yt_dlp/extractor/chaturbate.py

   1 import re
   2
   3 from .common import InfoExtractor
   4 from ..utils import (
   5     ExtractorError,
   6     lowercase_escape,
   7     url_or_none,
   8     urlencode_postdata,
   9 )
  10
  11
  12 class ChaturbateIE(InfoExtractor):
  13     _VALID_URL = r'https?://(?:[^/]+\.)?chaturbate\.(?P<tld>com|eu|global)/(?:fullvideo/?\?.*?\bb=)?(?P<id>[^/?&#]+)'
  14     _TESTS = [{
  15         'url': 'https://www.chaturbate.com/siswet19/',
  16         'info_dict': {
  17             'id': 'siswet19',
  18             'ext': 'mp4',
  19             'title': 're:^siswet19 [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
  20             'age_limit': 18,
  21             'is_live': True,
  22         },
  23         'params': {
  24             'skip_download': True,
  25         },
  26         'skip': 'Room is offline',
  27     }, {
  28         'url': 'https://chaturbate.com/fullvideo/?b=caylin',
  29         'only_matching': True,
  30     }, {
  31         'url': 'https://en.chaturbate.com/siswet19/',
  32         'only_matching': True,
  33     }, {
  34         'url': 'https://chaturbate.eu/siswet19/',
  35         'only_matching': True,
  36     }, {
  37         'url': 'https://chaturbate.eu/fullvideo/?b=caylin',
  38         'only_matching': True,
  39     }, {
  40         'url': 'https://chaturbate.global/siswet19/',
  41         'only_matching': True,
  42     }]
  43
  44     _ERROR_MAP = {
  45         'offline': 'Room is currently offline',
  46         'private': 'Room is currently in a private show',
  47         'away': 'Performer is currently away',
  48         'password protected': 'Room is password protected',
  49         'hidden': 'Hidden session in progress',
  50     }
  51
  52     def _extract_from_api(self, video_id, tld):
  53         response = self._download_json(
  54             f'https://chaturbate.{tld}/get_edge_hls_url_ajax/', video_id,
  55             data=urlencode_postdata({'room_slug': video_id}),
  56             headers={
  57                 **self.geo_verification_headers(),
  58                 'X-Requested-With': 'XMLHttpRequest',
  59                 'Accept': 'application/json',
  60             }, fatal=False, impersonate=True) or {}
  61
  62         m3u8_url = response.get('url')
  63         if not m3u8_url:
  64             status = response.get('room_status')
  65             if error := self._ERROR_MAP.get(status):
  66                 raise ExtractorError(error, expected=True)
  67             if status == 'public':
  68                 self.raise_geo_restricted()
  69             self.report_warning(f'Got status "{status}" from API; falling back to webpage extraction')
  70             return None
  71
  72         return {
  73             'id': video_id,
  74             'title': video_id,
  75             'thumbnail': f'https://roomimg.stream.highwebmedia.com/ri/{video_id}.jpg',
  76             'is_live': True,
  77             'age_limit': 18,
  78             'formats': self._extract_m3u8_formats(m3u8_url, video_id, ext='mp4', live=True),
  79         }
  80
  81     def _extract_from_html(self, video_id, tld):
  82         webpage = self._download_webpage(
  83             f'https://chaturbate.{tld}/{video_id}/', video_id,
  84             headers=self.geo_verification_headers(), impersonate=True)
  85
  86         found_m3u8_urls = []
  87
  88         data = self._parse_json(
  89             self._search_regex(
  90                 r'initialRoomDossier\s*=\s*(["\'])(?P<value>(?:(?!\1).)+)\1',
  91                 webpage, 'data', default='{}', group='value'),
  92             video_id, transform_source=lowercase_escape, fatal=False)
  93         if data:
  94             m3u8_url = url_or_none(data.get('hls_source'))
  95             if m3u8_url:
  96                 found_m3u8_urls.append(m3u8_url)
  97
  98         if not found_m3u8_urls:
  99             for m in re.finditer(
 100                     r'(\\u002[27])(?P<url>http.+?\.m3u8.*?)\1', webpage):
 101                 found_m3u8_urls.append(lowercase_escape(m.group('url')))
 102
 103         if not found_m3u8_urls:
 104             for m in re.finditer(
 105                     r'(["\'])(?P<url>http.+?\.m3u8.*?)\1', webpage):
 106                 found_m3u8_urls.append(m.group('url'))
 107
 108         m3u8_urls = []
 109         for found_m3u8_url in found_m3u8_urls:
 110             m3u8_fast_url, m3u8_no_fast_url = found_m3u8_url, found_m3u8_url.replace('_fast', '')
 111             for m3u8_url in (m3u8_fast_url, m3u8_no_fast_url):
 112                 if m3u8_url not in m3u8_urls:
 113                     m3u8_urls.append(m3u8_url)
 114
 115         if not m3u8_urls:
 116             error = self._search_regex(
 117                 [r'<span[^>]+class=(["\'])desc_span\1[^>]*>(?P<error>[^<]+)</span>',
 118                  r'<div[^>]+id=(["\'])defchat\1[^>]*>\s*<p><strong>(?P<error>[^<]+)<'],
 119                 webpage, 'error', group='error', default=None)
 120             if not error:
 121                 if any(p in webpage for p in (
 122                         self._ERROR_MAP['offline'], 'offline_tipping', 'tip_offline')):
 123                     error = self._ERROR_MAP['offline']
 124             if error:
 125                 raise ExtractorError(error, expected=True)
 126             raise ExtractorError('Unable to find stream URL')
 127
 128         formats = []
 129         for m3u8_url in m3u8_urls:
 130             for known_id in ('fast', 'slow'):
 131                 if f'_{known_id}' in m3u8_url:
 132                     m3u8_id = known_id
 133                     break
 134             else:
 135                 m3u8_id = None
 136             formats.extend(self._extract_m3u8_formats(
 137                 m3u8_url, video_id, ext='mp4',
 138                 # ffmpeg skips segments for fast m3u8
 139                 preference=-10 if m3u8_id == 'fast' else None,
 140                 m3u8_id=m3u8_id, fatal=False, live=True))
 141
 142         return {
 143             'id': video_id,
 144             'title': video_id,
 145             'thumbnail': f'https://roomimg.stream.highwebmedia.com/ri/{video_id}.jpg',
 146             'age_limit': self._rta_search(webpage),
 147             'is_live': True,
 148             'formats': formats,
 149         }
 150
 151     def _real_extract(self, url):
 152         video_id, tld = self._match_valid_url(url).group('id', 'tld')
 153         return self._extract_from_api(video_id, tld) or self._extract_from_html(video_id, tld)