_ Git - youtube-dl/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 )
  15
  16
  17 class EpornerIE(InfoExtractor):
  18     _VALID_URL = r'https?://(?:www\.)?eporner\.com/hd-porn/(?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             'duration': 1838,
  28             'view_count': int,
  29             'age_limit': 18,
  30         },
  31     }, {
  32         # New (May 2016) URL layout
  33         'url': 'http://www.eporner.com/hd-porn/3YRUtzMcWn0/Star-Wars-XXX-Parody/',
  34         'only_matching': True,
  35     }, {
  36         'url': 'http://www.eporner.com/hd-porn/3YRUtzMcWn0',
  37         'only_matching': True,
  38     }]
  39
  40     def _real_extract(self, url):
  41         mobj = re.match(self._VALID_URL, url)
  42         video_id = mobj.group('id')
  43         display_id = mobj.group('display_id') or video_id
  44
  45         webpage, urlh = self._download_webpage_handle(url, display_id)
  46
  47         video_id = self._match_id(compat_str(urlh.geturl()))
  48
  49         hash = self._search_regex(
  50             r'hash\s*:\s*["\']([\da-f]{32})', webpage, 'hash')
  51
  52         title = self._og_search_title(webpage, default=None) or self._html_search_regex(
  53             r'<title>(.+?) - EPORNER', webpage, 'title')
  54
  55         # Reverse engineered from vjs.js
  56         def calc_hash(s):
  57             return ''.join((encode_base_n(int(s[lb:lb + 8], 16), 36) for lb in range(0, 32, 8)))
  58
  59         video = self._download_json(
  60             'http://www.eporner.com/xhr/video/%s' % video_id,
  61             display_id, note='Downloading video JSON',
  62             query={
  63                 'hash': calc_hash(hash),
  64                 'device': 'generic',
  65                 'domain': 'www.eporner.com',
  66                 'fallback': 'false',
  67             })
  68
  69         if video.get('available') is False:
  70             raise ExtractorError(
  71                 '%s said: %s' % (self.IE_NAME, video['message']), expected=True)
  72
  73         sources = video['sources']
  74
  75         formats = []
  76         for kind, formats_dict in sources.items():
  77             if not isinstance(formats_dict, dict):
  78                 continue
  79             for format_id, format_dict in formats_dict.items():
  80                 if not isinstance(format_dict, dict):
  81                     continue
  82                 src = format_dict.get('src')
  83                 if not isinstance(src, compat_str) or not src.startswith('http'):
  84                     continue
  85                 if kind == 'hls':
  86                     formats.extend(self._extract_m3u8_formats(
  87                         src, display_id, 'mp4', entry_protocol='m3u8_native',
  88                         m3u8_id=kind, fatal=False))
  89                 else:
  90                     height = int_or_none(self._search_regex(
  91                         r'(\d+)[pP]', format_id, 'height', default=None))
  92                     fps = int_or_none(self._search_regex(
  93                         r'(\d+)fps', format_id, 'fps', default=None))
  94
  95                     formats.append({
  96                         'url': src,
  97                         'format_id': format_id,
  98                         'height': height,
  99                         'fps': fps,
 100                     })
 101         self._sort_formats(formats)
 102
 103         duration = parse_duration(self._html_search_meta('duration', webpage))
 104         view_count = str_to_int(self._search_regex(
 105             r'id="cinemaviews">\s*([0-9,]+)\s*<small>views',
 106             webpage, 'view count', fatal=False))
 107
 108         return {
 109             'id': video_id,
 110             'display_id': display_id,
 111             'title': title,
 112             'duration': duration,
 113             'view_count': view_count,
 114             'formats': formats,
 115             'age_limit': 18,
 116         }