yt_dlp/extractor/vice.py

   1 import functools
   2 import hashlib
   3 import json
   4 import random
   5 import time
   6
   7 from .adobepass import AdobePassIE
   8 from .common import InfoExtractor
   9 from .youtube import YoutubeIE
  10 from ..compat import (
  11     compat_HTTPError,
  12     compat_str,
  13 )
  14 from ..utils import (
  15     clean_html,
  16     ExtractorError,
  17     int_or_none,
  18     OnDemandPagedList,
  19     parse_age_limit,
  20     str_or_none,
  21     try_get,
  22 )
  23
  24
  25 class ViceBaseIE(InfoExtractor):
  26     def _call_api(self, resource, resource_key, resource_id, locale, fields, args=''):
  27         return self._download_json(
  28             'https://video.vice.com/api/v1/graphql', resource_id, query={
  29                 'query': '''{
  30   %s(locale: "%s", %s: "%s"%s) {
  31     %s
  32   }
  33 }''' % (resource, locale, resource_key, resource_id, args, fields),
  34             })['data'][resource]
  35
  36
  37 class ViceIE(ViceBaseIE, AdobePassIE):
  38     IE_NAME = 'vice'
  39     _VALID_URL = r'https?://(?:(?:video|vms)\.vice|(?:www\.)?vice(?:land|tv))\.com/(?P<locale>[^/]+)/(?:video/[^/]+|embed)/(?P<id>[\da-f]{24})'
  40     _EMBED_REGEX = [r'<iframe\b[^>]+\bsrc=["\'](?P<url>(?:https?:)?//video\.vice\.com/[^/]+/embed/[\da-f]{24})']
  41     _TESTS = [{
  42         'url': 'https://video.vice.com/en_us/video/pet-cremator/58c69e38a55424f1227dc3f7',
  43         'info_dict': {
  44             'id': '58c69e38a55424f1227dc3f7',
  45             'ext': 'mp4',
  46             'title': '10 Questions You Always Wanted To Ask: Pet Cremator',
  47             'description': 'md5:fe856caacf61fe0e74fab15ce2b07ca5',
  48             'uploader': 'vice',
  49             'uploader_id': '57a204088cb727dec794c67b',
  50             'timestamp': 1489664942,
  51             'upload_date': '20170316',
  52             'age_limit': 14,
  53         },
  54         'params': {
  55             # m3u8 download
  56             'skip_download': True,
  57         },
  58     }, {
  59         # geo restricted to US
  60         'url': 'https://video.vice.com/en_us/video/the-signal-from-tolva/5816510690b70e6c5fd39a56',
  61         'info_dict': {
  62             'id': '5816510690b70e6c5fd39a56',
  63             'ext': 'mp4',
  64             'uploader': 'vice',
  65             'title': 'The Signal From Tölva',
  66             'description': 'md5:3927e3c79f9e8094606a2b3c5b5e55d5',
  67             'uploader_id': '57a204088cb727dec794c67b',
  68             'timestamp': 1477941983,
  69             'upload_date': '20161031',
  70         },
  71         'params': {
  72             # m3u8 download
  73             'skip_download': True,
  74         },
  75     }, {
  76         'url': 'https://video.vice.com/alps/video/ulfs-wien-beruchtigste-grafitti-crew-part-1/581b12b60a0e1f4c0fb6ea2f',
  77         'info_dict': {
  78             'id': '581b12b60a0e1f4c0fb6ea2f',
  79             'ext': 'mp4',
  80             'title': 'ULFs - Wien berüchtigste Grafitti Crew - Part 1',
  81             'description': 'Zwischen Hinterzimmer-Tattoos und U-Bahnschächten erzählen uns die Ulfs, wie es ist, "süchtig nach Sachbeschädigung" zu sein.',
  82             'uploader': 'vice',
  83             'uploader_id': '57a204088cb727dec794c67b',
  84             'timestamp': 1485368119,
  85             'upload_date': '20170125',
  86             'age_limit': 14,
  87         },
  88         'params': {
  89             # AES-encrypted m3u8
  90             'skip_download': True,
  91         },
  92     }, {
  93         'url': 'https://video.vice.com/en_us/video/pizza-show-trailer/56d8c9a54d286ed92f7f30e4',
  94         'only_matching': True,
  95     }, {
  96         'url': 'https://video.vice.com/en_us/embed/57f41d3556a0a80f54726060',
  97         'only_matching': True,
  98     }, {
  99         'url': 'https://vms.vice.com/en_us/video/preplay/58c69e38a55424f1227dc3f7',
 100         'only_matching': True,
 101     }, {
 102         'url': 'https://www.viceland.com/en_us/video/thursday-march-1-2018/5a8f2d7ff1cdb332dd446ec1',
 103         'only_matching': True,
 104     }]
 105
 106     def _real_extract(self, url):
 107         locale, video_id = self._match_valid_url(url).groups()
 108
 109         video = self._call_api('videos', 'id', video_id, locale, '''body
 110     locked
 111     rating
 112     thumbnail_url
 113     title''')[0]
 114         title = video['title'].strip()
 115         rating = video.get('rating')
 116
 117         query = {}
 118         if video.get('locked'):
 119             resource = self._get_mvpd_resource(
 120                 'VICELAND', title, video_id, rating)
 121             query['tvetoken'] = self._extract_mvpd_auth(
 122                 url, video_id, 'VICELAND', resource)
 123
 124         # signature generation algorithm is reverse engineered from signatureGenerator in
 125         # webpack:///../shared/~/vice-player/dist/js/vice-player.js in
 126         # https://www.viceland.com/assets/common/js/web.vendor.bundle.js
 127         # new JS is located here https://vice-web-statics-cdn.vice.com/vice-player/player-embed.js
 128         exp = int(time.time()) + 1440
 129
 130         query.update({
 131             'exp': exp,
 132             'sign': hashlib.sha512(('%s:GET:%d' % (video_id, exp)).encode()).hexdigest(),
 133             'skipadstitching': 1,
 134             'platform': 'desktop',
 135             'rn': random.randint(10000, 100000),
 136         })
 137
 138         try:
 139             preplay = self._download_json(
 140                 'https://vms.vice.com/%s/video/preplay/%s' % (locale, video_id),
 141                 video_id, query=query)
 142         except ExtractorError as e:
 143             if isinstance(e.cause, compat_HTTPError) and e.cause.code in (400, 401):
 144                 error = json.loads(e.cause.read().decode())
 145                 error_message = error.get('error_description') or error['details']
 146                 raise ExtractorError('%s said: %s' % (
 147                     self.IE_NAME, error_message), expected=True)
 148             raise
 149
 150         video_data = preplay['video']
 151         formats = self._extract_m3u8_formats(
 152             preplay['playURL'], video_id, 'mp4', 'm3u8_native')
 153         episode = video_data.get('episode') or {}
 154         channel = video_data.get('channel') or {}
 155         season = video_data.get('season') or {}
 156
 157         subtitles = {}
 158         for subtitle in preplay.get('subtitleURLs', []):
 159             cc_url = subtitle.get('url')
 160             if not cc_url:
 161                 continue
 162             language_code = try_get(subtitle, lambda x: x['languages'][0]['language_code'], compat_str) or 'en'
 163             subtitles.setdefault(language_code, []).append({
 164                 'url': cc_url,
 165             })
 166
 167         return {
 168             'formats': formats,
 169             'id': video_id,
 170             'title': title,
 171             'description': clean_html(video.get('body')),
 172             'thumbnail': video.get('thumbnail_url'),
 173             'duration': int_or_none(video_data.get('video_duration')),
 174             'timestamp': int_or_none(video_data.get('created_at'), 1000),
 175             'age_limit': parse_age_limit(video_data.get('video_rating') or rating),
 176             'series': try_get(video_data, lambda x: x['show']['base']['display_title'], compat_str),
 177             'episode_number': int_or_none(episode.get('episode_number')),
 178             'episode_id': str_or_none(episode.get('id') or video_data.get('episode_id')),
 179             'season_number': int_or_none(season.get('season_number')),
 180             'season_id': str_or_none(season.get('id') or video_data.get('season_id')),
 181             'uploader': channel.get('name'),
 182             'uploader_id': str_or_none(channel.get('id')),
 183             'subtitles': subtitles,
 184         }
 185
 186
 187 class ViceShowIE(ViceBaseIE):
 188     IE_NAME = 'vice:show'
 189     _VALID_URL = r'https?://(?:video\.vice|(?:www\.)?vice(?:land|tv))\.com/(?P<locale>[^/]+)/show/(?P<id>[^/?#&]+)'
 190     _PAGE_SIZE = 25
 191     _TESTS = [{
 192         'url': 'https://video.vice.com/en_us/show/fck-thats-delicious',
 193         'info_dict': {
 194             'id': '57a2040c8cb727dec794c901',
 195             'title': 'F*ck, That’s Delicious',
 196             'description': 'The life and eating habits of rap’s greatest bon vivant, Action Bronson.',
 197         },
 198         'playlist_mincount': 64,
 199     }, {
 200         'url': 'https://www.vicetv.com/en_us/show/fck-thats-delicious',
 201         'only_matching': True,
 202     }]
 203
 204     def _fetch_page(self, locale, show_id, page):
 205         videos = self._call_api('videos', 'show_id', show_id, locale, '''body
 206     id
 207     url''', ', page: %d, per_page: %d' % (page + 1, self._PAGE_SIZE))
 208         for video in videos:
 209             yield self.url_result(
 210                 video['url'], ViceIE.ie_key(), video.get('id'))
 211
 212     def _real_extract(self, url):
 213         locale, display_id = self._match_valid_url(url).groups()
 214         show = self._call_api('shows', 'slug', display_id, locale, '''dek
 215     id
 216     title''')[0]
 217         show_id = show['id']
 218
 219         entries = OnDemandPagedList(
 220             functools.partial(self._fetch_page, locale, show_id),
 221             self._PAGE_SIZE)
 222
 223         return self.playlist_result(
 224             entries, show_id, show.get('title'), show.get('dek'))
 225
 226
 227 class ViceArticleIE(ViceBaseIE):
 228     IE_NAME = 'vice:article'
 229     _VALID_URL = r'https://(?:www\.)?vice\.com/(?P<locale>[^/]+)/article/(?:[0-9a-z]{6}/)?(?P<id>[^?#]+)'
 230
 231     _TESTS = [{
 232         'url': 'https://www.vice.com/en_us/article/on-set-with-the-woman-making-mormon-porn-in-utah',
 233         'info_dict': {
 234             'id': '58dc0a3dee202d2a0ccfcbd8',
 235             'ext': 'mp4',
 236             'title': 'Mormon War on Porn',
 237             'description': 'md5:1c5d91fe25fa8aa304f9def118b92dbf',
 238             'uploader': 'vice',
 239             'uploader_id': '57a204088cb727dec794c67b',
 240             'timestamp': 1491883129,
 241             'upload_date': '20170411',
 242             'age_limit': 17,
 243         },
 244         'params': {
 245             # AES-encrypted m3u8
 246             'skip_download': True,
 247         },
 248         'add_ie': [ViceIE.ie_key()],
 249     }, {
 250         'url': 'https://www.vice.com/en_us/article/how-to-hack-a-car',
 251         'md5': '13010ee0bc694ea87ec40724397c2349',
 252         'info_dict': {
 253             'id': '3jstaBeXgAs',
 254             'ext': 'mp4',
 255             'title': 'How to Hack a Car: Phreaked Out (Episode 2)',
 256             'description': 'md5:ee95453f7ff495db8efe14ae8bf56f30',
 257             'uploader': 'Motherboard',
 258             'uploader_id': 'MotherboardTV',
 259             'upload_date': '20140529',
 260         },
 261         'add_ie': [YoutubeIE.ie_key()],
 262     }, {
 263         'url': 'https://www.vice.com/en_us/article/znm9dx/karley-sciortino-slutever-reloaded',
 264         'md5': 'a7ecf64ee4fa19b916c16f4b56184ae2',
 265         'info_dict': {
 266             'id': '57f41d3556a0a80f54726060',
 267             'ext': 'mp4',
 268             'title': "Making The World's First Male Sex Doll",
 269             'description': 'md5:19b00b215b99961cf869c40fbe9df755',
 270             'uploader': 'vice',
 271             'uploader_id': '57a204088cb727dec794c67b',
 272             'timestamp': 1476919911,
 273             'upload_date': '20161019',
 274             'age_limit': 17,
 275         },
 276         'params': {
 277             'skip_download': True,
 278         },
 279         'add_ie': [ViceIE.ie_key()],
 280     }, {
 281         'url': 'https://www.vice.com/en_us/article/cowboy-capitalists-part-1',
 282         'only_matching': True,
 283     }, {
 284         'url': 'https://www.vice.com/ru/article/big-night-out-ibiza-clive-martin-229',
 285         'only_matching': True,
 286     }]
 287
 288     def _real_extract(self, url):
 289         locale, display_id = self._match_valid_url(url).groups()
 290
 291         article = self._call_api('articles', 'slug', display_id, locale, '''body
 292     embed_code''')[0]
 293         body = article['body']
 294
 295         def _url_res(video_url, ie_key):
 296             return {
 297                 '_type': 'url_transparent',
 298                 'url': video_url,
 299                 'display_id': display_id,
 300                 'ie_key': ie_key,
 301             }
 302
 303         vice_url = ViceIE._extract_url(body)
 304         if vice_url:
 305             return _url_res(vice_url, ViceIE.ie_key())
 306
 307         embed_code = self._search_regex(
 308             r'embedCode=([^&\'"]+)', body,
 309             'ooyala embed code', default=None)
 310         if embed_code:
 311             return _url_res('ooyala:%s' % embed_code, 'Ooyala')
 312
 313         youtube_url = YoutubeIE._extract_url(body)
 314         if youtube_url:
 315             return _url_res(youtube_url, YoutubeIE.ie_key())
 316
 317         video_url = self._html_search_regex(
 318             r'data-video-url="([^"]+)"',
 319             article['embed_code'], 'video URL')
 320
 321         return _url_res(video_url, ViceIE.ie_key())