Quick-and-dirty script to get some data about a repo

analyze-repo

#!/usr/bin/env python

"""
Usage::
    pip install gitpython
    cd <repo>
    analyze-repo
"""

from __future__ import absolute_import, division, print_function, unicode_literals

import os

import git


def main():
    repo = git.Repo(os.getcwd())
    commits = repo.git.rev_list(all=True).splitlines()
    tested_commits = []

    # Use git show to list the files changed in each commit
    for commit in commits:
        # We could get more precise with a regex, but 'in' is good enough & faster.
        if 'test' in repo.git.show(commit, pretty='format:', name_only=True):
            tested_commits.append(commit)

    total_commits = len(commits)
    commits_tested = len(tested_commits)

    print('Total no. of commits:            {0:>6,}'.format(total_commits));
    print('No. of commits that touch tests: {0:>6,}'.format(commits_tested));
    print('% of commits that touch tests:   {0:>10.2%}'.format(commits_tested / total_commits))


if __name__ == '__main__':
    main()