Parse nginx access.log and access.log.gz files into namedtuples

nginx.py

import csv
from collections import namedtuple
import gzip
from os import listdir
from os.path import join


Entry = namedtuple("Entry", "ip, blank, user, time, offset, request, status, sent, referer, agent")


def main(curdir="."):
    entries = []

    for f in listdir(curdir):
        if not f.startswith("access.log"):
            continue

        opener = gzip.open if f.endswith(".gz") else open
        full = join(curdir, f)
        with opener(full, "rt") as f:
            entries.extend(map(Entry._make, csv.reader(f, delimiter=" ")))


if __name__ == "__main__":
    main()