Python script for plotting LoC statistics

git_loc_stats.py

#!/usr/bin/python

"""
Display the per-commit size of the current git branch.
source: http://stackoverflow.com/questions/23907/how-can-i-graph-the-lines-of-code-history-for-git-repo
"""

import subprocess
import re
import sys

import matplotlib.pyplot as plt
import numpy as np


def main(argv):
    lines_added = []
    lines_removed = []

    git = subprocess.Popen(["git", "log", "--shortstat", "--reverse",
                            "--pretty=oneline"], stdout=subprocess.PIPE)
    out, err = git.communicate()
    total_files, total_insertions, total_deletions = 0, 0, 0
    for line in out.split('\n'):
        if not line.strip():
            continue

        if line[0] != ' ':
            # This is a description line
            hash, desc = line.split(" ", 1)
        else:
            # This is a stat line
            if "deletions" and "insertions" in line:
                data = re.findall(
                    '(\d+) file[s]? changed, (\d+) insertion[s]?\(\+\), (\d+) deletion[s]?\(-\)',
                    line)

                try:
                    files, insertions, deletions = (int(x) for x in data[0])
                except:
                    print '!', line, data
            elif "insertions" in line:
                data = re.findall(
                    '(\d+) file[s]? changed, (\d+) insertion[s]?\(\+\)',
                    line)

                try:
                    files, insertions = (int(x) for x in data[0])
                except:
                    print '!', line, data
                deletions = 0
            elif "deletions" in line:
                data = re.findall(
                    '(\d+) file[s]? changed, (\d+) deletion[s]?\(\+\)',
                    line)

                try:
                    files, deletions = (int(x) for x in data[0])
                except:
                    print '!', line, data
                insertions = 0

            total_files += files

            lines_added.append(insertions)
            lines_removed.append(deletions)

    print("{}: {} files, {} lines".format(hash, total_files,
                                          sum(lines_added)-sum(lines_removed)))

    lines_added = np.array(lines_added)
    lines_removed = np.array(lines_removed)

    plt.subplot(1, 2, 1)
    plt.title('Absolute statistics')
    plt.plot(lines_added, label='added')
    plt.plot(lines_removed, label='removed')
    plt.plot(lines_added-lines_removed, label='difference')
    plt.legend(loc='best', prop={'size': 10})

    plt.subplot(1, 2, 2)
    plt.title('Cumulative statistics')
    plt.plot(lines_added.cumsum(), label='added')
    plt.plot(lines_removed.cumsum(), label='removed')
    plt.plot(lines_added.cumsum()-lines_removed.cumsum(), label='difference')
    plt.legend(loc='best', prop={'size': 10})

    plt.savefig('stats.png', dpi=400)

if __name__ == '__main__':
    sys.exit(main(sys.argv))