Пример написания парсера на Python: парсинг ссылок на профили пользователей MyShows.ru

Для разбора DOM был использован модуль PyQuery.

https://gist.github.com/3038443

	# -*- coding: utf-8 -*-
	from tornado import ioloop
	from tornado import web
	from tornado import httpclient
	from tornado.escape import (json, json_decode, json_encode)
	from pyquery import PyQuery
	class ParseUsersHandler(web.RequestHandler):
	def get(self):
	client = httpclient.HTTPClient()
	try:
	response = client.fetch("http://myshows.ru/search/users/")
	jQuery = PyQuery(response.body)
	links = []
	jQuery("ul.users.users-quad > li > a").each(lambda i, el: links.append(PyQuery(el).attr("href")))
	self.write(json_encode(links))
	except httpclient.HTTPError, e:
	print "Error:", e
	application = web.Application(
	[
	(r"/rest/parser/users", ParseUsersHandler)
	],
	cookie_secret="QQ#j(leMvennf^&$(J000KIKh23373b#FFF$e",
	debug=True
	)
	if __name__ == '__main__':
	import logging
	logging.getLogger().setLevel(logging.DEBUG)
	application.listen(6848)
	ioloop.IOLoop.instance().start()

view raw myshows_parser.py hosted with ❤ by GitHub