Script for sorting csv file with a list of a domains with a Page Rank

pr_sort.py

#!/usr/bin/env python
# -*- coding: utf-8 -*-

"""Page Rank CSV sorter

Usage:
    pr_sort.py /path/to/csv-file.csv [/path/to/output-file.csv]

    Output filename is not required, and if not supplied, output goes to the stdout.
"""

import sys
import os.path
import urlparse

def parse(input, output):

    urls = {}

    # Parse initial file
    for line in open(input):
        url, rank = line.strip().split(',')
        url = urlparse.urlsplit(url).netloc
        urls[url] = int(rank)

    results = {}

    # Iterate over parsed data and remove duplicates
    for url in urls.iterkeys():

        # Determine both www-parted and non-www-parted URLs
        if url.startswith('www.'):
            www_url = url
            non_www_url = url[4:]
        else:
            www_url = 'www.%s' % url
            non_www_url = url

        # Get Page Ranks for them
        www_rank = urls.get(www_url)
        non_www_rank = urls.get(non_www_url)

        # Compare now
        if www_rank == non_www_rank:
            results[www_url] = www_rank
        elif www_rank > non_www_rank:
            results[www_url] = www_rank
        else:
            results[non_www_url] = non_www_rank

    if output:
        result_file = open(output, 'w')
        for url, rank in results.iteritems():
            result_file.write('http://%s,%s\n' % (url, rank))
        result_file.close()
    else:
        for url, rank in results.iteritems():
            print 'http://%s,%s' % (url, rank)

if __name__ == '__main__':
    if len(sys.argv) < 2:
        print 'Usage: %s /path/to/csv-file.csv [/path/to/output-file.csv]' % sys.argv[0]
        sys.exit(-1)

    if not os.path.exists(sys.argv[1]):
        print 'File `%s` doesnt exists' % sys.argv[1]
        sys.exit(-2)

    try:
        output = sys.argv[2]
    except IndexError:
        output = None

    parse(sys.argv[1], output)