youtube_dl/extractor/youjizz.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from .common import InfoExtractor
   6 from ..utils import (
   7     determine_ext,
   8     int_or_none,
   9     parse_duration,
  10     url_or_none,
  11 )
  12
  13
  14 class YouJizzIE(InfoExtractor):
  15     _VALID_URL = r'https?://(?:\w+\.)?youjizz\.com/videos/(?:[^/#?]*-(?P<id>\d+)\.html|embed/(?P<embed_id>\d+))'
  16     _TESTS = [{
  17         'url': 'http://www.youjizz.com/videos/zeichentrick-1-2189178.html',
  18         'md5': 'b1e1dfaa8bb9537d8b84eeda9cf4acf4',
  19         'info_dict': {
  20             'id': '2189178',
  21             'ext': 'mp4',
  22             'title': 'Zeichentrick 1',
  23             'age_limit': 18,
  24             'duration': 2874,
  25         }
  26     }, {
  27         'url': 'http://www.youjizz.com/videos/-2189178.html',
  28         'only_matching': True,
  29     }, {
  30         'url': 'https://www.youjizz.com/videos/embed/31991001',
  31         'only_matching': True,
  32     }]
  33
  34     def _real_extract(self, url):
  35         mobj = re.match(self._VALID_URL, url)
  36         video_id = mobj.group('id') or mobj.group('embed_id')
  37
  38         webpage = self._download_webpage(url, video_id)
  39
  40         title = self._html_search_regex(
  41             r'<title>(.+?)</title>', webpage, 'title')
  42
  43         formats = []
  44
  45         encodings = self._parse_json(
  46             self._search_regex(
  47                 r'[Ee]ncodings\s*=\s*(\[.+?\]);\n', webpage, 'encodings',
  48                 default='[]'),
  49             video_id, fatal=False)
  50         for encoding in encodings:
  51             if not isinstance(encoding, dict):
  52                 continue
  53             format_url = url_or_none(encoding.get('filename'))
  54             if not format_url:
  55                 continue
  56             if determine_ext(format_url) == 'm3u8':
  57                 formats.extend(self._extract_m3u8_formats(
  58                     format_url, video_id, 'mp4', entry_protocol='m3u8_native',
  59                     m3u8_id='hls', fatal=False))
  60             else:
  61                 format_id = encoding.get('name') or encoding.get('quality')
  62                 height = int_or_none(self._search_regex(
  63                     r'^(\d+)[pP]', format_id, 'height', default=None))
  64                 formats.append({
  65                     'url': format_url,
  66                     'format_id': format_id,
  67                     'height': height,
  68                 })
  69
  70         if formats:
  71             info_dict = {
  72                 'formats': formats,
  73             }
  74         else:
  75             # YouJizz's HTML5 player has invalid HTML
  76             webpage = webpage.replace('"controls', '" controls')
  77             info_dict = self._parse_html5_media_entries(
  78                 url, webpage, video_id)[0]
  79
  80         duration = parse_duration(self._search_regex(
  81             r'<strong>Runtime:</strong>([^<]+)', webpage, 'duration',
  82             default=None))
  83         uploader = self._search_regex(
  84             r'<strong>Uploaded By:.*?<a[^>]*>([^<]+)', webpage, 'uploader',
  85             default=None)
  86
  87         info_dict.update({
  88             'id': video_id,
  89             'title': title,
  90             'age_limit': self._rta_search(webpage),
  91             'duration': duration,
  92             'uploader': uploader,
  93         })
  94
  95         return info_dict