def test_deserialisationGroups(self):
print "start"
symbols = []
nbSymbol = random.randint(2, 50)
for iSymbol in range(0, nbSymbol):
# We create 6 messages of 2 group
originalSymbol = Symbol(str(uuid.uuid4()), "TestSymbol", None)
# group1
message1 = RawMessage(str(uuid.uuid4()), str(time.time()), TypeConvertor.stringToNetzobRaw("bonjour " + self.generateRandomString(20, 30) + " comment vas-tu ?"))
message2 = RawMessage(str(uuid.uuid4()), str(time.time()), TypeConvertor.stringToNetzobRaw("bonjour " + self.generateRandomString(20, 30) + " comment vas-tu ?"))
message3 = RawMessage(str(uuid.uuid4()), str(time.time()), TypeConvertor.stringToNetzobRaw("bonjour " + self.generateRandomString(20, 30) + " comment vas-tu ?"))
# group2
message4 = RawMessage(str(uuid.uuid4()), str(time.time()), TypeConvertor.stringToNetzobRaw("salut à toi " + self.generateRandomString(10, 15) + " what's up ?"))
message5 = RawMessage(str(uuid.uuid4()), str(time.time()), TypeConvertor.stringToNetzobRaw("salut à toi " + self.generateRandomString(10, 15) + " what's up ?"))
message6 = RawMessage(str(uuid.uuid4()), str(time.time()), TypeConvertor.stringToNetzobRaw("salut à toi " + self.generateRandomString(10, 15) + " what's up ?"))
originalSymbol.addMessage(message1)
originalSymbol.addMessage(message2)
originalSymbol.addMessage(message3)
originalSymbol.addMessage(message4)
originalSymbol.addMessage(message5)
originalSymbol.addMessage(message6)
symbols.append(originalSymbol)
# Start the clustering
clusteringSolution = UPGMA(None, [originalSymbol], True, 100, 90, True)
result = clusteringSolution.deserializeGroups(symbols)
self.assertEqual(result, len(symbols))
def test_deserialisationGroups(self):
print("start")
symbols = []
nbSymbol = random.randint(2, 50)
for iSymbol in range(0, nbSymbol):
# We create 6 messages of 2 group
originalSymbol = Symbol(str(uuid.uuid4()), "TestSymbol", None)
# group1
message1 = RawMessage(str(uuid.uuid4()), str(time.time()), TypeConvertor.stringToNetzobRaw("bonjour " + self.generateRandomString(20, 30) + " comment vas-tu ?"))
message2 = RawMessage(str(uuid.uuid4()), str(time.time()), TypeConvertor.stringToNetzobRaw("bonjour " + self.generateRandomString(20, 30) + " comment vas-tu ?"))
message3 = RawMessage(str(uuid.uuid4()), str(time.time()), TypeConvertor.stringToNetzobRaw("bonjour " + self.generateRandomString(20, 30) + " comment vas-tu ?"))
# group2
message4 = RawMessage(str(uuid.uuid4()), str(time.time()), TypeConvertor.stringToNetzobRaw("salut à toi " + self.generateRandomString(10, 15) + " what's up ?"))
message5 = RawMessage(str(uuid.uuid4()), str(time.time()), TypeConvertor.stringToNetzobRaw("salut à toi " + self.generateRandomString(10, 15) + " what's up ?"))
message6 = RawMessage(str(uuid.uuid4()), str(time.time()), TypeConvertor.stringToNetzobRaw("salut à toi " + self.generateRandomString(10, 15) + " what's up ?"))
originalSymbol.addMessage(message1)
originalSymbol.addMessage(message2)
originalSymbol.addMessage(message3)
originalSymbol.addMessage(message4)
originalSymbol.addMessage(message5)
originalSymbol.addMessage(message6)
symbols.append(originalSymbol)
# Start the clustering
clusteringSolution = UPGMA(None, [originalSymbol], True, 100, 90, True)
result = clusteringSolution.deserializeGroups(symbols)
self.assertEqual(result, len(symbols))
def test_executingClusteringWithOrphanReduction(self):
# We create 6 messages of 2 group
# group1
originalSymbol1 = Symbol(uuid.uuid4(), "TestSymbol", None)
message1 = RawMessage(uuid.uuid4(), str(time.time()), TypeConvertor.stringToNetzobRaw("bonjour " + self.generateRandomString(200, 1000)))
originalSymbol1.addMessage(message1)
originalSymbol2 = Symbol(uuid.uuid4(), "TestSymbol2", None)
message2 = RawMessage(uuid.uuid4(), str(time.time()), TypeConvertor.stringToNetzobRaw("bonjour " + self.generateRandomString(200, 1000)))
originalSymbol2.addMessage(message2)
originalSymbol3 = Symbol(uuid.uuid4(), "TestSymbol3", None)
message3 = RawMessage(uuid.uuid4(), str(time.time()), TypeConvertor.stringToNetzobRaw("bonjour " + self.generateRandomString(200, 1000)))
originalSymbol3.addMessage(message3)
# group2
originalSymbol4 = Symbol(uuid.uuid4(), "TestSymbol4", None)
message4 = RawMessage(uuid.uuid4(), str(time.time()), TypeConvertor.stringToNetzobRaw("salut " + self.generateRandomString(200, 1000)))
originalSymbol4.addMessage(message4)
originalSymbol5 = Symbol(uuid.uuid4(), "TestSymbol5", None)
message5 = RawMessage(uuid.uuid4(), str(time.time()), TypeConvertor.stringToNetzobRaw("salut " + self.generateRandomString(200, 1000)))
originalSymbol5.addMessage(message5)
originalSymbol6 = Symbol(uuid.uuid4(), "TestSymbol6", None)
message6 = RawMessage(uuid.uuid4(), str(time.time()), TypeConvertor.stringToNetzobRaw("salut " + self.generateRandomString(200, 1000)))
originalSymbol6.addMessage(message6)
symbols = [originalSymbol1, originalSymbol2, originalSymbol3, originalSymbol4, originalSymbol5, originalSymbol6]
# Start the clustering
clusteringSolution = UPGMA(None, symbols, True, 100, 80, True, Format.ASCII)
resultBeforeOrphan = clusteringSolution.executeClustering()
resultAfterOrphan = clusteringSolution.executeOrphanReduction()
if (len(resultAfterOrphan) < len(resultBeforeOrphan)) :
print "Before Orphan Reduction : "
for symbol in resultBeforeOrphan :
print "Symbol : " + str(symbol.getName())
for m in symbol.getMessages() :
print " + " + str(m.getStringData())
print "After Orphan Reduction : "
for symbol in resultAfterOrphan :
print "Symbol : " + str(symbol.getName())
for m in symbol.getMessages() :
print " + " + str(m.getStringData())
self.assertGreaterEqual(len(resultBeforeOrphan), len(resultAfterOrphan))
def test_executingClusteringWithOrphanReduction(self):
# We create 6 messages of 2 group
# group1
originalSymbol1 = Symbol(str(uuid.uuid4()), "TestSymbol", None)
message1 = RawMessage(str(uuid.uuid4()), str(time.time()), TypeConvertor.stringToNetzobRaw("bonjour " + self.generateRandomString(200, 1000)))
originalSymbol1.addMessage(message1)
originalSymbol2 = Symbol(str(uuid.uuid4()), "TestSymbol2", None)
message2 = RawMessage(str(uuid.uuid4()), str(time.time()), TypeConvertor.stringToNetzobRaw("bonjour " + self.generateRandomString(200, 1000)))
originalSymbol2.addMessage(message2)
originalSymbol3 = Symbol(str(uuid.uuid4()), "TestSymbol3", None)
message3 = RawMessage(str(uuid.uuid4()), str(time.time()), TypeConvertor.stringToNetzobRaw("bonjour " + self.generateRandomString(200, 1000)))
originalSymbol3.addMessage(message3)
# group2
originalSymbol4 = Symbol(str(uuid.uuid4()), "TestSymbol4", None)
message4 = RawMessage(str(uuid.uuid4()), str(time.time()), TypeConvertor.stringToNetzobRaw("salut " + self.generateRandomString(200, 1000)))
originalSymbol4.addMessage(message4)
originalSymbol5 = Symbol(str(uuid.uuid4()), "TestSymbol5", None)
message5 = RawMessage(str(uuid.uuid4()), str(time.time()), TypeConvertor.stringToNetzobRaw("salut " + self.generateRandomString(200, 1000)))
originalSymbol5.addMessage(message5)
originalSymbol6 = Symbol(str(uuid.uuid4()), "TestSymbol6", None)
message6 = RawMessage(str(uuid.uuid4()), str(time.time()), TypeConvertor.stringToNetzobRaw("salut " + self.generateRandomString(200, 1000)))
originalSymbol6.addMessage(message6)
symbols = [originalSymbol1, originalSymbol2, originalSymbol3, originalSymbol4, originalSymbol5, originalSymbol6]
# Start the clustering
clusteringSolution = UPGMA(None, symbols, True, 100, 80, True, Format.ASCII)
resultBeforeOrphan = clusteringSolution.executeClustering()
resultAfterOrphan = clusteringSolution.executeOrphanReduction()
if (len(resultAfterOrphan) < len(resultBeforeOrphan)):
print "Before Orphan Reduction: "
for symbol in resultBeforeOrphan:
print "Symbol: " + str(symbol.getName())
for m in symbol.getMessages():
print " + " + str(m.getStringData())
print "After Orphan Reduction: "
for symbol in resultAfterOrphan:
print "Symbol: " + str(symbol.getName())
for m in symbol.getMessages():
print " + " + str(m.getStringData())
self.assertGreaterEqual(len(resultBeforeOrphan), len(resultAfterOrphan))
def test_executingClustering(self):
# We create 6 messages of 2 group
# group1
originalSymbol1 = Symbol(uuid.uuid4(), "TestSymbol", None)
message1 = RawMessage(uuid.uuid4(), str(time.time()), TypeConvertor.stringToNetzobRaw("bonjour " + self.generateRandomString(20, 30) + " comment vas-tu ?"))
originalSymbol1.addMessage(message1)
originalSymbol2 = Symbol(uuid.uuid4(), "TestSymbol2", None)
message2 = RawMessage(uuid.uuid4(), str(time.time()), TypeConvertor.stringToNetzobRaw("bonjour " + self.generateRandomString(20, 30) + " comment vas-tu ?"))
originalSymbol2.addMessage(message2)
originalSymbol3 = Symbol(uuid.uuid4(), "TestSymbol3", None)
message3 = RawMessage(uuid.uuid4(), str(time.time()), TypeConvertor.stringToNetzobRaw("bonjour " + self.generateRandomString(20, 30) + " comment vas-tu ?"))
originalSymbol3.addMessage(message3)
# group2
originalSymbol4 = Symbol(uuid.uuid4(), "TestSymbol4", None)
message4 = RawMessage(uuid.uuid4(), str(time.time()), TypeConvertor.stringToNetzobRaw("salut à toi " + self.generateRandomString(10, 15) + " what's up ?"))
originalSymbol4.addMessage(message4)
originalSymbol5 = Symbol(uuid.uuid4(), "TestSymbol5", None)
message5 = RawMessage(uuid.uuid4(), str(time.time()), TypeConvertor.stringToNetzobRaw("salut à toi " + self.generateRandomString(10, 15) + " what's up ?"))
originalSymbol5.addMessage(message5)
originalSymbol6 = Symbol(uuid.uuid4(), "TestSymbol6", None)
message6 = RawMessage(uuid.uuid4(), str(time.time()), TypeConvertor.stringToNetzobRaw("salut à toi " + self.generateRandomString(10, 15) + " what's up ?"))
originalSymbol6.addMessage(message6)
symbols = [originalSymbol1, originalSymbol2, originalSymbol3, originalSymbol4, originalSymbol5, originalSymbol6]
# Start the clustering
clusteringSolution = UPGMA(None, symbols, True, 100, 90, True, Format.ASCII)
result = clusteringSolution.executeClustering()
for symbol in result :
print "Symbol : " + str(symbol.getName())
for m in symbol.getMessages() :
print " + " + str(m.getStringData())
def test_executingClustering(self):
# We create 6 messages of 2 group
# group1
originalSymbol1 = Symbol(str(uuid.uuid4()), "TestSymbol", None)
message1 = RawMessage(str(uuid.uuid4()), str(time.time()), TypeConvertor.stringToNetzobRaw("bonjour " + self.generateRandomString(20, 30) + " comment vas-tu ?"))
originalSymbol1.addMessage(message1)
originalSymbol2 = Symbol(str(uuid.uuid4()), "TestSymbol2", None)
message2 = RawMessage(str(uuid.uuid4()), str(time.time()), TypeConvertor.stringToNetzobRaw("bonjour " + self.generateRandomString(20, 30) + " comment vas-tu ?"))
originalSymbol2.addMessage(message2)
originalSymbol3 = Symbol(str(uuid.uuid4()), "TestSymbol3", None)
message3 = RawMessage(str(uuid.uuid4()), str(time.time()), TypeConvertor.stringToNetzobRaw("bonjour " + self.generateRandomString(20, 30) + " comment vas-tu ?"))
originalSymbol3.addMessage(message3)
# group2
originalSymbol4 = Symbol(str(uuid.uuid4()), "TestSymbol4", None)
message4 = RawMessage(str(uuid.uuid4()), str(time.time()), TypeConvertor.stringToNetzobRaw("salut à toi " + self.generateRandomString(10, 15) + " what's up ?"))
originalSymbol4.addMessage(message4)
originalSymbol5 = Symbol(str(uuid.uuid4()), "TestSymbol5", None)
message5 = RawMessage(str(uuid.uuid4()), str(time.time()), TypeConvertor.stringToNetzobRaw("salut à toi " + self.generateRandomString(10, 15) + " what's up ?"))
originalSymbol5.addMessage(message5)
originalSymbol6 = Symbol(str(uuid.uuid4()), "TestSymbol6", None)
message6 = RawMessage(str(uuid.uuid4()), str(time.time()), TypeConvertor.stringToNetzobRaw("salut à toi " + self.generateRandomString(10, 15) + " what's up ?"))
originalSymbol6.addMessage(message6)
symbols = [originalSymbol1, originalSymbol2, originalSymbol3, originalSymbol4, originalSymbol5, originalSymbol6]
# Start the clustering
clusteringSolution = UPGMA(None, symbols, True, 100, 90, True, Format.ASCII)
result = clusteringSolution.executeClustering()
for symbol in result:
print "Symbol: " + str(symbol.getName())
for m in symbol.getMessages():
print " + " + str(m.getStringData())
def alignSymbols(self, symbols, project):
from netzob.Inference.Vocabulary.Alignment.UPGMA import UPGMA
self.result = []
preResults = []
# First we add in results, the symbols which wont be aligned
for symbol in project.getVocabulary().getSymbols():
found = False
for s in symbols:
if str(symbol.getID()) == str(s.getID()):
found = True
if not found:
logging.debug("Symbol " + str(symbol.getName()) + "[" + str(symbol.getID()) + "]] wont be aligned")
preResults.append(symbol)
# Then we retrieve all the parameters of the CLUSTERING / ALIGNMENT
defaultFormat = project.getConfiguration().getVocabularyInferenceParameter(ProjectConfiguration.VOCABULARY_GLOBAL_FORMAT)
nbIteration = project.getConfiguration().getVocabularyInferenceParameter(ProjectConfiguration.VOCABULARY_NB_ITERATION)
minEquivalence = project.getConfiguration().getVocabularyInferenceParameter(ProjectConfiguration.VOCABULARY_EQUIVALENCE_THRESHOLD)
doInternalSlick = project.getConfiguration().getVocabularyInferenceParameter(ProjectConfiguration.VOCABULARY_DO_INTERNAL_SLICK)
doOrphanReduction = project.getConfiguration().getVocabularyInferenceParameter(ProjectConfiguration.VOCABULARY_ORPHAN_REDUCTION)
# We try to cluster each symbol
listEqu = [] # list of thresholds recorded
emptySymbols = [] # list of all empty symbols
clusteringSolution = UPGMA(project, symbols, True, nbIteration, minEquivalence, doInternalSlick, defaultFormat, self.unitSize, self.cb_status)
t1 = time.time()
self.result = clusteringSolution.executeClustering()
# We optionally handle orphans
if doOrphanReduction:
self.result = clusteringSolution.executeOrphanReduction()
self.result.extend(preResults)
self.result.extend(emptySymbols) # Add the empty symbols (To discuss: can we delete them before?)
logging.info("Time of clustering : " + str(time.time() - t1))