git-blame-stats.py

#!/usr/bin/env python
#-*- coding: utf-8 -*-
import os
import sys
import re

def main(additional=''):
    re_author = re.compile(r'^.*?\((.*?)\s*\d{4}-\d{2}-\d{2}.*')
    authors = {}
    for filename in os.popen('git ls-tree --name-only -r HEAD'):
        sys.stderr.write('.')
        for line in os.popen('git blame %s HEAD %s' % ( additional, filename )):
            if line.startswith('^'):
                continue
            m = re_author.match(line)
            author = m.groups()[0]
            if not authors.has_key(author):
                authors[author] = 0
            authors[author] += 1
    m = sum(authors.values())
    print '\n%s lines' % m
    for name, lines in authors.items():
        print '%s: %02.02f (%s lines)' % (name, lines/float(m)*100, lines)


if __name__ == "__main__":
    since = ''
    if len( sys.argv ) > 1:
        since = '--since=%s' % sys.argv[1]
    main(since)