def get_dataframe_from_repo(repo, num=100):
"""Create pandas dataframe of contributors by country.
Args:
repo - a full GitHub repo URL
num - number of contributors to analyze per repo
Returns:
df - a pandas dataframe of contributors by country
num_contributors - total number of contributors
"""
# get contributors
repo_ending_string = extract_github_owner_and_repo(repo)
contributors = get_contributors(repo_ending_string, num)
num_contributors = len(contributors)
# get count of countries
country_list = []
for contributor in contributors:
location = get_contributor_location(contributor)
country = get_country_from_location(location)
country_list.append(country)
country_counter = Counter(country_list)
# convert counter to pandas dataframe
df = pd.DataFrame.from_records(country_counter.most_common(),
columns=["country", "contributor_count"])
return df, num_contributors
def scan_single_repo(repo, summary, output_csv, num=100):
"""Print location results for single GitHub repository.
Printing can either be by contributor or by country.
Output can optionally be stored as a csv.
Args:
repo - URL of repo
summary - whether to print results by country, i.e. summary.
output_csv - whether to store output in csv (default: false)
num - max number of contributors to analyze
Returns:
null
"""
repo_ending_string = extract_github_owner_and_repo(repo)
contributors = get_contributors(repo_ending_string, num)
print("-----------------")
print("GITHUB REPO: {}".format(repo_ending_string))
print("-----------------")
if summary:
print_by_country(contributors)
else:
print_by_contributor(repo_ending_string, contributors, output_csv)
def test_print_by_country(capsys):
"""Unit test for print_by_country() for networml python package."""
repo = "https://www.github.com/iqtlabs/networkml"
repo_ending_string = extract_github_owner_and_repo(repo)
contributors = get_contributors(repo_ending_string)
print_by_country(contributors)
captured = capsys.readouterr() # capture outpt printed to date
# dedent removes spacing, using the spacing width from the first line
output_text = textwrap.dedent(""" COUNTRY | # OF CONTRIBUTORS
---------------------------
None 11
United States 4
New Zealand 2\n""")
assert captured.out == output_text
def scan_single_repo(repo, summary):
"""Print location results for single GitHub repository
Printing can either be by contributor or by country
Args:
repo - URL of repo on GitHub
Returns:
null
"""
repo_ending_string = extract_github_owner_and_repo(repo)
contributors = get_contributors(repo_ending_string)
print("-----------------")
print("GITHUB REPO: {}".format(repo_ending_string))
print("-----------------")
if summary:
print_by_country(contributors)
else:
print_by_contributor(contributors)
def scan_multiple_repos(input_file="repos.txt", num=100):
"""Create csv of data for multiple repos.
Scan through repos provided in repos.txt and create a single csv that
stores all contributor-related data for each contributor in each repo.
Args:
input_file - file containing repo list
num - max number of contributors to analyze per repo
Returns:
None
"""
# create csv to store multi-repo scan results
timestamp = time.strftime("%Y%m%d-%H%M%S")
create_csv("multirepo", timestamp)
# open file that contains repos to scan and append contributors for each
# repo to csv. Also, repos.txt must contain repo names, one repo per line.
with open(input_file, "r") as input_repos:
for repo in input_repos:
# Skip blank lines
if repo == "":
continue
# strip blank space before extracting owner and repo name
repo_ending_string = extract_github_owner_and_repo(repo.strip())
contributors = get_contributors(repo_ending_string, num)
for contributor in contributors:
location = get_contributor_location(contributor)
country = get_country_from_location(location)
add_committer_to_csv(
"multirepo",
repo_ending_string,
timestamp,
contributor,
location,
country,
)
def test_extract_github_owner_and_repo(self):
"""Unit test for extract_github_owner_and_repo()."""
owner_and_repo = extract_github_owner_and_repo(
"www.github.com/psf/requests")
assert owner_and_repo == "psf/requests"