def get_words_per_comment_in_discussion(self):
"""
Get number of words of all comments on an issue or pull request
:return: sum of all words of all comments of an issue or pull request, per pull request or issue
:rtype: list
"""
print("#### Words/Comments in Discussions ####")
comments_in_discussion, words_in_discussion = self._get_comments_in_discussion(
)
words_per_comments_in_discussion = [[
'issue', 'number_words_per_comment'
]]
for key in comments_in_discussion.keys():
# print(str(key) + ': ' + str(comments_per_issue[key]))
words_per_comments_in_discussion.append([
key,
str(words_in_discussion[key] / comments_in_discussion[key])
])
csv = CSVHandler()
csv.write_csv(self.path + '/' + self.project + '/metrics/',
self.project + '_words_per_comments_in_discussion.csv',
words_per_comments_in_discussion)
return words_per_comments_in_discussion
def opened_employee_or_temporary(self):
"""
Collect the status of the user that opened the issue/pull request. Employee if has an author_association OWNER, MEMBER, COLLABORATOR or CONTRIBUTOR. Temporary else.
:return: list of the status of the user that opened the issue/pull request by issue/pull request
:rtype: list
"""
print("#### Opened by Employee or Temporary ####")
path = self.path + '/' + self.project
json = JSONHandler(path + '/')
issues = json.open_json(self.project + '_issues.json')
pulls = json.open_json(self.project + '_pulls.json')
opened_by = [['number', 'status', 'user']]
for issue in issues:
if 'author_association' in issue.keys():
# print(issue['author_association'])
opened_by.append(
[issue['issue_number'], self._employee_or_temporary(issue['author_association']), issue['user']])
for pull in pulls:
if 'author_association' in pull.keys():
opened_by.append(
[pull['pull_request_number'], self._employee_or_temporary(pull['author_association']),
pull['user']])
csv = CSVHandler()
csv.write_csv(self.path + '/' + self.project + '/metrics/',
self.project + '_opened_by.csv',
opened_by)
return opened_by
def number_of_core_devs(self):
"""
Collect the amount of users with author_association equals to MEMBER and OWNER on the comments
:return: list of amount of users with author_association equal to MEMBER and OWNER per issue/pull request
:rtype: list
"""
print("#### Number of Core Members ####")
core = [['id', 'count']]
for k in self.users_comments.keys():
count = 0
if k in self.users_comments.keys():
for association in self.users_comments[k]:
if association == 'MEMBER' or association == 'OWNER':
count += 1
if k in self.users_issues.keys():
for association in self.users_issues[k]:
if association == 'MEMBER' or association == 'OWNER':
count += 1
core.append([k, count])
csv = CSVHandler()
csv.write_csv(self.path + '/' + self.project + '/metrics/',
self.project + '_number_of_core_developers.csv', core)
return core
def number_of_contributors(self):
"""
Collect the amount of users with author_association equals to CONTRIBUTOR and COLLABORATOR on the comments
:return: list of amount of users with author_association equal to CONTRIBUTOR and COLLABORATOR per issue/pull request
:rtype: list
"""
print("#### Number of Contributors ####")
contributors = [['id', 'count']]
for k in self.users_comments.keys():
count = 0
if k in self.users_comments.keys():
for association in self.users_comments[k]:
if association == 'CONTRIBUTOR' or association == 'COLLABORATOR':
count += 1
if k in self.users_issues.keys():
for association in self.users_issues[k]:
if association == 'CONTRIBUTOR' or association == 'COLLABORATOR':
count += 1
contributors.append([k, count])
csv = CSVHandler()
csv.write_csv(self.path + '/' + self.project + '/metrics/',
self.project + '_number_of_contributors.csv',
contributors)
return contributors
def get_number_of_patches(self):
"""
Collects the number of snippets inside each comment of issues and pull requests.
:return: list of the number of snippets per issue or pull request
:rtype: list
"""
print('#### Number of Snippets ####')
mypath = self.path + self.project + '/comments/individual/'
json = JSONHandler(mypath)
onlyfiles = [f for f in listdir(mypath) if isfile(join(mypath, f))]
patches_in_discussion = {}
for file in onlyfiles:
comments = json.open_json(file)
for comment in comments:
if 'issue_url' in comment.keys():
issue = comment['issue_url'].split('/')
issue = issue[len(issue) - 1]
if issue not in patches_in_discussion.keys():
patches_in_discussion[issue] = 0
if '```' in comment['body']:
patches = comment['body'].split('```')
count = 0
aux = 0
if issue not in self.patches_size.keys():
self.patches_size[issue] = 0
for patch in patches:
if len(patches) != 1:
aux += 1
if aux % 2 != 0:
continue
self.patches_size[issue] += len(patch)
count += 1
patches_in_discussion[issue] += count
number_of_patches_in_discussion = [['issue', 'number_patches']]
for key in patches_in_discussion.keys():
number_of_patches_in_discussion.append(
[key, patches_in_discussion[key]])
csv = CSVHandler()
csv.write_csv(self.path + '/' + self.project + '/metrics/',
self.project + '_patches_in_discussion.csv',
number_of_patches_in_discussion)
return number_of_patches_in_discussion
def mean_time_between_replies(self):
"""
Collect the mean time between comments inside an issue or pull request
:return: list if mean time between comments per issue/pull request
:rtype: list
"""
print('#### Mean Time Between Comments ####')
mypath = self.path + self.project + '/comments/individual/'
json = JSONHandler(mypath)
onlyfiles = [f for f in listdir(mypath) if isfile(join(mypath, f))]
comments_per_issue = {}
for file in onlyfiles:
comments = json.open_json(file)
for comment in comments:
issue = comment['issue_url'].split('/')
issue = issue[len(issue) - 1]
if issue not in comments_per_issue.keys():
comments_per_issue[issue] = []
comments_per_issue[issue].append(comment['created_at'])
date_utils = DateUtils()
mean_time = [['issue', 'mean_time']]
for key in comments_per_issue.keys():
days_between = []
sorted_dates = date_utils.sort_dates(comments_per_issue[key])
aux = None
for date in sorted_dates:
if not aux:
aux = date
continue
days = date_utils.get_days_between_dates(aux, date)
days_between.append(days)
aux = date
length = len(days_between)
length += 1
sum_days = sum(days_between)
mean_days = sum_days / length
mean_time.append([key, mean_days])
csv = CSVHandler()
csv.write_csv(self.path + '/' + self.project + '/metrics/',
self.project + '_mean_time_between_replies.csv',
mean_time)
return mean_time
def get_words_in_discussion(self):
_, words_in_discussion = self.get_comments_in_discussion()
words_per_discussion = [['issue', 'number_words']]
for key in words_in_discussion.keys():
words_per_discussion.append([key, words_in_discussion[key]])
csv = CSVHandler()
csv.write_csv(self.path + '/' + self.project + '/metrics/',
self.project + '_words_in_discussion.csv',
words_per_discussion)
def get_words_per_comment_in_discussion(self):
comments_in_discussion, words_in_discussion = self.get_comments_in_discussion(
)
words_per_comments_in_discussion = [[
'issue', 'number_words_per_comment'
]]
for key in comments_in_discussion.keys():
# print(str(key) + ': ' + str(comments_per_issue[key]))
words_per_comments_in_discussion.append([
key,
str(words_in_discussion[key] / comments_in_discussion[key])
])
csv = CSVHandler()
csv.write_csv(self.path + '/' + self.project + '/metrics/',
self.project + '_words_per_comments_in_discussion.csv',
words_per_comments_in_discussion)
def get_time_in_days_between_open_and_close(self):
"""
Collects the time in days between the day an issue or pull request was opened and the day it was closed.
:return: list of time in days per issue/pull request.
:rtype: list
"""
print('#### Discussion Length ####')
path = self.path + '/' + self.project
json = JSONHandler(path + '/')
issues = json.open_json(self.project + '_issues.json')
pulls = json.open_json(self.project + '_pulls.json')
days_between = [['number', 'status']]
date_utils = DateUtils()
for issue in issues:
days = 0
if 'closed' in issue['state']:
days = date_utils.get_days_between_dates(
issue['created_at'], issue['closed_at'])
# print(issue['author_association'])
days_between.append([issue['issue_number'], days])
for pull in pulls:
days = 0
if 'closed' in pull['state']:
if pull['merged_at']:
days = date_utils.get_days_between_dates(
pull['created_at'], pull['merged_at'])
else:
days = date_utils.get_days_between_dates(
pull['created_at'], pull['closed_at'])
days_between.append([pull['pull_request_number'], days])
csv = CSVHandler()
csv.write_csv(self.path + '/' + self.project + '/metrics/',
self.project + '_discussion_length.csv', days_between)
return days_between
def number_of_pull_requests(self):
# Number of pull requests that contained changes to a class.
mypath = self.path + self.project + '/'
csv = CSVHandler()
pulls_commits = csv.open_csv(mypath + 'pulls_commits/' + self.project +
'_commits_by_pull_request.csv')
commits_classes = csv.open_csv(mypath + self.project + '.csv')
commits_in_pulls = {}
for pull_commit in pulls_commits:
if pull_commit[1] not in commits_in_pulls.keys():
commits_in_pulls[pull_commit[1]] = []
commits_in_pulls[pull_commit[1]].append(pull_commit[0])
renames = _get_renames(commits_classes)
commits_by_class = _get_commits_by_class(renames, commits_classes)
class_ids = _get_class_ids(commits_classes)
pulls_by_class = {}
for commit_class_key in commits_by_class.keys():
commit_list = commits_by_class[commit_class_key]
for commit in commit_list:
if commit in commits_in_pulls.keys():
if commit_class_key not in pulls_by_class.keys():
pulls_by_class[commit_class_key] = []
pulls_by_class[commit_class_key].extend(
commits_in_pulls[commit])
num_pulls = [['class_id', 'num_pulls']]
for key in pulls_by_class.keys():
pulls = pulls_by_class[key]
num_pulls.append([class_ids[key], len(set(pulls))])
# TODO Save in CSV
return commits_in_pulls, renames, commits_by_class, pulls_by_class, class_ids, num_pulls
def get_patch_size(self):
"""
Collects the size of the snippets of each comment in issues and pull requests.
:return: list with the sizes of the patches per issue and pull request
:rtype: list
"""
print('#### Snippets Size ####')
size_of_patches_in_discussion = [['issue', 'size_patches']]
for key in self.patches_size.keys():
#print(str(key) + ': ' + str(self.patches_size[key]))
size_of_patches_in_discussion.append([key, self.patches_size[key]])
csv = CSVHandler()
csv.write_csv(self.path + '/' + self.project + '/metrics/',
self.project + '_patches_size_in_discussion.csv',
size_of_patches_in_discussion)
return size_of_patches_in_discussion
def number_associated_issues(self):
# Number of issues associated to pull requests that contained changes to a class.
mypath = self.path + self.project + '/'
_, _, _, pulls_by_class, class_ids, _ = self.number_of_pull_requests()
csv = CSVHandler()
pulls_of_issues = csv.open_csv(mypath + self.project +
'_pulls_of_issues.csv')
associated_issues = {}
for row in pulls_of_issues:
if row[1] not in associated_issues.keys():
associated_issues[row[1]] = set()
associated_issues[row[1]].add(row[0])
issues_by_class = {}
for pull_class in pulls_by_class:
pulls = pulls_by_class[pull_class]
for pull in pulls:
if pull_class not in issues_by_class.keys():
issues_by_class[pull_class] = 0
if pull in associated_issues.keys():
issues_by_class[pull_class] += len(associated_issues[pull])
else:
issues_by_class[pull_class] += 1
issues_num = [['class_id', 'issue_num']]
for key in issues_by_class.keys():
issues = issues_by_class[key]
issues_num.append([class_ids[key], issues])
# TODO Save in CSV
return issues_num
def get_median_of_number_of_comments(self):
"""
Collects the median of the number of comments inside an issue or pull requests
:return: list with the median of the number of comments per issue or pull request
:rtype: list
"""
print("#### Median Comments ####")
mypath = self.path + self.project + '/comments/individual/'
json = JSONHandler(mypath)
onlyfiles = [f for f in listdir(mypath) if isfile(join(mypath, f))]
comments_per_issue = {}
for file in onlyfiles:
comments = json.open_json(file)
for comment in comments:
if 'issue_url' in comment.keys():
issue = comment['issue_url'].split('/')
issue = issue[len(issue) - 1]
if int(issue) not in comments_per_issue:
comments_per_issue[int(issue)] = 0
comments_per_issue[int(issue)] = comments_per_issue[int(issue)] + 1
values = []
median_comments = [['issue', 'median_comments']]
number_comments = [['id', 'number_comments']]
for key in sorted(comments_per_issue):
#print(str(key) + ': ' + str(comments_per_issue[key]))
values.append(comments_per_issue[key])
median_comments.append([key, median(values)])
number_comments.append([key, comments_per_issue[key]])
csv = CSVHandler()
csv.write_csv(self.path + '/' + self.project + '/metrics/',
self.project + '_median_comments.csv',
median_comments)
csv.write_csv(self.path + '/' + self.project + '/metrics/',
self.project + '_number_comments.csv',
number_comments)
return number_comments