youtube_dl/extractor/openload.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..compat import compat_chr
   8 from ..utils import (
   9     encode_base_n,
  10     ExtractorError,
  11 )
  12
  13
  14 class OpenloadIE(InfoExtractor):
  15     _VALID_URL = r'https://openload.(?:co|io)/(?:f|embed)/(?P<id>[a-zA-Z0-9-]+)'
  16
  17     _TESTS = [{
  18         'url': 'https://openload.co/f/kUEfGclsU9o',
  19         'md5': 'bf1c059b004ebc7a256f89408e65c36e',
  20         'info_dict': {
  21             'id': 'kUEfGclsU9o',
  22             'ext': 'mp4',
  23             'title': 'skyrim_no-audio_1080.mp4',
  24             'thumbnail': 're:^https?://.*\.jpg$',
  25         },
  26     }, {
  27         'url': 'https://openload.co/embed/kUEfGclsU9o/skyrim_no-audio_1080.mp4',
  28         'only_matching': True,
  29     }, {
  30         'url': 'https://openload.io/f/ZAn6oz-VZGE/',
  31         'only_matching': True,
  32     }]
  33
  34     @staticmethod
  35     def openload_level2_debase(m):
  36         radix, num = int(m.group(1)) + 27, int(m.group(2))
  37         return '"' + encode_base_n(num, radix) + '"'
  38
  39     @classmethod
  40     def openload_level2(cls, txt):
  41         # The function name is ǃ \u01c3
  42         # Using escaped unicode literals does not work in Python 3.2
  43         return re.sub(r'ǃ\((\d+),(\d+)\)', cls.openload_level2_debase, txt, re.UNICODE).replace('"+"', '')
  44
  45     # Openload uses a variant of aadecode
  46     # openload_decode and related functions are originally written by
  47     # vitas@matfyz.cz and released with public domain
  48     # See https://github.com/rg3/youtube-dl/issues/8489
  49     @classmethod
  50     def openload_decode(cls, txt):
  51         symbol_table = [
  52             ('_', '(ﾟДﾟ) [ﾟΘﾟ]'),
  53             ('a', '(ﾟДﾟ) [ﾟωﾟﾉ]'),
  54             ('b', '(ﾟДﾟ) [ﾟΘﾟﾉ]'),
  55             ('c', '(ﾟДﾟ) [\'c\']'),
  56             ('d', '(ﾟДﾟ) [ﾟｰﾟﾉ]'),
  57             ('e', '(ﾟДﾟ) [ﾟДﾟﾉ]'),
  58             ('f', '(ﾟДﾟ) [1]'),
  59
  60             ('o', '(ﾟДﾟ) [\'o\']'),
  61             ('u', '(oﾟｰﾟo)'),
  62             ('c', '(ﾟДﾟ) [\'c\']'),
  63
  64             ('7', '((ﾟｰﾟ) + (o^_^o))'),
  65             ('6', '((o^_^o) +(o^_^o) +(c^_^o))'),
  66             ('5', '((ﾟｰﾟ) + (ﾟΘﾟ))'),
  67             ('4', '(-~3)'),
  68             ('3', '(-~-~1)'),
  69             ('2', '(-~1)'),
  70             ('1', '(-~0)'),
  71             ('0', '((c^_^o)-(c^_^o))'),
  72         ]
  73         delim = '(ﾟДﾟ)[ﾟεﾟ]+'
  74         ret = ''
  75         for aachar in txt.split(delim):
  76             for val, pat in symbol_table:
  77                 aachar = aachar.replace(pat, val)
  78             aachar = aachar.replace('+ ', '')
  79             m = re.match(r'^\d+', aachar)
  80             if m:
  81                 ret += compat_chr(int(m.group(0), 8))
  82             else:
  83                 m = re.match(r'^u([\da-f]+)', aachar)
  84                 if m:
  85                     ret += compat_chr(int(m.group(1), 16))
  86         return cls.openload_level2(ret)
  87
  88     def _real_extract(self, url):
  89         video_id = self._match_id(url)
  90         webpage = self._download_webpage(url, video_id)
  91
  92         if 'File not found' in webpage:
  93             raise ExtractorError('File not found', expected=True)
  94
  95         code = self._search_regex(
  96             r'<video[^>]+>\s*<script[^>]+>([^<]+)</script>',
  97             webpage, 'JS code')
  98
  99         video_url = self._search_regex(
 100             r'return\s+"(https?://[^"]+)"', self.openload_decode(code), 'video URL')
 101
 102         return {
 103             'id': video_id,
 104             'title': self._og_search_title(webpage),
 105             'thumbnail': self._og_search_thumbnail(webpage),
 106             'url': video_url,
 107         }