Raphaël G. Git Repositories - youtubedl/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     encode_dict,
   9     int_or_none,
  10     sanitized_Request,
  11     ExtractorError,
  12     urlencode_postdata
  13 )
  14
  15
  16 class FunimationIE(InfoExtractor):
  17     _VALID_URL = r'https?://(?:www\.)?funimation\.com/shows/[^/]+/videos/(?:official|promotional)/(?P<id>[^/?#&]+)'
  18
  19     _NETRC_MACHINE = 'funimation'
  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_page = self._download_webpage(
  66             login_request, None, 'Logging in as %s' % username)
  67         if any(p in login_page for p in ('funimation.com/logout', '>Log Out<')):
  68             return
  69         error = self._html_search_regex(
  70             r'(?s)<div[^>]+id=["\']errorMessages["\'][^>]*>(.+?)</div>',
  71             login_page, 'error messages', default=None)
  72         if error:
  73             raise ExtractorError('Unable to login: %s' % error, expected=True)
  74         raise ExtractorError('Unable to log in')
  75
  76     def _real_initialize(self):
  77         self._login()
  78
  79     def _real_extract(self, url):
  80         display_id = self._match_id(url)
  81
  82         errors = []
  83         formats = []
  84
  85         ERRORS_MAP = {
  86             'ERROR_MATURE_CONTENT_LOGGED_IN': 'matureContentLoggedIn',
  87             'ERROR_MATURE_CONTENT_LOGGED_OUT': 'matureContentLoggedOut',
  88             'ERROR_SUBSCRIPTION_LOGGED_OUT': 'subscriptionLoggedOut',
  89             'ERROR_VIDEO_EXPIRED': 'videoExpired',
  90             'ERROR_TERRITORY_UNAVAILABLE': 'territoryUnavailable',
  91             'SVODBASIC_SUBSCRIPTION_IN_PLAYER': 'basicSubscription',
  92             'SVODNON_SUBSCRIPTION_IN_PLAYER': 'nonSubscription',
  93             'ERROR_PLAYER_NOT_RESPONDING': 'playerNotResponding',
  94             'ERROR_UNABLE_TO_CONNECT_TO_CDN': 'unableToConnectToCDN',
  95             'ERROR_STREAM_NOT_FOUND': 'streamNotFound',
  96         }
  97
  98         USER_AGENTS = (
  99             # PC UA is served with m3u8 that provides some bonus lower quality formats
 100             ('pc', 'Mozilla/5.0 (Windows NT 5.2; WOW64; rv:42.0) Gecko/20100101 Firefox/42.0'),
 101             # Mobile UA allows to extract direct links and also does not fail when
 102             # PC UA fails with hulu error (e.g.
 103             # http://www.funimation.com/shows/hacksign/videos/official/role-play)
 104             ('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'),
 105         )
 106
 107         for kind, user_agent in USER_AGENTS:
 108             request = sanitized_Request(url)
 109             request.add_header('User-Agent', user_agent)
 110             webpage = self._download_webpage(
 111                 request, display_id, 'Downloading %s webpage' % kind)
 112
 113             playlist = self._parse_json(
 114                 self._search_regex(
 115                     r'var\s+playersData\s*=\s*(\[.+?\]);\n',
 116                     webpage, 'players data'),
 117                 display_id)[0]['playlist']
 118
 119             items = next(item['items'] for item in playlist if item.get('items'))
 120             item = next(item for item in items if item.get('itemAK') == display_id)
 121
 122             error_messages = {}
 123             video_error_messages = self._search_regex(
 124                 r'var\s+videoErrorMessages\s*=\s*({.+?});\n',
 125                 webpage, 'error messages', default=None)
 126             if video_error_messages:
 127                 error_messages_json = self._parse_json(video_error_messages, display_id, fatal=False)
 128                 if error_messages_json:
 129                     for _, error in error_messages_json.items():
 130                         type_ = error.get('type')
 131                         description = error.get('description')
 132                         content = error.get('content')
 133                         if type_ == 'text' and description and content:
 134                             error_message = ERRORS_MAP.get(description)
 135                             if error_message:
 136                                 error_messages[error_message] = content
 137
 138             for video in item.get('videoSet', []):
 139                 auth_token = video.get('authToken')
 140                 if not auth_token:
 141                     continue
 142                 funimation_id = video.get('FUNImationID') or video.get('videoId')
 143                 preference = 1 if video.get('languageMode') == 'dub' else 0
 144                 if not auth_token.startswith('?'):
 145                     auth_token = '?%s' % auth_token
 146                 for quality, height in (('sd', 480), ('hd', 720), ('hd1080', 1080)):
 147                     format_url = video.get('%sUrl' % quality)
 148                     if not format_url:
 149                         continue
 150                     if not format_url.startswith(('http', '//')):
 151                         errors.append(format_url)
 152                         continue
 153                     if determine_ext(format_url) == 'm3u8':
 154                         formats.extend(self._extract_m3u8_formats(
 155                             format_url + auth_token, display_id, 'mp4', entry_protocol='m3u8_native',
 156                             preference=preference, m3u8_id='%s-hls' % funimation_id, fatal=False))
 157                     else:
 158                         tbr = int_or_none(self._search_regex(
 159                             r'-(\d+)[Kk]', format_url, 'tbr', default=None))
 160                         formats.append({
 161                             'url': format_url + auth_token,
 162                             'format_id': '%s-http-%dp' % (funimation_id, height),
 163                             'height': height,
 164                             'tbr': tbr,
 165                             'preference': preference,
 166                         })
 167
 168         if not formats and errors:
 169             raise ExtractorError(
 170                 '%s returned error: %s'
 171                 % (self.IE_NAME, clean_html(error_messages.get(errors[0], errors[0]))),
 172                 expected=True)
 173
 174         self._sort_formats(formats)
 175
 176         title = item['title']
 177         artist = item.get('artist')
 178         if artist:
 179             title = '%s - %s' % (artist, title)
 180         description = self._og_search_description(webpage) or item.get('description')
 181         thumbnail = self._og_search_thumbnail(webpage) or item.get('posterUrl')
 182         video_id = item.get('itemId') or display_id
 183
 184         return {
 185             'id': video_id,
 186             'display_id': display_id,
 187             'title': title,
 188             'description': description,
 189             'thumbnail': thumbnail,
 190             'formats': formats,
 191         }