-
-
Save timka/1131123ea18ef3c676f7 to your computer and use it in GitHub Desktop.
Программа не учитвает ситуации, когда внутри слова есть тэг (Log<b>Jam</b>) ибо не охота дальше усложнять. PEP8 — федеральный закон, Python Zen — конституция, где readability does count
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
# coding: utf-8 | |
# Программа не учитвает ситуации, когда внутри слова есть тэг (Log<b>Jam</b>) | |
# ибо не охота дальше усложнять. | |
# PEP8 — федеральный закон, Python Zen — конституция, где readability does count | |
import argparse | |
import webbrowser | |
import multiprocessing | |
import re | |
import sys | |
import SocketServer | |
import SimpleHTTPServer | |
from StringIO import StringIO | |
from HTMLParser import HTMLParser | |
from xml.sax.saxutils import escape, quoteattr | |
import requests | |
# Make setdefaultencoding available | |
sys = reload(sys) | |
class Proxy(SimpleHTTPServer.SimpleHTTPRequestHandler): | |
site = None | |
def do_GET(self): | |
try: | |
response = requests.get(self.site + self.path) | |
except requests.exceptions.Timeout: | |
self.send_error(504) | |
return | |
except requests.exceptions.ConnectionError: | |
self.send_error(502) | |
return | |
if 'html' not in response.headers['Content-Type']: | |
self.copyfile(response.raw, self.wfile) | |
return | |
sys.setdefaultencoding(response.encoding) | |
out = StringIO() | |
mangler = HTMLMangler(out) | |
try: | |
mangler.feed(response.text) | |
mangler.close() | |
except: | |
self.send_error(500) | |
raise | |
self.send_response(200) | |
for k, v in response.headers.items(): | |
if k in ['content-encoding', 'transfer-encoding', 'connection']: | |
continue | |
self.send_header(k, v) | |
out.seek(0) | |
self.copyfile(out, self.wfile) | |
self.end_headers() | |
class HTMLMangler(HTMLParser): | |
skip_tags = 'script style'.split() | |
pattern = re.compile(r'\b(\w{6})\b', re.UNICODE | re.IGNORECASE) | |
def __init__(self, out): | |
HTMLParser.__init__(self) | |
self.out = out | |
self.in_tag = None | |
def handle_startendtag(self, tag, attrs): | |
self.out.write('<%s' % tag) | |
self.handle_attrs(attrs) | |
self.out.write('/>') | |
def handle_starttag(self, tag, attrs): | |
self.out.write('<%s' % tag) | |
self.handle_attrs(attrs) | |
self.out.write('>') | |
self.in_tag = tag | |
def handle_attrs(self, attrs): | |
for attr, value in attrs: | |
if value is not None: | |
self.out.write(' %s=%s' % (attr, quoteattr(value))) | |
else: | |
self.out.write(' %s' % attr) | |
def handle_data(self, data): | |
if self.in_tag not in self.skip_tags: | |
data = self.pattern.sub(unicode(r'\1™'), data) | |
self.out.write(escape(data)) | |
def handle_endtag(self, tag): | |
self.out.write('</%s>' % tag) | |
self.in_tag = None | |
def handle_charref(self, name): | |
self.out.write('&#%s;' % name) | |
def handle_entityref(self, name): | |
self.out.write('&%s;' % name) | |
def handle_comment(self, data): | |
self.out.write('<!-- %s -->' % data) | |
def handle_decl(self, decl): | |
self.out.write('<!%s>' % decl) | |
unknown_decl = handle_decl | |
def handle_pi(self, data): | |
self.out.write('<?%s>' % data) | |
if __name__ == '__main__': | |
parser = argparse.ArgumentParser() | |
parser.add_argument("--host", help="bind host", default='localhost') | |
parser.add_argument("--port", help="bind port", type=int, default=1234) | |
parser.add_argument("--site", help="site url", default='http://habrahabr.ru') | |
args = parser.parse_args() | |
Proxy.site = args.site | |
httpd = SocketServer.ForkingTCPServer((args.host, args.port), Proxy) | |
p = multiprocessing.Process(target=httpd.serve_forever, args=()) | |
p.start() | |
host = 'localhost' if args.host == '0.0.0.0' else args.host | |
webbrowser.open_new_tab('http://' + host + ':' + str(args.port)) | |
raw_input() |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment