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 ..compat import compat_HTTPError
   6 from ..utils import (
   7     determine_ext,
   8     int_or_none,
   9     js_to_json,
  10     ExtractorError,
  11     urlencode_postdata
  12 )
  13
  14
  15 class FunimationIE(InfoExtractor):
  16     _VALID_URL = r'https?://(?:www\.)?funimation(?:\.com|now\.uk)/shows/[^/]+/(?P<id>[^/?#&]+)'
  17
  18     _NETRC_MACHINE = 'funimation'
  19     _TOKEN = None
  20
  21     _TESTS = [{
  22         'url': 'https://www.funimation.com/shows/hacksign/role-play/',
  23         'info_dict': {
  24             'id': '91144',
  25             'display_id': 'role-play',
  26             'ext': 'mp4',
  27             'title': '.hack//SIGN - Role Play',
  28             'description': 'md5:b602bdc15eef4c9bbb201bb6e6a4a2dd',
  29             'thumbnail': r're:https?://.*\.jpg',
  30         },
  31         'params': {
  32             # m3u8 download
  33             'skip_download': True,
  34         },
  35     }, {
  36         'url': 'https://www.funimation.com/shows/attack-on-titan-junior-high/broadcast-dub-preview/',
  37         'info_dict': {
  38             'id': '210051',
  39             'display_id': 'broadcast-dub-preview',
  40             'ext': 'mp4',
  41             'title': 'Attack on Titan: Junior High - Broadcast Dub Preview',
  42             'thumbnail': r're:https?://.*\.(?:jpg|png)',
  43         },
  44         'params': {
  45             # m3u8 download
  46             'skip_download': True,
  47         },
  48     }, {
  49         'url': 'https://www.funimationnow.uk/shows/puzzle-dragons-x/drop-impact/simulcast/',
  50         'only_matching': True,
  51     }]
  52
  53     def _login(self):
  54         username, password = self._get_login_info()
  55         if username is None:
  56             return
  57         try:
  58             data = self._download_json(
  59                 'https://prod-api-funimationnow.dadcdigital.com/api/auth/login/',
  60                 None, 'Logging in', data=urlencode_postdata({
  61                     'username': username,
  62                     'password': password,
  63                 }))
  64             self._TOKEN = data['token']
  65         except ExtractorError as e:
  66             if isinstance(e.cause, compat_HTTPError) and e.cause.code == 401:
  67                 error = self._parse_json(e.cause.read().decode(), None)['error']
  68                 raise ExtractorError(error, expected=True)
  69             raise
  70
  71     def _real_initialize(self):
  72         self._login()
  73
  74     def _real_extract(self, url):
  75         display_id = self._match_id(url)
  76         webpage = self._download_webpage(url, display_id)
  77
  78         def _search_kane(name):
  79             return self._search_regex(
  80                 r"KANE_customdimensions\.%s\s*=\s*'([^']+)';" % name,
  81                 webpage, name, default=None)
  82
  83         title_data = self._parse_json(self._search_regex(
  84             r'TITLE_DATA\s*=\s*({[^}]+})',
  85             webpage, 'title data', default=''),
  86             display_id, js_to_json, fatal=False) or {}
  87
  88         video_id = title_data.get('id') or self._search_regex([
  89             r"KANE_customdimensions.videoID\s*=\s*'(\d+)';",
  90             r'<iframe[^>]+src="/player/(\d+)"',
  91         ], webpage, 'video_id', default=None)
  92         if not video_id:
  93             player_url = self._html_search_meta([
  94                 'al:web:url',
  95                 'og:video:url',
  96                 'og:video:secure_url',
  97             ], webpage, fatal=True)
  98             video_id = self._search_regex(r'/player/(\d+)', player_url, 'video id')
  99
 100         title = episode = title_data.get('title') or _search_kane('videoTitle') or self._og_search_title(webpage)
 101         series = _search_kane('showName')
 102         if series:
 103             title = '%s - %s' % (series, title)
 104         description = self._html_search_meta(['description', 'og:description'], webpage, fatal=True)
 105
 106         try:
 107             headers = {}
 108             if self._TOKEN:
 109                 headers['Authorization'] = 'Token %s' % self._TOKEN
 110             sources = self._download_json(
 111                 'https://prod-api-funimationnow.dadcdigital.com/api/source/catalog/video/%s/signed/' % video_id,
 112                 video_id, headers=headers)['items']
 113         except ExtractorError as e:
 114             if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403:
 115                 error = self._parse_json(e.cause.read(), video_id)['errors'][0]
 116                 raise ExtractorError('%s said: %s' % (
 117                     self.IE_NAME, error.get('detail') or error.get('title')), expected=True)
 118             raise
 119
 120         formats = []
 121         for source in sources:
 122             source_url = source.get('src')
 123             if not source_url:
 124                 continue
 125             source_type = source.get('videoType') or determine_ext(source_url)
 126             if source_type == 'm3u8':
 127                 formats.extend(self._extract_m3u8_formats(
 128                     source_url, video_id, 'mp4',
 129                     m3u8_id='hls', fatal=False))
 130             else:
 131                 formats.append({
 132                     'format_id': source_type,
 133                     'url': source_url,
 134                 })
 135         self._sort_formats(formats)
 136
 137         return {
 138             'id': video_id,
 139             'display_id': display_id,
 140             'title': title,
 141             'description': description,
 142             'thumbnail': self._og_search_thumbnail(webpage),
 143             'series': series,
 144             'season_number': int_or_none(title_data.get('seasonNum') or _search_kane('season')),
 145             'episode_number': int_or_none(title_data.get('episodeNum')),
 146             'episode': episode,
 147             'season_id': title_data.get('seriesId'),
 148             'formats': formats,
 149         }