A neat way to cripple your network

crawler.py

from lxml.html import fromstring
from httplib import HTTPConnection

class Crawler(object):
    max_depth = 2

    def __init__(self, href=None, depth=0):
        if depth == self.max_depth:
            return

        self.depth = depth
        self.href = href or 'www.google.ca'
        for x in ['https://', 'http://']:
            self.href = self.href.replace(x, '')


        self.params = self.href[self.href.index('?'):] if '?' in self.href else ""
        if self.params:
            self.href = self.href.replace(self.params, '')
        self.resource = self.href[self.href.index('/'):] if '/' in self.href else '/index.html'
        if self.resource:
            self.href = self.href.replace(self.resource, '')
        if self.resource == '/': self.resource = '/index.html'

        self.run()

    def run(self):
        conn = HTTPConnection(self.href, 80)
        conn.request("GET", "%s%s" % (self.resource, self.params))
        res = conn.getresponse()
        data = fromstring(res.read())
        for link in data.cssselect('a'):
            if link.get('href') and link.get('href').startswith('http'):
                print link.get('href')
                Crawler(link.get('href'), self.depth + 1)


if __name__ == '__main__':
    c = Crawler()