_ 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
  24     _TESTS = [{
  25         'url': 'https://www.funimation.com/shows/hacksign/role-play/',
  26         'info_dict': {
  27             'id': '91144',
  28             'display_id': 'role-play',
  29             'ext': 'mp4',
  30             'title': '.hack//SIGN - Role Play',
  31             'description': 'md5:b602bdc15eef4c9bbb201bb6e6a4a2dd',
  32             'thumbnail': r're:https?://.*\.jpg',
  33         },
  34         'params': {
  35             # m3u8 download
  36             'skip_download': True,
  37         },
  38     }, {
  39         'url': 'https://www.funimation.com/shows/attack-on-titan-junior-high/broadcast-dub-preview/',
  40         'info_dict': {
  41             'id': '9635',
  42             'display_id': 'broadcast-dub-preview',
  43             'ext': 'mp4',
  44             'title': 'Attack on Titan: Junior High - Broadcast Dub Preview',
  45             'description': 'md5:f8ec49c0aff702a7832cd81b8a44f803',
  46             'thumbnail': r're:https?://.*\.(?:jpg|png)',
  47         },
  48         'skip': 'Access without user interaction is forbidden by CloudFlare',
  49     }, {
  50         'url': 'https://www.funimationnow.uk/shows/puzzle-dragons-x/drop-impact/simulcast/',
  51         'only_matching': True,
  52     }]
  53
  54     _LOGIN_URL = 'http://www.funimation.com/login'
  55
  56     def _extract_cloudflare_session_ua(self, url):
  57         ci_session_cookie = self._get_cookies(url).get('ci_session')
  58         if ci_session_cookie:
  59             ci_session = compat_urllib_parse_unquote_plus(ci_session_cookie.value)
  60             # ci_session is a string serialized by PHP function serialize()
  61             # This case is simple enough to use regular expressions only
  62             return self._search_regex(
  63                 r'"user_agent";s:\d+:"([^"]+)"', ci_session, 'user agent',
  64                 default=None)
  65
  66     def _login(self):
  67         (username, password) = self._get_login_info()
  68         if username is None:
  69             return
  70         data = urlencode_postdata({
  71             'email_field': username,
  72             'password_field': password,
  73         })
  74         user_agent = self._extract_cloudflare_session_ua(self._LOGIN_URL)
  75         if not user_agent:
  76             user_agent = 'Mozilla/5.0 (Windows NT 5.2; WOW64; rv:42.0) Gecko/20100101 Firefox/42.0'
  77         login_request = sanitized_Request(self._LOGIN_URL, data, headers={
  78             'User-Agent': user_agent,
  79             'Content-Type': 'application/x-www-form-urlencoded'
  80         })
  81         login_page = self._download_webpage(
  82             login_request, None, 'Logging in as %s' % username)
  83         if any(p in login_page for p in ('funimation.com/logout', '>Log Out<')):
  84             return
  85         error = self._html_search_regex(
  86             r'(?s)<div[^>]+id=["\']errorMessages["\'][^>]*>(.+?)</div>',
  87             login_page, 'error messages', default=None)
  88         if error:
  89             raise ExtractorError('Unable to login: %s' % error, expected=True)
  90         raise ExtractorError('Unable to log in')
  91
  92     def _real_initialize(self):
  93         self._login()
  94
  95     def _real_extract(self, url):
  96         display_id = self._match_id(url)
  97         webpage = self._download_webpage(url, display_id)
  98
  99         def _search_kane(name):
 100             return self._search_regex(
 101                 r"KANE_customdimensions\.%s\s*=\s*'([^']+)';" % name,
 102                 webpage, name, default=None)
 103
 104         title_data = self._parse_json(self._search_regex(
 105             r'TITLE_DATA\s*=\s*({[^}]+})',
 106             webpage, 'title data', default=''),
 107             display_id, js_to_json, fatal=False) or {}
 108
 109         video_id = title_data.get('id') or self._search_regex([
 110             r"KANE_customdimensions.videoID\s*=\s*'(\d+)';",
 111             r'<iframe[^>]+src="/player/(\d+)"',
 112         ], webpage, 'video_id', default=None)
 113         if not video_id:
 114             player_url = self._html_search_meta([
 115                 'al:web:url',
 116                 'og:video:url',
 117                 'og:video:secure_url',
 118             ], webpage, fatal=True)
 119             video_id = self._search_regex(r'/player/(\d+)', player_url, 'video id')
 120
 121         title = episode = title_data.get('title') or _search_kane('videoTitle') or self._og_search_title(webpage)
 122         series = _search_kane('showName')
 123         if series:
 124             title = '%s - %s' % (series, title)
 125         description = self._html_search_meta(['description', 'og:description'], webpage, fatal=True)
 126
 127         try:
 128             sources = self._download_json(
 129                 'https://prod-api-funimationnow.dadcdigital.com/api/source/catalog/video/%s/signed/' % video_id,
 130                 video_id)['items']
 131         except ExtractorError as e:
 132             if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403:
 133                 error = self._parse_json(e.cause.read(), video_id)['errors'][0]
 134                 raise ExtractorError('%s said: %s' % (
 135                     self.IE_NAME, error.get('detail') or error.get('title')), expected=True)
 136             raise
 137
 138         formats = []
 139         for source in sources:
 140             source_url = source.get('src')
 141             if not source_url:
 142                 continue
 143             source_type = source.get('videoType') or determine_ext(source_url)
 144             if source_type == 'm3u8':
 145                 formats.extend(self._extract_m3u8_formats(
 146                     source_url, video_id, 'mp4',
 147                     m3u8_id='hls', fatal=False))
 148             else:
 149                 formats.append({
 150                     'format_id': source_type,
 151                     'url': source_url,
 152                 })
 153         self._sort_formats(formats)
 154
 155         return {
 156             'id': video_id,
 157             'display_id': display_id,
 158             'title': title,
 159             'description': description,
 160             'thumbnail': self._og_search_thumbnail(webpage),
 161             'series': series,
 162             'season_number': int_or_none(title_data.get('seasonNum') or _search_kane('season')),
 163             'episode_number': int_or_none(title_data.get('episodeNum')),
 164             'episode': episode,
 165             'season_id': title_data.get('seriesId'),
 166             'formats': formats,
 167         }