Raphaël G. Git Repositories - youtubedl/blob - 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     parse_duration,
  13     str_to_int,
  14     url_or_none,
  15 )
  16
  17
  18 class EpornerIE(InfoExtractor):
  19     _VALID_URL = r'https?://(?:www\.)?eporner\.com/(?:hd-porn|embed)/(?P<id>\w+)(?:/(?P<display_id>[\w-]+))?'
  20     _TESTS = [{
  21         'url': 'http://www.eporner.com/hd-porn/95008/Infamous-Tiffany-Teen-Strip-Tease-Video/',
  22         'md5': '39d486f046212d8e1b911c52ab4691f8',
  23         'info_dict': {
  24             'id': 'qlDUmNsj6VS',
  25             'display_id': 'Infamous-Tiffany-Teen-Strip-Tease-Video',
  26             'ext': 'mp4',
  27             'title': 'Infamous Tiffany Teen Strip Tease Video',
  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/hd-porn/3YRUtzMcWn0',
  41         'only_matching': True,
  42     }]
  43
  44     def _real_extract(self, url):
  45         mobj = re.match(self._VALID_URL, url)
  46         video_id = mobj.group('id')
  47         display_id = mobj.group('display_id') or video_id
  48
  49         webpage, urlh = self._download_webpage_handle(url, display_id)
  50
  51         video_id = self._match_id(compat_str(urlh.geturl()))
  52
  53         hash = self._search_regex(
  54             r'hash\s*:\s*["\']([\da-f]{32})', webpage, 'hash')
  55
  56         title = self._og_search_title(webpage, default=None) or self._html_search_regex(
  57             r'<title>(.+?) - EPORNER', webpage, 'title')
  58
  59         # Reverse engineered from vjs.js
  60         def calc_hash(s):
  61             return ''.join((encode_base_n(int(s[lb:lb + 8], 16), 36) for lb in range(0, 32, 8)))
  62
  63         video = self._download_json(
  64             'http://www.eporner.com/xhr/video/%s' % video_id,
  65             display_id, note='Downloading video JSON',
  66             query={
  67                 'hash': calc_hash(hash),
  68                 'device': 'generic',
  69                 'domain': 'www.eporner.com',
  70                 'fallback': 'false',
  71             })
  72
  73         if video.get('available') is False:
  74             raise ExtractorError(
  75                 '%s said: %s' % (self.IE_NAME, video['message']), expected=True)
  76
  77         sources = video['sources']
  78
  79         formats = []
  80         for kind, formats_dict in sources.items():
  81             if not isinstance(formats_dict, dict):
  82                 continue
  83             for format_id, format_dict in formats_dict.items():
  84                 if not isinstance(format_dict, dict):
  85                     continue
  86                 src = url_or_none(format_dict.get('src'))
  87                 if not src or not src.startswith('http'):
  88                     continue
  89                 if kind == 'hls':
  90                     formats.extend(self._extract_m3u8_formats(
  91                         src, display_id, 'mp4', entry_protocol='m3u8_native',
  92                         m3u8_id=kind, fatal=False))
  93                 else:
  94                     height = int_or_none(self._search_regex(
  95                         r'(\d+)[pP]', format_id, 'height', default=None))
  96                     fps = int_or_none(self._search_regex(
  97                         r'(\d+)fps', format_id, 'fps', default=None))
  98
  99                     formats.append({
 100                         'url': src,
 101                         'format_id': format_id,
 102                         'height': height,
 103                         'fps': fps,
 104                     })
 105         self._sort_formats(formats)
 106
 107         duration = parse_duration(self._html_search_meta('duration', webpage))
 108         view_count = str_to_int(self._search_regex(
 109             r'id="cinemaviews">\s*([0-9,]+)\s*<small>views',
 110             webpage, 'view count', fatal=False))
 111
 112         return {
 113             'id': video_id,
 114             'display_id': display_id,
 115             'title': title,
 116             'duration': duration,
 117             'view_count': view_count,
 118             'formats': formats,
 119             'age_limit': 18,
 120         }