youtube_dl/extractor/eporner.py

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