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 ..networking.exceptions import HTTPError
  11 from ..utils import (
  12     ExtractorError,
  13     OnDemandPagedList,
  14     clean_html,
  15     int_or_none,
  16     parse_age_limit,
  17     str_or_none,
  18     try_get,
  19 )
  20
  21
  22 class ViceBaseIE(InfoExtractor):
  23     def _call_api(self, resource, resource_key, resource_id, locale, fields, args=''):
  24         return self._download_json(
  25             'https://video.vice.com/api/v1/graphql', resource_id, query={
  26                 'query': '''{
  27   %s(locale: "%s", %s: "%s"%s) {
  28     %s
  29   }
  30 }''' % (resource, locale, resource_key, resource_id, args, fields),  # noqa: UP031
  31             })['data'][resource]
  32
  33
  34 class ViceIE(ViceBaseIE, AdobePassIE):
  35     IE_NAME = 'vice'
  36     _VALID_URL = r'https?://(?:(?:video|vms)\.vice|(?:www\.)?vice(?:land|tv))\.com/(?P<locale>[^/]+)/(?:video/[^/]+|embed)/(?P<id>[\da-f]{24})'
  37     _EMBED_REGEX = [r'<iframe\b[^>]+\bsrc=["\'](?P<url>(?:https?:)?//video\.vice\.com/[^/]+/embed/[\da-f]{24})']
  38     _TESTS = [{
  39         'url': 'https://video.vice.com/en_us/video/pet-cremator/58c69e38a55424f1227dc3f7',
  40         'info_dict': {
  41             'id': '58c69e38a55424f1227dc3f7',
  42             'ext': 'mp4',
  43             'title': '10 Questions You Always Wanted To Ask: Pet Cremator',
  44             'description': 'md5:fe856caacf61fe0e74fab15ce2b07ca5',
  45             'uploader': 'vice',
  46             'uploader_id': '57a204088cb727dec794c67b',
  47             'timestamp': 1489664942,
  48             'upload_date': '20170316',
  49             'age_limit': 14,
  50         },
  51         'params': {
  52             # m3u8 download
  53             'skip_download': True,
  54         },
  55     }, {
  56         # geo restricted to US
  57         'url': 'https://video.vice.com/en_us/video/the-signal-from-tolva/5816510690b70e6c5fd39a56',
  58         'info_dict': {
  59             'id': '5816510690b70e6c5fd39a56',
  60             'ext': 'mp4',
  61             'uploader': 'vice',
  62             'title': 'The Signal From Tölva',
  63             'description': 'md5:3927e3c79f9e8094606a2b3c5b5e55d5',
  64             'uploader_id': '57a204088cb727dec794c67b',
  65             'timestamp': 1477941983,
  66             'upload_date': '20161031',
  67         },
  68         'params': {
  69             # m3u8 download
  70             'skip_download': True,
  71         },
  72     }, {
  73         'url': 'https://video.vice.com/alps/video/ulfs-wien-beruchtigste-grafitti-crew-part-1/581b12b60a0e1f4c0fb6ea2f',
  74         'info_dict': {
  75             'id': '581b12b60a0e1f4c0fb6ea2f',
  76             'ext': 'mp4',
  77             'title': 'ULFs - Wien berüchtigste Grafitti Crew - Part 1',
  78             'description': 'Zwischen Hinterzimmer-Tattoos und U-Bahnschächten erzählen uns die Ulfs, wie es ist, "süchtig nach Sachbeschädigung" zu sein.',
  79             'uploader': 'vice',
  80             'uploader_id': '57a204088cb727dec794c67b',
  81             'timestamp': 1485368119,
  82             'upload_date': '20170125',
  83             'age_limit': 14,
  84         },
  85         'params': {
  86             # AES-encrypted m3u8
  87             'skip_download': True,
  88         },
  89     }, {
  90         'url': 'https://video.vice.com/en_us/video/pizza-show-trailer/56d8c9a54d286ed92f7f30e4',
  91         'only_matching': True,
  92     }, {
  93         'url': 'https://video.vice.com/en_us/embed/57f41d3556a0a80f54726060',
  94         'only_matching': True,
  95     }, {
  96         'url': 'https://vms.vice.com/en_us/video/preplay/58c69e38a55424f1227dc3f7',
  97         'only_matching': True,
  98     }, {
  99         'url': 'https://www.viceland.com/en_us/video/thursday-march-1-2018/5a8f2d7ff1cdb332dd446ec1',
 100         'only_matching': True,
 101     }]
 102
 103     def _real_extract(self, url):
 104         locale, video_id = self._match_valid_url(url).groups()
 105
 106         video = self._call_api('videos', 'id', video_id, locale, '''body
 107     locked
 108     rating
 109     thumbnail_url
 110     title''')[0]
 111         title = video['title'].strip()
 112         rating = video.get('rating')
 113
 114         query = {}
 115         if video.get('locked'):
 116             resource = self._get_mvpd_resource(
 117                 'VICELAND', title, video_id, rating)
 118             query['tvetoken'] = self._extract_mvpd_auth(
 119                 url, video_id, 'VICELAND', resource)
 120
 121         # signature generation algorithm is reverse engineered from signatureGenerator in
 122         # webpack:///../shared/~/vice-player/dist/js/vice-player.js in
 123         # https://www.viceland.com/assets/common/js/web.vendor.bundle.js
 124         # new JS is located here https://vice-web-statics-cdn.vice.com/vice-player/player-embed.js
 125         exp = int(time.time()) + 1440
 126
 127         query.update({
 128             'exp': exp,
 129             'sign': hashlib.sha512(f'{video_id}:GET:{exp}'.encode()).hexdigest(),
 130             'skipadstitching': 1,
 131             'platform': 'desktop',
 132             'rn': random.randint(10000, 100000),
 133         })
 134
 135         try:
 136             preplay = self._download_json(
 137                 f'https://vms.vice.com/{locale}/video/preplay/{video_id}',
 138                 video_id, query=query)
 139         except ExtractorError as e:
 140             if isinstance(e.cause, HTTPError) and e.cause.status in (400, 401):
 141                 error = json.loads(e.cause.response.read().decode())
 142                 error_message = error.get('error_description') or error['details']
 143                 raise ExtractorError(f'{self.IE_NAME} said: {error_message}', expected=True)
 144             raise
 145
 146         video_data = preplay['video']
 147         formats = self._extract_m3u8_formats(
 148             preplay['playURL'], video_id, 'mp4', 'm3u8_native')
 149         episode = video_data.get('episode') or {}
 150         channel = video_data.get('channel') or {}
 151         season = video_data.get('season') or {}
 152
 153         subtitles = {}
 154         for subtitle in preplay.get('subtitleURLs', []):
 155             cc_url = subtitle.get('url')
 156             if not cc_url:
 157                 continue
 158             language_code = try_get(subtitle, lambda x: x['languages'][0]['language_code'], str) or 'en'
 159             subtitles.setdefault(language_code, []).append({
 160                 'url': cc_url,
 161             })
 162
 163         return {
 164             'formats': formats,
 165             'id': video_id,
 166             'title': title,
 167             'description': clean_html(video.get('body')),
 168             'thumbnail': video.get('thumbnail_url'),
 169             'duration': int_or_none(video_data.get('video_duration')),
 170             'timestamp': int_or_none(video_data.get('created_at'), 1000),
 171             'age_limit': parse_age_limit(video_data.get('video_rating') or rating),
 172             'series': try_get(video_data, lambda x: x['show']['base']['display_title'], str),
 173             'episode_number': int_or_none(episode.get('episode_number')),
 174             'episode_id': str_or_none(episode.get('id') or video_data.get('episode_id')),
 175             'season_number': int_or_none(season.get('season_number')),
 176             'season_id': str_or_none(season.get('id') or video_data.get('season_id')),
 177             'uploader': channel.get('name'),
 178             'uploader_id': str_or_none(channel.get('id')),
 179             'subtitles': subtitles,
 180         }
 181
 182
 183 class ViceShowIE(ViceBaseIE):
 184     IE_NAME = 'vice:show'
 185     _VALID_URL = r'https?://(?:video\.vice|(?:www\.)?vice(?:land|tv))\.com/(?P<locale>[^/]+)/show/(?P<id>[^/?#&]+)'
 186     _PAGE_SIZE = 25
 187     _TESTS = [{
 188         'url': 'https://video.vice.com/en_us/show/fck-thats-delicious',
 189         'info_dict': {
 190             'id': '57a2040c8cb727dec794c901',
 191             'title': 'F*ck, That’s Delicious',
 192             'description': 'The life and eating habits of rap’s greatest bon vivant, Action Bronson.',
 193         },
 194         'playlist_mincount': 64,
 195     }, {
 196         'url': 'https://www.vicetv.com/en_us/show/fck-thats-delicious',
 197         'only_matching': True,
 198     }]
 199
 200     def _fetch_page(self, locale, show_id, page):
 201         videos = self._call_api('videos', 'show_id', show_id, locale, '''body
 202     id
 203     url''', f', page: {page + 1}, per_page: {self._PAGE_SIZE}')
 204         for video in videos:
 205             yield self.url_result(
 206                 video['url'], ViceIE.ie_key(), video.get('id'))
 207
 208     def _real_extract(self, url):
 209         locale, display_id = self._match_valid_url(url).groups()
 210         show = self._call_api('shows', 'slug', display_id, locale, '''dek
 211     id
 212     title''')[0]
 213         show_id = show['id']
 214
 215         entries = OnDemandPagedList(
 216             functools.partial(self._fetch_page, locale, show_id),
 217             self._PAGE_SIZE)
 218
 219         return self.playlist_result(
 220             entries, show_id, show.get('title'), show.get('dek'))
 221
 222
 223 class ViceArticleIE(ViceBaseIE):
 224     IE_NAME = 'vice:article'
 225     _VALID_URL = r'https?://(?:www\.)?vice\.com/(?P<locale>[^/]+)/article/(?:[0-9a-z]{6}/)?(?P<id>[^?#]+)'
 226
 227     _TESTS = [{
 228         'url': 'https://www.vice.com/en_us/article/on-set-with-the-woman-making-mormon-porn-in-utah',
 229         'info_dict': {
 230             'id': '58dc0a3dee202d2a0ccfcbd8',
 231             'ext': 'mp4',
 232             'title': 'Mormon War on Porn',
 233             'description': 'md5:1c5d91fe25fa8aa304f9def118b92dbf',
 234             'uploader': 'vice',
 235             'uploader_id': '57a204088cb727dec794c67b',
 236             'timestamp': 1491883129,
 237             'upload_date': '20170411',
 238             'age_limit': 17,
 239         },
 240         'params': {
 241             # AES-encrypted m3u8
 242             'skip_download': True,
 243         },
 244         'add_ie': [ViceIE.ie_key()],
 245     }, {
 246         'url': 'https://www.vice.com/en_us/article/how-to-hack-a-car',
 247         'md5': '13010ee0bc694ea87ec40724397c2349',
 248         'info_dict': {
 249             'id': '3jstaBeXgAs',
 250             'ext': 'mp4',
 251             'title': 'How to Hack a Car: Phreaked Out (Episode 2)',
 252             'description': 'md5:ee95453f7ff495db8efe14ae8bf56f30',
 253             'uploader': 'Motherboard',
 254             'uploader_id': 'MotherboardTV',
 255             'upload_date': '20140529',
 256         },
 257         'add_ie': [YoutubeIE.ie_key()],
 258     }, {
 259         'url': 'https://www.vice.com/en_us/article/znm9dx/karley-sciortino-slutever-reloaded',
 260         'md5': 'a7ecf64ee4fa19b916c16f4b56184ae2',
 261         'info_dict': {
 262             'id': '57f41d3556a0a80f54726060',
 263             'ext': 'mp4',
 264             'title': "Making The World's First Male Sex Doll",
 265             'description': 'md5:19b00b215b99961cf869c40fbe9df755',
 266             'uploader': 'vice',
 267             'uploader_id': '57a204088cb727dec794c67b',
 268             'timestamp': 1476919911,
 269             'upload_date': '20161019',
 270             'age_limit': 17,
 271         },
 272         'params': {
 273             'skip_download': True,
 274         },
 275         'add_ie': [ViceIE.ie_key()],
 276     }, {
 277         'url': 'https://www.vice.com/en_us/article/cowboy-capitalists-part-1',
 278         'only_matching': True,
 279     }, {
 280         'url': 'https://www.vice.com/ru/article/big-night-out-ibiza-clive-martin-229',
 281         'only_matching': True,
 282     }]
 283
 284     def _real_extract(self, url):
 285         locale, display_id = self._match_valid_url(url).groups()
 286
 287         article = self._call_api('articles', 'slug', display_id, locale, '''body
 288     embed_code''')[0]
 289         body = article['body']
 290
 291         def _url_res(video_url, ie_key):
 292             return {
 293                 '_type': 'url_transparent',
 294                 'url': video_url,
 295                 'display_id': display_id,
 296                 'ie_key': ie_key,
 297             }
 298
 299         vice_url = ViceIE._extract_url(body)
 300         if vice_url:
 301             return _url_res(vice_url, ViceIE.ie_key())
 302
 303         youtube_url = YoutubeIE._extract_url(body)
 304         if youtube_url:
 305             return _url_res(youtube_url, YoutubeIE.ie_key())
 306
 307         video_url = self._html_search_regex(
 308             r'data-video-url="([^"]+)"',
 309             article['embed_code'], 'video URL')
 310
 311         return _url_res(video_url, ViceIE.ie_key())