_ 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 ..utils import (
   8     clean_html,
   9     determine_ext,
  10     encode_dict,
  11     int_or_none,
  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|promotional)/(?P<id>[^/?#&]+)'
  20
  21     _TESTS = [{
  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         'url': 'http://www.funimation.com/shows/hacksign/videos/official/role-play',
  33         'info_dict': {
  34             'id': '31128',
  35             'display_id': 'role-play',
  36             'ext': 'mp4',
  37             'title': '.hack//SIGN - 1 - Role Play',
  38             'description': 'md5:b602bdc15eef4c9bbb201bb6e6a4a2dd',
  39             'thumbnail': 're:https?://.*\.jpg',
  40         },
  41     }, {
  42         'url': 'http://www.funimation.com/shows/attack-on-titan-junior-high/videos/promotional/broadcast-dub-preview',
  43         'info_dict': {
  44             'id': '9635',
  45             'display_id': 'broadcast-dub-preview',
  46             'ext': 'mp4',
  47             'title': 'Attack on Titan: Junior High - Broadcast Dub Preview',
  48             'description': 'md5:f8ec49c0aff702a7832cd81b8a44f803',
  49             'thumbnail': 're:https?://.*\.(?:jpg|png)',
  50         },
  51     }]
  52
  53     def _login(self):
  54         (username, password) = self._get_login_info()
  55         if username is None:
  56             return
  57         data = urlencode_postdata(encode_dict({
  58             'email_field': username,
  59             'password_field': password,
  60         }))
  61         login_request = sanitized_Request('http://www.funimation.com/login', data, headers={
  62             'User-Agent': 'Mozilla/5.0 (Windows NT 5.2; WOW64; rv:42.0) Gecko/20100101 Firefox/42.0',
  63             'Content-Type': 'application/x-www-form-urlencoded'
  64         })
  65         login = self._download_webpage(
  66             login_request, None, 'Logging in as %s' % username)
  67         if re.search(r'<meta property="og:url" content="http://www.funimation.com/login"/>', login) is not None:
  68             raise ExtractorError('Unable to login, wrong username or password.', expected=True)
  69
  70     def _real_initialize(self):
  71         self._login()
  72
  73     def _real_extract(self, url):
  74         display_id = self._match_id(url)
  75
  76         errors = []
  77         formats = []
  78
  79         ERRORS_MAP = {
  80             'ERROR_MATURE_CONTENT_LOGGED_IN': 'matureContentLoggedIn',
  81             'ERROR_MATURE_CONTENT_LOGGED_OUT': 'matureContentLoggedOut',
  82             'ERROR_SUBSCRIPTION_LOGGED_OUT': 'subscriptionLoggedOut',
  83             'ERROR_VIDEO_EXPIRED': 'videoExpired',
  84             'ERROR_TERRITORY_UNAVAILABLE': 'territoryUnavailable',
  85             'SVODBASIC_SUBSCRIPTION_IN_PLAYER': 'basicSubscription',
  86             'SVODNON_SUBSCRIPTION_IN_PLAYER': 'nonSubscription',
  87             'ERROR_PLAYER_NOT_RESPONDING': 'playerNotResponding',
  88             'ERROR_UNABLE_TO_CONNECT_TO_CDN': 'unableToConnectToCDN',
  89             'ERROR_STREAM_NOT_FOUND': 'streamNotFound',
  90         }
  91
  92         USER_AGENTS = (
  93             # PC UA is served with m3u8 that provides some bonus lower quality formats
  94             ('pc', 'Mozilla/5.0 (Windows NT 5.2; WOW64; rv:42.0) Gecko/20100101 Firefox/42.0'),
  95             # Mobile UA allows to extract direct links and also does not fail when
  96             # PC UA fails with hulu error (e.g.
  97             # http://www.funimation.com/shows/hacksign/videos/official/role-play)
  98             ('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'),
  99         )
 100
 101         for kind, user_agent in USER_AGENTS:
 102             request = sanitized_Request(url)
 103             request.add_header('User-Agent', user_agent)
 104             webpage = self._download_webpage(
 105                 request, display_id, 'Downloading %s webpage' % kind)
 106
 107             playlist = self._parse_json(
 108                 self._search_regex(
 109                     r'var\s+playersData\s*=\s*(\[.+?\]);\n',
 110                     webpage, 'players data'),
 111                 display_id)[0]['playlist']
 112
 113             items = next(item['items'] for item in playlist if item.get('items'))
 114             item = next(item for item in items if item.get('itemAK') == display_id)
 115
 116             error_messages = {}
 117             video_error_messages = self._search_regex(
 118                 r'var\s+videoErrorMessages\s*=\s*({.+?});\n',
 119                 webpage, 'error messages', default=None)
 120             if video_error_messages:
 121                 error_messages_json = self._parse_json(video_error_messages, display_id, fatal=False)
 122                 if error_messages_json:
 123                     for _, error in error_messages_json.items():
 124                         type_ = error.get('type')
 125                         description = error.get('description')
 126                         content = error.get('content')
 127                         if type_ == 'text' and description and content:
 128                             error_message = ERRORS_MAP.get(description)
 129                             if error_message:
 130                                 error_messages[error_message] = content
 131
 132             for video in item.get('videoSet', []):
 133                 auth_token = video.get('authToken')
 134                 if not auth_token:
 135                     continue
 136                 funimation_id = video.get('FUNImationID') or video.get('videoId')
 137                 preference = 1 if video.get('languageMode') == 'dub' else 0
 138                 if not auth_token.startswith('?'):
 139                     auth_token = '?%s' % auth_token
 140                 for quality, height in (('sd', 480), ('hd', 720), ('hd1080', 1080)):
 141                     format_url = video.get('%sUrl' % quality)
 142                     if not format_url:
 143                         continue
 144                     if not format_url.startswith(('http', '//')):
 145                         errors.append(format_url)
 146                         continue
 147                     if determine_ext(format_url) == 'm3u8':
 148                         m3u8_formats = self._extract_m3u8_formats(
 149                             format_url + auth_token, display_id, 'mp4', entry_protocol='m3u8_native',
 150                             preference=preference, m3u8_id='%s-hls' % funimation_id, fatal=False)
 151                         if m3u8_formats:
 152                             formats.extend(m3u8_formats)
 153                     else:
 154                         tbr = int_or_none(self._search_regex(
 155                             r'-(\d+)[Kk]', format_url, 'tbr', default=None))
 156                         formats.append({
 157                             'url': format_url + auth_token,
 158                             'format_id': '%s-http-%dp' % (funimation_id, height),
 159                             'height': height,
 160                             'tbr': tbr,
 161                             'preference': preference,
 162                         })
 163
 164         if not formats and errors:
 165             raise ExtractorError(
 166                 '%s returned error: %s'
 167                 % (self.IE_NAME, clean_html(error_messages.get(errors[0], errors[0]))),
 168                 expected=True)
 169
 170         self._sort_formats(formats)
 171
 172         title = item['title']
 173         artist = item.get('artist')
 174         if artist:
 175             title = '%s - %s' % (artist, title)
 176         description = self._og_search_description(webpage) or item.get('description')
 177         thumbnail = self._og_search_thumbnail(webpage) or item.get('posterUrl')
 178         video_id = item.get('itemId') or display_id
 179
 180         return {
 181             'id': video_id,
 182             'display_id': display_id,
 183             'title': title,
 184             'description': description,
 185             'thumbnail': thumbnail,
 186             'formats': formats,
 187         }