yt_dlp/extractor/threespeak.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     try_get,
   9     unified_strdate,
  10 )
  11
  12
  13 class ThreeSpeakIE(InfoExtractor):
  14     _VALID_URL = r'https?://(?:www\.)?3speak\.tv/watch\?v\=[^/]+/(?P<id>[^/$&#?]+)'
  15
  16     _TESTS = [{
  17         'url': 'https://3speak.tv/watch?v=dannyshine/wjgoxyfy',
  18         'info_dict': {
  19             'id': 'wjgoxyfy',
  20             'ext': 'mp4',
  21             'title': 'Can People who took the Vax think Critically',
  22             'uploader': 'dannyshine',
  23             'description': 'md5:181aa7ccb304afafa089b5af3bca7a10',
  24             'tags': ['sex', 'covid', 'antinatalism', 'comedy', 'vaccines'],
  25             'thumbnail': 'https://img.3speakcontent.co/wjgoxyfy/thumbnails/default.png',
  26             'upload_date': '20211021',
  27             'duration': 2703.867833,
  28             'filesize': 1620054781,
  29         },
  30         'params': {'skip_download': True}
  31     }]
  32
  33     def _real_extract(self, url):
  34         id = self._match_id(url)
  35         webpage = self._download_webpage(url, id)
  36         json_str = self._html_search_regex(r'JSON\.parse\(\'([^\']+)\'\)', webpage, 'json')
  37         # The json string itself is escaped. Hence the double parsing
  38         data_json = self._parse_json(self._parse_json(f'"{json_str}"', id), id)
  39         video_json = self._parse_json(data_json['json_metadata'], id)
  40         formats, subtitles = [], {}
  41         og_m3u8 = self._html_search_regex(r'<meta\s?property=\"ogvideo\"\s?content=\"([^\"]+)\">', webpage, 'og m3u8', fatal=False)
  42         if og_m3u8:
  43             https_frmts, https_subs = self._extract_m3u8_formats_and_subtitles(og_m3u8, id, fatal=False, m3u8_id='https')
  44             formats.extend(https_frmts)
  45             subtitles = self._merge_subtitles(subtitles, https_subs)
  46         ipfs_m3u8 = try_get(video_json, lambda x: x['video']['info']['ipfs'])
  47         if ipfs_m3u8:
  48             ipfs_frmts, ipfs_subs = self._extract_m3u8_formats_and_subtitles(f'https://ipfs.3speak.tv/ipfs/{ipfs_m3u8}',
  49                                                                              id, fatal=False, m3u8_id='ipfs')
  50             formats.extend(ipfs_frmts)
  51             subtitles = self._merge_subtitles(subtitles, ipfs_subs)
  52         mp4_file = try_get(video_json, lambda x: x['video']['info']['file'])
  53         if mp4_file:
  54             formats.append({
  55                 'url': f'https://threespeakvideo.b-cdn.net/{id}/{mp4_file}',
  56                 'ext': 'mp4',
  57                 'format_id': 'https-mp4',
  58                 'duration': try_get(video_json, lambda x: x['video']['info']['duration']),
  59                 'filesize': try_get(video_json, lambda x: x['video']['info']['filesize']),
  60                 'quality': 11,
  61                 'format_note': 'Original file',
  62             })
  63         self._sort_formats(formats)
  64         return {
  65             'id': id,
  66             'title': data_json.get('title') or data_json.get('root_title'),
  67             'uploader': data_json.get('author'),
  68             'description': try_get(video_json, lambda x: x['video']['content']['description']),
  69             'tags': try_get(video_json, lambda x: x['video']['content']['tags']),
  70             'thumbnail': try_get(video_json, lambda x: x['image'][0]),
  71             'upload_date': unified_strdate(data_json.get('created')),
  72             'formats': formats,
  73             'subtitles': subtitles,
  74         }
  75
  76
  77 class ThreeSpeakUserIE(InfoExtractor):
  78     _VALID_URL = r'https?://(?:www\.)?3speak\.tv/user/(?P<id>[^/$&?#]+)'
  79
  80     _TESTS = [{
  81         'url': 'https://3speak.tv/user/theycallmedan',
  82         'info_dict': {
  83             'id': 'theycallmedan',
  84         },
  85         'playlist_mincount': 115,
  86     }]
  87
  88     def _real_extract(self, url):
  89         id = self._match_id(url)
  90         webpage = self._download_webpage(url, id)
  91         entries = [
  92             self.url_result(
  93                 'https://3speak.tv/watch?v=%s' % video,
  94                 ie=ThreeSpeakIE.ie_key())
  95             for video in re.findall(r'data-payout\s?\=\s?\"([^\"]+)\"', webpage) if video
  96         ]
  97         return self.playlist_result(entries, id)