Creates a fingerprint of a PHP server by analyzing the output of the credits Easter egg

phpfingerprint.py

#!/usr/bin/env python

import pprint
import requests
import hashlib
from bs4 import BeautifulSoup

def normalize(s):
    return hashlib.md5(s.strip().encode('utf-8')).hexdigest()

def parse_credits_page(base_url):
    # Construct the Easter egg URL
    url = base_url
    if '?' in url:
        url = base_url[:base_url.index('?')]
    url += '?=PHPB8B5F2A0-3C92-11d3-A3A9-4C7B08C10000'
        
    # Request the contents of the page
    r = requests.get(url)
    if not '<h1>PHP Credits</h1>' in r.text:
        raise Exception('Server does not respond to Easter egg')
    
    # Parse it into a structure
    struct = { }
    
    soup = BeautifulSoup(r.text)
    for table in soup.find_all('table'):
        header = table.find('th').text
        
        struct[header] = { }
        for row in table.find_all('tr'):
            if 'class' in row and row['class'] == 'h': continue
            if row.find('th'): continue
            e, v = row.find('td', 'e'), row.find('td', 'v')
            if v is None:
                struct[header] = normalize(e.text)
            else:
                struct[header][e.text.strip()] = normalize(v.text)
    
    return struct
    

if __name__ == '__main__':
    pprint.pprint(parse_credits_page('http://en.wikipedia.org/'))