yt_dlp/extractor/threespeak.py

   1 import re
   2
   3 from .common import InfoExtractor
   4 from ..utils import (
   5     try_get,
   6     unified_strdate,
   7 )
   8
   9
  10 class ThreeSpeakIE(InfoExtractor):
  11     _VALID_URL = r'https?://(?:www\.)?3speak\.tv/watch\?v\=[^/]+/(?P<id>[^/$&#?]+)'
  12
  13     _TESTS = [{
  14         'url': 'https://3speak.tv/watch?v=dannyshine/wjgoxyfy',
  15         'info_dict': {
  16             'id': 'wjgoxyfy',
  17             'ext': 'mp4',
  18             'title': 'Can People who took the Vax think Critically',
  19             'uploader': 'dannyshine',
  20             'description': 'md5:181aa7ccb304afafa089b5af3bca7a10',
  21             'tags': ['sex', 'covid', 'antinatalism', 'comedy', 'vaccines'],
  22             'thumbnail': 'https://img.3speakcontent.co/wjgoxyfy/thumbnails/default.png',
  23             'upload_date': '20211021',
  24             'duration': 2703.867833,
  25             'filesize': 1620054781,
  26         },
  27         'params': {'skip_download': True},
  28     }]
  29
  30     def _real_extract(self, url):
  31         video_id = self._match_id(url)
  32         webpage = self._download_webpage(url, video_id)
  33         json_str = self._html_search_regex(r'JSON\.parse\(\'([^\']+)\'\)', webpage, 'json')
  34         # The json string itself is escaped. Hence the double parsing
  35         data_json = self._parse_json(self._parse_json(f'"{json_str}"', video_id), video_id)
  36         video_json = self._parse_json(data_json['json_metadata'], video_id)
  37         formats, subtitles = [], {}
  38         og_m3u8 = self._html_search_regex(r'<meta\s?property=\"ogvideo\"\s?content=\"([^\"]+)\">', webpage, 'og m3u8', fatal=False)
  39         if og_m3u8:
  40             https_frmts, https_subs = self._extract_m3u8_formats_and_subtitles(og_m3u8, video_id, fatal=False, m3u8_id='https')
  41             formats.extend(https_frmts)
  42             subtitles = self._merge_subtitles(subtitles, https_subs)
  43         ipfs_m3u8 = try_get(video_json, lambda x: x['video']['info']['ipfs'])
  44         if ipfs_m3u8:
  45             ipfs_frmts, ipfs_subs = self._extract_m3u8_formats_and_subtitles(
  46                 f'https://ipfs.3speak.tv/ipfs/{ipfs_m3u8}', video_id, fatal=False, m3u8_id='ipfs')
  47             formats.extend(ipfs_frmts)
  48             subtitles = self._merge_subtitles(subtitles, ipfs_subs)
  49         mp4_file = try_get(video_json, lambda x: x['video']['info']['file'])
  50         if mp4_file:
  51             formats.append({
  52                 'url': f'https://threespeakvideo.b-cdn.net/{video_id}/{mp4_file}',
  53                 'ext': 'mp4',
  54                 'format_id': 'https-mp4',
  55                 'duration': try_get(video_json, lambda x: x['video']['info']['duration']),
  56                 'filesize': try_get(video_json, lambda x: x['video']['info']['filesize']),
  57                 'quality': 11,
  58                 'format_note': 'Original file',
  59             })
  60         return {
  61             'id': video_id,
  62             'title': data_json.get('title') or data_json.get('root_title'),
  63             'uploader': data_json.get('author'),
  64             'description': try_get(video_json, lambda x: x['video']['content']['description']),
  65             'tags': try_get(video_json, lambda x: x['video']['content']['tags']),
  66             'thumbnail': try_get(video_json, lambda x: x['image'][0]),
  67             'upload_date': unified_strdate(data_json.get('created')),
  68             'formats': formats,
  69             'subtitles': subtitles,
  70         }
  71
  72
  73 class ThreeSpeakUserIE(InfoExtractor):
  74     _VALID_URL = r'https?://(?:www\.)?3speak\.tv/user/(?P<id>[^/$&?#]+)'
  75
  76     _TESTS = [{
  77         'url': 'https://3speak.tv/user/theycallmedan',
  78         'info_dict': {
  79             'id': 'theycallmedan',
  80         },
  81         'playlist_mincount': 115,
  82     }]
  83
  84     def _real_extract(self, url):
  85         playlist_id = self._match_id(url)
  86         webpage = self._download_webpage(url, playlist_id)
  87         entries = [
  88             self.url_result(
  89                 f'https://3speak.tv/watch?v={video}',
  90                 ie=ThreeSpeakIE.ie_key())
  91             for video in re.findall(r'data-payout\s?\=\s?\"([^\"]+)\"', webpage) if video
  92         ]
  93         return self.playlist_result(entries, playlist_id)