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