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     determine_ext,
  10     encode_base_n,
  11     ExtractorError,
  12     mimetype2ext,
  13 )
  14
  15
  16 class OpenloadIE(InfoExtractor):
  17     _VALID_URL = r'https://openload.(?:co|io)/(?:f|embed)/(?P<id>[a-zA-Z0-9-]+)'
  18
  19     _TESTS = [{
  20         'url': 'https://openload.co/f/kUEfGclsU9o',
  21         'md5': 'bf1c059b004ebc7a256f89408e65c36e',
  22         'info_dict': {
  23             'id': 'kUEfGclsU9o',
  24             'ext': 'mp4',
  25             'title': 'skyrim_no-audio_1080.mp4',
  26             'thumbnail': 're:^https?://.*\.jpg$',
  27         },
  28     }, {
  29         'url': 'https://openload.co/embed/kUEfGclsU9o/skyrim_no-audio_1080.mp4',
  30         'only_matching': True,
  31     }, {
  32         'url': 'https://openload.io/f/ZAn6oz-VZGE/',
  33         'only_matching': True,
  34     }, {
  35         # unavailable via https://openload.co/f/Sxz5sADo82g/, different layout
  36         # for title and ext
  37         'url': 'https://openload.co/embed/Sxz5sADo82g/',
  38         'only_matching': True,
  39     }]
  40
  41     @staticmethod
  42     def openload_level2_debase(m):
  43         radix, num = int(m.group(1)) + 27, int(m.group(2))
  44         return '"' + encode_base_n(num, radix) + '"'
  45
  46     @classmethod
  47     def openload_level2(cls, txt):
  48         # The function name is ǃ \u01c3
  49         # Using escaped unicode literals does not work in Python 3.2
  50         return re.sub(r'ǃ\((\d+),(\d+)\)', cls.openload_level2_debase, txt, re.UNICODE).replace('"+"', '')
  51
  52     # Openload uses a variant of aadecode
  53     # openload_decode and related functions are originally written by
  54     # vitas@matfyz.cz and released with public domain
  55     # See https://github.com/rg3/youtube-dl/issues/8489
  56     @classmethod
  57     def openload_decode(cls, txt):
  58         symbol_table = [
  59             ('_', '(ﾟДﾟ) [ﾟΘﾟ]'),
  60             ('a', '(ﾟДﾟ) [ﾟωﾟﾉ]'),
  61             ('b', '(ﾟДﾟ) [ﾟΘﾟﾉ]'),
  62             ('c', '(ﾟДﾟ) [\'c\']'),
  63             ('d', '(ﾟДﾟ) [ﾟｰﾟﾉ]'),
  64             ('e', '(ﾟДﾟ) [ﾟДﾟﾉ]'),
  65             ('f', '(ﾟДﾟ) [1]'),
  66
  67             ('o', '(ﾟДﾟ) [\'o\']'),
  68             ('u', '(oﾟｰﾟo)'),
  69             ('c', '(ﾟДﾟ) [\'c\']'),
  70
  71             ('7', '((ﾟｰﾟ) + (o^_^o))'),
  72             ('6', '((o^_^o) +(o^_^o) +(c^_^o))'),
  73             ('5', '((ﾟｰﾟ) + (ﾟΘﾟ))'),
  74             ('4', '(-~3)'),
  75             ('3', '(-~-~1)'),
  76             ('2', '(-~1)'),
  77             ('1', '(-~0)'),
  78             ('0', '((c^_^o)-(c^_^o))'),
  79         ]
  80         delim = '(ﾟДﾟ)[ﾟεﾟ]+'
  81         ret = ''
  82         for aachar in txt.split(delim):
  83             for val, pat in symbol_table:
  84                 aachar = aachar.replace(pat, val)
  85             aachar = aachar.replace('+ ', '')
  86             m = re.match(r'^\d+', aachar)
  87             if m:
  88                 ret += compat_chr(int(m.group(0), 8))
  89             else:
  90                 m = re.match(r'^u([\da-f]+)', aachar)
  91                 if m:
  92                     ret += compat_chr(int(m.group(1), 16))
  93         return cls.openload_level2(ret)
  94
  95     def _real_extract(self, url):
  96         video_id = self._match_id(url)
  97         webpage = self._download_webpage(url, video_id)
  98
  99         if 'File not found' in webpage:
 100             raise ExtractorError('File not found', expected=True)
 101
 102         code = self._search_regex(
 103             r'</video>\s*</div>\s*<script[^>]+>([^<]+)</script>',
 104             webpage, 'JS code')
 105
 106         decoded = self.openload_decode(code)
 107
 108         video_url = self._search_regex(
 109             r'return\s+"(https?://[^"]+)"', decoded, 'video URL')
 110
 111         title = self._og_search_title(webpage, default=None) or self._search_regex(
 112             r'<span[^>]+class=["\']title["\'][^>]*>([^<]+)', webpage,
 113             'title', default=None) or self._html_search_meta(
 114             'description', webpage, 'title', fatal=True)
 115
 116         ext = mimetype2ext(self._search_regex(
 117             r'window\.vt\s*=\s*(["\'])(?P<mimetype>.+?)\1', decoded,
 118             'mimetype', default=None, group='mimetype')) or determine_ext(
 119             video_url, 'mp4')
 120
 121         return {
 122             'id': video_id,
 123             'title': title,
 124             'ext': ext,
 125             'thumbnail': self._og_search_thumbnail(webpage, default=None),
 126             'url': video_url,
 127         }