youtube_dl/extractor/openload.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals, division
   3
   4 from .common import InfoExtractor
   5 from ..compat import (
   6     compat_chr,
   7     compat_ord,
   8 )
   9 from ..utils import (
  10     determine_ext,
  11     ExtractorError,
  12 )
  13
  14
  15 class OpenloadIE(InfoExtractor):
  16     _VALID_URL = r'https?://openload\.(?:co|io)/(?:f|embed)/(?P<id>[a-zA-Z0-9-_]+)'
  17
  18     _TESTS = [{
  19         'url': 'https://openload.co/f/kUEfGclsU9o',
  20         'md5': 'bf1c059b004ebc7a256f89408e65c36e',
  21         'info_dict': {
  22             'id': 'kUEfGclsU9o',
  23             'ext': 'mp4',
  24             'title': 'skyrim_no-audio_1080.mp4',
  25             'thumbnail': 're:^https?://.*\.jpg$',
  26         },
  27     }, {
  28         'url': 'https://openload.co/embed/kUEfGclsU9o/skyrim_no-audio_1080.mp4',
  29         'only_matching': True,
  30     }, {
  31         'url': 'https://openload.io/f/ZAn6oz-VZGE/',
  32         'only_matching': True,
  33     }, {
  34         'url': 'https://openload.co/f/_-ztPaZtMhM/',
  35         'only_matching': True,
  36     }, {
  37         # unavailable via https://openload.co/f/Sxz5sADo82g/, different layout
  38         # for title and ext
  39         'url': 'https://openload.co/embed/Sxz5sADo82g/',
  40         'only_matching': True,
  41     }]
  42
  43     def _real_extract(self, url):
  44         video_id = self._match_id(url)
  45         webpage = self._download_webpage('https://openload.co/embed/%s/' % video_id, video_id)
  46
  47         if 'File not found' in webpage or 'deleted by the owner' in webpage:
  48             raise ExtractorError('File not found', expected=True)
  49
  50         # The following decryption algorithm is written by @yokrysty and
  51         # declared to be freely used in youtube-dl
  52         # See https://github.com/rg3/youtube-dl/issues/10408
  53         enc_data = self._html_search_regex(
  54             r'<span[^>]*>([^<]+)</span>\s*<span[^>]*>[^<]+</span>\s*<span[^>]+id="streamurl"',
  55             webpage, 'encrypted data')
  56
  57         video_url_chars = []
  58
  59         for idx, c in enumerate(enc_data):
  60             j = compat_ord(c)
  61             if j >= 33 and j <= 126:
  62                 j = ((j + 14) % 94) + 33
  63             if idx == len(enc_data) - 1:
  64                 j += 2
  65             video_url_chars += compat_chr(j)
  66
  67         video_url = 'https://openload.co/stream/%s?mime=true' % ''.join(video_url_chars)
  68
  69         title = self._og_search_title(webpage, default=None) or self._search_regex(
  70             r'<span[^>]+class=["\']title["\'][^>]*>([^<]+)', webpage,
  71             'title', default=None) or self._html_search_meta(
  72             'description', webpage, 'title', fatal=True)
  73
  74         return {
  75             'id': video_id,
  76             'title': title,
  77             'thumbnail': self._og_search_thumbnail(webpage, default=None),
  78             'url': video_url,
  79             # Seems all videos have extensions in their titles
  80             'ext': determine_ext(title),
  81         }