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         # normal, multi-format video
  13         'url': 'http://www.moviefap.com/videos/be9867c9416c19f54a4a/experienced-milf-amazing-handjob.html',
  14         'md5': '26624b4e2523051b550067d547615906',
  15         'info_dict': {
  16             'id': 'be9867c9416c19f54a4a',
  17             'ext': 'mp4',
  18             'title': 'Experienced MILF Amazing Handjob',
  19             'description': 'Experienced MILF giving an Amazing Handjob',
  20             'thumbnail': 'http://img.moviefap.com/a16:9w990r/thumbs/be/322032-20l.jpg',
  21             'uploader_id': 'darvinfred06',
  22             'display_id': 'experienced-milf-amazing-handjob',
  23             'categories': ['Amateur', 'Masturbation', 'Mature', 'Flashing']
  24         }
  25     }, {
  26         # quirky single-format case where the extension is given as fid, but the video is really an flv
  27         'url': 'http://www.moviefap.com/videos/e5da0d3edce5404418f5/jeune-couple-russe.html',
  28         'md5': 'fa56683e291fc80635907168a743c9ad',
  29         'info_dict': {
  30             'id': 'e5da0d3edce5404418f5',
  31             'ext': 'flv',
  32             'title': 'Jeune Couple Russe',
  33             'description': 'Amateur',
  34             'thumbnail': 'http://pic.moviefap.com/thumbs/e5/949-18l.jpg',
  35             'uploader_id': 'whiskeyjar',
  36             'display_id': 'jeune-couple-russe',
  37             'categories': ['Amateur', 'Teen']
  38         }
  39     }]
  40
  41     @staticmethod
  42     def __get_thumbnail_data(xml):
  43
  44         """
  45         Constructs a list of video thumbnails from timeline preview images.
  46         :param xml: the information XML document to parse
  47         """
  48
  49         timeline = xml.find('timeline')
  50         if timeline is None:
  51             # not all videos have the data - ah well
  52             return []
  53
  54         # get the required information from the XML
  55         width = str_to_int(timeline.find('imageWidth').text)
  56         height = str_to_int(timeline.find('imageHeight').text)
  57         first = str_to_int(timeline.find('imageFirst').text)
  58         last = str_to_int(timeline.find('imageLast').text)
  59         pattern = timeline.find('imagePattern').text
  60
  61         # generate the list of thumbnail information dicts
  62         thumbnails = []
  63         for i in range(first, last + 1):
  64             thumbnails.append({
  65                 'url': pattern.replace('#', str(i)),
  66                 'width': width,
  67                 'height': height
  68             })
  69         return thumbnails
  70
  71     def _real_extract(self, url):
  72
  73         video_id = self._match_id(url)
  74         webpage = self._download_webpage(url, video_id)
  75
  76         # find and retrieve the XML document detailing video download URLs
  77         info_url = self._html_search_regex(r'flashvars\.config = escape\("(.+?)"', webpage, 'player parameters')
  78         xml = self._download_xml(info_url, video_id)
  79
  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', fatal=False),
  87             'uploader_id': self._html_search_regex(r'name="username" value="(.*?)"', webpage, 'uploader_id', fatal=False),
  88             'view_count': str_to_int(self._html_search_regex(r'<br>Views <strong>([0-9]+)</strong>', webpage, 'view_count, fatal=False')),
  89             'average_rating': float(self._html_search_regex(r'Current Rating<br> <strong>(.*?)</strong>', webpage, 'average_rating', fatal=False)),
  90             'comment_count': str_to_int(self._html_search_regex(r'<span id="comCount">([0-9]+)</span>', webpage, 'comment_count', fatal=False)),
  91             'age_limit': 18,
  92             'webpage_url': self._html_search_regex(r'name="link" value="(.*?)"', webpage, 'webpage_url', fatal=False),
  93             'categories': self._html_search_regex(r'</div>\s*(.*?)\s*<br>', webpage, 'categories', fatal=False).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