habraproxy.py

habraproxy.py

# -*- coding: utf-8 -*-
import BaseHTTPServer
import requests
from bs4 import BeautifulSoup
import string


HOST = ''
PORT = 8232


def main():
    server_class = BaseHTTPServer.HTTPServer
    server_address = (HOST, PORT)
    httpd = server_class(server_address, Handler)
    httpd.serve_forever()


class Handler(BaseHTTPServer.BaseHTTPRequestHandler):
    site = 'habrahabr.ru'

    def do_GET(self):
        url = 'http://' + self.site + self.path
        req = requests.get(url)
        content = req.content.replace('http://' + self.site, '')
        headers = req.headers
        self.send_response(200)
        self.send_header('Content-type', headers['content-type'])
        self.end_headers()
        if headers['content-type'].startswith('text/html'):
            content = transform_html(content)
        self.wfile.write(content)


def transform_html(html):
    skip = ('style', 'script', '[document]', 'head', 'title')
    soup = BeautifulSoup(html)
    for tag in soup.find_all(text=True):
        if tag.parent.name in skip:
            continue
        s = tag.string.strip()
        if s:
            tag.string.replace_with(transform(s))
    return soup.prettify().encode('utf-8')


def is_tm_word(s):
    return len(s) == 6 and not s.endswith(string.punctuation)


def transform(s):
    return u' '.join(x + u'™' if is_tm_word(x) else x for x in s.split())


if __name__ == '__main__':
    main()