
parser.set_text("/first/")
automat = b_Automaton()
automat._automaton = parser.get_nfa()
automat.remove_epsilons()
cp = automat.get_automaton()
result = nfa_data().load_from_file(
"test_data/(1)pcre_get_nfa.nfa_data")
self.assertTrue(
sorted(cp.states.keys()) == sorted(result.states.keys()))
self.assertTrue(cp.alphabet == result.alphabet)
self.assertTrue(cp.start == result.start)
self.assertTrue(cp.final == result.final)
self.assertTrue(cp.transitions == result.transitions)
self.assertTrue(cp.Flags == result.Flags)
# 9) concatenation, with create_eof_symbols = True, $
parser = pcre_parser(create_eof_symbols=True)
parser.set_text("/first$/")
automat = b_Automaton()
automat._automaton = parser.get_nfa()
automat.remove_epsilons()
cp = automat.get_automaton()
result = nfa_data().load_from_file(
"test_data/(9)pcre_get_nfa.nfa_data")
self.assertTrue(
sorted(cp.states.keys()) == sorted(result.states.keys()))
self.assertTrue(cp.alphabet == result.alphabet)
self.assertTrue(cp.start == result.start)
self.assertTrue(cp.final == result.final)
self.assertTrue(cp.transitions == result.transitions)
self.assertTrue(cp.Flags == result.Flags)
# 10) branch, iteration +, harder counting constraint
# create_eof_symbols = True, create_cnt_constr = True
parser = pcre_parser(create_eof_symbols=True, create_cnt_constr=True)
parser.set_text("/a[bc]+d{2,3}$/")
automat = b_Automaton()
automat._automaton = parser.get_nfa()
automat.remove_epsilons()
cp = automat.get_automaton()
result = nfa_data().load_from_file(
"test_data/(10)pcre_get_nfa.nfa_data")
self.assertTrue(
sorted(cp.states.keys()) == sorted(result.states.keys()))
self.assertTrue(cp.alphabet == result.alphabet)
self.assertTrue(cp.start == result.start)
self.assertTrue(cp.final == result.final)
self.assertTrue(cp.transitions == result.transitions)
self.assertTrue(cp.Flags == result.Flags)
from netbench.pattern_match import sym_char, sym_char_class
from netbench.pattern_match import pcre_parser
from netbench.pattern_match.b_automaton import b_Automaton
from netbench.pattern_match.nfa_data import nfa_data
#from netbench.pattern_match.b_dfa import b_dfa
from phf_dfa import PHF_DFA
#FileName= "../../rules/Moduly/web-client.rules.pcre"
FileName = "../../rules/Moduly/web-php.rules.pcre"
parser = pcre_parser.pcre_parser()
parser.load_file(FileName)
print("Joining...")
N_Automaton = b_Automaton()
N_Automaton.create_from_nfa_data(parser.get_nfa())
#while (parser.next_line()):
# N_Automaton.join(parser.get_nfa())
N_Automaton.get_automaton().Show("test_NFA.dot")
print("Automata joined")
D_Automaton = PHF_DFA()
D_Automaton.create_from_nfa_data(N_Automaton.get_automaton())
print("Determinising...")
D_Automaton.determinise(states_limit=10000)
print("Minimising...")
D_Automaton.minimise() #Co to vlastne vypisuje?
D_Automaton.get_automaton().Show("test_min_dfa.dot")
D_Automaton.get_automaton().SaveToFile("temp_automaton")
def test_get_nfa(self):
"""get_nfa()"""
# If attribute _position < 0, check returning None.
parser = pcre_parser()
self.assertTrue(parser._position < 0)
self.assertTrue(parser.get_nfa() == None)
# Try method on a few regular expressions.
# The results obtained compare with the manually completed machines.
# (Recommend to compare after the elimination of epsilon transition)
# 1) concatenation
parser = pcre_parser()
parser.set_text("/first/")
automat = b_Automaton()
automat._automaton = parser.get_nfa()
automat.remove_epsilons()
cp = automat.get_automaton()
result = nfa_data().load_from_file("test_data/(1)pcre_get_nfa.nfa_data")
self.assertTrue(sorted(cp.states.keys()) == sorted(result.states.keys()))
self.assertTrue(cp.alphabet == result.alphabet)
self.assertTrue(cp.start == result.start)
self.assertTrue(cp.final == result.final)
self.assertTrue(cp.transitions == result.transitions)
self.assertTrue(cp.Flags == result.Flags)
# 2) branch (automat create char class), iteration *
parser = pcre_parser()
parser.set_text("/[ab]cd*/")
automat = b_Automaton()
automat._automaton = parser.get_nfa()
automat.remove_epsilons()
cp = automat.get_automaton()
result = nfa_data().load_from_file("test_data/(2)pcre_get_nfa.nfa_data")
self.assertTrue(sorted(cp.states.keys()) == sorted(result.states.keys()))
self.assertTrue(cp.alphabet == result.alphabet)
self.assertTrue(cp.start == result.start)
self.assertTrue(cp.final == result.final)
self.assertTrue(cp.transitions == result.transitions)
self.assertTrue(cp.Flags == result.Flags)
# 3) try second RE (move to next line)
parser = pcre_parser()
parser.set_text("/abc/\n/ABC/\n")
parser.next_line()
automat = b_Automaton()
automat._automaton = parser.get_nfa()
automat.remove_epsilons()
cp = automat.get_automaton()
result = nfa_data().load_from_file("test_data/(3)pcre_get_nfa.nfa_data")
self.assertTrue(sorted(cp.states.keys()) == sorted(result.states.keys()))
self.assertTrue(cp.alphabet == result.alphabet)
self.assertTrue(cp.start == result.start)
self.assertTrue(cp.final == result.final)
self.assertTrue(cp.transitions == result.transitions)
self.assertTrue(cp.Flags == result.Flags)
# 4) basic counting constratin
parser = pcre_parser()
parser.set_text("/ab{5}c/")
automat = b_Automaton()
automat._automaton = parser.get_nfa()
automat.remove_epsilons()
cp = automat.get_automaton()
result = nfa_data().load_from_file("test_data/(4)pcre_get_nfa.nfa_data")
self.assertTrue(sorted(cp.states.keys()) == sorted(result.states.keys()))
self.assertTrue(cp.alphabet == result.alphabet)
self.assertTrue(cp.start == result.start)
self.assertTrue(cp.final == result.final)
self.assertTrue(cp.transitions == result.transitions)
self.assertTrue(cp.Flags == result.Flags)
# 5) branch, iteration +, harder counting constraint
parser = pcre_parser()
parser.set_text("/a[bc]+d{2,3}/")
automat = b_Automaton()
automat._automaton = parser.get_nfa()
automat.remove_epsilons()
cp = automat.get_automaton()
result = nfa_data().load_from_file("test_data/(5)pcre_get_nfa.nfa_data")
self.assertTrue(sorted(cp.states.keys()) == sorted(result.states.keys()))
self.assertTrue(cp.alphabet == result.alphabet)
self.assertTrue(cp.start == result.start)
self.assertTrue(cp.final == result.final)
self.assertTrue(cp.transitions == result.transitions)
self.assertTrue(cp.Flags == result.Flags)
# 6) basic counting constratin, use param create_cnt_constr = True
parser = pcre_parser(create_cnt_constr = True)
parser.set_text("/ab{5}c/")
automat = b_Automaton()
automat._automaton = parser.get_nfa()
automat.remove_epsilons()
cp = automat.get_automaton()
result = nfa_data().load_from_file("test_data/(6)pcre_get_nfa.nfa_data")
self.assertTrue(sorted(cp.states.keys()) == sorted(result.states.keys()))
self.assertTrue(cp.alphabet == result.alphabet)
self.assertTrue(cp.start == result.start)
self.assertTrue(cp.final == result.final)
self.assertTrue(cp.transitions == result.transitions)
self.assertTrue(cp.Flags == result.Flags)
# 7) branch, iteration +, harder counting constraint,
# use param create_cnt_constr = True
parser = pcre_parser(create_cnt_constr = True)
parser.set_text("/a[bc]+d{2,3}/")
automat = b_Automaton()
automat._automaton = parser.get_nfa()
automat.remove_epsilons()
cp = automat.get_automaton()
result = nfa_data().load_from_file("test_data/(7)pcre_get_nfa.nfa_data")
self.assertTrue(sorted(cp.states.keys()) == sorted(result.states.keys()))
self.assertTrue(cp.alphabet == result.alphabet)
self.assertTrue(cp.start == result.start)
self.assertTrue(cp.final == result.final)
self.assertTrue(cp.transitions == result.transitions)
self.assertTrue(cp.Flags == result.Flags)
# 8) concatenation, with create_eof_symbols = True, no $
parser = pcre_parser(create_eof_symbols = True)
parser.set_text("/first/")
automat = b_Automaton()
automat._automaton = parser.get_nfa()
automat.remove_epsilons()
cp = automat.get_automaton()
result = nfa_data().load_from_file("test_data/(1)pcre_get_nfa.nfa_data")
self.assertTrue(sorted(cp.states.keys()) == sorted(result.states.keys()))
self.assertTrue(cp.alphabet == result.alphabet)
self.assertTrue(cp.start == result.start)
self.assertTrue(cp.final == result.final)
self.assertTrue(cp.transitions == result.transitions)
self.assertTrue(cp.Flags == result.Flags)
# 9) concatenation, with create_eof_symbols = True, $
parser = pcre_parser(create_eof_symbols = True)
parser.set_text("/first$/")
automat = b_Automaton()
automat._automaton = parser.get_nfa()
automat.remove_epsilons()
cp = automat.get_automaton()
result = nfa_data().load_from_file("test_data/(9)pcre_get_nfa.nfa_data")
self.assertTrue(sorted(cp.states.keys()) == sorted(result.states.keys()))
self.assertTrue(cp.alphabet == result.alphabet)
self.assertTrue(cp.start == result.start)
self.assertTrue(cp.final == result.final)
self.assertTrue(cp.transitions == result.transitions)
self.assertTrue(cp.Flags == result.Flags)
# 10) branch, iteration +, harder counting constraint
# create_eof_symbols = True, create_cnt_constr = True
parser = pcre_parser(create_eof_symbols = True, create_cnt_constr = True)
parser.set_text("/a[bc]+d{2,3}$/")
automat = b_Automaton()
automat._automaton = parser.get_nfa()
automat.remove_epsilons()
cp = automat.get_automaton()
result = nfa_data().load_from_file("test_data/(10)pcre_get_nfa.nfa_data")
self.assertTrue(sorted(cp.states.keys()) == sorted(result.states.keys()))
self.assertTrue(cp.alphabet == result.alphabet)
self.assertTrue(cp.start == result.start)
self.assertTrue(cp.final == result.final)
self.assertTrue(cp.transitions == result.transitions)
self.assertTrue(cp.Flags == result.Flags)
def _replace_length_restriction_with_a_closure(self, NFA):
"""
The first step in this construction replaces the length
restriction with a closure, and constructs the H-FA, with
the closure represented by a flag in the history buffer.
:param NFA: NFA
:type NFA: nfa_data
:returns: NFA without counting constraint
:rtype: nfa_data
"""
# identify counting transitions with exactly X counting
cnt_transitions = list()
for t in NFA.transitions:
if NFA.alphabet[t[1]].ctype == io_mapper["b_Sym_cnt_constr"]:
if NFA.alphabet[t[1]].m == NFA.alphabet[t[1]].n:
cnt_transitions.append(t)
# remove founded counting transtions
# and replace them with loop transitions
# and add epsilon tran. to next state
for t in cnt_transitions:
NFA.transitions.remove(t)
cnt_symbol = NFA.alphabet[t[1]]
self.flags_cnt[t[0]] = str(cnt_symbol.m)
NFA.transitions.add(
(t[0],
t[1],
t[0])
)
NFA.transitions.add(
(t[0],
-1,
t[2])
)
# replace cnt symbol at char or char class
# and add epsilon symbol into alphabet if does not exist
for t in cnt_transitions:
symbolID = t[1]
cnt_symbol = copy.deepcopy(NFA.alphabet[symbolID])
if cnt_symbol.ctype == io_mapper["b_Sym_cnt_constr"]:
if isinstance(cnt_symbol.symbol, str):
NFA.alphabet[symbolID] = b_Sym_char_class(
new_text = cnt_symbol._text,
charClass = set([cnt_symbol.symbol]),
new_id = cnt_symbol._id)
else :
NFA.alphabet[symbolID] = b_Sym_char_class(
new_text = cnt_symbol._text,
charClass = cnt_symbol.symbol,
new_id = cnt_symbol._id)
epsilonID = -1
if not epsilonID in NFA.alphabet:
NFA.alphabet[epsilonID] = b_Sym_char("Epsilon", "", -1)
# remove epsilons
aut = b_Automaton()
aut._automaton = NFA
aut.remove_epsilons()
return aut._automaton