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

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..compat import compat_urllib_request
   8 from ..utils import (
   9     clean_html,
  10     determine_ext,
  11     encode_dict,
  12     sanitized_Request,
  13     ExtractorError,
  14     urlencode_postdata
  15 )
  16
  17
  18 class FunimationIE(InfoExtractor):
  19     _VALID_URL = r'https?://(?:www\.)?funimation\.com/shows/[^/]+/videos/official/(?P<id>[^?]+)'
  20
  21     _TEST = {
  22         'url': 'http://www.funimation.com/shows/air/videos/official/breeze',
  23         'info_dict': {
  24             'id': '658',
  25             'display_id': 'breeze',
  26             'ext': 'mp4',
  27             'title': 'Air - 1 - Breeze',
  28             'description': 'md5:1769f43cd5fc130ace8fd87232207892',
  29             'thumbnail': 're:https?://.*\.jpg',
  30         },
  31     }
  32
  33     def _download_webpage(self, url_or_request, video_id, note='Downloading webpage'):
  34         HEADERS = {
  35             'User-Agent': 'Mozilla/5.0 (Windows NT 5.2; WOW64; rv:42.0) Gecko/20100101 Firefox/42.0',
  36         }
  37         if isinstance(url_or_request, compat_urllib_request.Request):
  38             for header, value in HEADERS.items():
  39                 url_or_request.add_header(header, value)
  40         else:
  41             url_or_request = sanitized_Request(url_or_request, headers=HEADERS)
  42         response = super(FunimationIE, self)._download_webpage(url_or_request, video_id, note)
  43         return response
  44
  45     def _login(self):
  46         (username, password) = self._get_login_info()
  47         if username is None:
  48             return
  49         data = urlencode_postdata(encode_dict({
  50             'email_field': username,
  51             'password_field': password,
  52         }))
  53         login_request = sanitized_Request('http://www.funimation.com/login', data, headers={
  54             'Content-Type': 'application/x-www-form-urlencoded'
  55         })
  56         login = self._download_webpage(
  57             login_request, None, 'Logging in as %s' % username)
  58         if re.search(r'<meta property="og:url" content="http://www.funimation.com/login"/>', login) is not None:
  59             raise ExtractorError('Unable to login, wrong username or password.', expected=True)
  60
  61     def _real_initialize(self):
  62         self._login()
  63
  64     def _real_extract(self, url):
  65         display_id = self._match_id(url)
  66
  67         webpage = self._download_webpage(url, display_id)
  68
  69         items = self._parse_json(
  70             self._search_regex(
  71                 r'var\s+playersData\s*=\s*(\[.+?\]);\n',
  72                 webpage, 'players data'),
  73             display_id)[0]['playlist'][0]['items']
  74
  75         item = next(item for item in items if item.get('itemAK') == display_id)
  76
  77         ERRORS_MAP = {
  78             'ERROR_MATURE_CONTENT_LOGGED_IN': 'matureContentLoggedIn',
  79             'ERROR_MATURE_CONTENT_LOGGED_OUT': 'matureContentLoggedOut',
  80             'ERROR_SUBSCRIPTION_LOGGED_OUT': 'subscriptionLoggedOut',
  81             'ERROR_VIDEO_EXPIRED': 'videoExpired',
  82             'ERROR_TERRITORY_UNAVAILABLE': 'territoryUnavailable',
  83             'SVODBASIC_SUBSCRIPTION_IN_PLAYER': 'basicSubscription',
  84             'SVODNON_SUBSCRIPTION_IN_PLAYER': 'nonSubscription',
  85             'ERROR_PLAYER_NOT_RESPONDING': 'playerNotResponding',
  86             'ERROR_UNABLE_TO_CONNECT_TO_CDN': 'unableToConnectToCDN',
  87             'ERROR_STREAM_NOT_FOUND': 'streamNotFound',
  88         }
  89
  90         error_messages = {}
  91         video_error_messages = self._search_regex(
  92             r'var\s+videoErrorMessages\s*=\s*({.+?});\n',
  93             webpage, 'error messages', default=None)
  94         if video_error_messages:
  95             error_messages_json = self._parse_json(video_error_messages, display_id, fatal=False)
  96             if error_messages_json:
  97                 for _, error in error_messages_json.items():
  98                     type_ = error.get('type')
  99                     description = error.get('description')
 100                     content = error.get('content')
 101                     if type_ == 'text' and description and content:
 102                         error_message = ERRORS_MAP.get(description)
 103                         if error_message:
 104                             error_messages[error_message] = content
 105
 106         errors = []
 107         formats = []
 108         for video in item['videoSet']:
 109             auth_token = video.get('authToken')
 110             if not auth_token:
 111                 continue
 112             funimation_id = video.get('FUNImationID') or video.get('videoId')
 113             if not auth_token.startswith('?'):
 114                 auth_token = '?%s' % auth_token
 115             for quality in ('sd', 'hd', 'hd1080'):
 116                 format_url = video.get('%sUrl' % quality)
 117                 if not format_url:
 118                     continue
 119                 if not format_url.startswith(('http', '//')):
 120                     errors.append(format_url)
 121                     continue
 122                 if determine_ext(format_url) == 'm3u8':
 123                     m3u8_formats = self._extract_m3u8_formats(
 124                         format_url + auth_token, display_id, 'mp4', entry_protocol='m3u8_native',
 125                         m3u8_id=funimation_id or 'hls', fatal=False)
 126                     if m3u8_formats:
 127                         formats.extend(m3u8_formats)
 128                 else:
 129                     formats.append({
 130                         'url': format_url + auth_token,
 131                     })
 132
 133         if not formats and errors:
 134             raise ExtractorError(
 135                 '%s returned error: %s'
 136                 % (self.IE_NAME, clean_html(error_messages.get(errors[0], errors[0]))),
 137                 expected=True)
 138
 139         title = item['title']
 140         artist = item.get('artist')
 141         if artist:
 142             title = '%s - %s' % (artist, title)
 143         description = self._og_search_description(webpage) or item.get('description')
 144         thumbnail = self._og_search_thumbnail(webpage) or item.get('posterUrl')
 145         video_id = item.get('itemId') or display_id
 146
 147         return {
 148             'id': video_id,
 149             'display_id': display_id,
 150             'title': title,
 151             'description': description,
 152             'thumbnail': thumbnail,
 153             'formats': formats,
 154         }