import csv import collections from operator import itemgetter class RepoLabelsClass: def __init__(self, r): self.repository = str(r) # Issue and Class size self.classifieldSize = '' # Labels usage flags self.isLabelingFirstPratice = False self.issuesAmount = 0 # Amount of issues labeled self.labeledIssueAmount = 0 self.issuesLabeledFirstAmount = 0 self.labelsRepo = [] self.whenLabelingOccurs = [] def repoClassification(self): if(self.issuesAmount > 142): if(self.issuesAmount > 546): self.classifieldSize = 'Biggest' else: self.classifieldSize = 'Bigger' else: if(self.issuesAmount < 35): self.classifieldSize = 'Small' else: self.classifieldSize = 'Smallest' def mountLabelInfo(self, label): type = '' if(label in ["bug", "documentation", "duplicate", "enhancement", "good first issue", "help wanted", "invalid", "question", "wontfix"]): type = 'standard' else: type = 'custom' return { "Label": str(label), "Amount": 0, "Title_Related": 0, "Body_Related": 0, "Comment_Related": 0, "Type": str(type) } def labelExistsInList(self, label): for l in self.labelsRepo: if(label in l['Label']): return True return False def addLabel(self, labels, flag): for label in labels: label_search = next((x for x in self.labelsRepo if x["Label"] == label.lower()), None) if(label_search is None): label_registry = self.mountLabelInfo(label.lower()) if(flag == 1): label_registry['Amount'] += 1 self.labelsRepo.append(label_registry) elif(label_search is not None): label_search['Amount'] += 1 def countLabelOccurrenceInIssue(self, label, component): label_search = next((x for x in self.labelsRepo if x["Label"] == label.lower()), None) if(component == 'Title'): label_search['Title_Related'] += 1 elif(component == 'Body'): label_search['Body_Related'] += 1 elif(component == 'Comment'): label_search['Comment_Related'] += 1 def mount_event(self, issue, pos): return {'Issue' : int(issue), 'Pos': int(pos)} def countFirstLabelingEvent(self, issue, issue_events): event_counter = 1 for e in issue_events: if(e['Event'] == 'labeled'): if(event_counter == 1): self.issuesLabeledFirstAmount += 1 #evnt = self.mount_event(issue, event_counter) self.whenLabelingOccurs.append(event_counter) event_counter += 1 def countLabelOccurrencies(self, labels_list , title, body, comments): occurrences_in_title = 0 occurrences_in_body = 0 occurrences_in_commentary = 0 for label in labels_list: label_search = next((x for x in self.labelsRepo if x["Label"] == label.lower()), None) if(title is not None): occurrences_in_title += title.count(label) if(body is not None): occurrences_in_body += body.count(label) if(comments is not None): for c in comments: occurrences_in_commentary += c.count(label) label_search['Title_Related'] += occurrences_in_title label_search['Body_Related'] += occurrences_in_body label_search['Comment_Related'] += occurrences_in_commentary def savingGeneralStatistics(self): repository = self.repository #[1] amountOfIssues = self.issuesAmount #[2] classifield = self.classifieldSize #[3] useLabels = 'No' #[4] if(self.labeledIssueAmount > 0): useLabels = 'Yes' amountOfIssuesLabeled = self.labeledIssueAmount #[5] useStandardLabel = 'No' #[6] amountOfStandardLabelsUsed = 0 #[7] StandardLabel_list = sorted(list(filter(lambda l: l['Type'] == 'standard', self.labelsRepo)), key=itemgetter('Amount'), reverse=True) standardLabel_mostUsed = next((x for x in StandardLabel_list), None) standardLabel_mostUsed_Label = '-' standardLabel_mostUsed_Amount = '0' if(standardLabel_mostUsed is not None): if(standardLabel_mostUsed['Amount'] > 0): standardLabel_mostUsed_Label = standardLabel_mostUsed['Label'] standardLabel_mostUsed_Amount = standardLabel_mostUsed['Amount'] amountOfStandardLabelsUsed = sum([x['Amount'] for x in StandardLabel_list]) if(amountOfStandardLabelsUsed > 0): useStandardLabel = 'Yes' # 9 labels AllStandardLabelsUsed = 'No' #[8] UsageOfStandardLabel = list(filter(lambda l: l['Amount'] > 0, StandardLabel_list)) if(UsageOfStandardLabel == 9): AllStandardLabelsUsed = 'Yes' createCustomLabels = 'No' #[9] CustomLabel_list = list(filter(lambda l: l['Type'] == 'custom', self.labelsRepo)) if(len(CustomLabel_list) > 0): createCustomLabels = 'Yes' amountOfCustomLabelsCreated = len(CustomLabel_list) #[10] useCustomLabels = 'No' #[11] amountOfCustomLabelsUsed = sum([x['Amount'] for x in CustomLabel_list]) #[12] CustomLabel_Usage = list(filter(lambda l: l['Amount'] > 0, CustomLabel_list)) if(len(CustomLabel_Usage) > 0): useCustomLabels = 'Yes' useAllCustomLabel = 'No' #[13] CustomLabelsUsed = len(CustomLabel_Usage) #[14] if(len(CustomLabel_list) == len(CustomLabel_Usage)): useAllCustomLabel = 'Yes' if(len(CustomLabel_list) == 0 and len(CustomLabel_Usage) == 0): useAllCustomLabel = 'No' repository_row = [repository, amountOfIssues, classifield, useLabels, amountOfIssuesLabeled, useStandardLabel, amountOfStandardLabelsUsed, AllStandardLabelsUsed, standardLabel_mostUsed_Label, standardLabel_mostUsed_Amount, createCustomLabels, amountOfCustomLabelsCreated, useCustomLabels, amountOfCustomLabelsUsed, useAllCustomLabel, CustomLabelsUsed ] with open('GeneralInfo.csv', mode='a') as GeneralInfo_csv: mount_generalInfo_csv = csv.writer(GeneralInfo_csv) mount_generalInfo_csv.writerow(repository_row) GeneralInfo_csv.close() def savingCustomLabelStatistics(self): CustomLabel_list = sorted(list(filter(lambda l: l['Type'] == 'custom', self.labelsRepo)), key=itemgetter('Amount'), reverse=True) registry_list = [] registry_list.append(self.repository) for l in CustomLabel_list[:5]: registry_list.append(str(l['Label'])) registry_list.append(str(l['Amount'])) with open('CustomLabelsInfo.csv', mode='a') as CustomLabelInfo_CSV: mount_customLabelInfo_csv = csv.writer(CustomLabelInfo_CSV) mount_customLabelInfo_csv.writerow(registry_list) CustomLabelInfo_CSV.close() def WhenLabelingOccursStatistics(self): repository = self.repository isLabelingFirstPratice = 'No' event_labeling_most_occurs = collections.Counter(self.whenLabelingOccurs) ev1 = '-' ev2 = '-' if(len(event_labeling_most_occurs) > 0): event_most_occors_pos = event_labeling_most_occurs.most_common(1)[0] if(event_most_occors_pos[0] == 1): isLabelingFirstPratice = 'Yes' ev1 = event_most_occors_pos[0] ev2 = event_most_occors_pos[1] registry_line = [repository, isLabelingFirstPratice, self.issuesLabeledFirstAmount, ev1, ev2] with open('WhenLabelingOccursInfo.csv', mode='a') as WhenOccursInfo_csv: mount_whenOccursInfo_csv = csv.writer(WhenOccursInfo_csv) mount_whenOccursInfo_csv.writerow(registry_line) WhenOccursInfo_csv.close() def LabelRelatedStatus(self): repository = self.repository isTitleRelated = 'No' isBodyRelated = 'No' isCommentRelated = 'No' Label_TitleRelated_list = list(filter(lambda l: l['Title_Related'] > 0, self.labelsRepo)) Label_BodyRelated_list = list(filter(lambda l: l['Body_Related'] > 0, self.labelsRepo)) Label_CommentRelated_list = list(filter(lambda l: l['Comment_Related'] > 0, self.labelsRepo)) amountOfTitleRelated = sum([x['Title_Related'] for x in Label_TitleRelated_list]) amountOfBodyRelated = sum([x['Body_Related'] for x in Label_BodyRelated_list]) amountOfCommentRelated = sum([x['Comment_Related'] for x in Label_CommentRelated_list]) Sorted_First_TitleRelated = sorted(Label_TitleRelated_list, key=itemgetter('Title_Related'), reverse=True) Sorted_First_BodyRelated = sorted(Label_BodyRelated_list, key=itemgetter('Body_Related'), reverse=True) Sorted_First_CommentRelated = sorted(Label_CommentRelated_list, key=itemgetter('Comment_Related'), reverse=True) First_TitleRelated = next((t for t in Sorted_First_TitleRelated), None) First_BodyRelated = next((b for b in Sorted_First_BodyRelated), None) First_CommentRelated = next((c for c in Sorted_First_CommentRelated), None) firstTitleLabel = '' firstTitleNumber = 0 firstBodyLabel = '' firstBodyNumber = 0 firstCommentsLabel = '' firstCommentsNumber = 0 if(First_TitleRelated is not None): firstTitleLabel = First_TitleRelated['Label'] firstTitleNumber = First_TitleRelated['Title_Related'] if(First_BodyRelated is not None): firstBodyLabel = First_BodyRelated['Label'] firstBodyNumber = First_BodyRelated['Body_Related'] if(First_CommentRelated is not None): firstCommentsLabel = First_CommentRelated['Label'] firstCommentsNumber = First_CommentRelated['Comment_Related'] if(len(Label_TitleRelated_list) > 0): isTitleRelated = 'Yes' if(len(Label_BodyRelated_list) > 0): isBodyRelated = 'Yes' if(len(Label_CommentRelated_list) > 0): isCommentRelated = 'Yes' registry_line = [repository, isTitleRelated, amountOfTitleRelated, firstTitleLabel, firstTitleNumber, isBodyRelated, amountOfBodyRelated, firstBodyLabel, firstBodyNumber, isCommentRelated, amountOfCommentRelated, firstCommentsLabel, firstCommentsNumber ] with open('LabelRelated.csv', mode='a') as LabelRelatedInfo_csv: mount_labelRelated_csv = csv.writer(LabelRelatedInfo_csv) mount_labelRelated_csv.writerow(registry_line) LabelRelatedInfo_csv.close()