compute histogram based on text file line length w/ csv optional output

line_length_hist.py

#!/usr/bin/env python3
import sys
import os
import argparse
import operator
import distutils.util


def generate_options():
    parser = argparse.ArgumentParser(description='histogram per file lenth')
    parser.add_argument(
        '--file',
        type=str,
        help='file that is new line delimited. i.e: --file=foo.json')
    parser.add_argument(
        '--csv',
        type=distutils.util.strtobool,
        default='true',
        help='print to stdout as csv')
    parser.add_argument(
        '--buckets',
        type=int,
        default=10,
        help='number of buckets to print. i.e: --buckets=10')

    return parser


def read_file(filename):
    with open(filename, 'r') as f:
        return f.readlines()


def histogram_for_dict(od, buckets):

    m = {}
    ks = sorted(od.keys())
    first_k = ks[0]
    last_k = ks[-1]
    step = int((last_k - first_k) / buckets)
    for i in range(first_k, last_k, step):
        m[i] = int(0)
        for k in ks:
            if k >= i and k < i + step:
                m[i] = m[i] + od[k]
    return m


def compute_frequencies(lines):
    m = {}
    for l in lines:
        s = len(l)
        if s == 0: continue
        v = 1 + m.get(s, 0)
        m[s] = v
    return m


def print_histogram(csv_print, hist):
    total = sum(hist.values())
    h = sorted(hist.keys())
    counter = 1
    v_counter = 0
    first_v = hist[h[0]]
    if csv_print:
        sys.stdout.write(
            "payload_size,number_of_requests,cummulative_requests,percentile\n"
        )
    for k in h:
        v = hist[k]
        v_counter = v_counter + v
        if csv_print:
            sys.stdout.write(
                "%s,%s,%s,%s\n" % (k, v, v_counter,
                                   round(100 * (v_counter / total), 5)))
        else:
            sys.stdout.write("%s:\t%s\t%s\t%s\t%s\n" % (
                counter, k, v, v_counter, round(100 * (v_counter / total), 5)))
        counter = counter + 1


def main():
    parser = generate_options()
    options, program_options = parser.parse_known_args()
    print_histogram(
        options.csv,
        histogram_for_dict(
            compute_frequencies(read_file(options.file)), options.buckets - 1))


if __name__ == '__main__':
    main()