Raphaël G. Git Repositories - youtubedl/blob - 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 ..compat import (
   8     compat_urllib_parse,
   9     compat_urllib_parse_urlparse,
  10     compat_urllib_request,
  11 )
  12 from ..utils import (
  13     ExtractorError,
  14     str_to_int,
  15 )
  16 from ..aes import (
  17     aes_decrypt_text
  18 )
  19
  20
  21 class PornHubIE(InfoExtractor):
  22     _VALID_URL = r'https?://(?:www\.)?pornhub\.com/view_video\.php\?viewkey=(?P<id>[0-9a-f]+)'
  23     _TEST = {
  24         'url': 'http://www.pornhub.com/view_video.php?viewkey=648719015',
  25         'md5': '882f488fa1f0026f023f33576004a2ed',
  26         'info_dict': {
  27             'id': '648719015',
  28             'ext': 'mp4',
  29             "uploader": "Babes",
  30             "title": "Seductive Indian beauty strips down and fingers her pink pussy",
  31             "age_limit": 18
  32         }
  33     }
  34
  35     def _extract_count(self, pattern, webpage, name):
  36         count = self._html_search_regex(pattern, webpage, '%s count' % name, fatal=False)
  37         if count:
  38             count = str_to_int(count)
  39         return count
  40
  41     def _real_extract(self, url):
  42         video_id = self._match_id(url)
  43
  44         req = compat_urllib_request.Request(url)
  45         req.add_header('Cookie', 'age_verified=1')
  46         webpage = self._download_webpage(req, video_id)
  47
  48         error_msg = self._html_search_regex(
  49             r'(?s)<div class="userMessageSection[^"]*".*?>(.*?)</div>',
  50             webpage, 'error message', default=None)
  51         if error_msg:
  52             error_msg = re.sub(r'\s+', ' ', error_msg)
  53             raise ExtractorError(
  54                 'PornHub said: %s' % error_msg,
  55                 expected=True, video_id=video_id)
  56
  57         video_title = self._html_search_regex(r'<h1 [^>]+>([^<]+)', webpage, 'title')
  58         video_uploader = self._html_search_regex(
  59             r'(?s)From:&nbsp;.+?<(?:a href="/users/|a href="/channels/|span class="username)[^>]+>(.+?)<',
  60             webpage, 'uploader', fatal=False)
  61         thumbnail = self._html_search_regex(r'"image_url":"([^"]+)', webpage, 'thumbnail', fatal=False)
  62         if thumbnail:
  63             thumbnail = compat_urllib_parse.unquote(thumbnail)
  64
  65         view_count = self._extract_count(r'<span class="count">([\d,\.]+)</span> views', webpage, 'view')
  66         like_count = self._extract_count(r'<span class="votesUp">([\d,\.]+)</span>', webpage, 'like')
  67         dislike_count = self._extract_count(r'<span class="votesDown">([\d,\.]+)</span>', webpage, 'dislike')
  68         comment_count = self._extract_count(
  69             r'All comments \(<var class="videoCommentCount">([\d,\.]+)</var>', webpage, 'comment')
  70
  71         video_urls = list(map(compat_urllib_parse.unquote, re.findall(r'"quality_[0-9]{3}p":"([^"]+)', webpage)))
  72         if webpage.find('"encrypted":true') != -1:
  73             password = compat_urllib_parse.unquote_plus(self._html_search_regex(r'"video_title":"([^"]+)', webpage, 'password'))
  74             video_urls = list(map(lambda s: aes_decrypt_text(s, password, 32).decode('utf-8'), video_urls))
  75
  76         formats = []
  77         for video_url in video_urls:
  78             path = compat_urllib_parse_urlparse(video_url).path
  79             extension = os.path.splitext(path)[1][1:]
  80             format = path.split('/')[5].split('_')[:2]
  81             format = "-".join(format)
  82
  83             m = re.match(r'^(?P<height>[0-9]+)P-(?P<tbr>[0-9]+)K$', format)
  84             if m is None:
  85                 height = None
  86                 tbr = None
  87             else:
  88                 height = int(m.group('height'))
  89                 tbr = int(m.group('tbr'))
  90
  91             formats.append({
  92                 'url': video_url,
  93                 'ext': extension,
  94                 'format': format,
  95                 'format_id': format,
  96                 'tbr': tbr,
  97                 'height': height,
  98             })
  99         self._sort_formats(formats)
 100
 101         return {
 102             'id': video_id,
 103             'uploader': video_uploader,
 104             'title': video_title,
 105             'thumbnail': thumbnail,
 106             'view_count': view_count,
 107             'like_count': like_count,
 108             'dislike_count': dislike_count,
 109             'comment_count': comment_count,
 110             'formats': formats,
 111             'age_limit': 18,
 112         }
 113
 114
 115 class PornHubPlaylistIE(InfoExtractor):
 116     _VALID_URL = r'https?://(?:www\.)?pornhub\.com/playlist/(?P<id>\d+)'
 117     _TESTS = [{
 118         'url': 'http://www.pornhub.com/playlist/6201671',
 119         'info_dict': {
 120             'id': '6201671',
 121             'title': 'P0p4',
 122         },
 123         'playlist_mincount': 35,
 124     }]
 125
 126     def _real_extract(self, url):
 127         playlist_id = self._match_id(url)
 128
 129         webpage = self._download_webpage(url, playlist_id)
 130
 131         entries = [
 132             self.url_result('http://www.pornhub.com/%s' % video_url, 'PornHub')
 133             for video_url in set(re.findall('href="/?(view_video\.php\?viewkey=\d+[^"]*)"', webpage))
 134         ]
 135
 136         playlist = self._parse_json(
 137             self._search_regex(
 138                 r'playlistObject\s*=\s*({.+?});', webpage, 'playlist'),
 139             playlist_id)
 140
 141         return self.playlist_result(
 142             entries, playlist_id, playlist.get('title'), playlist.get('description'))