Python parsing log file for IP address and Protocol

Question

this is my first question asked here at stackoverflow and am really looking forward to being part of this community. I am new to program and python was the most recommended first program by many people.

Anyways . I have a log file which looks like this:

"No.","Time","Source","Destination","Protocol","Info"
"1","0.000000","120.107.103.180","172.16.112.50","TELNET","Telnet Data ..." 
"2","0.000426","172.16.112.50","172.16.113.168","TELNET","Telnet Data ..." 
"3","0.019849","172.16.113.168","172.16.112.50","TCP","21582 > telnet [ACK]" 
"4","0.530125","172.16.113.168","172.16.112.50","TELNET","Telnet Data ..." 
"5","0.530634","172.16.112.50","172.16.113.168","TELNET","Telnet Data ..."

And I wanted to parse the log file using Python to make it look like this as the result:

From IP 135.13.216.191 Protocol Count: (IMF 1) (SMTP 38) (TCP 24) (Total: 63)

I would really like some help on what path to take to tackle this problem should I use lists and loop through it or dictionaries/tuples?

Thanks in advance for your help!

Answer 1

You can parse the file using the csv module :

import csv

with open('logfile.txt') as logfile:
     for row in csv.reader(logfile):
         no, time, source, dest, protocol, info = row
         # do stuff with these

I can't quite tell what you're asking, but I think you want:

import csv
from collections import defaultdict

# A dictionary whose values are by default (a
# dictionary whose values are by default 0)
bySource = defaultdict(lambda: defaultdict(lambda: 0))

with open('logfile.txt') as logfile:
     for row in csv.DictReader(logfile):
         bySource[row["Source"]][row["Protocol"]] += 1

for source, protocols in bySource.iteritems():
    protocols['Total'] = sum(protocols.values())

    print "From IP %s Protocol Count: %s" % (
        source,
        ' '.join("(%s: %d)" % item for item in protocols.iteritems())
    )

Answer 2

I would begin by first reading the file into a list:

contents = []
with open("file_path") as f:
    contents = f.readlines()

Then you can split each line into a list of it's own:

ips = [l[1:-1].split('","') for l in contents]

We can then map these into a dict:

sourceIps = {}
for ip in ips:
    try:
       sourceIps[ip[2]].append(ip)
    except:
       sourceIps[ip[2]] = [ip]

And finally print out the result:

for ip, stuff in sourceIps.iteritems():
   print "From {0} ... ".format(ip, ...)

Answer 3

First you'll want to read in the text file

# Open the file
file = open('log_file.csv')
# readlines() will return the data as a list of strings, one for each line
log_data = file.readlines()
# close the log file
file.close()

Set up a dictionary to hold your results

results = {}

Now iterate over your data, one line at a time, and record the protocol in the dictionary

for entry in log_data:
    entry_data = entry.split(',')
    # We are going to have a separate entry for each source ip
    # If we haven't already seen this ip, we need to make an entry for it
    if entry_data[2] not in results:
        results[entry_data[2]] = {'total':0}
    # Now check to see if we've seen the protocol for this ip before
    # If we haven't, add a new entry set to 0
    if entry_data[4] not in results[entry_data[2]]:
         results[entry_data[2]][entry_data[4]] = 0
    # Now we increment the count for this protocol
    results[entry_data[2]][entry_data[4]] += 1
    # And we increment the total count
    results[entry_data[2]]['total'] += 1

Once you've counted everything, just iterate over your counts and print out the results

for ip in results:
    # Here we're printing a string with placeholders. the {0}, {1} and {2} will be filled
    # in by the call to format
    print "from: IP {0} Protocol Count: {1})".format(
        ip,
        # And finally create the value for the protocol counts with another format call
        # The square braces with the for statement inside create a list with one entry
        # for each entry, in this case, one entry for each protocol
        # We use ' '.join to join each of the counts with a string
        ' '.join(["({0}: {1})".format(protocol, results[ip][protocol] for protocol in results[ip])]))