convHIST + cleanup

advisors/detection/Decor/Decor_blob_replication.py

@@ -0,0 +1,100 @@
+from __future__ import division
+
+import csv
+import sys
+
+sys.path.insert(0, '../')
+import evaluate
+
+''' This file is just used to obtain DECOR detection results on the test set'''
+
+def blob(systemName):
+	decorMetricsFile = '../../results/Decor/Blob/' + systemName + '.csv'
+
+	smells = []
+	with open(decorMetricsFile, 'rb') as csvfile:
+		reader = csv.DictReader(csvfile, delimiter=';')
+		for row in reader:
+			nmdNad      = float(row['NMD+NAD'])
+			nmdNadBound = float(row['nmdNadBound'])
+			lcom5       = float(row['LCOM5'])
+			lcom5Bound  = float(row['lcom5Bound'])
+			cc          = int(row['ControllerClass'] )
+			nbDataClass = int(row['nbDataClass'])
+
+			if nbDataClass > 0:
+				if nmdNad/nmdNadBound > 1:
+					smells.append(row['ClassName'])
+
+				elif lcom5/lcom5Bound > 1:
+					smells.append(row['ClassName']) 
+
+				elif cc == 1:
+					smells.append(row['ClassName'])
+
+	return smells
+
+
+
+
+def test(systemName):
+	print(systemName)
+	trueFile = '../../../data/labels/Blob/hand_validated/' + systemName + '.csv'
+
+	# Get Smells occurences
+	true = []
+	with open(trueFile, 'rb') as csvfile:
+		reader = csv.reader(csvfile, delimiter=';')
+
+		for row in reader:
+			true.append(row[0])
+
+	# Get classes detected as God Classes
+	detected = blob(systemName)
+
+	pre = evaluate.precision(detected, true)
+	rec = evaluate.recall(detected, true)
+	f_m = evaluate.f_mesure(detected, true)
+
+	print('Precision :' + str(pre))
+	print('Recall    :' + str(rec))
+	print('F-Mesure  :' + str(f_m))
+
+	return f_m
+
+def overall(systems):
+
+	overallSmells = []
+	overallTrue = []
+	for system in systems:
+		trueFile = '../../../data/labels/Blob/hand_validated/' + system + '.csv'
+
+		with open(trueFile, 'rb') as csvfile:
+			reader = csv.reader(csvfile, delimiter=';')
+
+			for row in reader:
+				overallTrue.append(row[0])
+
+		smells = blob(system)
+
+		for s in smells:
+			overallSmells.append(s)
+
+	pre = evaluate.precision(overallSmells, overallTrue)
+	rec = evaluate.recall(overallSmells, overallTrue)
+	f_m = evaluate.f_mesure(overallSmells, overallTrue)
+
+	print('Precision :' + str(pre))
+	print('Recall    :' + str(rec))
+	print('F-Mesure  :' + str(f_m))
+
+
+if __name__ == "__main__":
+	systems = ['apache-tomcat', 'jedit', 'android-platform-support']
+
+	for system in systems:
+		r = test(system)
+		print("")
+
+	print("OVERALL")
+	overall(systems)

advisors/detection/Hist/hist_blob_replication.py

@@ -0,0 +1,182 @@
+from __future__ import division
+
+import numpy as np
+import matplotlib.pyplot as plt
+
+import csv
+import progressbar
+import sys
+
+sys.path.insert(0, '../')
+import evaluate
+
+sys.path.insert(0, '../../../')
+import reader as rdr
+
+''' This file is just used to obtain DECOR detection results on the test set'''
+
+
+def blob(systemName, alpha=8.0):
+	historyFile = '../../../data/history/class_changes/' + systemName + '.csv'
+	systemClassesFile = '../../../data/instances/classes/' + systemName + '.csv'
+
+	# Get the name of all the classes in the system's current version (the version used for the study)
+	classes = []
+	with open(systemClassesFile, 'rb') as csvfile:
+		reader = csv.reader(csvfile, delimiter=';')
+
+		for row in reader:
+			classes.append(row[0])
+
+	reverseDictionnary = {classes[i]: i for i in xrange(len(classes))}
+
+
+
+	# Create an history list containing the names of the classes that changed in each commit.
+	# For example, if class1 and class3 changed in the first commit, 
+	# and class1, class2, class3 changed in the second commit, etc ...
+	# The history list will be [[class1, Class3], [class1, class2, class3], ...]
+
+	history_dict = rdr.readHistory(historyFile, "C")
+
+	history = []
+	commit = []
+	snapshot = history_dict[0]['Snapshot']
+	for i, change in enumerate(history_dict):
+		if snapshot != change['Snapshot']:
+			history.append(list(set(commit)))
+			commit = []
+			snapshot = change['Snapshot']
+
+		commit.append(change['Class'])
+
+		if i == len(history_dict)-1:
+			history.append(list(set(commit)))
+
+
+	# Compute for each class in classes, the number of commit involving at least another class,
+	# and the number of occurences in this set of commit.
+	nbCommit = np.array([0 for _ in range(len(classes))])
+	occurences = [0 for _ in range(len(classes))]
+	for commit in history:
+		nbCommit = [n+1 for n in nbCommit]
+
+		if len(commit) > 1:
+			for className in commit:
+				if className in classes:
+					idx = reverseDictionnary[className]
+					occurences[idx] = occurences[idx] + 1
+
+		else:
+			className = list(commit)[0]
+			if className in classes:
+				idx = reverseDictionnary[className]
+				nbCommit[idx] = nbCommit[idx] - 1
+
+
+	# Return all the classes 'involved in more than alpha% of commits involving at least another class'
+	smells = []
+	for i, nbOcc in enumerate(occurences):
+		threshold = nbCommit[i]*alpha/100
+		if nbOcc > threshold:
+			smells.append(classes[i])
+
+
+	return smells
+
+
+
+def test(systemName, alpha):
+	print(systemName, alpha)
+	trueFile = '../../../data/labels/Blob/hand_validated/' + systemName + '.csv'
+
+	# Get Smells occurences
+	true = []
+	with open(trueFile, 'rb') as csvfile:
+		reader = csv.reader(csvfile, delimiter=';')
+
+		for row in reader:
+			true.append(row[0])
+
+	# Get classes detected as God Classes
+	detected = blob(systemName, alpha)
+	print(detected)
+
+	pre = evaluate.precision(detected, true)
+	rec = evaluate.recall(detected, true)
+	f_m = evaluate.f_mesure(detected, true)
+
+	print('Precision :' + str(pre))
+	print('Recall    :' + str(rec))
+	print('F-Mesure  :' + str(f_m))
+
+	return f_m
+
+def overall(systems):
+
+	overallSmells = []
+	overallTrue = []
+	for system in systems:
+		trueFile = '../../../data/labels/Blob/hand_validated/' + system + '.csv'
+
+		with open(trueFile, 'rb') as csvfile:
+			reader = csv.reader(csvfile, delimiter=';')
+
+			for row in reader:
+				overallTrue.append(row[0])
+
+		smells = blob(system)
+
+		for s in smells:
+			overallSmells.append(s)
+
+	pre = evaluate.precision(overallSmells, overallTrue)
+	rec = evaluate.recall(overallSmells, overallTrue)
+	f_m = evaluate.f_mesure(overallSmells, overallTrue)
+
+	print('Precision :' + str(pre))
+	print('Recall    :' + str(rec))
+	print('F-Mesure  :' + str(f_m))
+
+
+
+if __name__ == "__main__":
+	systems = ['apache-tomcat', 'jedit', 'android-platform-support']
+
+	for system in systems:
+		r = test(system, 8.0)
+		print("")
+
+	print("OVERALL")
+	overall(systems)
+
+
+
+	# Uncomment to perform alpha calibration. 
+	# Best alpha: 10.0
+	'''
+	num_tests = 200
+
+	bar = progressbar.ProgressBar(maxval=num_tests, \
+		widgets=['Performing calibration of alpha: ' ,progressbar.Percentage()])
+	bar.start()
+
+	alphas = 0 + 0.1*np.array(range(num_tests))
+	f_m = []
+	std = []
+	i = 0
+	for i, alpha in enumerate(alphas):
+		bar.update(i)
+
+		results = []
+		for systemName in systems:
+			results.append(test(systemName, alpha))
+
+		f_m.append(np.mean(np.array(results)))
+		std.append(np.std(np.array(results)))
+
+	bar.finish()
+
+	plt.plot(alphas, f_m, 'ro', alphas, std)
+	plt.show()'''
+