yt_dlp/extractor/eporner.py

   1 from .common import InfoExtractor
   2 from ..utils import (
   3     encode_base_n,
   4     ExtractorError,
   5     int_or_none,
   6     merge_dicts,
   7     parse_duration,
   8     str_to_int,
   9     url_or_none,
  10 )
  11
  12
  13 class EpornerIE(InfoExtractor):
  14     _VALID_URL = r'https?://(?:www\.)?eporner\.com/(?:(?:hd-porn|embed)/|video-)(?P<id>\w+)(?:/(?P<display_id>[\w-]+))?'
  15     _TESTS = [{
  16         'url': 'http://www.eporner.com/hd-porn/95008/Infamous-Tiffany-Teen-Strip-Tease-Video/',
  17         'md5': '39d486f046212d8e1b911c52ab4691f8',
  18         'info_dict': {
  19             'id': 'qlDUmNsj6VS',
  20             'display_id': 'Infamous-Tiffany-Teen-Strip-Tease-Video',
  21             'ext': 'mp4',
  22             'title': 'Infamous Tiffany Teen Strip Tease Video',
  23             'description': 'md5:764f39abf932daafa37485eb46efa152',
  24             'timestamp': 1232520922,
  25             'upload_date': '20090121',
  26             'duration': 1838,
  27             'view_count': int,
  28             'age_limit': 18,
  29         },
  30         'params': {
  31             'proxy': '127.0.0.1:8118'
  32         }
  33     }, {
  34         # New (May 2016) URL layout
  35         'url': 'http://www.eporner.com/hd-porn/3YRUtzMcWn0/Star-Wars-XXX-Parody/',
  36         'only_matching': True,
  37     }, {
  38         'url': 'http://www.eporner.com/hd-porn/3YRUtzMcWn0',
  39         'only_matching': True,
  40     }, {
  41         'url': 'http://www.eporner.com/embed/3YRUtzMcWn0',
  42         'only_matching': True,
  43     }, {
  44         'url': 'https://www.eporner.com/video-FJsA19J3Y3H/one-of-the-greats/',
  45         'only_matching': True,
  46     }]
  47
  48     def _real_extract(self, url):
  49         mobj = self._match_valid_url(url)
  50         video_id = mobj.group('id')
  51         display_id = mobj.group('display_id') or video_id
  52
  53         webpage, urlh = self._download_webpage_handle(url, display_id)
  54
  55         video_id = self._match_id(urlh.url)
  56
  57         hash = self._search_regex(
  58             r'hash\s*[:=]\s*["\']([\da-f]{32})', webpage, 'hash')
  59
  60         title = self._og_search_title(webpage, default=None) or self._html_search_regex(
  61             r'<title>(.+?) - EPORNER', webpage, 'title')
  62
  63         # Reverse engineered from vjs.js
  64         def calc_hash(s):
  65             return ''.join((encode_base_n(int(s[lb:lb + 8], 16), 36) for lb in range(0, 32, 8)))
  66
  67         video = self._download_json(
  68             'http://www.eporner.com/xhr/video/%s' % video_id,
  69             display_id, note='Downloading video JSON',
  70             query={
  71                 'hash': calc_hash(hash),
  72                 'device': 'generic',
  73                 'domain': 'www.eporner.com',
  74                 'fallback': 'false',
  75             })
  76
  77         if video.get('available') is False:
  78             raise ExtractorError(
  79                 '%s said: %s' % (self.IE_NAME, video['message']), expected=True)
  80
  81         sources = video['sources']
  82
  83         formats = []
  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
 110         json_ld = self._search_json_ld(webpage, display_id, default={})
 111
 112         duration = parse_duration(self._html_search_meta(
 113             'duration', webpage, default=None))
 114         view_count = str_to_int(self._search_regex(
 115             r'id=["\']cinemaviews1["\'][^>]*>\s*([0-9,]+)',
 116             webpage, 'view count', default=None))
 117
 118         return merge_dicts(json_ld, {
 119             'id': video_id,
 120             'display_id': display_id,
 121             'title': title,
 122             'duration': duration,
 123             'view_count': view_count,
 124             'formats': formats,
 125             'age_limit': 18,
 126         })