add to all words > 6 letters 'tm'

habraproxy.py

#!/usr/bin/env python
# -*- coding: utf-8 -*-

import string
import webbrowser
import re
from BaseHTTPServer import BaseHTTPRequestHandler, HTTPServer

# pip install requests
# pip install beautifulsoup4
# pip install lxml
import requests
from bs4 import BeautifulSoup, Comment
from bs4.dammit import EntitySubstitution
from bs4.element import NavigableString

SITE = 'https://habrahabr.ru'
PORT = 8232

# Regexp pattern for word with 6 letters
REGEXP = re.compile(ur'\b(\w{6})\b', re.UNICODE)

# list of tags which content can not be replaced
NOT_ALLOWED_TAGS = [
    'body',
    'head',
    'link',
    'meta',
    'style',
    'title',
    'script',
]


def all_but_comments(text):
    """
    What happens in comments stays in comments
    Check comment for new searchbox :)
    """
    return re.search(REGEXP, text) and not isinstance(text, Comment)


def add_trademark(word):
    word_length = len(word.strip(string.punctuation + string.whitespace))
    return word + u'\u2122' if word_length == 6 else word


def transform_content(page_tree):
    """
    Adds trademark symbol to word if word length equals 6
    page_tree is BeautifulSoup object
    """
    for content in page_tree.find_all(string=all_but_comments):
        splitted = re.split(REGEXP, unicode(content))
        transformed_text = u''.join(add_trademark(word) for word in splitted)
        if content.parent.name not in NOT_ALLOWED_TAGS:
            content.replace_with(transformed_text)
    return page_tree.encode('utf-8', formatter=habraformatter)


def habraformatter(entity):
    """
    Format html only if it's in <code> tag. Helps with xml entities
    which is invisible unless converted into valid html.
    Otherwise just return entity - we don't want to escape all the things
    (plus signs on frontpage for example)
    """
    if isinstance(entity, NavigableString) and entity.parent.name == 'code':
        return EntitySubstitution.substitute_xml(entity)
    else:
        return entity


class HttpProcessor(BaseHTTPRequestHandler):

    def do_GET(self):
        r = requests.get(SITE + self.path)

        # Replace url for better navigation
        request_content = r.content.replace(SITE, '')
        self.send_response(r.status_code)
        self.send_header('Content-type', r.headers['content-type'])
        self.end_headers()

        if 'text/html' in r.headers['content-type']:
            page_tree = BeautifulSoup(request_content, 'lxml')
            request_content = transform_content(page_tree)

        self.wfile.write(request_content)


def main():
    server = HTTPServer(("localhost", PORT), HttpProcessor)
    webbrowser.open_new('http://localhost:%s' % PORT)
    server.serve_forever()


if __name__ == '__main__':
    main()