Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/mitele.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4 import json
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     compat_urllib_parse,
   9     compat_urlparse,
  10     get_element_by_attribute,
  11     parse_duration,
  12     strip_jsonp,
  13 )
  14
  15
  16 class MiTeleIE(InfoExtractor):
  17     IE_NAME = 'mitele.es'
  18     _VALID_URL = r'http://www\.mitele\.es/[^/]+/[^/]+/[^/]+/(?P<episode>[^/]+)/'
  19
  20     _TEST = {
  21         'url': 'http://www.mitele.es/programas-tv/diario-de/la-redaccion/programa-144/',
  22         'md5': '6a75fe9d0d3275bead0cb683c616fddb',
  23         'info_dict': {
  24             'id': '0fce117d',
  25             'ext': 'mp4',
  26             'title': 'Programa 144 - Tor, la web invisible',
  27             'description': 'md5:3b6fce7eaa41b2d97358726378d9369f',
  28             'display_id': 'programa-144',
  29             'duration': 2913,
  30         },
  31     }
  32
  33     def _real_extract(self, url):
  34         mobj = re.match(self._VALID_URL, url)
  35         episode = mobj.group('episode')
  36         webpage = self._download_webpage(url, episode)
  37         embed_data_json = self._search_regex(
  38             r'MSV\.embedData\[.*?\]\s*=\s*({.*?});', webpage, 'embed data',
  39             flags=re.DOTALL
  40         ).replace('\'', '"')
  41         embed_data = json.loads(embed_data_json)
  42
  43         domain = embed_data['mediaUrl']
  44         if not domain.startswith('http'):
  45             # only happens in telecinco.es videos
  46             domain = 'http://' + domain
  47         info_url = compat_urlparse.urljoin(
  48             domain,
  49             compat_urllib_parse.unquote(embed_data['flashvars']['host'])
  50         )
  51         info_el = self._download_xml(info_url, episode).find('./video/info')
  52
  53         video_link = info_el.find('videoUrl/link').text
  54         token_query = compat_urllib_parse.urlencode({'id': video_link})
  55         token_info = self._download_json(
  56             embed_data['flashvars']['ov_tk'] + '?' + token_query,
  57             episode,
  58             transform_source=strip_jsonp
  59         )
  60
  61         return {
  62             'id': embed_data['videoId'],
  63             'display_id': episode,
  64             'title': info_el.find('title').text,
  65             'url': token_info['tokenizedUrl'],
  66             'description': get_element_by_attribute('class', 'text', webpage),
  67             'thumbnail': info_el.find('thumb').text,
  68             'duration': parse_duration(info_el.find('duration').text),
  69         }