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
   9 from .adobepass import AdobePassIE
  10 from .common import InfoExtractor
  11 from ..compat import compat_HTTPError
  12 from ..utils import (
  13     int_or_none,
  14     parse_age_limit,
  15     str_or_none,
  16     parse_duration,
  17     ExtractorError,
  18     extract_attributes,
  19 )
  20
  21
  22 class ViceBaseIE(AdobePassIE):
  23     def _extract_preplay_video(self, url, locale, webpage):
  24         watch_hub_data = extract_attributes(self._search_regex(
  25             r'(?s)(<watch-hub\s*.+?</watch-hub>)', webpage, 'watch hub'))
  26         video_id = watch_hub_data['vms-id']
  27         title = watch_hub_data['video-title']
  28
  29         query = {}
  30         is_locked = watch_hub_data.get('video-locked') == '1'
  31         if is_locked:
  32             resource = self._get_mvpd_resource(
  33                 'VICELAND', title, video_id,
  34                 watch_hub_data.get('video-rating'))
  35             query['tvetoken'] = self._extract_mvpd_auth(
  36                 url, video_id, 'VICELAND', resource)
  37
  38         # signature generation algorithm is reverse engineered from signatureGenerator in
  39         # webpack:///../shared/~/vice-player/dist/js/vice-player.js in
  40         # https://www.viceland.com/assets/common/js/web.vendor.bundle.js
  41         exp = int(time.time()) + 14400
  42         query.update({
  43             'exp': exp,
  44             'sign': hashlib.sha512(('%s:GET:%d' % (video_id, exp)).encode()).hexdigest(),
  45         })
  46
  47         try:
  48             host = 'www.viceland' if is_locked else self._PREPLAY_HOST
  49             preplay = self._download_json(
  50                 'https://%s.com/%s/preplay/%s' % (host, locale, video_id),
  51                 video_id, query=query)
  52         except ExtractorError as e:
  53             if isinstance(e.cause, compat_HTTPError) and e.cause.code == 400:
  54                 error = json.loads(e.cause.read().decode())
  55                 raise ExtractorError('%s said: %s' % (
  56                     self.IE_NAME, error['details']), expected=True)
  57             raise
  58
  59         video_data = preplay['video']
  60         base = video_data['base']
  61         uplynk_preplay_url = preplay['preplayURL']
  62         episode = video_data.get('episode', {})
  63         channel = video_data.get('channel', {})
  64
  65         subtitles = {}
  66         cc_url = preplay.get('ccURL')
  67         if cc_url:
  68             subtitles['en'] = [{
  69                 'url': cc_url,
  70             }]
  71
  72         return {
  73             '_type': 'url_transparent',
  74             'url': uplynk_preplay_url,
  75             'id': video_id,
  76             'title': title,
  77             'description': base.get('body') or base.get('display_body'),
  78             'thumbnail': watch_hub_data.get('cover-image') or watch_hub_data.get('thumbnail'),
  79             'duration': int_or_none(video_data.get('video_duration')) or parse_duration(watch_hub_data.get('video-duration')),
  80             'timestamp': int_or_none(video_data.get('created_at'), 1000),
  81             'age_limit': parse_age_limit(video_data.get('video_rating')),
  82             'series': video_data.get('show_title') or watch_hub_data.get('show-title'),
  83             'episode_number': int_or_none(episode.get('episode_number') or watch_hub_data.get('episode')),
  84             'episode_id': str_or_none(episode.get('id') or video_data.get('episode_id')),
  85             'season_number': int_or_none(watch_hub_data.get('season')),
  86             'season_id': str_or_none(episode.get('season_id')),
  87             'uploader': channel.get('base', {}).get('title') or watch_hub_data.get('channel-title'),
  88             'uploader_id': str_or_none(channel.get('id')),
  89             'subtitles': subtitles,
  90             'ie_key': 'UplynkPreplay',
  91         }
  92
  93
  94 class ViceIE(ViceBaseIE):
  95     IE_NAME = 'vice'
  96     _VALID_URL = r'https?://(?:.+?\.)?vice\.com/(?:(?P<locale>[^/]+)/)?videos?/(?P<id>[^/?#&]+)'
  97
  98     _TESTS = [{
  99         'url': 'https://news.vice.com/video/experimenting-on-animals-inside-the-monkey-lab',
 100         'md5': '7d3ae2f9ba5f196cdd9f9efd43657ac2',
 101         'info_dict': {
 102             'id': 'N2bzkydjraWDGwnt8jAttCF6Y0PDv4Zj',
 103             'ext': 'flv',
 104             'title': 'Monkey Labs of Holland',
 105             'description': 'md5:92b3c7dcbfe477f772dd4afa496c9149',
 106         },
 107         'add_ie': ['Ooyala'],
 108     }, {
 109         'url': 'https://video.vice.com/en_us/video/the-signal-from-tolva/5816510690b70e6c5fd39a56',
 110         'info_dict': {
 111             'id': '5816510690b70e6c5fd39a56',
 112             'ext': 'mp4',
 113             'uploader': 'Waypoint',
 114             'title': 'The Signal From Tölva',
 115             'description': 'md5:3927e3c79f9e8094606a2b3c5b5e55d5',
 116             'uploader_id': '57f7d621e05ca860fa9ccaf9',
 117             'timestamp': 1477941983,
 118             'upload_date': '20161031',
 119         },
 120         'params': {
 121             # m3u8 download
 122             'skip_download': True,
 123         },
 124         'add_ie': ['UplynkPreplay'],
 125     }, {
 126         'url': 'https://video.vice.com/alps/video/ulfs-wien-beruchtigste-grafitti-crew-part-1/581b12b60a0e1f4c0fb6ea2f',
 127         'info_dict': {
 128             'id': '581b12b60a0e1f4c0fb6ea2f',
 129             'ext': 'mp4',
 130             'title': 'ULFs - Wien berüchtigste Grafitti Crew - Part 1',
 131             'description': '<p>Zwischen Hinterzimmer-Tattoos und U-Bahnschächten erzählen uns die Ulfs, wie es ist, "süchtig nach Sachbeschädigung" zu sein.</p>',
 132             'uploader': 'VICE',
 133             'uploader_id': '57a204088cb727dec794c67b',
 134             'timestamp': 1485368119,
 135             'upload_date': '20170125',
 136             'age_limit': 14,
 137         },
 138         'params': {
 139             # AES-encrypted m3u8
 140             'skip_download': True,
 141         },
 142         'add_ie': ['UplynkPreplay'],
 143     }, {
 144         'url': 'https://video.vice.com/en_us/video/pizza-show-trailer/56d8c9a54d286ed92f7f30e4',
 145         'only_matching': True,
 146     }]
 147     _PREPLAY_HOST = 'video.vice'
 148
 149     def _real_extract(self, url):
 150         locale, video_id = re.match(self._VALID_URL, url).groups()
 151         webpage, urlh = self._download_webpage_handle(url, video_id)
 152         embed_code = self._search_regex(
 153             r'embedCode=([^&\'"]+)', webpage,
 154             'ooyala embed code', default=None)
 155         if embed_code:
 156             return self.url_result('ooyala:%s' % embed_code, 'Ooyala')
 157         youtube_id = self._search_regex(
 158             r'data-youtube-id="([^"]+)"', webpage, 'youtube id', default=None)
 159         if youtube_id:
 160             return self.url_result(youtube_id, 'Youtube')
 161         return self._extract_preplay_video(urlh.geturl(), locale, webpage)
 162
 163
 164 class ViceShowIE(InfoExtractor):
 165     IE_NAME = 'vice:show'
 166     _VALID_URL = r'https?://(?:.+?\.)?vice\.com/(?:[^/]+/)?show/(?P<id>[^/?#&]+)'
 167
 168     _TEST = {
 169         'url': 'https://munchies.vice.com/en/show/fuck-thats-delicious-2',
 170         'info_dict': {
 171             'id': 'fuck-thats-delicious-2',
 172             'title': "Fuck, That's Delicious",
 173             'description': 'Follow the culinary adventures of rapper Action Bronson during his ongoing world tour.',
 174         },
 175         'playlist_count': 17,
 176     }
 177
 178     def _real_extract(self, url):
 179         show_id = self._match_id(url)
 180         webpage = self._download_webpage(url, show_id)
 181
 182         entries = [
 183             self.url_result(video_url, ViceIE.ie_key())
 184             for video_url, _ in re.findall(
 185                 r'<h2[^>]+class="article-title"[^>]+data-id="\d+"[^>]*>\s*<a[^>]+href="(%s.*?)"'
 186                 % ViceIE._VALID_URL, webpage)]
 187
 188         title = self._search_regex(
 189             r'<title>(.+?)</title>', webpage, 'title', default=None)
 190         if title:
 191             title = re.sub(r'(.+)\s*\|\s*.+$', r'\1', title).strip()
 192         description = self._html_search_meta(
 193             'description', webpage, 'description')
 194
 195         return self.playlist_result(entries, show_id, title, description)
 196
 197
 198 class ViceArticleIE(InfoExtractor):
 199     IE_NAME = 'vice:article'
 200     _VALID_URL = r'https://www.vice.com/[^/]+/article/(?P<id>[^?#]+)'
 201
 202     _TESTS = [{
 203         'url': 'https://www.vice.com/en_us/article/on-set-with-the-woman-making-mormon-porn-in-utah',
 204         'info_dict': {
 205             'id': '58dc0a3dee202d2a0ccfcbd8',
 206             'ext': 'mp4',
 207             'title': 'Mormon War on Porn ',
 208             'description': 'md5:ad396a2481e7f8afb5ed486878421090',
 209             'uploader': 'VICE',
 210             'uploader_id': '57a204088cb727dec794c693',
 211             'timestamp': 1489160690,
 212             'upload_date': '20170310',
 213         },
 214         'params': {
 215             # AES-encrypted m3u8
 216             'skip_download': True,
 217         },
 218         'add_ie': ['UplynkPreplay'],
 219     }, {
 220         'url': 'https://www.vice.com/en_us/article/how-to-hack-a-car',
 221         'md5': 'a7ecf64ee4fa19b916c16f4b56184ae2',
 222         'info_dict': {
 223             'id': '3jstaBeXgAs',
 224             'ext': 'mp4',
 225             'title': 'How to Hack a Car: Phreaked Out (Episode 2)',
 226             'description': 'md5:ee95453f7ff495db8efe14ae8bf56f30',
 227             'uploader_id': 'MotherboardTV',
 228             'uploader': 'Motherboard',
 229             'upload_date': '20140529',
 230         },
 231         'add_ie': ['Youtube'],
 232     }, {
 233         'url': 'https://www.vice.com/en_us/article/cowboy-capitalists-part-1',
 234         'only_matching': True,
 235     }, {
 236         'url': 'https://www.vice.com/ru/article/big-night-out-ibiza-clive-martin-229',
 237         'only_matching': True,
 238     }]
 239
 240     def _real_extract(self, url):
 241         display_id = self._match_id(url)
 242
 243         webpage = self._download_webpage(url, display_id)
 244
 245         prefetch_data = self._parse_json(self._search_regex(
 246             r'window\.__PREFETCH_DATA\s*=\s*({.*});',
 247             webpage, 'prefetch data'), display_id)
 248         body = prefetch_data['body']
 249
 250         def _url_res(video_url, ie_key):
 251             return {
 252                 '_type': 'url_transparent',
 253                 'url': video_url,
 254                 'display_id': display_id,
 255                 'ie_key': ie_key,
 256             }
 257
 258         embed_code = self._search_regex(
 259             r'embedCode=([^&\'"]+)', body,
 260             'ooyala embed code', default=None)
 261         if embed_code:
 262             return _url_res('ooyala:%s' % embed_code, 'Ooyala')
 263
 264         youtube_url = self._html_search_regex(
 265             r'<iframe[^>]+src="(.*youtube\.com/.*)"',
 266             body, 'YouTube URL', default=None)
 267         if youtube_url:
 268             return _url_res(youtube_url, 'Youtube')
 269
 270         video_url = self._html_search_regex(
 271             r'data-video-url="([^"]+)"',
 272             prefetch_data['embed_code'], 'video URL')
 273
 274         return _url_res(video_url, ViceIE.ie_key())