def issueDurationDetector():
zeroCount=0
values=gt.launchDump()['duration']
print (values)
print ("Method 1")
values.sort()
for i in values:
if i <=0.1:
zeroCount+=1
else:
break
hv=base.high(values)['value']
hp=base.high(values)['pos']
mv=base.median(values)['value']
mp=base.median(values)['pos']
lv=base.low(values)['value']
lp=base.low(values)['pos']
print ('low val',lv,'median val',mv,'high val',hv)
print ('low pos',lp,'median pos',mp,'high pos',hp)
k= float((hv-mv)*(mp-lp)/((hp-mp)*(mv-lv)))
if k>5 or 20*zeroCount>len(values):
print ('BadSmell found in issue Duration')
elif k==5:
print ('Probably have smells in issue length')
else:
print ('Issue Duration Detector Passed')
def earlyIssuesByWeekDetector():
issues_nums = []
items = gt.launchDump()
issues_nums = gt.divideByTime(items['create_at'])
avg_val = np.mean(issues_nums[0:4])
std_dev = np.std(issues_nums[0:4])
val = std_dev
data = issues_nums[0:4]
print(data)
sum = 0
for i in range(len(data)):
res = cmp(
data[i], avg_val + val
) # when isssues number is greater than avg + 2 * stand deviation or less than avg - 2 * stand deviation
res2 = cmp(data[i], avg_val - val)
if (i > 0):
sum = issues_nums[i - 1] + issues_nums[i]
if sum < avg_val and i > 0:
print('Badsmells: This week has too less issues.')
sum = 0
elif res > 0:
print('Badsmells: This week has too many issues.')
elif res2 < 0:
print('Badsmells: This week has too less issues.')
else:
print('Issues numbers are normal')
def issueDurationDetector2():
print ("Method 2")
zeroCount=0
data=gt.launchDump()['duration']
data.sort()
for i in data:
if i <=0.1:
zeroCount+=1
else:
break
data=data[1:len(data)-1]
stddeviation= std(data)
avg= mean(data)
high=avg+stddeviation
if avg-stddeviation>0:
low=avg-stddeviation
else:
low=0
lowCount=0
highCount=0
for i in data:
if i < low:
lowCount+=1
if i>high:
highCount+=1
print ("avg",avg,"std",stddeviation)
print ("strange small count=",lowCount,"strange large count=",highCount,"none issue count",zeroCount)
if 20*(lowCount+highCount) > len(data) or 20*zeroCount>len(data):
print ("BadSmell found in issue Duration")
else:
print ("Issue Duration Detector Passed")
def earlyIssuesByWeekDetector():
issues_nums=[]
items = gt.launchDump()
issues_nums = gt.divideByTime(items['create_at'])
avg_val = np.mean(issues_nums[0:4])
std_dev = np.std(issues_nums[0:4])
val = std_dev
data = issues_nums[0:4]
print(data)
sum = 0
for i in range(len(data)):
res = cmp(data[i], avg_val + val) # when isssues number is greater than avg + 2 * stand deviation or less than avg - 2 * stand deviation
res2 = cmp(data[i], avg_val - val)
if (i > 0):
sum = issues_nums[i-1] + issues_nums[i]
if sum < avg_val and i > 0:
print ('Badsmells: This week has too less issues.')
sum = 0
elif res > 0:
print ('Badsmells: This week has too many issues.')
elif res2 < 0:
print ('Badsmells: This week has too less issues.')
else:
print ('Issues numbers are normal')
def labelUseDetector():
data=gt.launchDump()['labels'].values()
# print (data)
data.sort()
data=data[0:len(data)-1]
stddeviation= std(data)
avg= mean(data)
high=avg+stddeviation
if avg-stddeviation>0:
low=avg-stddeviation
else:
low=0
lowCount=0
highCount=0
print ("avg",avg)
print ("std",stddeviation)
for i in data:
if i < low:
lowCount+=1
if i>high:
highCount+=1
print ("number of strange low data",lowCount)
print ("number of strange high data",highCount)
if 10*(lowCount+highCount)/len(data) > 1:
print ("BadSmell found in Label Use")
else:
print ("Label Use Detector Passed")
def issueDurationDetector():
zeroCount = 0
values = gt.launchDump()['duration']
print(values)
print("Method 1")
values.sort()
for i in values:
if i <= 0.1:
zeroCount += 1
else:
break
hv = base.high(values)['value']
hp = base.high(values)['pos']
mv = base.median(values)['value']
mp = base.median(values)['pos']
lv = base.low(values)['value']
lp = base.low(values)['pos']
print('low val', lv, 'median val', mv, 'high val', hv)
print('low pos', lp, 'median pos', mp, 'high pos', hp)
k = float((hv - mv) * (mp - lp) / ((hp - mp) * (mv - lv)))
if k > 5 or 20 * zeroCount > len(values):
print('BadSmell found in issue Duration')
elif k == 5:
print('Probably have smells in issue length')
else:
print('Issue Duration Detector Passed')
def issueDurationDetector2():
print("Method 2")
zeroCount = 0
data = gt.launchDump()['duration']
data.sort()
for i in data:
if i <= 0.1:
zeroCount += 1
else:
break
data = data[1:len(data) - 1]
stddeviation = std(data)
avg = mean(data)
high = avg + stddeviation
if avg - stddeviation > 0:
low = avg - stddeviation
else:
low = 0
lowCount = 0
highCount = 0
for i in data:
if i < low:
lowCount += 1
if i > high:
highCount += 1
print("avg", avg, "std", stddeviation)
print("strange small count=", lowCount, "strange large count=", highCount,
"none issue count", zeroCount)
if 20 * (lowCount + highCount) > len(data) or 20 * zeroCount > len(data):
print("BadSmell found in issue Duration")
else:
print("Issue Duration Detector Passed")
def labelUseDetector():
data = gt.launchDump()['labels'].values()
# print (data)
data.sort()
data = data[0:len(data) - 1]
stddeviation = std(data)
avg = mean(data)
high = avg + stddeviation
if avg - stddeviation > 0:
low = avg - stddeviation
else:
low = 0
lowCount = 0
highCount = 0
print("avg", avg)
print("std", stddeviation)
for i in data:
if i < low:
lowCount += 1
if i > high:
highCount += 1
print("number of strange low data", lowCount)
print("number of strange high data", highCount)
if 10 * (lowCount + highCount) / len(data) > 1:
print("BadSmell found in Label Use")
else:
print("Label Use Detector Passed")
def process_issues(features):
issues = gitable.launchDump()
weekly_issues = issues['week']
del issues['week']
author_issues = dict()
events_issues =dict()
comments_issues = dict()
for issue, events in issues.iteritems():
#print("ISSUE " + str(issue))
dict_add(author_issues, events[0].user)
dict_add(comments_issues, events[0].comments)
dict_add(events_issues, len(events[1:]))
draw_bar(author_issues.values(), "issues number posted by person","issues", range(len(author_issues.keys())),"person",0.35)
author_filter = filter.filter(author_issues)
large_author = author_filter.large()
features['issues_person'] = author_filter
if(len(large_author) >0 ):
features['large issues post by single user'] = large_author
small_author = author_filter.small(delta=2)
if(len(small_author) >0 ):
features['small issues post by single user'] = small_author
draw_bar(comments_issues.values(),"issues number with same comments number","issues",comments_issues.keys(),"comments number",0.35)
events_filter= filter.filter(events_issues)
features['events_issues'] = events_filter
if len(events_filter.large())>0:
features['large issues with same events'] = events_filter.large()
comments_filter = filter.filter(comments_issues)
features['comments_issues'] = comments_filter
large_comments = comments_filter.large()
if(len(large_comments)>0):
features['large issues with same comments'] = large_comments
draw_bar(events_issues.values(),"issues number with same events number", "issues",events_issues.keys(),"events number",0.35)
single_user = filter.filter(author_issues)
single_user.large(5)
def earlyDetector():
issues_nums=[]
items = gt.launchDump()
issues_nums = gt.divideByTime(items['create_at'])
avg_val = np.mean(issues_nums[0:3])
std_dev = np.std(issues_nums[0:3])
val = 1.2*std_dev
for x in issues_nums[0:3]:
res = cmp(x, avg_val + val) # when isssues number is greater than avg + 2 * stand deviation or less than avg - 2 * stand deviation
res2 = cmp(x, avg_val - val)
if res > 0:
print ('Badsmells: This week has too many issues.')
elif res2 < 0:
print ('Badsmells: This week has too less issues.')
else:
print ('Issues numbers are normal')
def labelUseDetector():
data=gt.launchDump()['labels'].values()
print (data)
data.sort()
data=data[1:len(data)-1]
stddeviation= std(data)
avg= mean(data)
high=avg+stddeviation
if avg-stddeviation>0:
low=avg-stddeviation
else:
low=0
lowCount=0
highCount=0
for i in data:
if i < low:
lowCount+=1
if i>high:
highCount+=1
if 8*lowCount/len(data) > 1 or 8*highCount/len(data)>1:
print ("BadSmell found in Label Use")
else:
print ("Label Use Detector Passed")
def issueDurationDetector():
values=gt.launchDump()['duration']
# values.sort()
print (values)