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 print_by_contributor(software_name,
contributors,
output_csv=False,
pypi_data=None):
"""Print location results by contributor.
Print contributors and countries to terminal window. If output csv is set
to true, then also output results to a csv file.
Args:
software_name - name of package or repo
contributors - a list of contributors
output_csv - whether to output a csv.
pypi_data - a pypi data object.
Returns:
null
"""
# create csv if output_csv specified
if output_csv:
# unique current time timestamp to create unique filename
timestamp = time.strftime("%Y%m%d-%H%M%S")
create_csv("contributor", timestamp)
print("CONTRIBUTOR, LOCATION")
if pypi_data is not None:
print("* indicates PyPI maintainer")
print("---------------------")
for contributor in contributors:
location = get_contributor_location(contributor)
country = get_country_from_location(location)
if output_csv:
add_committer_to_csv("contributor", software_name, timestamp,
contributor, location, country)
try:
# Check if pypi_data is not None, indicating a PyPI package scan
if pypi_data is not None and contributor in pypi_data[
"pypi_maintainers"]:
print(contributor, "*", "|", location, "|", country)
else:
print(contributor, "|", location, "|", country)
except UnicodeEncodeError:
print(contributor, "| error")
def print_by_country(contributors):
"""Print contributors aggregated by country.
Print contributor county by country to terminal window.
Args:
contributors: a list of contributors
Returns:
null
"""
print("COUNTRY | # OF CONTRIBUTORS")
print("---------------------------")
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)
for country, count in country_counter.most_common():
print(country, count)
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_get_contributor_location(self):
"""Unit test for get_contributor_location()."""
assert get_contributor_location("anarkiwi") == "Wellington, New Zealand"