IBM Web Server Log Parse

parse.py

from hashlib import md5 as xx
from collections import namedtuple
import os
import re

# Environment parametes
COLUMN_SEPARATOR = chr(os.getenv('SEPARATOR', 449))
FILE_NAME = os.getenv('FILE', 'hashed.csv')

# Regular expression to capture JSESSIONID
JSESSION_PATTERN = r'JSESSIONID=([a-zA-Z0-9_\-]+)?[;\s]?'

# Data Structure for log entries
AccessLogRow = namedtuple('AccessLogRow', 'ip date method code url referer agent cookie version id')
with open("logs.csv") as reader, open(FILE_NAME, "w") as writer:
  for line in reader:
    # Convert log entry to data structure
    log_entry = AccessLogRow(*line.split(COLUMN_SEPARATOR))
    
    # Hashing Fields
    haship = xx(log_entry.ip.encode('utf-8')).hexdigest()

    # Default value for jsession or pattern matched
    jsessionid = 'EMPTY'
    if "JSESSIONID" in log_entry.cookie:
        jsessionid = re.search(JSESSION_PATTERN, log_entry.cookie).group(1)

    writer.write(COLUMN_SEPARATOR.join(
        [haship] +
        [log_entry.date] +
        [log_entry.method] +
        [log_entry.code] +
        [log_entry.url] +
        [log_entry.referer] +
        [log_entry.agent] +
        [jsessionid] + 
        [log_entry.version] +
        [log_entry.id]))