mgaughan/24_deb_pkg_gov
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

updating pr_grabber again

Browse Source
This commit is contained in:
Matthew Gaughan 2024-03-12 08:57:41 -05:00
parent 6bdc4fa6ef
commit 99fb547920
1 changed files with 56 additions and 19 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

75
pr_data_get.py
View File

@ -24,6 +24,8 @@ def file_get_pr(upstream_vcs_link, me_read):
#print(upstream_vcs_link.split('/')[4])
project_dict = {}
project_dict['upstream_vcs_link'] = upstream_vcs_link
upstream_vcs_link = "/".join(upstream_vcs_link.split("/")[0:5])
print(upstream_vcs_link)
full_temp_path = temp_dir + upstream_vcs_link.split('/')[4] + ".git"
repo = Git(uri=upstream_vcs_link, gitpath=full_temp_path)
try:
@ -39,13 +41,13 @@ def file_get_pr(upstream_vcs_link, me_read):
first_date_readme = ""
for commit in commits:
if "Merge" in commit['data'].keys():
commit_list.append([commit['data']['CommitDate'], True])
commit_list.append([commit['data']['CommitDate'], True, commit['data']['Author'], commit['data']['Commit']])
if has_contributing:
merge_post_cont += 1
else:
merge_pre_cont += 1
else:
commit_list.append([commit['data']['CommitDate'], False])
commit_list.append([commit['data']['CommitDate'], False, commit['data']['Author'], commit['data']['Commit']])
files = commit['data']['files']
#print(commit['data']['CommitDate'])
#print(type(dateutil.parser.parse(commit['data']['CommitDate'])))
@ -61,40 +63,74 @@ def file_get_pr(upstream_vcs_link, me_read):
return {}
if me_read:
project_dict['first_readme'] = first_date_readme
before_read = pr_count(first_date_readme+ dt.timedelta(days=-window, hours=0), first_date_readme, commit_list)
project_dict['b6w_prs_read'] = before_read[0]
project_dict['b6w_mrg_read'] = before_read[1]
after_read = pr_count(first_date_readme, first_date_readme + dt.timedelta(days=window, hours=0), commit_list)
project_dict['a6w_prs_read'] = after_read[0]
project_dict['a6w_mrg_read'] = after_read[1]
before_read = pr_count(first_date_readme+ dt.timedelta(days=-window, hours=0), first_date_readme, commit_list, [], [])
if before_read != None:
project_dict['b6w_prs_read'] = before_read[0]
project_dict['b6w_mrg_read'] = before_read[1]
project_dict['b6w_auth_read'] = before_read[2]
project_dict['b6w_commit_read'] = before_read[3]
else:
return {}
after_read = pr_count(first_date_readme, first_date_readme + dt.timedelta(days=window, hours=0), commit_list, before_read[4], before_read[5])
if after_read != None:
project_dict['a6w_prs_read'] = after_read[0]
project_dict['a6w_mrg_read'] = after_read[1]
project_dict['a6w_auth_read'] = after_read[2]
project_dict['a6w_commit_read'] = after_read[3]
else:
return {}
else:
project_dict['first_contributing'] = first_date_contributing
before_cont = pr_count(first_date_contributing + dt.timedelta(days=-window, hours=0), first_date_contributing, commit_list)
project_dict['b6w_prs_cont'] = before_cont[0]
project_dict['b6w_mrg_cont'] = before_cont[1]
after_cont = pr_count(first_date_contributing, first_date_contributing + dt.timedelta(days=window, hours=0), commit_list)
project_dict['a6w_prs_cont'] = after_cont[0]
project_dict['a6w_mrg_cont'] = after_cont[1]
before_cont = pr_count(first_date_contributing + dt.timedelta(days=-window, hours=0), first_date_contributing, commit_list, [], [])
if before_cont != None:
project_dict['b6w_prs_cont'] = before_cont[0]
project_dict['b6w_mrg_cont'] = before_cont[1]
project_dict['b6w_auth_cont'] = before_cont[2]
project_dict['b6w_commit_cont'] = before_cont[3]
else:
return {}
after_cont = pr_count(first_date_contributing, first_date_contributing + dt.timedelta(days=window, hours=0), commit_list, before_cont[4], before_cont[5])
if after_cont != None:
project_dict['a6w_prs_cont'] = after_cont[0]
project_dict['a6w_mrg_cont'] = after_cont[1]
project_dict['a6w_auth_cont'] = after_cont[2]
project_dict['a6w_commit_cont'] = after_cont[3]
else:
return {}
print(project_dict)
return project_dict
#TODO: pr_count should return an array of values for weekly/6mo
def pr_count(start, end, commits):
def pr_count(start, end, commits, author_roster, commit_roster):
count = 0
merge_count = 0
by_week = [0] * 27
by_week_merge =[0] * 27
current_week = 0
new_authors = 0
new_committers = 0
for commit in tqdm(commits):
if dateutil.parser.parse(commit[0]) <= start:
if commit[2] not in author_roster:
author_roster.append(commit[2])
if commit[1] and commit[3] not in commit_roster:
commit_roster.append(commit[3])
if dateutil.parser.parse(commit[0]) > start:
if math.floor((dateutil.parser.parse(commit[0]) - start).days / 7) <= 26:
by_week[math.floor((dateutil.parser.parse(commit[0]) - start).days / 7)] += 1
if commit[1]:
by_week_merge[math.floor((dateutil.parser.parse(commit[0]) - start).days / 7)] += 1
if commit[3] not in commit_roster:
new_committers += 1
#remaining question of whether to make this the author of the merge commit[2] or the committer of the merge commit[3]
commit_roster.append(commit[3])
if commit[2] not in author_roster:
new_authors += 1
author_roster.append(commit[2])
if dateutil.parser.parse(commit[0]) > end:
print(len(by_week))
return [by_week, by_week_merge]
return [by_week, by_week_merge, new_authors, new_committers, author_roster, commit_roster]
#TODO: need to do this for all files in the dataset of readme or contributing
def for_files():
@ -106,16 +142,17 @@ def for_files():
keys = ['upstream_vcs_link', "first_readme", "b6w_prs_read", "b6w_mrg_read", "a6w_prs_read", "a6w_mrg_read"]
dict_writer = csv.DictWriter(writing_file, keys)
dict_writer.writeheader()
#training wheels on right now
for row in csv_reader:
count += 1
print(row['upstream_vcs_link'])
# this would have to get switched to false for the cont dataset
dict_row = file_get_pr(row['upstream_vcs_link'].strip(), True)
dict_writer.writerow(dict_row)
if __name__ == "__main__":
for_files()
#file_get_pr("https://github.com/tqdm/tqdm", False)
#file_get_pr("https://github.com/GameServerManagers/LinuxGSM")
#file_get_pr("https://github.com/tqdm/tqdm", True)
#file_get_pr("https://github.com/GameServerManagers/LinuxGSM", True)
#file_get_pr("https://github.com/walling/unorm/issues/new/", True)