#!/usr/bin/env python3
"""
Analyze Segwit2x contributors via git history.

To use:

  git clone git@github.com:bitcoin/bitcoin.git
  cd bitcoin
  git remote add 2x git@github.com:btc1/bitcoin.git
  git fetch 2x
  [wget $THIS_URL, chmod +x, whatever]
  git log origin/master..2x/segwit2x-dev --shortstat --pretty="%cE"  | ./analyze2x.py

"""
import fileinput
import re
from collections import namedtuple, defaultdict

Stat = namedtuple('Stat', 'adds rms files')

changeline_regexp = (
    r'(?P<files>\d+) file.? changed, '
    r'((?P<adds>\d+) insertion.*)?'
    r'((?P<rms>\d+) delet.*)?')


definitely_not_a_2x_dev = {
    'pieter.wuille@gmail.com',
    'practicalswift@users.noreply.github.com',
    'laanwj@gmail.com',
    'noreply@github.com',
    'sdaftuar@gmail.com',
    'dev@jonasschnelli.ch',
    'practicalswift@users.noreply.github.com',
    'russ@yanofsky.org',
    'morcos@chaincode.com',
    'luke-jr+git@utopios.org',
    'sdaftuar@chaincode.com',
    'git@bluematt.me',
    'cory-nospam-@coryfields.com',
    'falke.marco@gmail.com',
    'greg@xiph.org',
    'dev@jonasschnelli.ch',
    'sdaftuar@gmail.com',
}


def main():
    authors_to_stats = defaultdict(list)
    authors_per_line = set()
    lines = fileinput.input()

    for line in lines:
        line = line.strip()

        m = re.search(changeline_regexp, line)
        if m:
            stat = Stat(**m.groupdict())
            for author in authors_per_line:
                authors_to_stats[author].append(stat)
            authors_per_line = set()
        elif line:
            if line not in definitely_not_a_2x_dev:
                authors_per_line.add(line)

    total_stat = aggregate_stat(
        stat for stats in authors_to_stats.values() for stat in stats)

    prettyprint(aggregate(authors_to_stats), total_stat)


def aggregate(authors_to_stats):
    out = {}

    for author, stats in authors_to_stats.items():
        out[author] = aggregate_stat(stats)

    return list(reversed(sorted(
        out.items(), key=lambda i: i[1].adds + i[1].rms)))


def aggregate_stat(stats):
    def to_int(a):
        return int(a) if a else 0

    return Stat(
         adds=sum(to_int(s.adds) for s in stats),
         rms=sum(to_int(s.rms) for s in stats),
         files=sum(to_int(s.files) for s in stats),
     )


def prettyprint(author_stats, total_stat):
    total_changes = total_stat.rms + total_stat.adds

    for author, stat in author_stats:
        perc = (stat.adds + stat.rms) / total_changes
        print("{:>6} {:<40} {:>8} {:>8} {:>20}".format(
            f'{int(perc * 100)}%',
            author,
            f'+{stat.adds}',
            f'-{stat.rms}',
            f'(over {stat.files} files)',
        ))


if __name__ == '__main__':
    main()