def sava_bugs_for_project(repo, jira_key):
if not os.path.exists(repo):
print "start git clone https://github.com/apache/{0}.git".format(
os.path.basename(repo))
return
if not "pom.xml" in os.listdir(repo):
return
versions = get_repo_versions(repo)
if len(versions) < 5:
return
save_bugs(
os.path.join(VERSIONS, jira_key) + ".csv", repo,
r"http://issues.apache.org/jira", jira_key, versions)
def get_repo_adata(repo_path, jira_key):
if not os.path.exists(repo_path):
print("start git clone https://github.com/apache/{0}.git".format(
os.path.basename(repo_path)))
return
versions = get_repo_versions(repo_path)
if len(versions) < 4:
print("Less than 5 versions for repo, aborting")
return
save_tickets_stats(
os.path.join(VERSIONS, jira_key) + "_STATS.csv", repo_path,
r"http://issues.apache.org/jira", jira_key, versions)
def get_versions_by_type(repo):
import re
all_versions = get_repo_versions(repo)
majors = []
minors = []
micros = []
SEPERATORS = ['\.', '\-', '\_']
template_base = [['([0-9])', '([0-9])([0-9])', '([0-9])$'],
['([0-9])', '([0-9])([0-9])$'],
['([0-9])', '([0-9])', '([0-9])([0-9])$'],
['([0-9])([0-9])', '([0-9])$'],
['([0-9])', '([0-9])', '([0-9])$'],
['([0-9])', '([0-9])$']]
templates = []
for base in template_base:
templates.extend(map(lambda sep: sep.join(base), SEPERATORS))
templates.extend(['([0-9])([0-9])([0-9])$', '([0-9])([0-9])$'])
for version in all_versions:
for template in templates:
values = re.findall(template, version._name)
if values:
values = map(int, values[0])
if len(values) == 4:
micros.append(version)
major, minor1, minor2, micro = values
minor = 10 * minor1 + minor2
elif len(values) == 3:
micros.append(version)
major, minor, micro = values
else:
major, minor = values
micro = 0
if micro == 0:
minors.append(version)
if minor == 0 and micro == 0:
majors.append(version)
break
return {
"all": all_versions,
"majors": majors,
"minors": minors,
"micros": micros
}.items()
def visualize_repo(repo_path):
if not os.path.exists(repo_path):
print("start git clone https://github.com/apache/{0}.git".format(
os.path.basename(repo_path)))
return
versions = get_repo_versions(repo_path)
if len(versions) < 4:
print("Less than 5 versions for repo, aborting")
return
version_tag_dates = [v._commit._commit_date for v in versions]
repo = git.Repo(repo_path)
print('getting master')
dates = []
num_files = []
num_additions = []
num_deletions = []
total_files = []
for git_commit in repo.iter_commits():
if not git_commit.parents:
text_file_stats = repo.git.diff(git_commit.hexsha,
'--',
'--name-status',
numstat=True)
else:
text_file_stats = repo.git.diff(git_commit.parents[0].hexsha,
git_commit.hexsha,
'--name-status',
numstat=True)
all_statuses = [
line.split("\t")[0] for line in text_file_stats.splitlines()
]
dates.append(git_commit.committed_datetime)
num_deletions.append(len([s for s in all_statuses if 'D' in s]) * -1)
num_additions.append(len([s for s in all_statuses if 'A' in s]))
num_files.append(len(all_statuses))
total_files.append(
len([
1 for item in git_commit.tree.traverse() if item.type == 'blob'
]))
#
#
# commits_dates = [c.committed_datetime for c in commits]
#
# print('got dates ', len(commits_dates))
#
# print('single commit stats ')
# print(dir(commits[0].stats))
#
# commits_num_files = [len(c.stats.files) for c in commits]
#
# print('got num files ', len(commits_num_files))
from matplotlib import pyplot as plt
fig, ax = plt.subplots(2, 1, sharex=True)
ax[0].plot(dates, num_additions, label='additions')
ax[0].plot(dates, num_deletions, label='deletions')
ax[1].plot(dates, num_files, label='commit files')
ax[1].plot(dates, total_files, label='total files')
ax[0].set(title=os.path.split(repo_path)[1])
ax[0].legend()
ax[1].set(xlabel="dates", ylabel="# files")
ax[1].legend()
# import plot_repos
# plot_repos.plot_line( dates, {'additions': num_additions,
# 'deletions': num_deletions,
# 'commit files': num_files,
# 'total files': total_files},
# "dates", "#files", os.path.split(repo_path)[1], num_subplots=3)
pass
ratio_bugged_commits,
datetime.fromtimestamp(
tag.tag._commit._commit_date).strftime("%Y-%m-%d")
])
def main(out_file, gitPath, jira_url, jira_project_name):
commits = get_data(jira_project_name, jira_url, gitPath)
with open(out_file, "wb") as f:
writer = csv.writer(f)
writer.writerows([c.to_list() for c in commits])
if __name__ == "__main__":
repo = git.Repo(r"c:\temp\tika")
versions = get_repo_versions(r"c:\temp\tika")
tags_commits = get_commits_between_versions(
map(lambda c: Commit.init_commit_by_git_commit(c, 0),
list(repo.iter_commits())[:1000]), versions)
tags = []
for tag in tags_commits:
tags.append(VersionInfo(tag, tags_commits[tag]))
import apache_repos
from caching import REPOSIROTY_DATA_DIR
VERSIONS = os.path.join(REPOSIROTY_DATA_DIR, r"apache_versions")
repos_and_jira = apache_repos.get_apache_repos_data()
for repo, jira_key in repos_and_jira:
if not os.path.exists(repo):
print "start git clone https://github.com/apache/{0}.git".format(
os.path.basename(repo))
continue