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

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import json
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     ExtractorError,
   9     parse_iso8601,
  10     sanitized_Request,
  11 )
  12
  13
  14 class VesselIE(InfoExtractor):
  15     _VALID_URL = r'https?://(?:www\.)?vessel\.com/videos/(?P<id>[0-9a-zA-Z]+)'
  16     _API_URL_TEMPLATE = 'https://www.vessel.com/api/view/items/%s'
  17     _LOGIN_URL = 'https://www.vessel.com/api/account/login'
  18     _NETRC_MACHINE = 'vessel'
  19     _TEST = {
  20         'url': 'https://www.vessel.com/videos/HDN7G5UMs',
  21         'md5': '455cdf8beb71c6dd797fd2f3818d05c4',
  22         'info_dict': {
  23             'id': 'HDN7G5UMs',
  24             'ext': 'mp4',
  25             'title': 'Nvidia GeForce GTX Titan X - The Best Video Card on the Market?',
  26             'thumbnail': 're:^https?://.*\.jpg$',
  27             'upload_date': '20150317',
  28             'description': 'Did Nvidia pull out all the stops on the Titan X, or does its performance leave something to be desired?',
  29             'timestamp': int,
  30         },
  31     }
  32
  33     @staticmethod
  34     def make_json_request(url, data):
  35         payload = json.dumps(data).encode('utf-8')
  36         req = sanitized_Request(url, payload)
  37         req.add_header('Content-Type', 'application/json; charset=utf-8')
  38         return req
  39
  40     @staticmethod
  41     def find_assets(data, asset_type, asset_id=None):
  42         for asset in data.get('assets', []):
  43             if not asset.get('type') == asset_type:
  44                 continue
  45             elif asset_id is not None and not asset.get('id') == asset_id:
  46                 continue
  47             else:
  48                 yield asset
  49
  50     def _check_access_rights(self, data):
  51         access_info = data.get('__view', {})
  52         if not access_info.get('allow_access', True):
  53             err_code = access_info.get('error_code') or ''
  54             if err_code == 'ITEM_PAID_ONLY':
  55                 raise ExtractorError(
  56                     'This video requires subscription.', expected=True)
  57             else:
  58                 raise ExtractorError(
  59                     'Access to this content is restricted. (%s said: %s)' % (self.IE_NAME, err_code), expected=True)
  60
  61     def _login(self):
  62         (username, password) = self._get_login_info()
  63         if username is None:
  64             return
  65         self.report_login()
  66         data = {
  67             'client_id': 'web',
  68             'type': 'password',
  69             'user_key': username,
  70             'password': password,
  71         }
  72         login_request = VesselIE.make_json_request(self._LOGIN_URL, data)
  73         self._download_webpage(login_request, None, False, 'Wrong login info')
  74
  75     def _real_initialize(self):
  76         self._login()
  77
  78     def _real_extract(self, url):
  79         video_id = self._match_id(url)
  80
  81         webpage = self._download_webpage(url, video_id)
  82         data = self._parse_json(self._search_regex(
  83             r'App\.bootstrapData\((.*?)\);', webpage, 'data'), video_id)
  84         asset_id = data['model']['data']['id']
  85
  86         req = VesselIE.make_json_request(
  87             self._API_URL_TEMPLATE % asset_id, {'client': 'web'})
  88         data = self._download_json(req, video_id)
  89         video_asset_id = data.get('main_video_asset')
  90
  91         self._check_access_rights(data)
  92
  93         try:
  94             video_asset = next(
  95                 VesselIE.find_assets(data, 'video', asset_id=video_asset_id))
  96         except StopIteration:
  97             raise ExtractorError('No video assets found')
  98
  99         formats = []
 100         for f in video_asset.get('sources', []):
 101             if f['name'] == 'hls-index':
 102                 formats.extend(self._extract_m3u8_formats(
 103                     f['location'], video_id, ext='mp4', m3u8_id='m3u8'))
 104             else:
 105                 formats.append({
 106                     'format_id': f['name'],
 107                     'tbr': f.get('bitrate'),
 108                     'height': f.get('height'),
 109                     'width': f.get('width'),
 110                     'url': f['location'],
 111                 })
 112         self._sort_formats(formats)
 113
 114         thumbnails = []
 115         for im_asset in VesselIE.find_assets(data, 'image'):
 116             thumbnails.append({
 117                 'url': im_asset['location'],
 118                 'width': im_asset.get('width', 0),
 119                 'height': im_asset.get('height', 0),
 120             })
 121
 122         return {
 123             'id': video_id,
 124             'title': data['title'],
 125             'formats': formats,
 126             'thumbnails': thumbnails,
 127             'description': data.get('short_description'),
 128             'duration': data.get('duration'),
 129             'comment_count': data.get('comment_count'),
 130             'like_count': data.get('like_count'),
 131             'view_count': data.get('view_count'),
 132             'timestamp': parse_iso8601(data.get('released_at')),
 133         }