yt_dlp/extractor/eporner.py

   1 from .common import InfoExtractor
   2 from ..utils import (
   3     ExtractorError,
   4     encode_base_n,
   5     get_elements_by_class,
   6     int_or_none,
   7     join_nonempty,
   8     merge_dicts,
   9     parse_duration,
  10     str_to_int,
  11     url_or_none,
  12 )
  13
  14
  15 class EpornerIE(InfoExtractor):
  16     _VALID_URL = r'https?://(?:www\.)?eporner\.com/(?:(?:hd-porn|embed)/|video-)(?P<id>\w+)(?:/(?P<display_id>[\w-]+))?'
  17     _TESTS = [{
  18         'url': 'http://www.eporner.com/hd-porn/95008/Infamous-Tiffany-Teen-Strip-Tease-Video/',
  19         'md5': '39d486f046212d8e1b911c52ab4691f8',
  20         'info_dict': {
  21             'id': 'qlDUmNsj6VS',
  22             'display_id': 'Infamous-Tiffany-Teen-Strip-Tease-Video',
  23             'ext': 'mp4',
  24             'title': 'Infamous Tiffany Teen Strip Tease Video',
  25             'description': 'md5:764f39abf932daafa37485eb46efa152',
  26             'timestamp': 1232520922,
  27             'upload_date': '20090121',
  28             'duration': 1838,
  29             'view_count': int,
  30             'age_limit': 18,
  31         },
  32     }, {
  33         # New (May 2016) URL layout
  34         'url': 'http://www.eporner.com/hd-porn/3YRUtzMcWn0/Star-Wars-XXX-Parody/',
  35         'only_matching': True,
  36     }, {
  37         'url': 'http://www.eporner.com/hd-porn/3YRUtzMcWn0',
  38         'only_matching': True,
  39     }, {
  40         'url': 'http://www.eporner.com/embed/3YRUtzMcWn0',
  41         'only_matching': True,
  42     }, {
  43         'url': 'https://www.eporner.com/video-FJsA19J3Y3H/one-of-the-greats/',
  44         'only_matching': True,
  45     }]
  46
  47     def _real_extract(self, url):
  48         mobj = self._match_valid_url(url)
  49         video_id = mobj.group('id')
  50         display_id = mobj.group('display_id') or video_id
  51
  52         webpage, urlh = self._download_webpage_handle(url, display_id)
  53
  54         video_id = self._match_id(urlh.url)
  55
  56         vid_hash = self._search_regex(
  57             r'hash\s*[:=]\s*["\']([\da-f]{32})', webpage, 'hash')
  58
  59         title = self._og_search_title(webpage, default=None) or self._html_search_regex(
  60             r'<title>(.+?) - EPORNER', webpage, 'title')
  61
  62         # Reverse engineered from vjs.js
  63         def calc_hash(s):
  64             return ''.join(encode_base_n(int(s[lb:lb + 8], 16), 36) for lb in range(0, 32, 8))
  65
  66         video = self._download_json(
  67             f'http://www.eporner.com/xhr/video/{video_id}',
  68             display_id, note='Downloading video JSON',
  69             query={
  70                 'hash': calc_hash(vid_hash),
  71                 'device': 'generic',
  72                 'domain': 'www.eporner.com',
  73                 'fallback': 'false',
  74             })
  75
  76         if video.get('available') is False:
  77             raise ExtractorError(
  78                 '{} said: {}'.format(self.IE_NAME, video['message']), expected=True)
  79
  80         sources = video['sources']
  81
  82         formats = []
  83         has_av1 = bool(get_elements_by_class('download-av1', webpage))
  84         for kind, formats_dict in sources.items():
  85             if not isinstance(formats_dict, dict):
  86                 continue
  87             for format_id, format_dict in formats_dict.items():
  88                 if not isinstance(format_dict, dict):
  89                     continue
  90                 src = url_or_none(format_dict.get('src'))
  91                 if not src or not src.startswith('http'):
  92                     continue
  93                 if kind == 'hls':
  94                     formats.extend(self._extract_m3u8_formats(
  95                         src, display_id, 'mp4', entry_protocol='m3u8_native',
  96                         m3u8_id=kind, fatal=False))
  97                 else:
  98                     height = int_or_none(self._search_regex(
  99                         r'(\d+)[pP]', format_id, 'height', default=None))
 100                     fps = int_or_none(self._search_regex(
 101                         r'(\d+)fps', format_id, 'fps', default=None))
 102
 103                     formats.append({
 104                         'url': src,
 105                         'format_id': format_id,
 106                         'height': height,
 107                         'fps': fps,
 108                     })
 109                     if has_av1:
 110                         formats.append({
 111                             'url': src.replace('.mp4', '-av1.mp4'),
 112                             'format_id': join_nonempty('av1', format_id),
 113                             'height': height,
 114                             'fps': fps,
 115                             'vcodec': 'av1',
 116                         })
 117
 118         json_ld = self._search_json_ld(webpage, display_id, default={})
 119
 120         duration = parse_duration(self._html_search_meta(
 121             'duration', webpage, default=None))
 122         view_count = str_to_int(self._search_regex(
 123             r'id=["\']cinemaviews1["\'][^>]*>\s*([0-9,]+)',
 124             webpage, 'view count', default=None))
 125
 126         return merge_dicts(json_ld, {
 127             'id': video_id,
 128             'display_id': display_id,
 129             'title': title,
 130             'duration': duration,
 131             'view_count': view_count,
 132             'formats': formats,
 133             'age_limit': 18,
 134         })