yt_dlp/extractor/hytale.py

   1 import re
   2
   3 from .cloudflarestream import CloudflareStreamIE
   4 from .common import InfoExtractor
   5 from ..utils.traversal import traverse_obj
   6
   7
   8 class HytaleIE(InfoExtractor):
   9     _VALID_URL = r'https?://(?:www\.)?hytale\.com/news/\d+/\d+/(?P<id>[a-z0-9-]+)'
  10     _TESTS = [{
  11         'url': 'https://hytale.com/news/2021/07/summer-2021-development-update',
  12         'info_dict': {
  13             'id': 'summer-2021-development-update',
  14             'title': 'Summer 2021 Development Update',
  15         },
  16         'playlist_count': 4,
  17         'playlist': [{
  18             'md5': '0854ebe347d233ee19b86ab7b2ead610',
  19             'info_dict': {
  20                 'id': 'ed51a2609d21bad6e14145c37c334999',
  21                 'ext': 'mp4',
  22                 'title': 'Avatar Personalization',
  23                 'thumbnail': r're:https://videodelivery\.net/\w+/thumbnails/thumbnail\.jpg',
  24             },
  25         }],
  26     }, {
  27         'url': 'https://www.hytale.com/news/2019/11/hytale-graphics-update',
  28         'info_dict': {
  29             'id': 'hytale-graphics-update',
  30             'title': 'Hytale graphics update',
  31         },
  32         'playlist_count': 2,
  33     }]
  34
  35     def _real_initialize(self):
  36         media_webpage = self._download_webpage(
  37             'https://hytale.com/media', None, note='Downloading list of media', fatal=False) or ''
  38
  39         clips_json = traverse_obj(
  40             self._search_json(
  41                 r'window\.__INITIAL_COMPONENTS_STATE__\s*=\s*\[',
  42                 media_webpage, 'clips json', None),
  43             ('media', 'clips')) or []
  44
  45         self._titles = {clip.get('src'): clip.get('caption') for clip in clips_json}
  46
  47     def _real_extract(self, url):
  48         playlist_id = self._match_id(url)
  49         webpage = self._download_webpage(url, playlist_id)
  50         entries = [
  51             self.url_result(
  52                 f'https://cloudflarestream.com/{video_hash}/manifest/video.mpd?parentOrigin=https%3A%2F%2Fhytale.com',
  53                 CloudflareStreamIE, title=self._titles.get(video_hash), url_transparent=True)
  54             for video_hash in re.findall(
  55                 r'<stream\s+class\s*=\s*"ql-video\s+cf-stream"\s+src\s*=\s*"([a-f0-9]{32})"',
  56                 webpage)
  57         ]
  58
  59         return self.playlist_result(entries, playlist_id, self._og_search_title(webpage))