Parse tcpdump output

tcpdump_parser.py

import pandas as pd
import re

space_splitter = re.compile("\s+")
regex = re.compile("\s*(.*)\s*>\s*(.*?):\s.*")

def parse_line(line):
    try:
        elements = space_splitter.split(line)
        source_dest = regex.match(" ".join(elements[4:])).groups()
    except Exception:
        return None
    return tuple(elements[:4]) + tuple(x.strip() for x in source_dest)


with open("./output.txt", "r") as fid:
    lines = fid.readlines()

items = [parsed + (idx,) for idx, line in enumerate(lines) if (parsed := parse_line(line))]
non_records = [idx for idx, line in enumerate(lines) if not parse_line(line)]

df = pd.DataFrame.from_records(items, columns=["time", "network", "direction", "class", "source", "target", "idx"])