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