def getMirantisDefectiveContent():
dataset_file="/Users/akond/Documents/AkondOneDrive/OneDrive/stvr/dataset/MIRANTIS_FULL_DATASET.csv"
str_ = ''
with open(dataset_file, 'rU') as f:
reader_ = csv.reader(f)
next(reader_, None)
for row in reader_:
defectStatus = int(row[-1]) ### need to convert to int , otherwise gives error for sklearn.are_under_roc
fileToRead = row[1]
if (defectStatus == 1):
if (os.path.exists(fileToRead)):
with open(fileToRead, 'rU') as file_:
data_str = file_.read()
str_ = str_ + fileToRead + '\n' + '='*100 + '\n' + data_str + '\n' + '='*100 + '\n' + '='*100 + '\n'
utility.dumpContentIntoFile(str_, '../../../output/MIR_FOR_OPS.txt')
def dumpPredPerfValuesToFile(iterations, predPerfVector, fileName):
str2write = ''
for cnt in xrange(iterations):
auc_ = predPerfVector[0][cnt]
prec_ = predPerfVector[1][cnt]
recal = predPerfVector[2][cnt]
str2write = str2write + str(auc_) + ',' + str(prec_) + ',' + str(
recal) + ',' + '\n'
bytes_ = utility.dumpContentIntoFile(str2write, fileName)
print "Created {} of {} bytes".format(fileName, bytes_)
def dumpPredPerfValuesToFile(iterations, predPerfVector, fileName):
str2write = ''
headerStr = 'AUC,PRECISION,RECALL,FMEASURE,'
for cnt in xrange(iterations):
auc_ = predPerfVector[0][cnt]
prec_ = predPerfVector[1][cnt]
recal = predPerfVector[2][cnt]
fmeas = predPerfVector[3][cnt]
str2write = str2write + str(auc_) + ',' + str(prec_) + ',' + str(
recal) + ',' + str(fmeas) + ',' + '\n'
str2write = headerStr + '\n' + str2write
bytes_ = utility.dumpContentIntoFile(str2write, fileName)
print "Created {} of {} bytes".format(fileName, bytes_)
print '*' * 25
print '=' * 50
print "Processed {} bug IDs, {} left".format(
counter, valid_lpad_cnt - counter)
print '=' * 50
except Exception as e:
print 'Caught exception:', e.message
'''
DUMP LIST AS STR
'''
str2write = ''
for id_ in bugID2Dump:
str2write = str2write + str(id_) + ',' + '\n'
utility.dumpContentIntoFile(
str2write,
'/Users/akond/Documents/AkondOneDrive/OneDrive/SecurityInIaC/output/REPORTIDS_WITH_SEC_OST.csv'
)
'''
DUMP REPORTS AS STR
'''
report_no = 0
for report_ in bugMSG2Dump:
report_no += 1
tokens = report_.split(' ')
report_ = ''
for token_ in tokens:
try:
token_ = token_.encode('ascii', 'ignore').decode('ascii')
token_ = token_.strip()
token_ = token_.lower()
except TypeError: