youtube_dl/extractor/pornhub.py

   1 from __future__ import unicode_literals
   2
   3 import os
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     compat_urllib_parse_urlparse,
   9     compat_urllib_request,
  10     compat_urllib_parse,
  11     str_to_int,
  12 )
  13 from ..aes import (
  14     aes_decrypt_text
  15 )
  16
  17
  18 class PornHubIE(InfoExtractor):
  19     _VALID_URL = r'^(?:https?://)?(?:www\.)?(?P<url>pornhub\.com/view_video\.php\?viewkey=(?P<videoid>[0-9a-f]+))'
  20     _TEST = {
  21         'url': 'http://www.pornhub.com/view_video.php?viewkey=648719015',
  22         'file': '648719015.mp4',
  23         'md5': '882f488fa1f0026f023f33576004a2ed',
  24         'info_dict': {
  25             "uploader": "BABES-COM",
  26             "title": "Seductive Indian beauty strips down and fingers her pink pussy",
  27             "age_limit": 18
  28         }
  29     }
  30
  31     def _extract_count(self, pattern, webpage, name):
  32         count = self._html_search_regex(pattern, webpage, '%s count' % name, fatal=False)
  33         if count:
  34             count = str_to_int(count)
  35         return count
  36
  37     def _real_extract(self, url):
  38         mobj = re.match(self._VALID_URL, url)
  39         video_id = mobj.group('videoid')
  40         url = 'http://www.' + mobj.group('url')
  41
  42         req = compat_urllib_request.Request(url)
  43         req.add_header('Cookie', 'age_verified=1')
  44         webpage = self._download_webpage(req, video_id)
  45
  46         video_title = self._html_search_regex(r'<h1 [^>]+>([^<]+)', webpage, 'title')
  47         video_uploader = self._html_search_regex(
  48             r'(?s)From:&nbsp;.+?<(?:a href="/users/|<span class="username)[^>]+>(.+?)<',
  49             webpage, 'uploader', fatal=False)
  50         thumbnail = self._html_search_regex(r'"image_url":"([^"]+)', webpage, 'thumbnail', fatal=False)
  51         if thumbnail:
  52             thumbnail = compat_urllib_parse.unquote(thumbnail)
  53
  54         view_count = self._extract_count(r'<span class="count">([\d,\.]+)</span> views', webpage, 'view')
  55         like_count = self._extract_count(r'<span class="votesUp">([\d,\.]+)</span>', webpage, 'like')
  56         dislike_count = self._extract_count(r'<span class="votesDown">([\d,\.]+)</span>', webpage, 'dislike')
  57         comment_count = self._extract_count(
  58             r'All comments \(<var class="videoCommentCount">([\d,\.]+)</var>', webpage, 'comment')
  59
  60         video_urls = list(map(compat_urllib_parse.unquote , re.findall(r'"quality_[0-9]{3}p":"([^"]+)', webpage)))
  61         if webpage.find('"encrypted":true') != -1:
  62             password = compat_urllib_parse.unquote_plus(self._html_search_regex(r'"video_title":"([^"]+)', webpage, 'password'))
  63             video_urls = list(map(lambda s: aes_decrypt_text(s, password, 32).decode('utf-8'), video_urls))
  64
  65         formats = []
  66         for video_url in video_urls:
  67             path = compat_urllib_parse_urlparse(video_url).path
  68             extension = os.path.splitext(path)[1][1:]
  69             format = path.split('/')[5].split('_')[:2]
  70             format = "-".join(format)
  71
  72             m = re.match(r'^(?P<height>[0-9]+)P-(?P<tbr>[0-9]+)K$', format)
  73             if m is None:
  74                 height = None
  75                 tbr = None
  76             else:
  77                 height = int(m.group('height'))
  78                 tbr = int(m.group('tbr'))
  79
  80             formats.append({
  81                 'url': video_url,
  82                 'ext': extension,
  83                 'format': format,
  84                 'format_id': format,
  85                 'tbr': tbr,
  86                 'height': height,
  87             })
  88         self._sort_formats(formats)
  89
  90         return {
  91             'id': video_id,
  92             'uploader': video_uploader,
  93             'title': video_title,
  94             'thumbnail': thumbnail,
  95             'view_count': view_count,
  96             'like_count': like_count,
  97             'dislike_count': dislike_count,
  98             'comment_count': comment_count,
  99             'formats': formats,
 100             'age_limit': 18,
 101         }