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, 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(url, video_id, 'VICELAND', resource)
  36
  37         # signature generation algorithm is reverse engineered from signatureGenerator in
  38         # webpack:///../shared/~/vice-player/dist/js/vice-player.js in
  39         # https://www.viceland.com/assets/common/js/web.vendor.bundle.js
  40         exp = int(time.time()) + 14400
  41         query.update({
  42             'exp': exp,
  43             'sign': hashlib.sha512(('%s:GET:%d' % (video_id, exp)).encode()).hexdigest(),
  44         })
  45
  46         try:
  47             host = 'www.viceland' if is_locked else self._PREPLAY_HOST
  48             preplay = self._download_json('https://%s.com/en_us/preplay/%s' % (host, video_id), video_id, query=query)
  49         except ExtractorError as e:
  50             if isinstance(e.cause, compat_HTTPError) and e.cause.code == 400:
  51                 error = json.loads(e.cause.read().decode())
  52                 raise ExtractorError('%s said: %s' % (self.IE_NAME, error['details']), expected=True)
  53             raise
  54
  55         video_data = preplay['video']
  56         base = video_data['base']
  57         uplynk_preplay_url = preplay['preplayURL']
  58         episode = video_data.get('episode', {})
  59         channel = video_data.get('channel', {})
  60
  61         subtitles = {}
  62         cc_url = preplay.get('ccURL')
  63         if cc_url:
  64             subtitles['en'] = [{
  65                 'url': cc_url,
  66             }]
  67
  68         return {
  69             '_type': 'url_transparent',
  70             'url': uplynk_preplay_url,
  71             'id': video_id,
  72             'title': title,
  73             'description': base.get('body') or base.get('display_body'),
  74             'thumbnail': watch_hub_data.get('cover-image') or watch_hub_data.get('thumbnail'),
  75             'duration': int_or_none(video_data.get('video_duration')) or parse_duration(watch_hub_data.get('video-duration')),
  76             'timestamp': int_or_none(video_data.get('created_at'), 1000),
  77             'age_limit': parse_age_limit(video_data.get('video_rating')),
  78             'series': video_data.get('show_title') or watch_hub_data.get('show-title'),
  79             'episode_number': int_or_none(episode.get('episode_number') or watch_hub_data.get('episode')),
  80             'episode_id': str_or_none(episode.get('id') or video_data.get('episode_id')),
  81             'season_number': int_or_none(watch_hub_data.get('season')),
  82             'season_id': str_or_none(episode.get('season_id')),
  83             'uploader': channel.get('base', {}).get('title') or watch_hub_data.get('channel-title'),
  84             'uploader_id': str_or_none(channel.get('id')),
  85             'subtitles': subtitles,
  86             'ie_key': 'UplynkPreplay',
  87         }
  88
  89
  90 class ViceIE(ViceBaseIE):
  91     _VALID_URL = r'https?://(?:.+?\.)?vice\.com/(?:[^/]+/)?videos?/(?P<id>[^/?#&]+)'
  92
  93     _TESTS = [{
  94         'url': 'http://www.vice.com/video/cowboy-capitalists-part-1',
  95         'md5': 'e9d77741f9e42ba583e683cd170660f7',
  96         'info_dict': {
  97             'id': '43cW1mYzpia9IlestBjVpd23Yu3afAfp',
  98             'ext': 'flv',
  99             'title': 'VICE_COWBOYCAPITALISTS_PART01_v1_VICE_WM_1080p.mov',
 100             'duration': 725.983,
 101         },
 102         'add_ie': ['Ooyala'],
 103     }, {
 104         'url': 'http://www.vice.com/video/how-to-hack-a-car',
 105         'md5': 'a7ecf64ee4fa19b916c16f4b56184ae2',
 106         'info_dict': {
 107             'id': '3jstaBeXgAs',
 108             'ext': 'mp4',
 109             'title': 'How to Hack a Car: Phreaked Out (Episode 2)',
 110             'description': 'md5:ee95453f7ff495db8efe14ae8bf56f30',
 111             'uploader_id': 'MotherboardTV',
 112             'uploader': 'Motherboard',
 113             'upload_date': '20140529',
 114         },
 115         'add_ie': ['Youtube'],
 116     }, {
 117         'url': 'https://video.vice.com/en_us/video/the-signal-from-tolva/5816510690b70e6c5fd39a56',
 118         'md5': '',
 119         'info_dict': {
 120             'id': '5816510690b70e6c5fd39a56',
 121             'ext': 'mp4',
 122             'uploader': 'Waypoint',
 123             'title': 'The Signal From Tölva',
 124             'uploader_id': '57f7d621e05ca860fa9ccaf9',
 125             'timestamp': 1477941983938,
 126         },
 127         'params': {
 128             # m3u8 download
 129             'skip_download': True,
 130         },
 131         'add_ie': ['UplynkPreplay'],
 132     }, {
 133         'url': 'https://news.vice.com/video/experimenting-on-animals-inside-the-monkey-lab',
 134         'only_matching': True,
 135     }, {
 136         'url': 'http://www.vice.com/ru/video/big-night-out-ibiza-clive-martin-229',
 137         'only_matching': True,
 138     }, {
 139         'url': 'https://munchies.vice.com/en/videos/watch-the-trailer-for-our-new-series-the-pizza-show',
 140         'only_matching': True,
 141     }]
 142     _PREPLAY_HOST = 'video.vice'
 143
 144     def _real_extract(self, url):
 145         video_id = self._match_id(url)
 146         webpage, urlh = self._download_webpage_handle(url, video_id)
 147         embed_code = self._search_regex(
 148             r'embedCode=([^&\'"]+)', webpage,
 149             'ooyala embed code', default=None)
 150         if embed_code:
 151             return self.url_result('ooyala:%s' % embed_code, 'Ooyala')
 152         youtube_id = self._search_regex(
 153             r'data-youtube-id="([^"]+)"', webpage, 'youtube id', default=None)
 154         if youtube_id:
 155             return self.url_result(youtube_id, 'Youtube')
 156         return self._extract_preplay_video(urlh.geturl(), webpage)
 157
 158
 159 class ViceShowIE(InfoExtractor):
 160     _VALID_URL = r'https?://(?:.+?\.)?vice\.com/(?:[^/]+/)?show/(?P<id>[^/?#&]+)'
 161
 162     _TEST = {
 163         'url': 'https://munchies.vice.com/en/show/fuck-thats-delicious-2',
 164         'info_dict': {
 165             'id': 'fuck-thats-delicious-2',
 166             'title': "Fuck, That's Delicious",
 167             'description': 'Follow the culinary adventures of rapper Action Bronson during his ongoing world tour.',
 168         },
 169         'playlist_count': 17,
 170     }
 171
 172     def _real_extract(self, url):
 173         show_id = self._match_id(url)
 174         webpage = self._download_webpage(url, show_id)
 175
 176         entries = [
 177             self.url_result(video_url, ViceIE.ie_key())
 178             for video_url, _ in re.findall(
 179                 r'<h2[^>]+class="article-title"[^>]+data-id="\d+"[^>]*>\s*<a[^>]+href="(%s.*?)"'
 180                 % ViceIE._VALID_URL, webpage)]
 181
 182         title = self._search_regex(
 183             r'<title>(.+?)</title>', webpage, 'title', default=None)
 184         if title:
 185             title = re.sub(r'(.+)\s*\|\s*.+$', r'\1', title).strip()
 186         description = self._html_search_meta('description', webpage, 'description')
 187
 188         return self.playlist_result(entries, show_id, title, description)