yt_dlp/extractor/pokemon.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4
   5 from .common import InfoExtractor
   6 from ..utils import (
   7     ExtractorError,
   8     extract_attributes,
   9     int_or_none,
  10     js_to_json,
  11     merge_dicts,
  12 )
  13
  14
  15 class PokemonIE(InfoExtractor):
  16     _VALID_URL = r'https?://(?:www\.)?pokemon\.com/[a-z]{2}(?:.*?play=(?P<id>[a-z0-9]{32})|/(?:[^/]+/)+(?P<display_id>[^/?#&]+))'
  17     _TESTS = [{
  18         'url': 'https://www.pokemon.com/us/pokemon-episodes/20_30-the-ol-raise-and-switch/',
  19         'md5': '2fe8eaec69768b25ef898cda9c43062e',
  20         'info_dict': {
  21             'id': 'afe22e30f01c41f49d4f1d9eab5cd9a4',
  22             'ext': 'mp4',
  23             'title': 'The Ol’ Raise and Switch!',
  24             'description': 'md5:7db77f7107f98ba88401d3adc80ff7af',
  25         },
  26         'add_id': ['LimelightMedia'],
  27     }, {
  28         # no data-video-title
  29         'url': 'https://www.pokemon.com/fr/episodes-pokemon/films-pokemon/pokemon-lascension-de-darkrai-2008',
  30         'info_dict': {
  31             'id': 'dfbaf830d7e54e179837c50c0c6cc0e1',
  32             'ext': 'mp4',
  33             'title': "Pokémon : L'ascension de Darkrai",
  34             'description': 'md5:d1dbc9e206070c3e14a06ff557659fb5',
  35         },
  36         'add_id': ['LimelightMedia'],
  37         'params': {
  38             'skip_download': True,
  39         },
  40     }, {
  41         'url': 'http://www.pokemon.com/uk/pokemon-episodes/?play=2e8b5c761f1d4a9286165d7748c1ece2',
  42         'only_matching': True,
  43     }, {
  44         'url': 'http://www.pokemon.com/fr/episodes-pokemon/18_09-un-hiver-inattendu/',
  45         'only_matching': True,
  46     }, {
  47         'url': 'http://www.pokemon.com/de/pokemon-folgen/01_20-bye-bye-smettbo/',
  48         'only_matching': True,
  49     }]
  50
  51     def _real_extract(self, url):
  52         video_id, display_id = self._match_valid_url(url).groups()
  53         webpage = self._download_webpage(url, video_id or display_id)
  54         video_data = extract_attributes(self._search_regex(
  55             r'(<[^>]+data-video-id="%s"[^>]*>)' % (video_id if video_id else '[a-z0-9]{32}'),
  56             webpage, 'video data element'))
  57         video_id = video_data['data-video-id']
  58         title = video_data.get('data-video-title') or self._html_search_meta(
  59             'pkm-title', webpage, ' title', default=None) or self._search_regex(
  60             r'<h1[^>]+\bclass=["\']us-title[^>]+>([^<]+)', webpage, 'title')
  61         return {
  62             '_type': 'url_transparent',
  63             'id': video_id,
  64             'url': 'limelight:media:%s' % video_id,
  65             'title': title,
  66             'description': video_data.get('data-video-summary'),
  67             'thumbnail': video_data.get('data-video-poster'),
  68             'series': 'Pokémon',
  69             'season_number': int_or_none(video_data.get('data-video-season')),
  70             'episode': title,
  71             'episode_number': int_or_none(video_data.get('data-video-episode')),
  72             'ie_key': 'LimelightMedia',
  73         }
  74
  75
  76 class PokemonWatchIE(InfoExtractor):
  77     _VALID_URL = r'https?://watch\.pokemon\.com/[a-z]{2}-[a-z]{2}/(?:#/)?player(?:\.html)?\?id=(?P<id>[a-z0-9]{32})'
  78     _API_URL = 'https://www.pokemon.com/api/pokemontv/v2/channels/{0:}'
  79     _TESTS = [{
  80         'url': 'https://watch.pokemon.com/en-us/player.html?id=8309a40969894a8e8d5bc1311e9c5667',
  81         'md5': '62833938a31e61ab49ada92f524c42ff',
  82         'info_dict': {
  83             'id': '8309a40969894a8e8d5bc1311e9c5667',
  84             'ext': 'mp4',
  85             'title': 'Lillier and the Staff!',
  86             'description': 'md5:338841b8c21b283d24bdc9b568849f04',
  87         }
  88     }, {
  89         'url': 'https://watch.pokemon.com/en-us/#/player?id=3fe7752ba09141f0b0f7756d1981c6b2',
  90         'only_matching': True
  91     }, {
  92         'url': 'https://watch.pokemon.com/de-de/player.html?id=b3c402e111a4459eb47e12160ab0ba07',
  93         'only_matching': True
  94     }]
  95
  96     def _extract_media(self, channel_array, video_id):
  97         for channel in channel_array:
  98             for media in channel.get('media'):
  99                 if media.get('id') == video_id:
 100                     return media
 101         return None
 102
 103     def _real_extract(self, url):
 104         video_id = self._match_id(url)
 105
 106         info = {
 107             '_type': 'url',
 108             'id': video_id,
 109             'url': 'limelight:media:%s' % video_id,
 110             'ie_key': 'LimelightMedia',
 111         }
 112
 113         # API call can be avoided entirely if we are listing formats
 114         if self.get_param('listformats', False):
 115             return info
 116
 117         webpage = self._download_webpage(url, video_id)
 118         build_vars = self._parse_json(self._search_regex(
 119             r'(?s)buildVars\s*=\s*({.*?})', webpage, 'build vars'),
 120             video_id, transform_source=js_to_json)
 121         region = build_vars.get('region')
 122         channel_array = self._download_json(self._API_URL.format(region), video_id)
 123         video_data = self._extract_media(channel_array, video_id)
 124
 125         if video_data is None:
 126             raise ExtractorError(
 127                 'Video %s does not exist' % video_id, expected=True)
 128
 129         info['_type'] = 'url_transparent'
 130         images = video_data.get('images')
 131
 132         return merge_dicts(info, {
 133             'title': video_data.get('title'),
 134             'description': video_data.get('description'),
 135             'thumbnail': images.get('medium') or images.get('small'),
 136             'series': 'Pokémon',
 137             'season_number': int_or_none(video_data.get('season')),
 138             'episode': video_data.get('title'),
 139             'episode_number': int_or_none(video_data.get('episode')),
 140         })