def getCommitLevelTokensForTokenization(categ_file_param, the_flag):
completeLabels = []
completeCorpus = [
] ## a list of lists with tokens from defected and non defected files
with open(categ_file_param, 'rU') as f:
reader_ = csv.reader(f)
next(reader_, None)
for row in reader_:
id_ = row[0]
repo_path = row[1]
catgeg_ = row[3]
if catgeg_ == 'N':
defectStatus = 0
else:
defectStatus = 1
if repo_path[-1] != '/':
repo_path = repo_path + '/'
fileToRead = repo_path + 'diffs/' + str(id_) + '.txt'
commitContentAsStr = utility.giveCommentFreeContentFromFile(
fileToRead, the_flag)
#print commitContentAsStr
#print "!"*75
filtered_str_from_one_commit = tokenization_preprocessor.processTokensOfOneFile(
commitContentAsStr)
#print filtered_str_from_one_commit
#print "="*75
completeCorpus.append(filtered_str_from_one_commit)
### after getting the text , getthe labels
completeLabels.append(defectStatus)
#print len(completeCorpus), len(completeLabels)
return completeCorpus, completeLabels
def getICSTFilesForExa(datasetParam, kw_param):
completeLabels = []
completeCorpus = [
] ## a list of lists with tokens from defected and non defected files
with open(datasetParam, 'rU') as f:
reader_ = csv.reader(f)
next(reader_, None)
for row in reader_:
# defectStatus = int(row[20]) ### need to convert to int , otherwise gives error for sklearn.are_under_roc
defectStatus = int(
row[-1]
) ### need to convert to int , otherwise gives error for sklearn.are_under_roc
fileToRead = row[1]
fileContentAsStr = utility.giveCommentFreeContentFromFile(
fileToRead)
#print "!"*75
#print fileContentAsStr
filtered_str_from_one_file = tokenization_preprocessor.processTokensOfOneFile(
fileContentAsStr)
#print len(filtered_str_from_one_file)
#print "="*75
if ((kw_param in filtered_str_from_one_file)
and (defectStatus == 1)):
print '-' * 25
print fileToRead
print '-' * 25
print filtered_str_from_one_file
print '-' * 25
def getTokensForTokenization(datasetParam):
completeLabels = []
completeCorpus = [
] ## a list of lists with tokens from defected and non defected files
'''
token holders for manual/topic modeling
'''
defective_tokens, non_defective_tokens = [], []
with open(datasetParam, '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]
fileContentAsStr = utility.giveCommentFreeContentFromFile(
fileToRead)
#print "!"*75
#print fileContentAsStr
filtered_str_from_one_file = tokenization_preprocessor.processTokensOfOneFile(
fileContentAsStr)
#print len(filtered_str_from_one_file)
#print "="*75
completeCorpus.append(filtered_str_from_one_file)
### after getting the text , getthe labels
completeLabels.append(defectStatus)
return completeCorpus, completeLabels
def getTokensForTokenization(datasetParam):
defectedCorpus = [] ## a list of lists with tokens from defected files
nonDefectedCorpus = [
] ## a list of lists with tokens from non defected files
completeCorpus = [] ## a list of lists with tokens from all files
with open(datasetParam, 'rU') as f:
reader_ = csv.reader(f)
next(reader_, None)
for row in reader_:
defectStatus = int(
row[20]
) ### need to convert to int , otherwise gives error for sklearn.are_under_roc
fileToRead = row[1]
fileContentAsStr = utility.giveCommentFreeFileContent(fileToRead)
#print "!"*75
#print fileContentAsStr
filtered_str_from_one_file = tokenization_preprocessor.processTokensOfOneFile(
fileContentAsStr)
#print len(filtered_str_from_one_file)
#print "="*75
completeCorpus.append(filtered_str_from_one_file)
if (defectStatus == 1):
defectedCorpus.append(filtered_str_from_one_file)
else:
nonDefectedCorpus.append(filtered_str_from_one_file)
return defectedCorpus, nonDefectedCorpus, completeCorpus
