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

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 from .common import InfoExtractor
   5 from ..utils import (
   6     clean_html,
   7     determine_ext,
   8     int_or_none,
   9     sanitized_Request,
  10     ExtractorError,
  11     urlencode_postdata
  12 )
  13
  14
  15 class FunimationIE(InfoExtractor):
  16     _VALID_URL = r'https?://(?:www\.)?funimation\.com/shows/[^/]+/videos/(?:official|promotional)/(?P<id>[^/?#&]+)'
  17
  18     _NETRC_MACHINE = 'funimation'
  19
  20     _TESTS = [{
  21         'url': 'http://www.funimation.com/shows/air/videos/official/breeze',
  22         'info_dict': {
  23             'id': '658',
  24             'display_id': 'breeze',
  25             'ext': 'mp4',
  26             'title': 'Air - 1 - Breeze',
  27             'description': 'md5:1769f43cd5fc130ace8fd87232207892',
  28             'thumbnail': 're:https?://.*\.jpg',
  29         },
  30     }, {
  31         'url': 'http://www.funimation.com/shows/hacksign/videos/official/role-play',
  32         'info_dict': {
  33             'id': '31128',
  34             'display_id': 'role-play',
  35             'ext': 'mp4',
  36             'title': '.hack//SIGN - 1 - Role Play',
  37             'description': 'md5:b602bdc15eef4c9bbb201bb6e6a4a2dd',
  38             'thumbnail': 're:https?://.*\.jpg',
  39         },
  40     }, {
  41         'url': 'http://www.funimation.com/shows/attack-on-titan-junior-high/videos/promotional/broadcast-dub-preview',
  42         'info_dict': {
  43             'id': '9635',
  44             'display_id': 'broadcast-dub-preview',
  45             'ext': 'mp4',
  46             'title': 'Attack on Titan: Junior High - Broadcast Dub Preview',
  47             'description': 'md5:f8ec49c0aff702a7832cd81b8a44f803',
  48             'thumbnail': 're:https?://.*\.(?:jpg|png)',
  49         },
  50     }]
  51
  52     def _login(self):
  53         (username, password) = self._get_login_info()
  54         if username is None:
  55             return
  56         data = urlencode_postdata({
  57             'email_field': username,
  58             'password_field': password,
  59         })
  60         login_request = sanitized_Request('http://www.funimation.com/login', data, headers={
  61             'User-Agent': 'Mozilla/5.0 (Windows NT 5.2; WOW64; rv:42.0) Gecko/20100101 Firefox/42.0',
  62             'Content-Type': 'application/x-www-form-urlencoded'
  63         })
  64         login_page = self._download_webpage(
  65             login_request, None, 'Logging in as %s' % username)
  66         if any(p in login_page for p in ('funimation.com/logout', '>Log Out<')):
  67             return
  68         error = self._html_search_regex(
  69             r'(?s)<div[^>]+id=["\']errorMessages["\'][^>]*>(.+?)</div>',
  70             login_page, 'error messages', default=None)
  71         if error:
  72             raise ExtractorError('Unable to login: %s' % error, expected=True)
  73         raise ExtractorError('Unable to log in')
  74
  75     def _real_initialize(self):
  76         self._login()
  77
  78     def _real_extract(self, url):
  79         display_id = self._match_id(url)
  80
  81         errors = []
  82         formats = []
  83
  84         ERRORS_MAP = {
  85             'ERROR_MATURE_CONTENT_LOGGED_IN': 'matureContentLoggedIn',
  86             'ERROR_MATURE_CONTENT_LOGGED_OUT': 'matureContentLoggedOut',
  87             'ERROR_SUBSCRIPTION_LOGGED_OUT': 'subscriptionLoggedOut',
  88             'ERROR_VIDEO_EXPIRED': 'videoExpired',
  89             'ERROR_TERRITORY_UNAVAILABLE': 'territoryUnavailable',
  90             'SVODBASIC_SUBSCRIPTION_IN_PLAYER': 'basicSubscription',
  91             'SVODNON_SUBSCRIPTION_IN_PLAYER': 'nonSubscription',
  92             'ERROR_PLAYER_NOT_RESPONDING': 'playerNotResponding',
  93             'ERROR_UNABLE_TO_CONNECT_TO_CDN': 'unableToConnectToCDN',
  94             'ERROR_STREAM_NOT_FOUND': 'streamNotFound',
  95         }
  96
  97         USER_AGENTS = (
  98             # PC UA is served with m3u8 that provides some bonus lower quality formats
  99             ('pc', 'Mozilla/5.0 (Windows NT 5.2; WOW64; rv:42.0) Gecko/20100101 Firefox/42.0'),
 100             # Mobile UA allows to extract direct links and also does not fail when
 101             # PC UA fails with hulu error (e.g.
 102             # http://www.funimation.com/shows/hacksign/videos/official/role-play)
 103             ('mobile', 'Mozilla/5.0 (Linux; Android 4.4.2; Nexus 4 Build/KOT49H) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/34.0.1847.114 Mobile Safari/537.36'),
 104         )
 105
 106         for kind, user_agent in USER_AGENTS:
 107             request = sanitized_Request(url)
 108             request.add_header('User-Agent', user_agent)
 109             webpage = self._download_webpage(
 110                 request, display_id, 'Downloading %s webpage' % kind)
 111
 112             playlist = self._parse_json(
 113                 self._search_regex(
 114                     r'var\s+playersData\s*=\s*(\[.+?\]);\n',
 115                     webpage, 'players data'),
 116                 display_id)[0]['playlist']
 117
 118             items = next(item['items'] for item in playlist if item.get('items'))
 119             item = next(item for item in items if item.get('itemAK') == display_id)
 120
 121             error_messages = {}
 122             video_error_messages = self._search_regex(
 123                 r'var\s+videoErrorMessages\s*=\s*({.+?});\n',
 124                 webpage, 'error messages', default=None)
 125             if video_error_messages:
 126                 error_messages_json = self._parse_json(video_error_messages, display_id, fatal=False)
 127                 if error_messages_json:
 128                     for _, error in error_messages_json.items():
 129                         type_ = error.get('type')
 130                         description = error.get('description')
 131                         content = error.get('content')
 132                         if type_ == 'text' and description and content:
 133                             error_message = ERRORS_MAP.get(description)
 134                             if error_message:
 135                                 error_messages[error_message] = content
 136
 137             for video in item.get('videoSet', []):
 138                 auth_token = video.get('authToken')
 139                 if not auth_token:
 140                     continue
 141                 funimation_id = video.get('FUNImationID') or video.get('videoId')
 142                 preference = 1 if video.get('languageMode') == 'dub' else 0
 143                 if not auth_token.startswith('?'):
 144                     auth_token = '?%s' % auth_token
 145                 for quality, height in (('sd', 480), ('hd', 720), ('hd1080', 1080)):
 146                     format_url = video.get('%sUrl' % quality)
 147                     if not format_url:
 148                         continue
 149                     if not format_url.startswith(('http', '//')):
 150                         errors.append(format_url)
 151                         continue
 152                     if determine_ext(format_url) == 'm3u8':
 153                         formats.extend(self._extract_m3u8_formats(
 154                             format_url + auth_token, display_id, 'mp4', entry_protocol='m3u8_native',
 155                             preference=preference, m3u8_id='%s-hls' % funimation_id, fatal=False))
 156                     else:
 157                         tbr = int_or_none(self._search_regex(
 158                             r'-(\d+)[Kk]', format_url, 'tbr', default=None))
 159                         formats.append({
 160                             'url': format_url + auth_token,
 161                             'format_id': '%s-http-%dp' % (funimation_id, height),
 162                             'height': height,
 163                             'tbr': tbr,
 164                             'preference': preference,
 165                         })
 166
 167         if not formats and errors:
 168             raise ExtractorError(
 169                 '%s returned error: %s'
 170                 % (self.IE_NAME, clean_html(error_messages.get(errors[0], errors[0]))),
 171                 expected=True)
 172
 173         self._sort_formats(formats)
 174
 175         title = item['title']
 176         artist = item.get('artist')
 177         if artist:
 178             title = '%s - %s' % (artist, title)
 179         description = self._og_search_description(webpage) or item.get('description')
 180         thumbnail = self._og_search_thumbnail(webpage) or item.get('posterUrl')
 181         video_id = item.get('itemId') or display_id
 182
 183         return {
 184             'id': video_id,
 185             'display_id': display_id,
 186             'title': title,
 187             'description': description,
 188             'thumbnail': thumbnail,
 189             'formats': formats,
 190         }