[mitele] Fix extraction (Closes #6414)

This commit is contained in:
Sergey M․ 2015-09-16 22:27:03 +06:00
parent 12bc242944
commit f84ce1ebaf
1 changed files with 58 additions and 47 deletions

View File

@ -1,17 +1,11 @@
from __future__ import unicode_literals
import json
from .common import InfoExtractor
from ..compat import (
compat_urllib_parse,
compat_urllib_parse_unquote,
compat_urlparse,
)
from ..compat import compat_urllib_parse
from ..utils import (
encode_dict,
get_element_by_attribute,
parse_duration,
strip_jsonp,
int_or_none,
)
@ -21,54 +15,71 @@ class MiTeleIE(InfoExtractor):
_TESTS = [{
'url': 'http://www.mitele.es/programas-tv/diario-de/la-redaccion/programa-144/',
'md5': 'ace7635b2a0b286aaa37d3ff192d2a8a',
'info_dict': {
'id': '0fce117d',
'ext': 'mp4',
'title': 'Programa 144 - Tor, la web invisible',
'description': 'md5:3b6fce7eaa41b2d97358726378d9369f',
'id': '0NF1jJnxS1Wu3pHrmvFyw2',
'display_id': 'programa-144',
'ext': 'flv',
'title': 'Tor, la web invisible',
'description': 'md5:3b6fce7eaa41b2d97358726378d9369f',
'thumbnail': 're:(?i)^https?://.*\.jpg$',
'duration': 2913,
},
'params': {
# m3u8 download
'skip_download': True,
},
}]
def _real_extract(self, url):
episode = self._match_id(url)
webpage = self._download_webpage(url, episode)
embed_data_json = self._search_regex(
r'(?s)MSV\.embedData\[.*?\]\s*=\s*({.*?});', webpage, 'embed data',
).replace('\'', '"')
embed_data = json.loads(embed_data_json)
display_id = self._match_id(url)
domain = embed_data['mediaUrl']
if not domain.startswith('http'):
# only happens in telecinco.es videos
domain = 'http://' + domain
info_url = compat_urlparse.urljoin(
domain,
compat_urllib_parse_unquote(embed_data['flashvars']['host'])
)
info_el = self._download_xml(info_url, episode).find('./video/info')
webpage = self._download_webpage(url, display_id)
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,
transform_source=strip_jsonp
)
formats = self._extract_m3u8_formats(
token_info['tokenizedUrl'], episode, ext='mp4')
config_url = self._search_regex(
r'data-config\s*=\s*"([^"]+)"', webpage, 'data config url')
config = self._download_json(
config_url, display_id, 'Downloading config JSON')
mmc = self._download_json(
config['services']['mmc'], display_id, 'Downloading mmc JSON')
formats = []
for location in mmc['locations']:
gat = self._proto_relative_url(location.get('gat'), 'http:')
bas = location.get('bas')
loc = location.get('loc')
ogn = location.get('ogn')
if None in (gat, bas, loc, ogn):
continue
token_data = {
'bas': bas,
'icd': loc,
'ogn': ogn,
'sta': '0',
}
media = self._download_json(
'%s/?%s' % (gat, compat_urllib_parse.urlencode(encode_dict(token_data)).encode('utf-8')),
display_id, 'Downloading %s JSON' % location['loc'])
file_ = media.get('file')
if not file_:
continue
formats.extend(self._extract_f4m_formats(
file_ + '&hdcore=3.2.0&plugin=aasp-3.2.0.77.18',
display_id, f4m_id=loc))
title = self._search_regex(
r'class="Destacado-text"[^>]*>\s*<strong>([^<]+)</strong>', webpage, 'title')
video_id = self._search_regex(
r'data-media-id\s*=\s*"([^"]+)"', webpage,
'data media id', default=None) or display_id
thumbnail = config.get('poster', {}).get('imageUrl')
duration = int_or_none(mmc.get('duration'))
return {
'id': embed_data['videoId'],
'display_id': episode,
'title': info_el.find('title').text,
'formats': formats,
'id': video_id,
'display_id': display_id,
'title': title,
'description': get_element_by_attribute('class', 'text', webpage),
'thumbnail': info_el.find('thumb').text,
'duration': parse_duration(info_el.find('duration').text),
'thumbnail': thumbnail,
'duration': duration,
'formats': formats,
}