yt_dlp/extractor/pokemon.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     ExtractorError,
   9     extract_attributes,
  10     int_or_none,
  11     js_to_json,
  12     merge_dicts,
  13 )
  14
  15
  16 class PokemonIE(InfoExtractor):
  17     _VALID_URL = r'https?://(?:www\.)?pokemon\.com/[a-z]{2}(?:.*?play=(?P<id>[a-z0-9]{32})|/(?:[^/]+/)+(?P<display_id>[^/?#&]+))'
  18     _TESTS = [{
  19         'url': 'https://www.pokemon.com/us/pokemon-episodes/20_30-the-ol-raise-and-switch/',
  20         'md5': '2fe8eaec69768b25ef898cda9c43062e',
  21         'info_dict': {
  22             'id': 'afe22e30f01c41f49d4f1d9eab5cd9a4',
  23             'ext': 'mp4',
  24             'title': 'The Ol’ Raise and Switch!',
  25             'description': 'md5:7db77f7107f98ba88401d3adc80ff7af',
  26         },
  27         'add_id': ['LimelightMedia'],
  28     }, {
  29         # no data-video-title
  30         'url': 'https://www.pokemon.com/fr/episodes-pokemon/films-pokemon/pokemon-lascension-de-darkrai-2008',
  31         'info_dict': {
  32             'id': 'dfbaf830d7e54e179837c50c0c6cc0e1',
  33             'ext': 'mp4',
  34             'title': "Pokémon : L'ascension de Darkrai",
  35             'description': 'md5:d1dbc9e206070c3e14a06ff557659fb5',
  36         },
  37         'add_id': ['LimelightMedia'],
  38         'params': {
  39             'skip_download': True,
  40         },
  41     }, {
  42         'url': 'http://www.pokemon.com/uk/pokemon-episodes/?play=2e8b5c761f1d4a9286165d7748c1ece2',
  43         'only_matching': True,
  44     }, {
  45         'url': 'http://www.pokemon.com/fr/episodes-pokemon/18_09-un-hiver-inattendu/',
  46         'only_matching': True,
  47     }, {
  48         'url': 'http://www.pokemon.com/de/pokemon-folgen/01_20-bye-bye-smettbo/',
  49         'only_matching': True,
  50     }]
  51
  52     def _real_extract(self, url):
  53         video_id, display_id = re.match(self._VALID_URL, url).groups()
  54         webpage = self._download_webpage(url, video_id or display_id)
  55         video_data = extract_attributes(self._search_regex(
  56             r'(<[^>]+data-video-id="%s"[^>]*>)' % (video_id if video_id else '[a-z0-9]{32}'),
  57             webpage, 'video data element'))
  58         video_id = video_data['data-video-id']
  59         title = video_data.get('data-video-title') or self._html_search_meta(
  60             'pkm-title', webpage, ' title', default=None) or self._search_regex(
  61             r'<h1[^>]+\bclass=["\']us-title[^>]+>([^<]+)', webpage, 'title')
  62         return {
  63             '_type': 'url_transparent',
  64             'id': video_id,
  65             'url': 'limelight:media:%s' % video_id,
  66             'title': title,
  67             'description': video_data.get('data-video-summary'),
  68             'thumbnail': video_data.get('data-video-poster'),
  69             'series': 'Pokémon',
  70             'season_number': int_or_none(video_data.get('data-video-season')),
  71             'episode': title,
  72             'episode_number': int_or_none(video_data.get('data-video-episode')),
  73             'ie_key': 'LimelightMedia',
  74         }
  75
  76
  77 class PokemonWatchIE(InfoExtractor):
  78     _VALID_URL = r'https?://watch\.pokemon\.com/[a-z]{2}-[a-z]{2}/(?:#/)?player(?:\.html)?\?id=(?P<id>[a-z0-9]{32})'
  79     _API_URL = 'https://www.pokemon.com/api/pokemontv/v2/channels/{0:}'
  80     _TESTS = [{
  81         'url': 'https://watch.pokemon.com/en-us/player.html?id=8309a40969894a8e8d5bc1311e9c5667',
  82         'md5': '62833938a31e61ab49ada92f524c42ff',
  83         'info_dict': {
  84             'id': '8309a40969894a8e8d5bc1311e9c5667',
  85             'ext': 'mp4',
  86             'title': 'Lillier and the Staff!',
  87             'description': 'md5:338841b8c21b283d24bdc9b568849f04',
  88         }
  89     }, {
  90         'url': 'https://watch.pokemon.com/en-us/#/player?id=3fe7752ba09141f0b0f7756d1981c6b2',
  91         'only_matching': True
  92     }, {
  93         'url': 'https://watch.pokemon.com/de-de/player.html?id=b3c402e111a4459eb47e12160ab0ba07',
  94         'only_matching': True
  95     }]
  96
  97     def _extract_media(self, channel_array, video_id):
  98         for channel in channel_array:
  99             for media in channel.get('media'):
 100                 if media.get('id') == video_id:
 101                     return media
 102         return None
 103
 104     def _real_extract(self, url):
 105         video_id = self._match_id(url)
 106
 107         info = {
 108             '_type': 'url',
 109             'id': video_id,
 110             'url': 'limelight:media:%s' % video_id,
 111             'ie_key': 'LimelightMedia',
 112         }
 113
 114         # API call can be avoided entirely if we are listing formats
 115         if self.get_param('listformats', False):
 116             return info
 117
 118         webpage = self._download_webpage(url, video_id)
 119         build_vars = self._parse_json(self._search_regex(
 120             r'(?s)buildVars\s*=\s*({.*?})', webpage, 'build vars'),
 121             video_id, transform_source=js_to_json)
 122         region = build_vars.get('region')
 123         channel_array = self._download_json(self._API_URL.format(region), video_id)
 124         video_data = self._extract_media(channel_array, video_id)
 125
 126         if video_data is None:
 127             raise ExtractorError(
 128                 'Video %s does not exist' % video_id, expected=True)
 129
 130         info['_type'] = 'url_transparent'
 131         images = video_data.get('images')
 132
 133         return merge_dicts(info, {
 134             'title': video_data.get('title'),
 135             'description': video_data.get('description'),
 136             'thumbnail': images.get('medium') or images.get('small'),
 137             'series': 'Pokémon',
 138             'season_number': int_or_none(video_data.get('season')),
 139             'episode': video_data.get('title'),
 140             'episode_number': int_or_none(video_data.get('episode')),
 141         })