report.py

#!/usr/bin/env python

import sys
import csv
import operator
import os
import pwd

"""
Very simple Python filter to collect stats from pwalk output.
use as a template to create other filters

Read csv file generated by pwalk and output the biggest users
""" 

_version_ = '1.0.0'
_author_ = 'jfdey@fredhutch.org'

# pwalk fields
fields = ['inode', 'parent-inode', 'directory-depth', 'filename', 'extension',
 'UID', 'GID', 'st_size', 'st_dev', 'st_blocks"', 'st_nlink', '"st_mode"',
 'atime', 'mtime', 'ctime', 'fcount', 'sum']

users = {}
with open(sys.argv[1]) as csv_file:
    csv_reader = csv.reader(csv_file, delimiter=',', quotechar='"')
    for row in csv_reader:
        uid = row[fields.index('UID')]
        if uid in users:
            users[uid]['sum'] += int(row[fields.index('st_size')])
            users[uid]['fcount'] += 1 
        else:
            users[uid] = {}
            users[uid]['sum'] = int(row[fields.index('st_size')])
            users[uid]['fcount'] = 1 

# create new hash to be sorted by size
summary = {}
for user in users:
    summary[user] = users[user]['sum']

# sorted_size is list of tuples (uid, size)
sorted_size = sorted(summary.items(), key=operator.itemgetter(1))
print(type(sorted_size))

for user in sorted_size:
    try:
        uname = pwd.getpwuid(int(user[0])).pw_name
    except KeyError:
        uname = user[0]
    print('{} UID: {} FileCount: {}'.format(user[1], uname, users[user[0]]['fcount']))