carbon_date.py

#!/usr/bin/env python
"""
Prints a histogram of the modifications dates of all the files
in the specified directory.

This is useful for determining the overall last modified date.

Author: David Foster (davidfstr)
Created: 2012-02-02
Environment: Python 2.7
"""

from collections import defaultdict
from datetime import datetime
import os
import os.path
import sys

HISTOGRAM_WIDTH = 80
IGNORED_EXTENSIONS = [
    '.class',
    '.pyc',
    '.pyo',
]

# Parse arguments
args = sys.argv[1:]
if len(args) != 1:
    exit('syntax: python carbon_date.py <dirpath>')
root_dirpath = args[0]

def is_ignored_file(filename):
    if filename == '.DS_Store':
        return True

    # Has an ignored extension?
    for ext in IGNORED_EXTENSIONS:
        if filename.endswith(ext):
            return True

    return False

def quarter_create(dt):
    return str(dt.year) + 'Q' + str(((dt.month-1) // 3) + 1)

def quarter_next(q):
    q_parts = map(int, q.split('Q'))
    q_parts[1] += 1
    if q_parts[1] == 5:
        q_parts = [q_parts[0]+1, 1]
    return 'Q'.join(map(str, q_parts))

bins = defaultdict(lambda: 0)
values = []
for (dirpath, dirnames, filenames) in os.walk(root_dirpath):
    for filename in filenames:
        if is_ignored_file(filename):
            continue
        filepath = os.path.join(dirpath, filename)
        if not os.path.exists(filepath):
            continue
        modified_dt = datetime.fromtimestamp(os.path.getmtime(filepath))

        bins[quarter_create(modified_dt)] += 1
        values.append(modified_dt)
values = sorted(values)

cur_bin = min(bins.keys())
max_bin = max(bins.keys())
max_freq = max(bins.values())
while cur_bin <= max_bin:
    cur_bar_length = bins[cur_bin] * HISTOGRAM_WIDTH // max_freq
    print '%s: %s%s|' % (
        cur_bin,
        '*' * cur_bar_length,
        ' ' * (HISTOGRAM_WIDTH - cur_bar_length),
    )

    cur_bin = quarter_next(cur_bin)
print

median_value = values[(len(values) - 1) // 2]
percentile_80 = values[int((len(values) - 1) * .80)]
percentile_90 = values[int((len(values) - 1) * .90)]
percentile_95 = values[int((len(values) - 1) * .95)]
print 'Median: %s' % median_value.ctime()
print '80th Percentile: %s' % percentile_80.ctime()
print '90th Percentile: %s' % percentile_90.ctime()
print '95th Percentile: %s' % percentile_95.ctime()