gwc_decode.py

#!/usr/bin/env python
# encoding: utf-8

"""
Reads the image URIs for the various graphs in Google's webmaster tools crawl
stats page, and pulls some data estimates from it.

$ python gwc_decode.py 'https://www.google.com/chart?...'
"""

import urllib
import urlparse
import sys

alphanum = 'ABCDEFGHIJKLMONPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789'

def readURL(url):
    url_parts = urlparse.urlparse(urllib.unquote(url))

    query = urlparse.parse_qs(url_parts[4])

    if 'chd' not in query or 'chxl' not in query:
        print 'URL is not of expected format. Should have keys "chd", "chxl"'
        return

    max = int(query['chxl'][0].split('|')[3].replace(',',''))
    data = query['chd'][0][2:]

    result = []
    for datum in data:
        index = alphanum.find(datum)
        if index is not -1:
            result.append(int((index + 1) / 62.0 * max))

    print result

for arg in sys.argv[1:]:
    readURL(arg)