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

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