simple http proxy

proxy.py

# -*- coding: utf-8 -*-

import argparse
import re
import requests
from bs4 import BeautifulSoup, Comment

from flask import Flask
from flask import Response

app = Flask(__name__)


@app.route('/')
@app.route('/<path:url>')
def proxy(url=''):
    url = '{}/{}'.format(args.target, url)
    req = requests.get(url)

    content_type = req.headers['content-type']
    content = req.content
    if 'text/html' in content_type:
        soup = BeautifulSoup(content, 'html5lib')
        for line in soup.find_all(string=re_marker):
            if line.parent.name in ('script', 'style', 'code', '[document]'):
                continue
            if line.parent(text=lambda x: isinstance(x, Comment)):
                continue
            text = re.sub(u'\\b(\\w{{{}}})\\b'.format(args.symbols), u'\\1™', line, flags=re.U)
            line.replace_with(text)
        for link in soup.find_all('a', href=re_source):
            link['href'] = link['href'].replace(args.target, 'http://{}:{}'.format(args.hostname, args.port))
        content = str(soup)
    return Response(content, content_type=req.headers['content-type'])


if __name__ == '__main__':
    parser = argparse.ArgumentParser(description='TM HTTP Proxy.')
    parser.add_argument('--hostname', default='127.0.0.1', help='Default: 127.0.0.1')
    parser.add_argument('--port', default='8232', help='Default: 8232')
    parser.add_argument('--target', default='https://habrahabr.ru', help='Default: https://habrahabr.ru')
    parser.add_argument('--symbols', default='6', help='Default: 6')

    args = parser.parse_args()
    re_source = re.compile(args.target, re.U)
    re_marker = re.compile(u'.*', re.U)
    print('Proxy for {}'.format(args.target))

    app.run(host=args.hostname, port=int(args.port))