_ 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': 'AIRENG0001',
  25             'title': 'Air - 1 - Breeze ',
  26             'ext': 'mp4',
  27             'thumbnail': 'http://www.funimation.com/admin/uploads/default/recap_thumbnails/7555590/home_spotlight/AIR0001.jpg',
  28             'description': 'Travelling puppeteer Yukito arrives in a small town where he hopes to earn money through the magic of his puppets. When a young girl named Misuzu lures him to her home with the promise of food, his life changes forever. ',
  29         }
  30     }
  31
  32     def _download_webpage(self, url_or_request, video_id, note='Downloading webpage'):
  33         HEADERS = {
  34             'User-Agent': 'Mozilla/5.0 (Windows NT 5.2; WOW64; rv:42.0) Gecko/20100101 Firefox/42.0',
  35         }
  36         if isinstance(url_or_request, compat_urllib_request.Request):
  37             for header, value in HEADERS.items():
  38                 url_or_request.add_header(header, value)
  39         else:
  40             url_or_request = sanitized_Request(url_or_request, headers=HEADERS)
  41         response = super(FunimationIE, self)._download_webpage(url_or_request, video_id, note)
  42         return response
  43
  44     def _login(self):
  45         (username, password) = self._get_login_info()
  46         if username is None:
  47             return
  48         data = urlencode_postdata(encode_dict({
  49             'email_field': username,
  50             'password_field': password,
  51         }))
  52         login_request = sanitized_Request('http://www.funimation.com/login', data, headers={
  53             'Content-Type': 'application/x-www-form-urlencoded'
  54         })
  55         login = self._download_webpage(
  56             login_request, None, 'Logging in as %s' % username)
  57         if re.search(r'<meta property="og:url" content="http://www.funimation.com/login"/>', login) is not None:
  58             raise ExtractorError('Unable to login, wrong username or password.', expected=True)
  59
  60     def _real_initialize(self):
  61         self._login()
  62
  63     def _real_extract(self, url):
  64         display_id = self._match_id(url)
  65
  66         webpage = self._download_webpage(url, display_id)
  67
  68         items = self._parse_json(
  69             self._search_regex(
  70                 r'var\s+playersData\s*=\s*(\[.+?\]);\n',
  71                 webpage, 'players data'),
  72             display_id)[0]['playlist'][0]['items']
  73
  74         item = next(item for item in items if item.get('itemAK') == display_id)
  75
  76         ERRORS_MAP = {
  77             'ERROR_MATURE_CONTENT_LOGGED_IN': 'matureContentLoggedIn',
  78             'ERROR_MATURE_CONTENT_LOGGED_OUT': 'matureContentLoggedOut',
  79             'ERROR_SUBSCRIPTION_LOGGED_OUT': 'subscriptionLoggedOut',
  80             'ERROR_VIDEO_EXPIRED': 'videoExpired',
  81             'ERROR_TERRITORY_UNAVAILABLE': 'territoryUnavailable',
  82             'SVODBASIC_SUBSCRIPTION_IN_PLAYER': 'basicSubscription',
  83             'SVODNON_SUBSCRIPTION_IN_PLAYER': 'nonSubscription',
  84             'ERROR_PLAYER_NOT_RESPONDING': 'playerNotResponding',
  85             'ERROR_UNABLE_TO_CONNECT_TO_CDN': 'unableToConnectToCDN',
  86             'ERROR_STREAM_NOT_FOUND': 'streamNotFound',
  87         }
  88
  89         error_messages = {}
  90         video_error_messages = self._search_regex(
  91             r'var\s+videoErrorMessages\s*=\s*({.+?});\n',
  92             webpage, 'error messages', default=None)
  93         if video_error_messages:
  94             error_messages_json = self._parse_json(video_error_messages, display_id, fatal=False)
  95             if error_messages_json:
  96                 for _, error in error_messages_json.items():
  97                     type_ = error.get('type')
  98                     description = error.get('description')
  99                     content = error.get('content')
 100                     if type_ == 'text' and description and content:
 101                         error_message = ERRORS_MAP.get(description)
 102                         if error_message:
 103                             error_messages[error_message] = content
 104
 105         errors = []
 106         formats = []
 107         for video in item['videoSet']:
 108             auth_token = video.get('authToken')
 109             if not auth_token:
 110                 continue
 111             funimation_id = video.get('FUNImationID') or video.get('videoId')
 112             if not auth_token.startswith('?'):
 113                 auth_token = '?%s' % auth_token
 114             for quality in ('sd', 'hd', 'hd1080'):
 115                 format_url = video.get('%sUrl' % quality)
 116                 if not format_url:
 117                     continue
 118                 if not format_url.startswith(('http', '//')):
 119                     errors.append(format_url)
 120                     continue
 121                 if determine_ext(format_url) == 'm3u8':
 122                     m3u8_formats = self._extract_m3u8_formats(
 123                         format_url + auth_token, display_id, 'mp4', entry_protocol='m3u8_native',
 124                         m3u8_id=funimation_id or 'hls', fatal=False)
 125                     if m3u8_formats:
 126                         formats.extend(m3u8_formats)
 127                 else:
 128                     formats.append({
 129                         'url': format_url + auth_token,
 130                     })
 131
 132         if not formats and errors:
 133             raise ExtractorError(
 134                 '%s returned error: %s'
 135                 % (self.IE_NAME, clean_html(error_messages.get(errors[0], errors[0]))),
 136                 expected=True)
 137
 138         title = item['title']
 139         artist = item.get('artist')
 140         if artist:
 141             title = '%s - %s' % (artist, title)
 142         description = self._og_search_description(webpage) or item.get('description')
 143         thumbnail = self._og_search_thumbnail(webpage) or item.get('posterUrl')
 144         video_id = item.get('itemId') or display_id
 145
 146         return {
 147             'id': video_id,
 148             'display_id': display_id,
 149             'title': title,
 150             'description': description,
 151             'thumbnail': thumbnail,
 152             'formats': formats,
 153         }