_ Git - youtube-dl/blob - youtube_dl/extractor/moviefap.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from .common import InfoExtractor
   6 from ..utils import str_to_int
   7
   8
   9 class MovieFapIE(InfoExtractor):
  10     _VALID_URL = r'https?://(?:www\.)?moviefap\.com/videos/(?P<id>[0-9a-f]+)/(?P<name>[a-z-_]+)'
  11     _TESTS = [{
  12         'url': 'http://www.moviefap.com/videos/e5da0d3edce5404418f5/jeune-couple-russe.html',
  13         'md5': 'fa56683e291fc80635907168a743c9ad',
  14         'info_dict': {
  15             'id': 'e5da0d3edce5404418f5',
  16             'ext': 'flv',
  17             'title': 'Jeune Couple Russe',
  18             'description': 'Amateur',
  19             'thumbnail': 'http://pic.moviefap.com/thumbs/e5/949-18l.jpg',
  20             'uploader_id': 'whiskeyjar',
  21             'display_id': 'jeune-couple-russe'
  22         }
  23     }, {
  24         'url': 'http://www.moviefap.com/videos/3080837f6712355015c2/busty-british-blonde-takes-backdoor-in-fake-taxi.html',
  25         'md5': 'bedef72cb23d27a20755fc430a6d7a0e',
  26         'info_dict': {
  27             'id': '3080837f6712355015c2',
  28             'ext': 'mp4',
  29             'title': 'Busty British blonde takes backdoor in fake taxi',
  30             'description': 'Big boobs British blonde flashing in fake taxi then giving titsjob and rimjob in the back seat before getting big cock up her tight ass',
  31             'thumbnail': 'http://img.moviefap.com/a16:9w990r/thumbs/30/322021-18l.jpg',
  32             'uploader_id': 'momcikoper',
  33             'display_id': 'busty-british-blonde-takes-backdoor-in-fake-taxi'
  34         }
  35     }]
  36
  37     @staticmethod
  38     def __get_thumbnail_data(xml):
  39
  40         """
  41         Constructs a list of video thumbnails from timeline preview images.
  42         :param xml: the information XML document to parse
  43         """
  44
  45         timeline = xml.find('timeline')
  46         if timeline is None:
  47             # not all videos have the data - ah well
  48             return []
  49
  50         # get the required information from the XML
  51         attrs = {attr: str_to_int(timeline.find(attr).text)
  52                  for attr in ['imageWidth', 'imageHeight', 'imageFirst', 'imageLast']}
  53         pattern = timeline.find('imagePattern').text
  54
  55         # generate the list of thumbnail information dicts
  56         thumbnails = []
  57         for i in range(attrs['imageFirst'], attrs['imageLast'] + 1):
  58             thumbnails.append({
  59                 'url': pattern.replace('#', str(i)),
  60                 'width': attrs['imageWidth'],
  61                 'height': attrs['imageHeight']
  62             })
  63         return thumbnails
  64
  65     def _real_extract(self, url):
  66
  67         # find the video ID
  68         video_id = self._match_id(url)
  69
  70         # retrieve the page HTML
  71         webpage = self._download_webpage(url, video_id)
  72
  73         # find the URL of the XML document detailing video download URLs
  74         info_url = self._html_search_regex(r'flashvars\.config = escape\("(.+?)"', webpage, 'player parameters')
  75
  76         # download that XML
  77         xml = self._download_xml(info_url, video_id)
  78
  79         # create dictionary of properties we know so far, or can find easily
  80         info = {
  81             'id': video_id,
  82             'title': self._html_search_regex(r'<div id="view_title"><h1>(.*?)</h1>', webpage, 'title'),
  83             'display_id': re.compile(self._VALID_URL).match(url).group('name'),
  84             'thumbnails': self.__get_thumbnail_data(xml),
  85             'thumbnail': xml.find('startThumb').text,
  86             'description': self._html_search_regex(r'name="description" value="(.*?)"', webpage, 'description'),
  87             'uploader_id': self._html_search_regex(r'name="username" value="(.*?)"', webpage, 'uploader_id'),
  88             'view_count': str_to_int(self._html_search_regex(r'<br>Views <strong>([0-9]+)</strong>', webpage, 'view_count')),
  89             'average_rating': float(self._html_search_regex(r'Current Rating<br> <strong>(.*?)</strong>', webpage, 'average_rating')),
  90             'comment_count': str_to_int(self._html_search_regex(r'<span id="comCount">([0-9]+)</span>', webpage, 'comment_count')),
  91             'age_limit': 18,
  92             'webpage_url': self._html_search_regex(r'name="link" value="(.*?)"', webpage, 'webpage_url'),
  93             'categories': self._html_search_regex(r'</div>\s*(.*?)\s*<br>', webpage, 'categories').split(', ')
  94         }
  95
  96         # find and add the format
  97         if xml.find('videoConfig') is not None:
  98             info['ext'] = xml.find('videoConfig').find('type').text
  99         else:
 100             info['ext'] = 'flv'  # guess...
 101
 102         # work out the video URL(s)
 103         if xml.find('videoLink') is not None:
 104             # single format available
 105             info['url'] = xml.find('videoLink').text
 106         else:
 107             # multiple formats available
 108             info['formats'] = []
 109
 110             # N.B. formats are already in ascending order of quality
 111             for item in xml.find('quality').findall('item'):
 112                 info['formats'].append({
 113                     'url': item.find('videoLink').text,
 114                     'resolution': item.find('res').text  # 480p etc.
 115                 })
 116
 117         return info