log extract 1

msgcount1.py

#!/usr/bin/env python
# Write a script which parses /var/log/messages and generates a CSV with two columns: minute, number_of_messages in sorted time order.
import re
from collections import defaultdict
msg_data = {}
for f in open('test.log'):
        rec = f.split()
        timestamp = ' '.join(rec[:3])
        timestamp = timestamp[:-3]
        msg_data[timestamp] += 1
outrecs = [(x[0],x[1]) for x in sorted(msg_data.items(),key=lambda tup: tup[1],reverse=True)]
for o in outrecs:
        print "%s,%s" % (o[0],o[1])

msgcount2.py

#!/usr/bin/env python
# Extract the program name from the field between the hostname and the log message and output those values in columns.
import re
from collections import defaultdict
msg_data = {}
found_programs = set()
for f in open('test.log'):
        rec = f.split()
        timestamp = ' '.join(rec[:3])
        timestamp = timestamp[:-3]
        m = re.match(r'[\w\-]+',rec[4])
        if m:
            program = m.group()
            found_programs.add(program)
        if timestamp not in msg_data:
            msg_data[timestamp] = defaultdict(int)
        msg_data[timestamp]['total_count'] += 1
        msg_data[timestamp][program] += 1

title = 'minute,total_count,' + ','.join(found_programs)
print title
for k in sorted(msg_data.keys()):
        output_line = "%s,%d," % (k, msg_data[k]['total_count'])
        program_data = []
        for p in found_programs:
                program_data.append(str(msg_data[k][p]))
        output_line += ','.join(program_data)
        print output_line