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

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5 import time
   6 import hashlib
   7 import json
   8 import random
   9
  10 from .adobepass import AdobePassIE
  11 from .youtube import YoutubeIE
  12 from .common import InfoExtractor
  13 from ..compat import (
  14     compat_HTTPError,
  15     compat_str,
  16 )
  17 from ..utils import (
  18     ExtractorError,
  19     int_or_none,
  20     parse_age_limit,
  21     str_or_none,
  22     try_get,
  23 )
  24
  25
  26 class ViceIE(AdobePassIE):
  27     IE_NAME = 'vice'
  28     _VALID_URL = r'https?://(?:(?:video|vms)\.vice|(?:www\.)?viceland)\.com/(?P<locale>[^/]+)/(?:video/[^/]+|embed)/(?P<id>[\da-f]+)'
  29     _TESTS = [{
  30         'url': 'https://video.vice.com/en_us/video/pet-cremator/58c69e38a55424f1227dc3f7',
  31         'info_dict': {
  32             'id': '5e647f0125e145c9aef2069412c0cbde',
  33             'ext': 'mp4',
  34             'title': '10 Questions You Always Wanted To Ask: Pet Cremator',
  35             'description': 'md5:fe856caacf61fe0e74fab15ce2b07ca5',
  36             'uploader': 'vice',
  37             'uploader_id': '57a204088cb727dec794c67b',
  38             'timestamp': 1489664942,
  39             'upload_date': '20170316',
  40             'age_limit': 14,
  41         },
  42         'params': {
  43             # m3u8 download
  44             'skip_download': True,
  45         },
  46         'add_ie': ['UplynkPreplay'],
  47     }, {
  48         # geo restricted to US
  49         'url': 'https://video.vice.com/en_us/video/the-signal-from-tolva/5816510690b70e6c5fd39a56',
  50         'info_dict': {
  51             'id': '930c0ad1f47141cc955087eecaddb0e2',
  52             'ext': 'mp4',
  53             'uploader': 'waypoint',
  54             'title': 'The Signal From Tölva',
  55             'description': 'md5:3927e3c79f9e8094606a2b3c5b5e55d5',
  56             'uploader_id': '57f7d621e05ca860fa9ccaf9',
  57             'timestamp': 1477941983,
  58             'upload_date': '20161031',
  59         },
  60         'params': {
  61             # m3u8 download
  62             'skip_download': True,
  63         },
  64         'add_ie': ['UplynkPreplay'],
  65     }, {
  66         'url': 'https://video.vice.com/alps/video/ulfs-wien-beruchtigste-grafitti-crew-part-1/581b12b60a0e1f4c0fb6ea2f',
  67         'info_dict': {
  68             'id': '581b12b60a0e1f4c0fb6ea2f',
  69             'ext': 'mp4',
  70             'title': 'ULFs - Wien berüchtigste Grafitti Crew - Part 1',
  71             'description': '<p>Zwischen Hinterzimmer-Tattoos und U-Bahnschächten erzählen uns die Ulfs, wie es ist, "süchtig nach Sachbeschädigung" zu sein.</p>',
  72             'uploader': 'VICE',
  73             'uploader_id': '57a204088cb727dec794c67b',
  74             'timestamp': 1485368119,
  75             'upload_date': '20170125',
  76             'age_limit': 14,
  77         },
  78         'params': {
  79             # AES-encrypted m3u8
  80             'skip_download': True,
  81             'proxy': '127.0.0.1:8118',
  82         },
  83         'add_ie': ['UplynkPreplay'],
  84     }, {
  85         'url': 'https://video.vice.com/en_us/video/pizza-show-trailer/56d8c9a54d286ed92f7f30e4',
  86         'only_matching': True,
  87     }, {
  88         'url': 'https://video.vice.com/en_us/embed/57f41d3556a0a80f54726060',
  89         'only_matching': True,
  90     }, {
  91         'url': 'https://vms.vice.com/en_us/video/preplay/58c69e38a55424f1227dc3f7',
  92         'only_matching': True,
  93     }, {
  94         'url': 'https://www.viceland.com/en_us/video/thursday-march-1-2018/5a8f2d7ff1cdb332dd446ec1',
  95         'only_matching': True,
  96     }]
  97     _PREPLAY_HOST = 'vms.vice'
  98
  99     @staticmethod
 100     def _extract_urls(webpage):
 101         return re.findall(
 102             r'<iframe\b[^>]+\bsrc=["\']((?:https?:)?//video\.vice\.com/[^/]+/embed/[\da-f]+)',
 103             webpage)
 104
 105     @staticmethod
 106     def _extract_url(webpage):
 107         urls = ViceIE._extract_urls(webpage)
 108         return urls[0] if urls else None
 109
 110     def _real_extract(self, url):
 111         locale, video_id = re.match(self._VALID_URL, url).groups()
 112
 113         webpage = self._download_webpage(
 114             'https://video.vice.com/%s/embed/%s' % (locale, video_id),
 115             video_id)
 116
 117         video = self._parse_json(
 118             self._search_regex(
 119                 r'PREFETCH_DATA\s*=\s*({.+?})\s*;\s*\n', webpage,
 120                 'app state'), video_id)['video']
 121         video_id = video.get('vms_id') or video.get('id') or video_id
 122         title = video['title']
 123         is_locked = video.get('locked')
 124         rating = video.get('rating')
 125         thumbnail = video.get('thumbnail_url')
 126         duration = int_or_none(video.get('duration'))
 127         series = try_get(
 128             video, lambda x: x['episode']['season']['show']['title'],
 129             compat_str)
 130         episode_number = try_get(
 131             video, lambda x: x['episode']['episode_number'])
 132         season_number = try_get(
 133             video, lambda x: x['episode']['season']['season_number'])
 134         uploader = None
 135
 136         query = {}
 137         if is_locked:
 138             resource = self._get_mvpd_resource(
 139                 'VICELAND', title, video_id, rating)
 140             query['tvetoken'] = self._extract_mvpd_auth(
 141                 url, video_id, 'VICELAND', resource)
 142
 143         # signature generation algorithm is reverse engineered from signatureGenerator in
 144         # webpack:///../shared/~/vice-player/dist/js/vice-player.js in
 145         # https://www.viceland.com/assets/common/js/web.vendor.bundle.js
 146         # new JS is located here https://vice-web-statics-cdn.vice.com/vice-player/player-embed.js
 147         exp = int(time.time()) + 1440
 148
 149         query.update({
 150             'exp': exp,
 151             'sign': hashlib.sha512(('%s:GET:%d' % (video_id, exp)).encode()).hexdigest(),
 152             '_ad_blocked': None,
 153             '_ad_unit': '',
 154             '_debug': '',
 155             'platform': 'desktop',
 156             'rn': random.randint(10000, 100000),
 157             'fbprebidtoken': '',
 158         })
 159
 160         try:
 161             host = 'www.viceland' if is_locked else self._PREPLAY_HOST
 162             preplay = self._download_json(
 163                 'https://%s.com/%s/video/preplay/%s' % (host, locale, video_id),
 164                 video_id, query=query)
 165         except ExtractorError as e:
 166             if isinstance(e.cause, compat_HTTPError) and e.cause.code in (400, 401):
 167                 error = json.loads(e.cause.read().decode())
 168                 error_message = error.get('error_description') or error['details']
 169                 raise ExtractorError('%s said: %s' % (
 170                     self.IE_NAME, error_message), expected=True)
 171             raise
 172
 173         video_data = preplay['video']
 174         base = video_data['base']
 175         uplynk_preplay_url = preplay['preplayURL']
 176         episode = video_data.get('episode', {})
 177         channel = video_data.get('channel', {})
 178
 179         subtitles = {}
 180         cc_url = preplay.get('ccURL')
 181         if cc_url:
 182             subtitles['en'] = [{
 183                 'url': cc_url,
 184             }]
 185
 186         return {
 187             '_type': 'url_transparent',
 188             'url': uplynk_preplay_url,
 189             'id': video_id,
 190             'title': title,
 191             'description': base.get('body') or base.get('display_body'),
 192             'thumbnail': thumbnail,
 193             'duration': int_or_none(video_data.get('video_duration')) or duration,
 194             'timestamp': int_or_none(video_data.get('created_at'), 1000),
 195             'age_limit': parse_age_limit(video_data.get('video_rating')),
 196             'series': video_data.get('show_title') or series,
 197             'episode_number': int_or_none(episode.get('episode_number') or episode_number),
 198             'episode_id': str_or_none(episode.get('id') or video_data.get('episode_id')),
 199             'season_number': int_or_none(season_number),
 200             'season_id': str_or_none(episode.get('season_id')),
 201             'uploader': channel.get('base', {}).get('title') or channel.get('name') or uploader,
 202             'uploader_id': str_or_none(channel.get('id')),
 203             'subtitles': subtitles,
 204             'ie_key': 'UplynkPreplay',
 205         }
 206
 207
 208 class ViceShowIE(InfoExtractor):
 209     IE_NAME = 'vice:show'
 210     _VALID_URL = r'https?://(?:.+?\.)?vice\.com/(?:[^/]+/)?show/(?P<id>[^/?#&]+)'
 211
 212     _TEST = {
 213         'url': 'https://munchies.vice.com/en/show/fuck-thats-delicious-2',
 214         'info_dict': {
 215             'id': 'fuck-thats-delicious-2',
 216             'title': "Fuck, That's Delicious",
 217             'description': 'Follow the culinary adventures of rapper Action Bronson during his ongoing world tour.',
 218         },
 219         'playlist_count': 17,
 220     }
 221
 222     def _real_extract(self, url):
 223         show_id = self._match_id(url)
 224         webpage = self._download_webpage(url, show_id)
 225
 226         entries = [
 227             self.url_result(video_url, ViceIE.ie_key())
 228             for video_url, _ in re.findall(
 229                 r'<h2[^>]+class="article-title"[^>]+data-id="\d+"[^>]*>\s*<a[^>]+href="(%s.*?)"'
 230                 % ViceIE._VALID_URL, webpage)]
 231
 232         title = self._search_regex(
 233             r'<title>(.+?)</title>', webpage, 'title', default=None)
 234         if title:
 235             title = re.sub(r'(.+)\s*\|\s*.+$', r'\1', title).strip()
 236         description = self._html_search_meta(
 237             'description', webpage, 'description')
 238
 239         return self.playlist_result(entries, show_id, title, description)
 240
 241
 242 class ViceArticleIE(InfoExtractor):
 243     IE_NAME = 'vice:article'
 244     _VALID_URL = r'https://www\.vice\.com/[^/]+/article/(?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': '41eae2a47b174a1398357cec55f1f6fc',
 250             'ext': 'mp4',
 251             'title': 'Mormon War on Porn ',
 252             'description': 'md5:6394a8398506581d0346b9ab89093fef',
 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': ['UplynkPreplay'],
 264     }, {
 265         'url': 'https://www.vice.com/en_us/article/how-to-hack-a-car',
 266         'md5': '7fe8ebc4fa3323efafc127b82bd821d9',
 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': ['Youtube'],
 277     }, {
 278         'url': 'https://www.vice.com/en_us/article/znm9dx/karley-sciortino-slutever-reloaded',
 279         'md5': 'a7ecf64ee4fa19b916c16f4b56184ae2',
 280         'info_dict': {
 281             'id': 'e2ed435eb67e43efb66e6ef9a6930a88',
 282             'ext': 'mp4',
 283             'title': "Making The World's First Male Sex Doll",
 284             'description': 'md5:916078ef0e032d76343116208b6cc2c4',
 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         },
 294         'add_ie': [ViceIE.ie_key()],
 295     }, {
 296         'url': 'https://www.vice.com/en_us/article/cowboy-capitalists-part-1',
 297         'only_matching': True,
 298     }, {
 299         'url': 'https://www.vice.com/ru/article/big-night-out-ibiza-clive-martin-229',
 300         'only_matching': True,
 301     }]
 302
 303     def _real_extract(self, url):
 304         display_id = self._match_id(url)
 305
 306         webpage = self._download_webpage(url, display_id)
 307
 308         prefetch_data = self._parse_json(self._search_regex(
 309             r'__APP_STATE\s*=\s*({.+?})(?:\s*\|\|\s*{}\s*)?;\s*\n',
 310             webpage, 'app state'), display_id)['pageData']
 311         body = prefetch_data['body']
 312
 313         def _url_res(video_url, ie_key):
 314             return {
 315                 '_type': 'url_transparent',
 316                 'url': video_url,
 317                 'display_id': display_id,
 318                 'ie_key': ie_key,
 319             }
 320
 321         vice_url = ViceIE._extract_url(webpage)
 322         if vice_url:
 323             return _url_res(vice_url, ViceIE.ie_key())
 324
 325         embed_code = self._search_regex(
 326             r'embedCode=([^&\'"]+)', body,
 327             'ooyala embed code', default=None)
 328         if embed_code:
 329             return _url_res('ooyala:%s' % embed_code, 'Ooyala')
 330
 331         youtube_url = YoutubeIE._extract_url(body)
 332         if youtube_url:
 333             return _url_res(youtube_url, YoutubeIE.ie_key())
 334
 335         video_url = self._html_search_regex(
 336             r'data-video-url="([^"]+)"',
 337             prefetch_data['embed_code'], 'video URL')
 338
 339         return _url_res(video_url, ViceIE.ie_key())