_ 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 ..compat import (
   6     compat_HTTPError,
   7     compat_urllib_parse_unquote_plus,
   8 )
   9 from ..utils import (
  10     determine_ext,
  11     int_or_none,
  12     js_to_json,
  13     sanitized_Request,
  14     ExtractorError,
  15     urlencode_postdata
  16 )
  17
  18
  19 class FunimationIE(InfoExtractor):
  20     _VALID_URL = r'https?://(?:www\.)?funimation(?:\.com|now\.uk)/shows/[^/]+/(?P<id>[^/?#&]+)'
  21
  22     _NETRC_MACHINE = 'funimation'
  23     _TOKEN = None
  24
  25     _TESTS = [{
  26         'url': 'https://www.funimation.com/shows/hacksign/role-play/',
  27         'info_dict': {
  28             'id': '91144',
  29             'display_id': 'role-play',
  30             'ext': 'mp4',
  31             'title': '.hack//SIGN - Role Play',
  32             'description': 'md5:b602bdc15eef4c9bbb201bb6e6a4a2dd',
  33             'thumbnail': r're:https?://.*\.jpg',
  34         },
  35         'params': {
  36             # m3u8 download
  37             'skip_download': True,
  38         },
  39     }, {
  40         'url': 'https://www.funimation.com/shows/attack-on-titan-junior-high/broadcast-dub-preview/',
  41         'info_dict': {
  42             'id': '9635',
  43             'display_id': 'broadcast-dub-preview',
  44             'ext': 'mp4',
  45             'title': 'Attack on Titan: Junior High - Broadcast Dub Preview',
  46             'description': 'md5:f8ec49c0aff702a7832cd81b8a44f803',
  47             'thumbnail': r're:https?://.*\.(?:jpg|png)',
  48         },
  49         'skip': 'Access without user interaction is forbidden by CloudFlare',
  50     }, {
  51         'url': 'https://www.funimationnow.uk/shows/puzzle-dragons-x/drop-impact/simulcast/',
  52         'only_matching': True,
  53     }]
  54
  55     _LOGIN_URL = 'http://www.funimation.com/login'
  56
  57     def _extract_cloudflare_session_ua(self, url):
  58         ci_session_cookie = self._get_cookies(url).get('ci_session')
  59         if ci_session_cookie:
  60             ci_session = compat_urllib_parse_unquote_plus(ci_session_cookie.value)
  61             # ci_session is a string serialized by PHP function serialize()
  62             # This case is simple enough to use regular expressions only
  63             return self._search_regex(
  64                 r'"user_agent";s:\d+:"([^"]+)"', ci_session, 'user agent',
  65                 default=None)
  66
  67     def _login(self):
  68         (username, password) = self._get_login_info()
  69         if username is None:
  70             return
  71         try:
  72             data = self._download_json(
  73                 'https://prod-api-funimationnow.dadcdigital.com/api/auth/login/',
  74                 None, 'Logging in as %s' % username, data=urlencode_postdata({
  75                     'username': username,
  76                     'password': password,
  77                 }))
  78             self._TOKEN = data['token']
  79         except ExtractorError as e:
  80             if isinstance(e.cause, compat_HTTPError) and e.cause.code == 401:
  81                 error = self._parse_json(e.cause.read().decode(), None)['error']
  82                 raise ExtractorError(error, expected=True)
  83             raise
  84
  85     def _real_initialize(self):
  86         self._login()
  87
  88     def _real_extract(self, url):
  89         display_id = self._match_id(url)
  90         webpage = self._download_webpage(url, display_id)
  91
  92         def _search_kane(name):
  93             return self._search_regex(
  94                 r"KANE_customdimensions\.%s\s*=\s*'([^']+)';" % name,
  95                 webpage, name, default=None)
  96
  97         title_data = self._parse_json(self._search_regex(
  98             r'TITLE_DATA\s*=\s*({[^}]+})',
  99             webpage, 'title data', default=''),
 100             display_id, js_to_json, fatal=False) or {}
 101
 102         video_id = title_data.get('id') or self._search_regex([
 103             r"KANE_customdimensions.videoID\s*=\s*'(\d+)';",
 104             r'<iframe[^>]+src="/player/(\d+)"',
 105         ], webpage, 'video_id', default=None)
 106         if not video_id:
 107             player_url = self._html_search_meta([
 108                 'al:web:url',
 109                 'og:video:url',
 110                 'og:video:secure_url',
 111             ], webpage, fatal=True)
 112             video_id = self._search_regex(r'/player/(\d+)', player_url, 'video id')
 113
 114         title = episode = title_data.get('title') or _search_kane('videoTitle') or self._og_search_title(webpage)
 115         series = _search_kane('showName')
 116         if series:
 117             title = '%s - %s' % (series, title)
 118         description = self._html_search_meta(['description', 'og:description'], webpage, fatal=True)
 119
 120         try:
 121             headers = {}
 122             if self._TOKEN:
 123                 headers['Authorization'] = 'Token %s' % self._TOKEN
 124             sources = self._download_json(
 125                 'https://prod-api-funimationnow.dadcdigital.com/api/source/catalog/video/%s/signed/' % video_id,
 126                 video_id, headers=headers)['items']
 127         except ExtractorError as e:
 128             if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403:
 129                 error = self._parse_json(e.cause.read(), video_id)['errors'][0]
 130                 raise ExtractorError('%s said: %s' % (
 131                     self.IE_NAME, error.get('detail') or error.get('title')), expected=True)
 132             raise
 133
 134         formats = []
 135         for source in sources:
 136             source_url = source.get('src')
 137             if not source_url:
 138                 continue
 139             source_type = source.get('videoType') or determine_ext(source_url)
 140             if source_type == 'm3u8':
 141                 formats.extend(self._extract_m3u8_formats(
 142                     source_url, video_id, 'mp4',
 143                     m3u8_id='hls', fatal=False))
 144             else:
 145                 formats.append({
 146                     'format_id': source_type,
 147                     'url': source_url,
 148                 })
 149         self._sort_formats(formats)
 150
 151         return {
 152             'id': video_id,
 153             'display_id': display_id,
 154             'title': title,
 155             'description': description,
 156             'thumbnail': self._og_search_thumbnail(webpage),
 157             'series': series,
 158             'season_number': int_or_none(title_data.get('seasonNum') or _search_kane('season')),
 159             'episode_number': int_or_none(title_data.get('episodeNum')),
 160             'episode': episode,
 161             'season_id': title_data.get('seriesId'),
 162             'formats': formats,
 163         }