yt_dlp/extractor/vice.py

   1 import functools
   2 import hashlib
   3 import json
   4 import random
   5 import re
   6 import time
   7
   8 from .adobepass import AdobePassIE
   9 from .common import InfoExtractor
  10 from .youtube import YoutubeIE
  11 from ..compat import (
  12     compat_HTTPError,
  13     compat_str,
  14 )
  15 from ..utils import (
  16     clean_html,
  17     ExtractorError,
  18     int_or_none,
  19     OnDemandPagedList,
  20     parse_age_limit,
  21     str_or_none,
  22     try_get,
  23 )
  24
  25
  26 class ViceBaseIE(InfoExtractor):
  27     def _call_api(self, resource, resource_key, resource_id, locale, fields, args=''):
  28         return self._download_json(
  29             'https://video.vice.com/api/v1/graphql', resource_id, query={
  30                 'query': '''{
  31   %s(locale: "%s", %s: "%s"%s) {
  32     %s
  33   }
  34 }''' % (resource, locale, resource_key, resource_id, args, fields),
  35             })['data'][resource]
  36
  37
  38 class ViceIE(ViceBaseIE, AdobePassIE):
  39     IE_NAME = 'vice'
  40     _VALID_URL = r'https?://(?:(?:video|vms)\.vice|(?:www\.)?vice(?:land|tv))\.com/(?P<locale>[^/]+)/(?:video/[^/]+|embed)/(?P<id>[\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     @staticmethod
 107     def _extract_urls(webpage):
 108         return re.findall(
 109             r'<iframe\b[^>]+\bsrc=["\']((?:https?:)?//video\.vice\.com/[^/]+/embed/[\da-f]{24})',
 110             webpage)
 111
 112     @staticmethod
 113     def _extract_url(webpage):
 114         urls = ViceIE._extract_urls(webpage)
 115         return urls[0] if urls else None
 116
 117     def _real_extract(self, url):
 118         locale, video_id = self._match_valid_url(url).groups()
 119
 120         video = self._call_api('videos', 'id', video_id, locale, '''body
 121     locked
 122     rating
 123     thumbnail_url
 124     title''')[0]
 125         title = video['title'].strip()
 126         rating = video.get('rating')
 127
 128         query = {}
 129         if video.get('locked'):
 130             resource = self._get_mvpd_resource(
 131                 'VICELAND', title, video_id, rating)
 132             query['tvetoken'] = self._extract_mvpd_auth(
 133                 url, video_id, 'VICELAND', resource)
 134
 135         # signature generation algorithm is reverse engineered from signatureGenerator in
 136         # webpack:///../shared/~/vice-player/dist/js/vice-player.js in
 137         # https://www.viceland.com/assets/common/js/web.vendor.bundle.js
 138         # new JS is located here https://vice-web-statics-cdn.vice.com/vice-player/player-embed.js
 139         exp = int(time.time()) + 1440
 140
 141         query.update({
 142             'exp': exp,
 143             'sign': hashlib.sha512(('%s:GET:%d' % (video_id, exp)).encode()).hexdigest(),
 144             'skipadstitching': 1,
 145             'platform': 'desktop',
 146             'rn': random.randint(10000, 100000),
 147         })
 148
 149         try:
 150             preplay = self._download_json(
 151                 'https://vms.vice.com/%s/video/preplay/%s' % (locale, video_id),
 152                 video_id, query=query)
 153         except ExtractorError as e:
 154             if isinstance(e.cause, compat_HTTPError) and e.cause.code in (400, 401):
 155                 error = json.loads(e.cause.read().decode())
 156                 error_message = error.get('error_description') or error['details']
 157                 raise ExtractorError('%s said: %s' % (
 158                     self.IE_NAME, error_message), expected=True)
 159             raise
 160
 161         video_data = preplay['video']
 162         formats = self._extract_m3u8_formats(
 163             preplay['playURL'], video_id, 'mp4', 'm3u8_native')
 164         self._sort_formats(formats)
 165         episode = video_data.get('episode') or {}
 166         channel = video_data.get('channel') or {}
 167         season = video_data.get('season') or {}
 168
 169         subtitles = {}
 170         for subtitle in preplay.get('subtitleURLs', []):
 171             cc_url = subtitle.get('url')
 172             if not cc_url:
 173                 continue
 174             language_code = try_get(subtitle, lambda x: x['languages'][0]['language_code'], compat_str) or 'en'
 175             subtitles.setdefault(language_code, []).append({
 176                 'url': cc_url,
 177             })
 178
 179         return {
 180             'formats': formats,
 181             'id': video_id,
 182             'title': title,
 183             'description': clean_html(video.get('body')),
 184             'thumbnail': video.get('thumbnail_url'),
 185             'duration': int_or_none(video_data.get('video_duration')),
 186             'timestamp': int_or_none(video_data.get('created_at'), 1000),
 187             'age_limit': parse_age_limit(video_data.get('video_rating') or rating),
 188             'series': try_get(video_data, lambda x: x['show']['base']['display_title'], compat_str),
 189             'episode_number': int_or_none(episode.get('episode_number')),
 190             'episode_id': str_or_none(episode.get('id') or video_data.get('episode_id')),
 191             'season_number': int_or_none(season.get('season_number')),
 192             'season_id': str_or_none(season.get('id') or video_data.get('season_id')),
 193             'uploader': channel.get('name'),
 194             'uploader_id': str_or_none(channel.get('id')),
 195             'subtitles': subtitles,
 196         }
 197
 198
 199 class ViceShowIE(ViceBaseIE):
 200     IE_NAME = 'vice:show'
 201     _VALID_URL = r'https?://(?:video\.vice|(?:www\.)?vice(?:land|tv))\.com/(?P<locale>[^/]+)/show/(?P<id>[^/?#&]+)'
 202     _PAGE_SIZE = 25
 203     _TESTS = [{
 204         'url': 'https://video.vice.com/en_us/show/fck-thats-delicious',
 205         'info_dict': {
 206             'id': '57a2040c8cb727dec794c901',
 207             'title': 'F*ck, That’s Delicious',
 208             'description': 'The life and eating habits of rap’s greatest bon vivant, Action Bronson.',
 209         },
 210         'playlist_mincount': 64,
 211     }, {
 212         'url': 'https://www.vicetv.com/en_us/show/fck-thats-delicious',
 213         'only_matching': True,
 214     }]
 215
 216     def _fetch_page(self, locale, show_id, page):
 217         videos = self._call_api('videos', 'show_id', show_id, locale, '''body
 218     id
 219     url''', ', page: %d, per_page: %d' % (page + 1, self._PAGE_SIZE))
 220         for video in videos:
 221             yield self.url_result(
 222                 video['url'], ViceIE.ie_key(), video.get('id'))
 223
 224     def _real_extract(self, url):
 225         locale, display_id = self._match_valid_url(url).groups()
 226         show = self._call_api('shows', 'slug', display_id, locale, '''dek
 227     id
 228     title''')[0]
 229         show_id = show['id']
 230
 231         entries = OnDemandPagedList(
 232             functools.partial(self._fetch_page, locale, show_id),
 233             self._PAGE_SIZE)
 234
 235         return self.playlist_result(
 236             entries, show_id, show.get('title'), show.get('dek'))
 237
 238
 239 class ViceArticleIE(ViceBaseIE):
 240     IE_NAME = 'vice:article'
 241     _VALID_URL = r'https://(?:www\.)?vice\.com/(?P<locale>[^/]+)/article/(?:[0-9a-z]{6}/)?(?P<id>[^?#]+)'
 242
 243     _TESTS = [{
 244         'url': 'https://www.vice.com/en_us/article/on-set-with-the-woman-making-mormon-porn-in-utah',
 245         'info_dict': {
 246             'id': '58dc0a3dee202d2a0ccfcbd8',
 247             'ext': 'mp4',
 248             'title': 'Mormon War on Porn',
 249             'description': 'md5:1c5d91fe25fa8aa304f9def118b92dbf',
 250             'uploader': 'vice',
 251             'uploader_id': '57a204088cb727dec794c67b',
 252             'timestamp': 1491883129,
 253             'upload_date': '20170411',
 254             'age_limit': 17,
 255         },
 256         'params': {
 257             # AES-encrypted m3u8
 258             'skip_download': True,
 259         },
 260         'add_ie': [ViceIE.ie_key()],
 261     }, {
 262         'url': 'https://www.vice.com/en_us/article/how-to-hack-a-car',
 263         'md5': '13010ee0bc694ea87ec40724397c2349',
 264         'info_dict': {
 265             'id': '3jstaBeXgAs',
 266             'ext': 'mp4',
 267             'title': 'How to Hack a Car: Phreaked Out (Episode 2)',
 268             'description': 'md5:ee95453f7ff495db8efe14ae8bf56f30',
 269             'uploader': 'Motherboard',
 270             'uploader_id': 'MotherboardTV',
 271             'upload_date': '20140529',
 272         },
 273         'add_ie': [YoutubeIE.ie_key()],
 274     }, {
 275         'url': 'https://www.vice.com/en_us/article/znm9dx/karley-sciortino-slutever-reloaded',
 276         'md5': 'a7ecf64ee4fa19b916c16f4b56184ae2',
 277         'info_dict': {
 278             'id': '57f41d3556a0a80f54726060',
 279             'ext': 'mp4',
 280             'title': "Making The World's First Male Sex Doll",
 281             'description': 'md5:19b00b215b99961cf869c40fbe9df755',
 282             'uploader': 'vice',
 283             'uploader_id': '57a204088cb727dec794c67b',
 284             'timestamp': 1476919911,
 285             'upload_date': '20161019',
 286             'age_limit': 17,
 287         },
 288         'params': {
 289             'skip_download': True,
 290         },
 291         'add_ie': [ViceIE.ie_key()],
 292     }, {
 293         'url': 'https://www.vice.com/en_us/article/cowboy-capitalists-part-1',
 294         'only_matching': True,
 295     }, {
 296         'url': 'https://www.vice.com/ru/article/big-night-out-ibiza-clive-martin-229',
 297         'only_matching': True,
 298     }]
 299
 300     def _real_extract(self, url):
 301         locale, display_id = self._match_valid_url(url).groups()
 302
 303         article = self._call_api('articles', 'slug', display_id, locale, '''body
 304     embed_code''')[0]
 305         body = article['body']
 306
 307         def _url_res(video_url, ie_key):
 308             return {
 309                 '_type': 'url_transparent',
 310                 'url': video_url,
 311                 'display_id': display_id,
 312                 'ie_key': ie_key,
 313             }
 314
 315         vice_url = ViceIE._extract_url(body)
 316         if vice_url:
 317             return _url_res(vice_url, ViceIE.ie_key())
 318
 319         embed_code = self._search_regex(
 320             r'embedCode=([^&\'"]+)', body,
 321             'ooyala embed code', default=None)
 322         if embed_code:
 323             return _url_res('ooyala:%s' % embed_code, 'Ooyala')
 324
 325         youtube_url = YoutubeIE._extract_url(body)
 326         if youtube_url:
 327             return _url_res(youtube_url, YoutubeIE.ie_key())
 328
 329         video_url = self._html_search_regex(
 330             r'data-video-url="([^"]+)"',
 331             article['embed_code'], 'video URL')
 332
 333         return _url_res(video_url, ViceIE.ie_key())