_ Git - youtube-dl/blob - youtube_dl/extractor/vice.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4 import time
   5 import hashlib
   6 import json
   7
   8 from .adobepass import AdobePassIE
   9 from .common import InfoExtractor
  10 from ..compat import compat_HTTPError
  11 from ..utils import (
  12     int_or_none,
  13     parse_age_limit,
  14     str_or_none,
  15     parse_duration,
  16     ExtractorError,
  17     extract_attributes,
  18 )
  19
  20
  21 class ViceBaseIE(AdobePassIE):
  22     def _extract_preplay_video(self, url, webpage):
  23         watch_hub_data = extract_attributes(self._search_regex(
  24             r'(?s)(<watch-hub\s*.+?</watch-hub>)', webpage, 'watch hub'))
  25         video_id = watch_hub_data['vms-id']
  26         title = watch_hub_data['video-title']
  27
  28         query = {}
  29         is_locked = watch_hub_data.get('video-locked') == '1'
  30         if is_locked:
  31             resource = self._get_mvpd_resource(
  32                 'VICELAND', title, video_id,
  33                 watch_hub_data.get('video-rating'))
  34             query['tvetoken'] = self._extract_mvpd_auth(url, video_id, 'VICELAND', resource)
  35
  36         # signature generation algorithm is reverse engineered from signatureGenerator in
  37         # webpack:///../shared/~/vice-player/dist/js/vice-player.js in
  38         # https://www.viceland.com/assets/common/js/web.vendor.bundle.js
  39         exp = int(time.time()) + 14400
  40         query.update({
  41             'exp': exp,
  42             'sign': hashlib.sha512(('%s:GET:%d' % (video_id, exp)).encode()).hexdigest(),
  43         })
  44
  45         try:
  46             host = 'www.viceland' if is_locked else self._PREPLAY_HOST
  47             preplay = self._download_json('https://%s.com/en_us/preplay/%s' % (host, video_id), video_id, query=query)
  48         except ExtractorError as e:
  49             if isinstance(e.cause, compat_HTTPError) and e.cause.code == 400:
  50                 error = json.loads(e.cause.read().decode())
  51                 raise ExtractorError('%s said: %s' % (self.IE_NAME, error['details']), expected=True)
  52             raise
  53
  54         video_data = preplay['video']
  55         base = video_data['base']
  56         uplynk_preplay_url = preplay['preplayURL']
  57         episode = video_data.get('episode', {})
  58         channel = video_data.get('channel', {})
  59
  60         subtitles = {}
  61         cc_url = preplay.get('ccURL')
  62         if cc_url:
  63             subtitles['en'] = [{
  64                 'url': cc_url,
  65             }]
  66
  67         return {
  68             '_type': 'url_transparent',
  69             'url': uplynk_preplay_url,
  70             'id': video_id,
  71             'title': title,
  72             'description': base.get('body'),
  73             'thumbnail': watch_hub_data.get('cover-image') or watch_hub_data.get('thumbnail'),
  74             'duration': parse_duration(video_data.get('video_duration') or watch_hub_data.get('video-duration')),
  75             'timestamp': int_or_none(video_data.get('created_at')),
  76             'age_limit': parse_age_limit(video_data.get('video_rating')),
  77             'series': video_data.get('show_title') or watch_hub_data.get('show-title'),
  78             'episode_number': int_or_none(episode.get('episode_number') or watch_hub_data.get('episode')),
  79             'episode_id': str_or_none(episode.get('id') or video_data.get('episode_id')),
  80             'season_number': int_or_none(watch_hub_data.get('season')),
  81             'season_id': str_or_none(episode.get('season_id')),
  82             'uploader': channel.get('base', {}).get('title') or watch_hub_data.get('channel-title'),
  83             'uploader_id': str_or_none(channel.get('id')),
  84             'subtitles': subtitles,
  85             'ie_key': 'UplynkPreplay',
  86         }
  87
  88
  89 class ViceIE(ViceBaseIE):
  90     _VALID_URL = r'https?://(?:.+?\.)?vice\.com/(?:[^/]+/)?videos?/(?P<id>[^/?#&]+)'
  91
  92     _TESTS = [{
  93         'url': 'http://www.vice.com/video/cowboy-capitalists-part-1',
  94         'md5': 'e9d77741f9e42ba583e683cd170660f7',
  95         'info_dict': {
  96             'id': '43cW1mYzpia9IlestBjVpd23Yu3afAfp',
  97             'ext': 'flv',
  98             'title': 'VICE_COWBOYCAPITALISTS_PART01_v1_VICE_WM_1080p.mov',
  99             'duration': 725.983,
 100         },
 101         'add_ie': ['Ooyala'],
 102     }, {
 103         'url': 'http://www.vice.com/video/how-to-hack-a-car',
 104         'md5': 'a7ecf64ee4fa19b916c16f4b56184ae2',
 105         'info_dict': {
 106             'id': '3jstaBeXgAs',
 107             'ext': 'mp4',
 108             'title': 'How to Hack a Car: Phreaked Out (Episode 2)',
 109             'description': 'md5:ee95453f7ff495db8efe14ae8bf56f30',
 110             'uploader_id': 'MotherboardTV',
 111             'uploader': 'Motherboard',
 112             'upload_date': '20140529',
 113         },
 114         'add_ie': ['Youtube'],
 115     }, {
 116         'url': 'https://video.vice.com/en_us/video/the-signal-from-tolva/5816510690b70e6c5fd39a56',
 117         'md5': '',
 118         'info_dict': {
 119             'id': '5816510690b70e6c5fd39a56',
 120             'ext': 'mp4',
 121             'uploader': 'Waypoint',
 122             'title': 'The Signal From Tölva',
 123             'uploader_id': '57f7d621e05ca860fa9ccaf9',
 124             'timestamp': 1477941983938,
 125         },
 126         'params': {
 127             # m3u8 download
 128             'skip_download': True,
 129         },
 130         'add_ie': ['UplynkPreplay'],
 131     }, {
 132         'url': 'https://news.vice.com/video/experimenting-on-animals-inside-the-monkey-lab',
 133         'only_matching': True,
 134     }, {
 135         'url': 'http://www.vice.com/ru/video/big-night-out-ibiza-clive-martin-229',
 136         'only_matching': True,
 137     }, {
 138         'url': 'https://munchies.vice.com/en/videos/watch-the-trailer-for-our-new-series-the-pizza-show',
 139         'only_matching': True,
 140     }]
 141     _PREPLAY_HOST = 'video.vice'
 142
 143     def _real_extract(self, url):
 144         video_id = self._match_id(url)
 145         webpage, urlh = self._download_webpage_handle(url, video_id)
 146         embed_code = self._search_regex(
 147             r'embedCode=([^&\'"]+)', webpage,
 148             'ooyala embed code', default=None)
 149         if embed_code:
 150             return self.url_result('ooyala:%s' % embed_code, 'Ooyala')
 151         youtube_id = self._search_regex(
 152             r'data-youtube-id="([^"]+)"', webpage, 'youtube id', default=None)
 153         if youtube_id:
 154             return self.url_result(youtube_id, 'Youtube')
 155         return self._extract_preplay_video(urlh.geturl(), webpage)
 156
 157
 158 class ViceShowIE(InfoExtractor):
 159     _VALID_URL = r'https?://(?:.+?\.)?vice\.com/(?:[^/]+/)?show/(?P<id>[^/?#&]+)'
 160
 161     _TEST = {
 162         'url': 'https://munchies.vice.com/en/show/fuck-thats-delicious-2',
 163         'info_dict': {
 164             'id': 'fuck-thats-delicious-2',
 165             'title': "Fuck, That's Delicious",
 166             'description': 'Follow the culinary adventures of rapper Action Bronson during his ongoing world tour.',
 167         },
 168         'playlist_count': 17,
 169     }
 170
 171     def _real_extract(self, url):
 172         show_id = self._match_id(url)
 173         webpage = self._download_webpage(url, show_id)
 174
 175         entries = [
 176             self.url_result(video_url, ViceIE.ie_key())
 177             for video_url, _ in re.findall(
 178                 r'<h2[^>]+class="article-title"[^>]+data-id="\d+"[^>]*>\s*<a[^>]+href="(%s.*?)"'
 179                 % ViceIE._VALID_URL, webpage)]
 180
 181         title = self._search_regex(
 182             r'<title>(.+?)</title>', webpage, 'title', default=None)
 183         if title:
 184             title = re.sub(r'(.+)\s*\|\s*.+$', r'\1', title).strip()
 185         description = self._html_search_meta('description', webpage, 'description')
 186
 187         return self.playlist_result(entries, show_id, title, description)