_ Git - youtube-dl/blob - youtube_dl/extractor/onionstudios.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     compat_str,
   9     int_or_none,
  10     js_to_json,
  11     parse_iso8601,
  12     try_get,
  13 )
  14
  15
  16 class OnionStudiosIE(InfoExtractor):
  17     _VALID_URL = r'https?://(?:www\.)?onionstudios\.com/(?:video(?:s/[^/]+-|/)|embed\?.*\bid=)(?P<id>\d+)(?!-)'
  18
  19     _TESTS = [{
  20         'url': 'http://www.onionstudios.com/videos/hannibal-charges-forward-stops-for-a-cocktail-2937',
  21         'md5': '5a118d466d62b5cd03647cf2c593977f',
  22         'info_dict': {
  23             'id': '2937',
  24             'ext': 'mp4',
  25             'title': 'Hannibal charges forward, stops for a cocktail',
  26             'description': 'md5:545299bda6abf87e5ec666548c6a9448',
  27             'thumbnail': r're:^https?://.*\.jpg$',
  28             'uploader': 'a.v. club',
  29             'upload_date': '20150619',
  30             'timestamp': 1434728546,
  31         },
  32     }, {
  33         'url': 'http://www.onionstudios.com/embed?id=2855&autoplay=true',
  34         'only_matching': True,
  35     }, {
  36         'url': 'http://www.onionstudios.com/video/6139.json',
  37         'only_matching': True,
  38     }]
  39
  40     @staticmethod
  41     def _extract_url(webpage):
  42         mobj = re.search(
  43             r'(?s)<(?:iframe|bulbs-video)[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?onionstudios\.com/(?:embed.+?|video/\d+\.json))\1', webpage)
  44         if mobj:
  45             return mobj.group('url')
  46
  47     def _real_extract(self, url):
  48         video_id = self._match_id(url)
  49
  50         webpage = self._download_webpage(
  51             'http://onionstudios.com/embed/dc94dc2899fe644c0e7241fa04c1b732.js',
  52             video_id)
  53         mcp_id = compat_str(self._parse_json(self._search_regex(
  54             r'window\.mcpMapping\s*=\s*({.+?});', webpage,
  55             'MCP Mapping'), video_id, js_to_json)[video_id]['mcp_id'])
  56         video_data = self._download_json(
  57             'https://api.vmh.univision.com/metadata/v1/content/' + mcp_id,
  58             mcp_id)['videoMetadata']
  59         iptc = video_data['photoVideoMetadataIPTC']
  60         title = iptc['title']['en']
  61         fmg = video_data.get('photoVideoMetadata_fmg') or {}
  62         tvss_domain = fmg.get('tvssDomain') or 'https://auth.univision.com'
  63         data = self._download_json(
  64             tvss_domain + '/api/v3/video-auth/url-signature-tokens',
  65             mcp_id, query={'mcpids': mcp_id})['data'][0]
  66         formats = []
  67
  68         rendition_url = data.get('renditionUrl')
  69         if rendition_url:
  70             formats = self._extract_m3u8_formats(
  71                 rendition_url, mcp_id, 'mp4',
  72                 'm3u8_native', m3u8_id='hls', fatal=False)
  73
  74         fallback_rendition_url = data.get('fallbackRenditionUrl')
  75         if fallback_rendition_url:
  76             formats.append({
  77                 'format_id': 'fallback',
  78                 'tbr': int_or_none(self._search_regex(
  79                     r'_(\d+)\.mp4', fallback_rendition_url,
  80                     'bitrate', default=None)),
  81                 'url': fallback_rendition_url,
  82             })
  83
  84         self._sort_formats(formats)
  85
  86         return {
  87             'id': video_id,
  88             'title': title,
  89             'thumbnail': try_get(iptc, lambda x: x['cloudinaryLink']['link'], compat_str),
  90             'uploader': fmg.get('network'),
  91             'duration': int_or_none(iptc.get('fileDuration')),
  92             'formats': formats,
  93             'description': try_get(iptc, lambda x: x['description']['en'], compat_str),
  94             'timestamp': parse_iso8601(iptc.get('dateReleased')),
  95         }