yt_dlp/extractor/eporner.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4
   5 from .common import InfoExtractor
   6 from ..utils import (
   7     encode_base_n,
   8     ExtractorError,
   9     int_or_none,
  10     merge_dicts,
  11     parse_duration,
  12     str_to_int,
  13     url_or_none,
  14 )
  15
  16
  17 class EpornerIE(InfoExtractor):
  18     _VALID_URL = r'https?://(?:www\.)?eporner\.com/(?:(?:hd-porn|embed)/|video-)(?P<id>\w+)(?:/(?P<display_id>[\w-]+))?'
  19     _TESTS = [{
  20         'url': 'http://www.eporner.com/hd-porn/95008/Infamous-Tiffany-Teen-Strip-Tease-Video/',
  21         'md5': '39d486f046212d8e1b911c52ab4691f8',
  22         'info_dict': {
  23             'id': 'qlDUmNsj6VS',
  24             'display_id': 'Infamous-Tiffany-Teen-Strip-Tease-Video',
  25             'ext': 'mp4',
  26             'title': 'Infamous Tiffany Teen Strip Tease Video',
  27             'description': 'md5:764f39abf932daafa37485eb46efa152',
  28             'timestamp': 1232520922,
  29             'upload_date': '20090121',
  30             'duration': 1838,
  31             'view_count': int,
  32             'age_limit': 18,
  33         },
  34         'params': {
  35             'proxy': '127.0.0.1:8118'
  36         }
  37     }, {
  38         # New (May 2016) URL layout
  39         'url': 'http://www.eporner.com/hd-porn/3YRUtzMcWn0/Star-Wars-XXX-Parody/',
  40         'only_matching': True,
  41     }, {
  42         'url': 'http://www.eporner.com/hd-porn/3YRUtzMcWn0',
  43         'only_matching': True,
  44     }, {
  45         'url': 'http://www.eporner.com/embed/3YRUtzMcWn0',
  46         'only_matching': True,
  47     }, {
  48         'url': 'https://www.eporner.com/video-FJsA19J3Y3H/one-of-the-greats/',
  49         'only_matching': True,
  50     }]
  51
  52     def _real_extract(self, url):
  53         mobj = self._match_valid_url(url)
  54         video_id = mobj.group('id')
  55         display_id = mobj.group('display_id') or video_id
  56
  57         webpage, urlh = self._download_webpage_handle(url, display_id)
  58
  59         video_id = self._match_id(urlh.geturl())
  60
  61         hash = self._search_regex(
  62             r'hash\s*[:=]\s*["\']([\da-f]{32})', webpage, 'hash')
  63
  64         title = self._og_search_title(webpage, default=None) or self._html_search_regex(
  65             r'<title>(.+?) - EPORNER', webpage, 'title')
  66
  67         # Reverse engineered from vjs.js
  68         def calc_hash(s):
  69             return ''.join((encode_base_n(int(s[lb:lb + 8], 16), 36) for lb in range(0, 32, 8)))
  70
  71         video = self._download_json(
  72             'http://www.eporner.com/xhr/video/%s' % video_id,
  73             display_id, note='Downloading video JSON',
  74             query={
  75                 'hash': calc_hash(hash),
  76                 'device': 'generic',
  77                 'domain': 'www.eporner.com',
  78                 'fallback': 'false',
  79             })
  80
  81         if video.get('available') is False:
  82             raise ExtractorError(
  83                 '%s said: %s' % (self.IE_NAME, video['message']), expected=True)
  84
  85         sources = video['sources']
  86
  87         formats = []
  88         for kind, formats_dict in sources.items():
  89             if not isinstance(formats_dict, dict):
  90                 continue
  91             for format_id, format_dict in formats_dict.items():
  92                 if not isinstance(format_dict, dict):
  93                     continue
  94                 src = url_or_none(format_dict.get('src'))
  95                 if not src or not src.startswith('http'):
  96                     continue
  97                 if kind == 'hls':
  98                     formats.extend(self._extract_m3u8_formats(
  99                         src, display_id, 'mp4', entry_protocol='m3u8_native',
 100                         m3u8_id=kind, fatal=False))
 101                 else:
 102                     height = int_or_none(self._search_regex(
 103                         r'(\d+)[pP]', format_id, 'height', default=None))
 104                     fps = int_or_none(self._search_regex(
 105                         r'(\d+)fps', format_id, 'fps', default=None))
 106
 107                     formats.append({
 108                         'url': src,
 109                         'format_id': format_id,
 110                         'height': height,
 111                         'fps': fps,
 112                     })
 113         self._sort_formats(formats)
 114
 115         json_ld = self._search_json_ld(webpage, display_id, default={})
 116
 117         duration = parse_duration(self._html_search_meta(
 118             'duration', webpage, default=None))
 119         view_count = str_to_int(self._search_regex(
 120             r'id=["\']cinemaviews1["\'][^>]*>\s*([0-9,]+)',
 121             webpage, 'view count', default=None))
 122
 123         return merge_dicts(json_ld, {
 124             'id': video_id,
 125             'display_id': display_id,
 126             'title': title,
 127             'duration': duration,
 128             'view_count': view_count,
 129             'formats': formats,
 130             'age_limit': 18,
 131         })