youtube_dl/extractor/xnxx.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     determine_ext,
   9     int_or_none,
  10     NO_DEFAULT,
  11     str_to_int,
  12 )
  13
  14
  15 class XNXXIE(InfoExtractor):
  16     _VALID_URL = r'https?://(?:video|www)\.xnxx\.com/video-?(?P<id>[0-9a-z]+)/'
  17     _TESTS = [{
  18         'url': 'http://www.xnxx.com/video-55awb78/skyrim_test_video',
  19         'md5': '7583e96c15c0f21e9da3453d9920fbba',
  20         'info_dict': {
  21             'id': '55awb78',
  22             'ext': 'mp4',
  23             'title': 'Skyrim Test Video',
  24             'thumbnail': r're:^https?://.*\.jpg',
  25             'duration': 469,
  26             'view_count': int,
  27             'age_limit': 18,
  28         },
  29     }, {
  30         'url': 'http://video.xnxx.com/video1135332/lida_naked_funny_actress_5_',
  31         'only_matching': True,
  32     }, {
  33         'url': 'http://www.xnxx.com/video-55awb78/',
  34         'only_matching': True,
  35     }]
  36
  37     def _real_extract(self, url):
  38         video_id = self._match_id(url)
  39
  40         webpage = self._download_webpage(url, video_id)
  41
  42         def get(meta, default=NO_DEFAULT, fatal=True):
  43             return self._search_regex(
  44                 r'set%s\s*\(\s*(["\'])(?P<value>(?:(?!\1).)+)\1' % meta,
  45                 webpage, meta, default=default, fatal=fatal, group='value')
  46
  47         title = self._og_search_title(
  48             webpage, default=None) or get('VideoTitle')
  49
  50         formats = []
  51         for mobj in re.finditer(
  52                 r'setVideo(?:Url(?P<id>Low|High)|HLS)\s*\(\s*(?P<q>["\'])(?P<url>(?:https?:)?//.+?)(?P=q)', webpage):
  53             format_url = mobj.group('url')
  54             if determine_ext(format_url) == 'm3u8':
  55                 formats.extend(self._extract_m3u8_formats(
  56                     format_url, video_id, 'mp4', entry_protocol='m3u8_native',
  57                     preference=1, m3u8_id='hls', fatal=False))
  58             else:
  59                 format_id = mobj.group('id')
  60                 if format_id:
  61                     format_id = format_id.lower()
  62                 formats.append({
  63                     'url': format_url,
  64                     'format_id': format_id,
  65                     'quality': -1 if format_id == 'low' else 0,
  66                 })
  67         self._sort_formats(formats)
  68
  69         thumbnail = self._og_search_thumbnail(webpage, default=None) or get(
  70             'ThumbUrl', fatal=False) or get('ThumbUrl169', fatal=False)
  71         duration = int_or_none(self._og_search_property('duration', webpage))
  72         view_count = str_to_int(self._search_regex(
  73             r'id=["\']nb-views-number[^>]+>([\d,.]+)', webpage, 'view count',
  74             default=None))
  75
  76         return {
  77             'id': video_id,
  78             'title': title,
  79             'thumbnail': thumbnail,
  80             'duration': duration,
  81             'view_count': view_count,
  82             'age_limit': 18,
  83             'formats': formats,
  84         }