get median of an Apache httpd log file

get_median.py

from flask import Flask
import re
import math
import redis 
import zlib
import ast

app = Flask(__name__)

LOG_FILE = '/var/log/access.log'

red = redis.Redis()

def get_sizes():
    regex='([(\\S\\.)]+) - - \\[(.*?)\\] "(.*?)" (\\d+) (\\d+)'
    sizes = {}
    with open(LOG_FILE) as f:
        for line in f:
            # In case the request has a - in place of size, replace it with 0
            if line[-2] == '-':
                line = line[:-2]+'0'
            log_groups=re.match(regex, line).groups()
            if log_groups[3] in sizes:
                sizes[log_groups[3]].append(int(log_groups[4]))
            else:
                sizes[log_groups[3]] = [int(log_groups[4])]
    return sizes

def get_median(resp_code):
    '''
    https://en.wikipedia.org/wiki/Median
    The median is the value separating the higher half of a data sample, a population, or a probability distribution, from the lower half. In simple terms, it may be thought of as the "middle" value of a data set.
    '''
    adler = zlib.adler32(open(LOG_FILE).read())
    median_str = red.get(adler)
    if median_str:
        median_dict = ast.literal_eval(median_str)
    else:
        sizes = get_sizes() 
        median_dict = {}
        for key in sizes:
            sizes[key].sort()
            length = len(sizes[key])
        # Remember that index of a list starts at 0.
            if length % 2 == 0:
                median = (sizes[resp_code][length/2 - 1] + sizes[resp_code][length/2])/2.0
            else:
                median = float(sizes[resp_code][int(math.floor(length/2))])
            median_dict[key] = median
        red.set(adler, str(median_dict))
    return median_dict[resp_code]

@app.route('/<resp_code>', methods=["GET"])
def get_response(resp_code):
    return str(get_median(resp_code))


if __name__ == '__main__':
    app.run(host="0.0.0.0", port=8888)