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

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import json
   5
   6 from .turner import TurnerBaseIE
   7 from ..utils import (
   8     determine_ext,
   9     ExtractorError,
  10     int_or_none,
  11     mimetype2ext,
  12     parse_duration,
  13     parse_iso8601,
  14     qualities,
  15 )
  16
  17
  18 class TeamcocoIE(TurnerBaseIE):
  19     _VALID_URL = r'https?://(?:\w+\.)?teamcoco\.com/(?P<id>([^/]+/)*[^/?#]+)'
  20     _TESTS = [
  21         {
  22             'url': 'http://teamcoco.com/video/mary-kay-remote',
  23             'md5': '55d532f81992f5c92046ad02fec34d7d',
  24             'info_dict': {
  25                 'id': '80187',
  26                 'ext': 'mp4',
  27                 'title': 'Conan Becomes A Mary Kay Beauty Consultant',
  28                 'description': 'Mary Kay is perhaps the most trusted name in female beauty, so of course Conan is a natural choice to sell their products.',
  29                 'duration': 495.0,
  30                 'upload_date': '20140402',
  31                 'timestamp': 1396407600,
  32             }
  33         }, {
  34             'url': 'http://teamcoco.com/video/louis-ck-interview-george-w-bush',
  35             'md5': 'cde9ba0fa3506f5f017ce11ead928f9a',
  36             'info_dict': {
  37                 'id': '19705',
  38                 'ext': 'mp4',
  39                 'description': 'Louis C.K. got starstruck by George W. Bush, so what? Part one.',
  40                 'title': 'Louis C.K. Interview Pt. 1 11/3/11',
  41                 'duration': 288,
  42                 'upload_date': '20111104',
  43                 'timestamp': 1320405840,
  44             }
  45         }, {
  46             'url': 'http://teamcoco.com/video/timothy-olyphant-drinking-whiskey',
  47             'info_dict': {
  48                 'id': '88748',
  49                 'ext': 'mp4',
  50                 'title': 'Timothy Olyphant Raises A Toast To “Justified”',
  51                 'description': 'md5:15501f23f020e793aeca761205e42c24',
  52                 'upload_date': '20150415',
  53                 'timestamp': 1429088400,
  54             },
  55             'params': {
  56                 'skip_download': True,  # m3u8 downloads
  57             }
  58         }, {
  59             'url': 'http://teamcoco.com/video/full-episode-mon-6-1-joel-mchale-jake-tapper-and-musical-guest-courtney-barnett?playlist=x;eyJ0eXBlIjoidGFnIiwiaWQiOjl9',
  60             'info_dict': {
  61                 'id': '89341',
  62                 'ext': 'mp4',
  63                 'title': 'Full Episode - Mon. 6/1 - Joel McHale, Jake Tapper, And Musical Guest Courtney Barnett',
  64                 'description': 'Guests: Joel McHale, Jake Tapper, And Musical Guest Courtney Barnett',
  65             },
  66             'params': {
  67                 'skip_download': True,  # m3u8 downloads
  68             },
  69             'skip': 'This video is no longer available.',
  70         }, {
  71             'url': 'http://teamcoco.com/video/the-conan-audiencey-awards-for-04/25/18',
  72             'only_matching': True,
  73         }, {
  74             'url': 'http://teamcoco.com/italy/conan-jordan-schlansky-hit-the-streets-of-florence',
  75             'only_matching': True,
  76         }, {
  77             'url': 'http://teamcoco.com/haiti/conan-s-haitian-history-lesson',
  78             'only_matching': True,
  79         }, {
  80             'url': 'http://teamcoco.com/israel/conan-hits-the-streets-beaches-of-tel-aviv',
  81             'only_matching': True,
  82         }, {
  83             'url': 'https://conan25.teamcoco.com/video/ice-cube-kevin-hart-conan-share-lyft',
  84             'only_matching': True,
  85         }
  86     ]
  87     _RECORD_TEMPL = '''id
  88         title
  89         teaser
  90         publishOn
  91         thumb {
  92           preview
  93         }
  94         tags {
  95           name
  96         }
  97         duration
  98         turnerMediaId
  99         turnerMediaAuthToken'''
 100
 101     def _graphql_call(self, query_template, object_type, object_id):
 102         find_object = 'find' + object_type
 103         return self._download_json(
 104             'https://teamcoco.com/graphql', object_id, data=json.dumps({
 105                 'query': query_template % (find_object, object_id)
 106             }).encode(), headers={
 107                 'Content-Type': 'application/json',
 108             })['data'][find_object]
 109
 110     def _real_extract(self, url):
 111         display_id = self._match_id(url)
 112
 113         response = self._graphql_call('''{
 114   %%s(slug: "%%s") {
 115     ... on RecordSlug {
 116       record {
 117         %s
 118       }
 119     }
 120     ... on PageSlug {
 121       child {
 122         id
 123       }
 124     }
 125     ... on NotFoundSlug {
 126       status
 127     }
 128   }
 129 }''' % self._RECORD_TEMPL, 'Slug', display_id)
 130         if response.get('status'):
 131             raise ExtractorError('This video is no longer available.', expected=True)
 132
 133         child = response.get('child')
 134         if child:
 135             record = self._graphql_call('''{
 136   %%s(id: "%%s") {
 137     ... on Video {
 138       %s
 139     }
 140   }
 141 }''' % self._RECORD_TEMPL, 'Record', child['id'])
 142         else:
 143             record = response['record']
 144         video_id = record['id']
 145
 146         info = {
 147             'id': video_id,
 148             'display_id': display_id,
 149             'title': record['title'],
 150             'thumbnail': record.get('thumb', {}).get('preview'),
 151             'description': record.get('teaser'),
 152             'duration': parse_duration(record.get('duration')),
 153             'timestamp': parse_iso8601(record.get('publishOn')),
 154         }
 155
 156         media_id = record.get('turnerMediaId')
 157         if media_id:
 158             self._initialize_geo_bypass({
 159                 'countries': ['US'],
 160             })
 161             info.update(self._extract_ngtv_info(media_id, {
 162                 'accessToken': record['turnerMediaAuthToken'],
 163                 'accessTokenType': 'jws',
 164             }))
 165         else:
 166             video_sources = self._download_json(
 167                 'https://teamcoco.com/_truman/d/' + video_id,
 168                 video_id)['meta']['src']
 169             if isinstance(video_sources, dict):
 170                 video_sources = video_sources.values()
 171
 172             formats = []
 173             get_quality = qualities(['low', 'sd', 'hd', 'uhd'])
 174             for src in video_sources:
 175                 if not isinstance(src, dict):
 176                     continue
 177                 src_url = src.get('src')
 178                 if not src_url:
 179                     continue
 180                 format_id = src.get('label')
 181                 ext = determine_ext(src_url, mimetype2ext(src.get('type')))
 182                 if format_id == 'hls' or ext == 'm3u8':
 183                     # compat_urllib_parse.urljoin does not work here
 184                     if src_url.startswith('/'):
 185                         src_url = 'http://ht.cdn.turner.com/tbs/big/teamcoco' + src_url
 186                     formats.extend(self._extract_m3u8_formats(
 187                         src_url, video_id, 'mp4', m3u8_id=format_id, fatal=False))
 188                 else:
 189                     if src_url.startswith('/mp4:protected/'):
 190                         # TODO Correct extraction for these files
 191                         continue
 192                     tbr = int_or_none(self._search_regex(
 193                         r'(\d+)k\.mp4', src_url, 'tbr', default=None))
 194
 195                     formats.append({
 196                         'url': src_url,
 197                         'ext': ext,
 198                         'tbr': tbr,
 199                         'format_id': format_id,
 200                         'quality': get_quality(format_id),
 201                     })
 202             self._sort_formats(formats)
 203             info['formats'] = formats
 204
 205         return info