youtube_dl/extractor/funimation.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import random
   5 import string
   6
   7 from .common import InfoExtractor
   8 from ..compat import compat_HTTPError
   9 from ..utils import (
  10     determine_ext,
  11     int_or_none,
  12     js_to_json,
  13     ExtractorError,
  14     urlencode_postdata
  15 )
  16
  17
  18 class FunimationIE(InfoExtractor):
  19     _VALID_URL = r'https?://(?:www\.)?funimation(?:\.com|now\.uk)/(?:[^/]+/)?shows/[^/]+/(?P<id>[^/?#&]+)'
  20
  21     _NETRC_MACHINE = 'funimation'
  22     _TOKEN = None
  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': '210051',
  42             'display_id': 'broadcast-dub-preview',
  43             'ext': 'mp4',
  44             'title': 'Attack on Titan: Junior High - Broadcast Dub Preview',
  45             'thumbnail': r're:https?://.*\.(?:jpg|png)',
  46         },
  47         'params': {
  48             # m3u8 download
  49             'skip_download': True,
  50         },
  51     }, {
  52         'url': 'https://www.funimationnow.uk/shows/puzzle-dragons-x/drop-impact/simulcast/',
  53         'only_matching': True,
  54     }, {
  55         # with lang code
  56         'url': 'https://www.funimation.com/en/shows/hacksign/role-play/',
  57         'only_matching': True,
  58     }]
  59
  60     def _login(self):
  61         username, password = self._get_login_info()
  62         if username is None:
  63             return
  64         try:
  65             data = self._download_json(
  66                 'https://prod-api-funimationnow.dadcdigital.com/api/auth/login/',
  67                 None, 'Logging in', data=urlencode_postdata({
  68                     'username': username,
  69                     'password': password,
  70                 }))
  71             self._TOKEN = data['token']
  72         except ExtractorError as e:
  73             if isinstance(e.cause, compat_HTTPError) and e.cause.code == 401:
  74                 error = self._parse_json(e.cause.read().decode(), None)['error']
  75                 raise ExtractorError(error, expected=True)
  76             raise
  77
  78     def _real_initialize(self):
  79         self._login()
  80
  81     def _real_extract(self, url):
  82         display_id = self._match_id(url)
  83         webpage = self._download_webpage(url, display_id)
  84
  85         def _search_kane(name):
  86             return self._search_regex(
  87                 r"KANE_customdimensions\.%s\s*=\s*'([^']+)';" % name,
  88                 webpage, name, default=None)
  89
  90         title_data = self._parse_json(self._search_regex(
  91             r'TITLE_DATA\s*=\s*({[^}]+})',
  92             webpage, 'title data', default=''),
  93             display_id, js_to_json, fatal=False) or {}
  94
  95         video_id = title_data.get('id') or self._search_regex([
  96             r"KANE_customdimensions.videoID\s*=\s*'(\d+)';",
  97             r'<iframe[^>]+src="/player/(\d+)',
  98         ], webpage, 'video_id', default=None)
  99         if not video_id:
 100             player_url = self._html_search_meta([
 101                 'al:web:url',
 102                 'og:video:url',
 103                 'og:video:secure_url',
 104             ], webpage, fatal=True)
 105             video_id = self._search_regex(r'/player/(\d+)', player_url, 'video id')
 106
 107         title = episode = title_data.get('title') or _search_kane('videoTitle') or self._og_search_title(webpage)
 108         series = _search_kane('showName')
 109         if series:
 110             title = '%s - %s' % (series, title)
 111         description = self._html_search_meta(['description', 'og:description'], webpage, fatal=True)
 112
 113         try:
 114             headers = {}
 115             if self._TOKEN:
 116                 headers['Authorization'] = 'Token %s' % self._TOKEN
 117             sources = self._download_json(
 118                 'https://www.funimation.com/api/showexperience/%s/' % video_id,
 119                 video_id, headers=headers, query={
 120                     'pinst_id': ''.join([random.choice(string.digits + string.ascii_letters) for _ in range(8)]),
 121                 })['items']
 122         except ExtractorError as e:
 123             if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403:
 124                 error = self._parse_json(e.cause.read(), video_id)['errors'][0]
 125                 raise ExtractorError('%s said: %s' % (
 126                     self.IE_NAME, error.get('detail') or error.get('title')), expected=True)
 127             raise
 128
 129         formats = []
 130         for source in sources:
 131             source_url = source.get('src')
 132             if not source_url:
 133                 continue
 134             source_type = source.get('videoType') or determine_ext(source_url)
 135             if source_type == 'm3u8':
 136                 formats.extend(self._extract_m3u8_formats(
 137                     source_url, video_id, 'mp4',
 138                     m3u8_id='hls', fatal=False))
 139             else:
 140                 formats.append({
 141                     'format_id': source_type,
 142                     'url': source_url,
 143                 })
 144         self._sort_formats(formats)
 145
 146         return {
 147             'id': video_id,
 148             'display_id': display_id,
 149             'title': title,
 150             'description': description,
 151             'thumbnail': self._og_search_thumbnail(webpage),
 152             'series': series,
 153             'season_number': int_or_none(title_data.get('seasonNum') or _search_kane('season')),
 154             'episode_number': int_or_none(title_data.get('episodeNum')),
 155             'episode': episode,
 156             'season_id': title_data.get('seriesId'),
 157             'formats': formats,
 158         }