def _test_search2(self):
re = "/a+b*c+d/\n/abc/"
delay_dfa = DELAY_DFA()
parser = pcre_parser()
parser.set_text(re)
delay_dfa.create_by_parser(parser)
delay_dfa.compute()
self.assertTrue(delay_dfa.get_compute())
aut = b_dfa()
parser = pcre_parser()
parser.set_text(re)
aut.create_by_parser(parser)
aut.compute()
self.assertEqual(delay_dfa.search("abcd"), aut.search("abcd"))
self.assertEqual(delay_dfa.search("abcd abc"), aut.search("abcd abc"))
self.assertEqual(delay_dfa.search("abc"), aut.search("abc"))
self.assertEqual(delay_dfa.search("acd"), aut.search("acd"))
self.assertEqual(delay_dfa.search("abd"), aut.search("abd"))
def _test_compute1(self):
delay_dfa = DELAY_DFA()
parser = pcre_parser()
parser.set_text("/^abcd/")
delay_dfa.create_by_parser(parser)
delay_dfa.compute()
self.assertTrue(delay_dfa.get_compute())
dfa = b_dfa()
dfa.create_by_parser(parser)
dfa.compute()
a = delay_dfa.get_automaton()
b = dfa.get_automaton()
l = len(b.alphabet.keys())
b.add_symbols(DEF_SYMBOLS("default", l))
self.assertEqual(a.states.keys(), b.states.keys())
self.assertEqual(a.start, b.start)
self.assertEqual(a.final, b.final)
self.assertEqual(a.alphabet, b.alphabet)
self.assertEqual(a.transitions, b.transitions)
self.assertTrue(a.Flags["Delay DFA"])
def _test_compute2(self):
delay_dfa = DELAY_DFA()
parser = pcre_parser()
parser.set_text("/^(a|b)+/")
delay_dfa.create_by_parser(parser)
delay_dfa.compute()
self.assertTrue(delay_dfa.get_compute())
a = delay_dfa.get_automaton()
b = nfa_data()
b.add_symbols(b_Sym_char("a", "a", 0))
b.add_symbols(b_Sym_char("b", "b", 1))
b.add_symbols(DEF_SYMBOLS("default", 2))
b.add_states(b_State(0, set()))
b.add_states(b_State(1, set([0])))
b.start = 0
b.final = set([1])
b.add_transitions((0, 0, 1))
b.add_transitions((0, 1, 1))
b.add_transitions((1, 2, 0))
self.assertEqual(a.states.keys(), b.states.keys())
self.assertEqual(a.start, b.start)
self.assertEqual(a.final, b.final)
self.assertEqual(a.alphabet, b.alphabet)
self.assertEqual(a.transitions, b.transitions)
self.assertTrue(a.Flags["Delay DFA"])
def _test_compute3(self):
# Get test directory
tdir = aux_func.getPatternMatchDir() + "/algorithms/delay_dfa/"
delay_dfa = DELAY_DFA()
nfaData = nfa_data().load_from_file(tdir +
"test_data/text_ddfa.nfa_data")
delay_dfa.create_from_nfa_data(nfaData)
delay_dfa.determinise()
delay_dfa.compute(False)
self.assertTrue(delay_dfa.get_compute())
a = delay_dfa.get_automaton()
b = nfa_data()
b.add_symbols(b_Sym_char("a", "a", 0))
b.add_symbols(b_Sym_char("b", "b", 1))
b.add_symbols(b_Sym_char("c", "c", 2))
b.add_symbols(b_Sym_char("d", "d", 3))
b.add_symbols(DEF_SYMBOLS("default", 4))
b.add_states(b_State(0, set()))
b.add_states(b_State(1, set([0])))
b.add_states(b_State(2, set()))
b.add_states(b_State(3, set([0])))
b.add_states(b_State(4, set([0])))
b.start = 0
b.final = set([1, 3, 4])
b.add_transitions((0, 2, 0))
b.add_transitions((0, 0, 1))
b.add_transitions((0, 1, 2))
b.add_transitions((0, 3, 3))
b.add_transitions((1, 4, 0))
b.add_transitions((2, 2, 4))
b.add_transitions((2, 4, 0))
b.add_transitions((3, 4, 0))
b.add_transitions((4, 4, 0))
self.assertEqual(a.states.keys(), b.states.keys())
self.assertEqual(a.start, b.start)
self.assertEqual(a.final, b.final)
self.assertEqual(a.alphabet, b.alphabet)
self.assertEqual(a.transitions, b.transitions)
self.assertTrue(a.Flags["Delay DFA"])
def _test_compute2(self):
delay_dfa = DELAY_DFA()
parser = pcre_parser()
parser.set_text("/^(a|b)+/")
delay_dfa.create_by_parser(parser)
delay_dfa.compute()
self.assertTrue(delay_dfa.get_compute())
a = delay_dfa.get_automaton()
b = nfa_data()
b.add_symbols(b_Sym_char("a","a",0))
b.add_symbols(b_Sym_char("b","b",1))
b.add_symbols(DEF_SYMBOLS("default", 2))
b.add_states(b_State(0,set()))
b.add_states(b_State(1,set([0])))
b.start = 0
b.final = set([1])
b.add_transitions( (0,0,1) )
b.add_transitions( (0,1,1) )
b.add_transitions( (1,2,0) )
self.assertEqual(a.states.keys(), b.states.keys())
self.assertEqual(a.start, b.start)
self.assertEqual(a.final, b.final)
self.assertEqual(a.alphabet, b.alphabet)
self.assertEqual(a.transitions, b.transitions)
self.assertTrue(a.Flags["Delay DFA"])
def _test_compute1(self):
delay_dfa = DELAY_DFA()
parser = pcre_parser()
parser.set_text("/^abcd/")
delay_dfa.create_by_parser(parser)
delay_dfa.compute()
self.assertTrue(delay_dfa.get_compute())
dfa = b_dfa()
dfa.create_by_parser(parser)
dfa.compute()
a = delay_dfa.get_automaton()
b = dfa.get_automaton()
l = len(b.alphabet.keys())
b.add_symbols(DEF_SYMBOLS("default", l))
self.assertEqual(a.states.keys(), b.states.keys())
self.assertEqual(a.start, b.start)
self.assertEqual(a.final, b.final)
self.assertEqual(a.alphabet, b.alphabet)
self.assertEqual(a.transitions, b.transitions)
self.assertTrue(a.Flags["Delay DFA"])
def get_ddfa(ruleset):
"""
Generate number of states, transitions and consumed memory for DDFA.
"""
# Create parser - use default parser
po = parser.parser()
# Parse input file
po.load_file(ruleset)
# Create Delay DFA object
DelayDfa = DELAY_DFA()
# Make automaton from RE which was in input file
DelayDfa.create_by_parser(po)
# Make Delay DFA
# Resolve alphabet
DelayDfa.resolve_alphabet()
# Create Delay DFA
DelayDfa.compute()
# Return experimental results
return ["Delay DFA", DelayDfa.get_state_num(), DelayDfa.get_trans_num(), DelayDfa.report_memory_optimal(), DelayDfa.report_memory_naive()]
def _test_compute3(self):
# Get test directory
tdir = aux_func.getPatternMatchDir() + "/algorithms/delay_dfa/"
delay_dfa = DELAY_DFA()
nfaData = nfa_data().load_from_file(tdir + "test_data/text_ddfa.nfa_data")
delay_dfa.create_from_nfa_data(nfaData)
delay_dfa.determinise()
delay_dfa.compute(False)
self.assertTrue(delay_dfa.get_compute())
a = delay_dfa.get_automaton()
b = nfa_data()
b.add_symbols(b_Sym_char("a","a",0))
b.add_symbols(b_Sym_char("b","b",1))
b.add_symbols(b_Sym_char("c","c",2))
b.add_symbols(b_Sym_char("d","d",3))
b.add_symbols(DEF_SYMBOLS("default", 4))
b.add_states(b_State(0,set()))
b.add_states(b_State(1,set([0])))
b.add_states(b_State(2,set()))
b.add_states(b_State(3,set([0])))
b.add_states(b_State(4,set([0])))
b.start = 0
b.final = set([1,3,4])
b.add_transitions( (0,2,0) )
b.add_transitions( (0,0,1) )
b.add_transitions( (0,1,2) )
b.add_transitions( (0,3,3) )
b.add_transitions( (1,4,0) )
b.add_transitions( (2,2,4) )
b.add_transitions( (2,4,0) )
b.add_transitions( (3,4,0) )
b.add_transitions( (4,4,0) )
self.assertEqual(a.states.keys(), b.states.keys())
self.assertEqual(a.start, b.start)
self.assertEqual(a.final, b.final)
self.assertEqual(a.alphabet, b.alphabet)
self.assertEqual(a.transitions, b.transitions)
self.assertTrue(a.Flags["Delay DFA"])
def _test_search1(self):
re = "/a+b*c.*a|bc+/"
delay_dfa = DELAY_DFA()
parser = pcre_parser()
parser.set_text(re)
delay_dfa.create_by_parser(parser)
delay_dfa.compute()
self.assertTrue(delay_dfa.get_compute())
aut = b_dfa()
parser = pcre_parser()
parser.set_text(re)
aut.create_by_parser(parser)
self.assertEqual(delay_dfa.search("ac123ac"), aut.search("ac123ac"))
self.assertEqual(delay_dfa.search("aacac"), aut.search("aacac"))
self.assertEqual(delay_dfa.search("abbb"), aut.search("abbb"))
def _test_search1(self):
re = "/a+b*c.*a|bc+/"
delay_dfa = DELAY_DFA()
parser = pcre_parser()
parser.set_text(re)
delay_dfa.create_by_parser(parser)
delay_dfa.compute()
self.assertTrue(delay_dfa.get_compute())
aut = b_dfa()
parser = pcre_parser()
parser.set_text(re)
aut.create_by_parser(parser)
self.assertEqual(delay_dfa.search("ac123ac"), aut.search("ac123ac"))
self.assertEqual(delay_dfa.search("aacac"), aut.search("aacac"))
self.assertEqual(delay_dfa.search("abbb"), aut.search("abbb"))
def get_ddfa(ruleset):
"""
Generate number of states, transitions and consumed memory for DDFA.
"""
# Create parser - use default parser
po = parser.parser()
# Parse input file
po.load_file(ruleset)
# Create Delay DFA object
DelayDfa = DELAY_DFA()
# Make automaton from RE which was in input file
DelayDfa.create_by_parser(po)
# Make Delay DFA
# Resolve alphabet
DelayDfa.resolve_alphabet()
# Create Delay DFA
DelayDfa.compute()
# Return experimental results
return [
"Delay DFA",
DelayDfa.get_state_num(),
DelayDfa.get_trans_num(),
DelayDfa.report_memory_optimal(),
DelayDfa.report_memory_naive()
]
def test_get_default_trans_num(self):
"""get_default_trans_num"""
#Tests with regular expressions from test_compute
delay_dfa1 = DELAY_DFA()
parser = pcre_parser()
parser.set_text("/^abcd/")
delay_dfa1.create_by_parser(parser)
delay_dfa1.compute()
self.assertTrue(delay_dfa1.get_compute())
delay_dfa2 = DELAY_DFA()
parser = pcre_parser()
parser.set_text("/^(a|b)+/")
delay_dfa2.create_by_parser(parser)
delay_dfa2.compute()
self.assertTrue(delay_dfa2.get_compute())
delay_dfa3 = DELAY_DFA()
# Get test directory
tdir = aux_func.getPatternMatchDir() + "/algorithms/delay_dfa/"
nfaData = nfa_data().load_from_file(tdir + "test_data/text_ddfa.nfa_data")
delay_dfa3.create_from_nfa_data(nfaData)
delay_dfa3.determinise()
delay_dfa3.compute(False)
self.assertTrue(delay_dfa3.get_compute())
self.assertEqual(delay_dfa1.get_default_trans_num(),0)
self.assertEqual(delay_dfa2.get_default_trans_num(),1)
self.assertEqual(delay_dfa3.get_default_trans_num(),4)
def test_set_bound(self):
"""set_bound()"""
#Tests manual control of the number of consecutive default
#transitions in larger automaton
re = "/a+b*c+/\n/b+c*d+/\n/c+d*e/"
delay_dfa = DELAY_DFA()
parser = pcre_parser()
parser.set_text(re)
delay_dfa.create_by_parser(parser)
delay_dfa.set_bound(0)
delay_dfa.compute()
self.assertTrue(delay_dfa.get_compute())
a = delay_dfa.get_automaton()
lenght1 = self._get_length_default_path(a)
delay_dfa = DELAY_DFA()
parser = pcre_parser()
parser.set_text(re)
delay_dfa.create_by_parser(parser)
delay_dfa.set_bound(1)
delay_dfa.compute()
self.assertTrue(delay_dfa.get_compute())
a = delay_dfa.get_automaton()
lenght2 = self._get_length_default_path(a)
delay_dfa = DELAY_DFA()
parser = pcre_parser()
parser.set_text(re)
delay_dfa.create_by_parser(parser)
delay_dfa.set_bound(2)
delay_dfa.compute()
self.assertTrue(delay_dfa.get_compute())
a = delay_dfa.get_automaton()
lenght3 = self._get_length_default_path(a)
delay_dfa = DELAY_DFA()
parser = pcre_parser()
parser.set_text(re)
delay_dfa.create_by_parser(parser)
delay_dfa.set_bound(3)
delay_dfa.compute()
self.assertTrue(delay_dfa.get_compute())
a = delay_dfa.get_automaton()
lenght4 = self._get_length_default_path(a)
self.assertEqual(lenght1, 3)
self.assertEqual(lenght2, 1)
self.assertEqual(lenght3, 2)
self.assertEqual(lenght4, 3)
def _test_search3(self):
re = "/a+b*c+/\n/b+c*d+/\n/c+d*e/"
delay_dfa = DELAY_DFA()
parser = pcre_parser()
parser.set_text(re)
delay_dfa.create_by_parser(parser)
delay_dfa.compute()
self.assertTrue(delay_dfa.get_compute())
aut = b_dfa()
parser = pcre_parser()
parser.set_text(re)
aut.create_by_parser(parser)
aut.compute()
self.assertEqual(delay_dfa.search("abcd"), aut.search("abcd"))
self.assertEqual(delay_dfa.search("abcd abc"), aut.search("abcd abc"))
self.assertEqual(delay_dfa.search("gabc"), aut.search("gabc"))
self.assertEqual(delay_dfa.search("ac bd ce"), aut.search("ac bd ce"))
self.assertEqual(delay_dfa.search("ab123a bcd cde"), aut.search("ab123a bcd cde"))
self.assertEqual(delay_dfa.search("bce"), aut.search("bce"))
self.assertEqual(delay_dfa.search("abe"), aut.search("abe"))
def _test_search2(self):
re = "/a+b*c+d/\n/abc/"
delay_dfa = DELAY_DFA()
parser = pcre_parser()
parser.set_text(re)
delay_dfa.create_by_parser(parser)
delay_dfa.compute()
self.assertTrue(delay_dfa.get_compute())
aut = b_dfa()
parser = pcre_parser()
parser.set_text(re)
aut.create_by_parser(parser)
aut.compute()
self.assertEqual(delay_dfa.search("abcd"), aut.search("abcd"))
self.assertEqual(delay_dfa.search("abcd abc"), aut.search("abcd abc"))
self.assertEqual(delay_dfa.search("abc"), aut.search("abc"))
self.assertEqual(delay_dfa.search("acd"), aut.search("acd"))
self.assertEqual(delay_dfa.search("abd"), aut.search("abd"))
def _test_search3(self):
re = "/a+b*c+/\n/b+c*d+/\n/c+d*e/"
delay_dfa = DELAY_DFA()
parser = pcre_parser()
parser.set_text(re)
delay_dfa.create_by_parser(parser)
delay_dfa.compute()
self.assertTrue(delay_dfa.get_compute())
aut = b_dfa()
parser = pcre_parser()
parser.set_text(re)
aut.create_by_parser(parser)
aut.compute()
self.assertEqual(delay_dfa.search("abcd"), aut.search("abcd"))
self.assertEqual(delay_dfa.search("abcd abc"), aut.search("abcd abc"))
self.assertEqual(delay_dfa.search("gabc"), aut.search("gabc"))
self.assertEqual(delay_dfa.search("ac bd ce"), aut.search("ac bd ce"))
self.assertEqual(delay_dfa.search("ab123a bcd cde"),
aut.search("ab123a bcd cde"))
self.assertEqual(delay_dfa.search("bce"), aut.search("bce"))
self.assertEqual(delay_dfa.search("abe"), aut.search("abe"))
def test_set_bound(self):
"""set_bound()"""
#Tests manual control of the number of consecutive default
#transitions in larger automaton
re = "/a+b*c+/\n/b+c*d+/\n/c+d*e/"
delay_dfa = DELAY_DFA()
parser = pcre_parser()
parser.set_text(re)
delay_dfa.create_by_parser(parser)
delay_dfa.set_bound(0)
delay_dfa.compute()
self.assertTrue(delay_dfa.get_compute())
a = delay_dfa.get_automaton()
lenght1 = self._get_length_default_path(a)
delay_dfa = DELAY_DFA()
parser = pcre_parser()
parser.set_text(re)
delay_dfa.create_by_parser(parser)
delay_dfa.set_bound(1)
delay_dfa.compute()
self.assertTrue(delay_dfa.get_compute())
a = delay_dfa.get_automaton()
lenght2 = self._get_length_default_path(a)
delay_dfa = DELAY_DFA()
parser = pcre_parser()
parser.set_text(re)
delay_dfa.create_by_parser(parser)
delay_dfa.set_bound(2)
delay_dfa.compute()
self.assertTrue(delay_dfa.get_compute())
a = delay_dfa.get_automaton()
lenght3 = self._get_length_default_path(a)
delay_dfa = DELAY_DFA()
parser = pcre_parser()
parser.set_text(re)
delay_dfa.create_by_parser(parser)
delay_dfa.set_bound(3)
delay_dfa.compute()
self.assertTrue(delay_dfa.get_compute())
a = delay_dfa.get_automaton()
lenght4 = self._get_length_default_path(a)
self.assertEqual(lenght1, 3)
self.assertEqual(lenght2, 1)
self.assertEqual(lenght3, 2)
self.assertEqual(lenght4, 3)
def test_get_default_trans_num(self):
"""get_default_trans_num"""
#Tests with regular expressions from test_compute
delay_dfa1 = DELAY_DFA()
parser = pcre_parser()
parser.set_text("/^abcd/")
delay_dfa1.create_by_parser(parser)
delay_dfa1.compute()
self.assertTrue(delay_dfa1.get_compute())
delay_dfa2 = DELAY_DFA()
parser = pcre_parser()
parser.set_text("/^(a|b)+/")
delay_dfa2.create_by_parser(parser)
delay_dfa2.compute()
self.assertTrue(delay_dfa2.get_compute())
delay_dfa3 = DELAY_DFA()
# Get test directory
tdir = aux_func.getPatternMatchDir() + "/algorithms/delay_dfa/"
nfaData = nfa_data().load_from_file(tdir +
"test_data/text_ddfa.nfa_data")
delay_dfa3.create_from_nfa_data(nfaData)
delay_dfa3.determinise()
delay_dfa3.compute(False)
self.assertTrue(delay_dfa3.get_compute())
self.assertEqual(delay_dfa1.get_default_trans_num(), 0)
self.assertEqual(delay_dfa2.get_default_trans_num(), 1)
self.assertEqual(delay_dfa3.get_default_trans_num(), 4)