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

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import functools
   5 import hashlib
   6 import json
   7 import random
   8 import re
   9 import time
  10
  11 from .adobepass import AdobePassIE
  12 from .common import InfoExtractor
  13 from .youtube import YoutubeIE
  14 from ..compat import (
  15     compat_HTTPError,
  16     compat_str,
  17 )
  18 from ..utils import (
  19     clean_html,
  20     ExtractorError,
  21     int_or_none,
  22     OnDemandPagedList,
  23     parse_age_limit,
  24     str_or_none,
  25     try_get,
  26 )
  27
  28
  29 class ViceBaseIE(InfoExtractor):
  30     def _call_api(self, resource, resource_key, resource_id, locale, fields, args=''):
  31         return self._download_json(
  32             'https://video.vice.com/api/v1/graphql', resource_id, query={
  33                 'query': '''{
  34   %s(locale: "%s", %s: "%s"%s) {
  35     %s
  36   }
  37 }''' % (resource, locale, resource_key, resource_id, args, fields),
  38             })['data'][resource]
  39
  40
  41 class ViceIE(ViceBaseIE, AdobePassIE):
  42     IE_NAME = 'vice'
  43     _VALID_URL = r'https?://(?:(?:video|vms)\.vice|(?:www\.)?vice(?:land|tv))\.com/(?P<locale>[^/]+)/(?:video/[^/]+|embed)/(?P<id>[\da-f]{24})'
  44     _TESTS = [{
  45         'url': 'https://video.vice.com/en_us/video/pet-cremator/58c69e38a55424f1227dc3f7',
  46         'info_dict': {
  47             'id': '58c69e38a55424f1227dc3f7',
  48             'ext': 'mp4',
  49             'title': '10 Questions You Always Wanted To Ask: Pet Cremator',
  50             'description': 'md5:fe856caacf61fe0e74fab15ce2b07ca5',
  51             'uploader': 'vice',
  52             'uploader_id': '57a204088cb727dec794c67b',
  53             'timestamp': 1489664942,
  54             'upload_date': '20170316',
  55             'age_limit': 14,
  56         },
  57         'params': {
  58             # m3u8 download
  59             'skip_download': True,
  60         },
  61     }, {
  62         # geo restricted to US
  63         'url': 'https://video.vice.com/en_us/video/the-signal-from-tolva/5816510690b70e6c5fd39a56',
  64         'info_dict': {
  65             'id': '5816510690b70e6c5fd39a56',
  66             'ext': 'mp4',
  67             'uploader': 'vice',
  68             'title': 'The Signal From Tölva',
  69             'description': 'md5:3927e3c79f9e8094606a2b3c5b5e55d5',
  70             'uploader_id': '57a204088cb727dec794c67b',
  71             'timestamp': 1477941983,
  72             'upload_date': '20161031',
  73         },
  74         'params': {
  75             # m3u8 download
  76             'skip_download': True,
  77         },
  78     }, {
  79         'url': 'https://video.vice.com/alps/video/ulfs-wien-beruchtigste-grafitti-crew-part-1/581b12b60a0e1f4c0fb6ea2f',
  80         'info_dict': {
  81             'id': '581b12b60a0e1f4c0fb6ea2f',
  82             'ext': 'mp4',
  83             'title': 'ULFs - Wien berüchtigste Grafitti Crew - Part 1',
  84             'description': 'Zwischen Hinterzimmer-Tattoos und U-Bahnschächten erzählen uns die Ulfs, wie es ist, "süchtig nach Sachbeschädigung" zu sein.',
  85             'uploader': 'vice',
  86             'uploader_id': '57a204088cb727dec794c67b',
  87             'timestamp': 1485368119,
  88             'upload_date': '20170125',
  89             'age_limit': 14,
  90         },
  91         'params': {
  92             # AES-encrypted m3u8
  93             'skip_download': True,
  94         },
  95     }, {
  96         'url': 'https://video.vice.com/en_us/video/pizza-show-trailer/56d8c9a54d286ed92f7f30e4',
  97         'only_matching': True,
  98     }, {
  99         'url': 'https://video.vice.com/en_us/embed/57f41d3556a0a80f54726060',
 100         'only_matching': True,
 101     }, {
 102         'url': 'https://vms.vice.com/en_us/video/preplay/58c69e38a55424f1227dc3f7',
 103         'only_matching': True,
 104     }, {
 105         'url': 'https://www.viceland.com/en_us/video/thursday-march-1-2018/5a8f2d7ff1cdb332dd446ec1',
 106         'only_matching': True,
 107     }]
 108
 109     @staticmethod
 110     def _extract_urls(webpage):
 111         return re.findall(
 112             r'<iframe\b[^>]+\bsrc=["\']((?:https?:)?//video\.vice\.com/[^/]+/embed/[\da-f]{24})',
 113             webpage)
 114
 115     @staticmethod
 116     def _extract_url(webpage):
 117         urls = ViceIE._extract_urls(webpage)
 118         return urls[0] if urls else None
 119
 120     def _real_extract(self, url):
 121         locale, video_id = re.match(self._VALID_URL, url).groups()
 122
 123         video = self._call_api('videos', 'id', video_id, locale, '''body
 124     locked
 125     rating
 126     thumbnail_url
 127     title''')[0]
 128         title = video['title'].strip()
 129         rating = video.get('rating')
 130
 131         query = {}
 132         if video.get('locked'):
 133             resource = self._get_mvpd_resource(
 134                 'VICELAND', title, video_id, rating)
 135             query['tvetoken'] = self._extract_mvpd_auth(
 136                 url, video_id, 'VICELAND', resource)
 137
 138         # signature generation algorithm is reverse engineered from signatureGenerator in
 139         # webpack:///../shared/~/vice-player/dist/js/vice-player.js in
 140         # https://www.viceland.com/assets/common/js/web.vendor.bundle.js
 141         # new JS is located here https://vice-web-statics-cdn.vice.com/vice-player/player-embed.js
 142         exp = int(time.time()) + 1440
 143
 144         query.update({
 145             'exp': exp,
 146             'sign': hashlib.sha512(('%s:GET:%d' % (video_id, exp)).encode()).hexdigest(),
 147             'skipadstitching': 1,
 148             'platform': 'desktop',
 149             'rn': random.randint(10000, 100000),
 150         })
 151
 152         try:
 153             preplay = self._download_json(
 154                 'https://vms.vice.com/%s/video/preplay/%s' % (locale, video_id),
 155                 video_id, query=query)
 156         except ExtractorError as e:
 157             if isinstance(e.cause, compat_HTTPError) and e.cause.code in (400, 401):
 158                 error = json.loads(e.cause.read().decode())
 159                 error_message = error.get('error_description') or error['details']
 160                 raise ExtractorError('%s said: %s' % (
 161                     self.IE_NAME, error_message), expected=True)
 162             raise
 163
 164         video_data = preplay['video']
 165         formats = self._extract_m3u8_formats(
 166             preplay['playURL'], video_id, 'mp4', 'm3u8_native')
 167         self._sort_formats(formats)
 168         episode = video_data.get('episode') or {}
 169         channel = video_data.get('channel') or {}
 170         season = video_data.get('season') or {}
 171
 172         subtitles = {}
 173         for subtitle in preplay.get('subtitleURLs', []):
 174             cc_url = subtitle.get('url')
 175             if not cc_url:
 176                 continue
 177             language_code = try_get(subtitle, lambda x: x['languages'][0]['language_code'], compat_str) or 'en'
 178             subtitles.setdefault(language_code, []).append({
 179                 'url': cc_url,
 180             })
 181
 182         return {
 183             'formats': formats,
 184             'id': video_id,
 185             'title': title,
 186             'description': clean_html(video.get('body')),
 187             'thumbnail': video.get('thumbnail_url'),
 188             'duration': int_or_none(video_data.get('video_duration')),
 189             'timestamp': int_or_none(video_data.get('created_at'), 1000),
 190             'age_limit': parse_age_limit(video_data.get('video_rating') or rating),
 191             'series': try_get(video_data, lambda x: x['show']['base']['display_title'], compat_str),
 192             'episode_number': int_or_none(episode.get('episode_number')),
 193             'episode_id': str_or_none(episode.get('id') or video_data.get('episode_id')),
 194             'season_number': int_or_none(season.get('season_number')),
 195             'season_id': str_or_none(season.get('id') or video_data.get('season_id')),
 196             'uploader': channel.get('name'),
 197             'uploader_id': str_or_none(channel.get('id')),
 198             'subtitles': subtitles,
 199         }
 200
 201
 202 class ViceShowIE(ViceBaseIE):
 203     IE_NAME = 'vice:show'
 204     _VALID_URL = r'https?://(?:video\.vice|(?:www\.)?vice(?:land|tv))\.com/(?P<locale>[^/]+)/show/(?P<id>[^/?#&]+)'
 205     _PAGE_SIZE = 25
 206     _TESTS = [{
 207         'url': 'https://video.vice.com/en_us/show/fck-thats-delicious',
 208         'info_dict': {
 209             'id': '57a2040c8cb727dec794c901',
 210             'title': 'F*ck, That’s Delicious',
 211             'description': 'The life and eating habits of rap’s greatest bon vivant, Action Bronson.',
 212         },
 213         'playlist_mincount': 64,
 214     }, {
 215         'url': 'https://www.vicetv.com/en_us/show/fck-thats-delicious',
 216         'only_matching': True,
 217     }]
 218
 219     def _fetch_page(self, locale, show_id, page):
 220         videos = self._call_api('videos', 'show_id', show_id, locale, '''body
 221     id
 222     url''', ', page: %d, per_page: %d' % (page + 1, self._PAGE_SIZE))
 223         for video in videos:
 224             yield self.url_result(
 225                 video['url'], ViceIE.ie_key(), video.get('id'))
 226
 227     def _real_extract(self, url):
 228         locale, display_id = re.match(self._VALID_URL, url).groups()
 229         show = self._call_api('shows', 'slug', display_id, locale, '''dek
 230     id
 231     title''')[0]
 232         show_id = show['id']
 233
 234         entries = OnDemandPagedList(
 235             functools.partial(self._fetch_page, locale, show_id),
 236             self._PAGE_SIZE)
 237
 238         return self.playlist_result(
 239             entries, show_id, show.get('title'), show.get('dek'))
 240
 241
 242 class ViceArticleIE(ViceBaseIE):
 243     IE_NAME = 'vice:article'
 244     _VALID_URL = r'https://(?:www\.)?vice\.com/(?P<locale>[^/]+)/article/(?:[0-9a-z]{6}/)?(?P<id>[^?#]+)'
 245
 246     _TESTS = [{
 247         'url': 'https://www.vice.com/en_us/article/on-set-with-the-woman-making-mormon-porn-in-utah',
 248         'info_dict': {
 249             'id': '58dc0a3dee202d2a0ccfcbd8',
 250             'ext': 'mp4',
 251             'title': 'Mormon War on Porn',
 252             'description': 'md5:1c5d91fe25fa8aa304f9def118b92dbf',
 253             'uploader': 'vice',
 254             'uploader_id': '57a204088cb727dec794c67b',
 255             'timestamp': 1491883129,
 256             'upload_date': '20170411',
 257             'age_limit': 17,
 258         },
 259         'params': {
 260             # AES-encrypted m3u8
 261             'skip_download': True,
 262         },
 263         'add_ie': [ViceIE.ie_key()],
 264     }, {
 265         'url': 'https://www.vice.com/en_us/article/how-to-hack-a-car',
 266         'md5': '13010ee0bc694ea87ec40724397c2349',
 267         'info_dict': {
 268             'id': '3jstaBeXgAs',
 269             'ext': 'mp4',
 270             'title': 'How to Hack a Car: Phreaked Out (Episode 2)',
 271             'description': 'md5:ee95453f7ff495db8efe14ae8bf56f30',
 272             'uploader': 'Motherboard',
 273             'uploader_id': 'MotherboardTV',
 274             'upload_date': '20140529',
 275         },
 276         'add_ie': [YoutubeIE.ie_key()],
 277     }, {
 278         'url': 'https://www.vice.com/en_us/article/znm9dx/karley-sciortino-slutever-reloaded',
 279         'md5': 'a7ecf64ee4fa19b916c16f4b56184ae2',
 280         'info_dict': {
 281             'id': '57f41d3556a0a80f54726060',
 282             'ext': 'mp4',
 283             'title': "Making The World's First Male Sex Doll",
 284             'description': 'md5:19b00b215b99961cf869c40fbe9df755',
 285             'uploader': 'vice',
 286             'uploader_id': '57a204088cb727dec794c67b',
 287             'timestamp': 1476919911,
 288             'upload_date': '20161019',
 289             'age_limit': 17,
 290         },
 291         'params': {
 292             'skip_download': True,
 293             'format': 'bestvideo',
 294         },
 295         'add_ie': [ViceIE.ie_key()],
 296     }, {
 297         'url': 'https://www.vice.com/en_us/article/cowboy-capitalists-part-1',
 298         'only_matching': True,
 299     }, {
 300         'url': 'https://www.vice.com/ru/article/big-night-out-ibiza-clive-martin-229',
 301         'only_matching': True,
 302     }]
 303
 304     def _real_extract(self, url):
 305         locale, display_id = re.match(self._VALID_URL, url).groups()
 306
 307         article = self._call_api('articles', 'slug', display_id, locale, '''body
 308     embed_code''')[0]
 309         body = article['body']
 310
 311         def _url_res(video_url, ie_key):
 312             return {
 313                 '_type': 'url_transparent',
 314                 'url': video_url,
 315                 'display_id': display_id,
 316                 'ie_key': ie_key,
 317             }
 318
 319         vice_url = ViceIE._extract_url(body)
 320         if vice_url:
 321             return _url_res(vice_url, ViceIE.ie_key())
 322
 323         embed_code = self._search_regex(
 324             r'embedCode=([^&\'"]+)', body,
 325             'ooyala embed code', default=None)
 326         if embed_code:
 327             return _url_res('ooyala:%s' % embed_code, 'Ooyala')
 328
 329         youtube_url = YoutubeIE._extract_url(body)
 330         if youtube_url:
 331             return _url_res(youtube_url, YoutubeIE.ie_key())
 332
 333         video_url = self._html_search_regex(
 334             r'data-video-url="([^"]+)"',
 335             article['embed_code'], 'video URL')
 336
 337         return _url_res(video_url, ViceIE.ie_key())