youtube-dl/youtube_dl/extractor/mitele.py

69 lines
2.2 KiB
Python
Raw Normal View History

2014-08-19 05:43:35 +09:00
from __future__ import unicode_literals
import json
from .common import InfoExtractor
from ..compat import (
2014-08-19 05:43:35 +09:00
compat_urllib_parse,
2014-10-24 04:26:48 +09:00
compat_urlparse,
)
from ..utils import (
2014-08-19 05:43:35 +09:00
get_element_by_attribute,
parse_duration,
strip_jsonp,
)
class MiTeleIE(InfoExtractor):
IE_NAME = 'mitele.es'
_VALID_URL = r'http://www\.mitele\.es/[^/]+/[^/]+/[^/]+/(?P<id>[^/]+)/'
2014-08-19 05:43:35 +09:00
_TESTS = [{
2014-08-19 05:43:35 +09:00
'url': 'http://www.mitele.es/programas-tv/diario-de/la-redaccion/programa-144/',
'md5': '6a75fe9d0d3275bead0cb683c616fddb',
'info_dict': {
'id': '0fce117d',
'ext': 'mp4',
'title': 'Programa 144 - Tor, la web invisible',
'description': 'md5:3b6fce7eaa41b2d97358726378d9369f',
'display_id': 'programa-144',
'duration': 2913,
},
}]
2014-08-19 05:43:35 +09:00
def _real_extract(self, url):
episode = self._match_id(url)
2014-08-19 05:43:35 +09:00
webpage = self._download_webpage(url, episode)
embed_data_json = self._search_regex(
r'(?s)MSV\.embedData\[.*?\]\s*=\s*({.*?});', webpage, 'embed data',
2014-08-19 05:43:35 +09:00
).replace('\'', '"')
embed_data = json.loads(embed_data_json)
domain = embed_data['mediaUrl']
if not domain.startswith('http'):
# only happens in telecinco.es videos
domain = 'http://' + domain
2014-10-24 04:26:48 +09:00
info_url = compat_urlparse.urljoin(
domain,
compat_urllib_parse.unquote(embed_data['flashvars']['host'])
)
2014-08-19 05:43:35 +09:00
info_el = self._download_xml(info_url, episode).find('./video/info')
video_link = info_el.find('videoUrl/link').text
token_query = compat_urllib_parse.urlencode({'id': video_link})
token_info = self._download_json(
embed_data['flashvars']['ov_tk'] + '?' + token_query,
episode,
2014-08-19 05:43:35 +09:00
transform_source=strip_jsonp
)
return {
'id': embed_data['videoId'],
'display_id': episode,
'title': info_el.find('title').text,
'url': token_info['tokenizedUrl'],
'description': get_element_by_attribute('class', 'text', webpage),
'thumbnail': info_el.find('thumb').text,
'duration': parse_duration(info_el.find('duration').text),
}