Raphaël G. Git Repositories - youtubedl/blob - 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         'url': 'https://openload.co/f/_-ztPaZtMhM/',
  36         'only_matching': True,
  37     }, {
  38         # unavailable via https://openload.co/f/Sxz5sADo82g/, different layout
  39         # for title and ext
  40         'url': 'https://openload.co/embed/Sxz5sADo82g/',
  41         'only_matching': True,
  42     }]
  43
  44     @staticmethod
  45     def openload_level2_debase(m):
  46         radix, num = int(m.group(1)) + 27, int(m.group(2))
  47         return '"' + encode_base_n(num, radix) + '"'
  48
  49     @classmethod
  50     def openload_level2(cls, txt):
  51         # The function name is ǃ \u01c3
  52         # Using escaped unicode literals does not work in Python 3.2
  53         return re.sub(r'ǃ\((\d+),(\d+)\)', cls.openload_level2_debase, txt, re.UNICODE).replace('"+"', '')
  54
  55     # Openload uses a variant of aadecode
  56     # openload_decode and related functions are originally written by
  57     # vitas@matfyz.cz and released with public domain
  58     # See https://github.com/rg3/youtube-dl/issues/8489
  59     @classmethod
  60     def openload_decode(cls, txt):
  61         symbol_table = [
  62             ('_', '(ﾟДﾟ) [ﾟΘﾟ]'),
  63             ('a', '(ﾟДﾟ) [ﾟωﾟﾉ]'),
  64             ('b', '(ﾟДﾟ) [ﾟΘﾟﾉ]'),
  65             ('c', '(ﾟДﾟ) [\'c\']'),
  66             ('d', '(ﾟДﾟ) [ﾟｰﾟﾉ]'),
  67             ('e', '(ﾟДﾟ) [ﾟДﾟﾉ]'),
  68             ('f', '(ﾟДﾟ) [1]'),
  69
  70             ('o', '(ﾟДﾟ) [\'o\']'),
  71             ('u', '(oﾟｰﾟo)'),
  72             ('c', '(ﾟДﾟ) [\'c\']'),
  73
  74             ('7', '((ﾟｰﾟ) + (o^_^o))'),
  75             ('6', '((o^_^o) +(o^_^o) +(c^_^o))'),
  76             ('5', '((ﾟｰﾟ) + (ﾟΘﾟ))'),
  77             ('4', '(-~3)'),
  78             ('3', '(-~-~1)'),
  79             ('2', '(-~1)'),
  80             ('1', '(-~0)'),
  81             ('0', '((c^_^o)-(c^_^o))'),
  82         ]
  83         delim = '(ﾟДﾟ)[ﾟεﾟ]+'
  84         ret = ''
  85         for aachar in txt.split(delim):
  86             for val, pat in symbol_table:
  87                 aachar = aachar.replace(pat, val)
  88             aachar = aachar.replace('+ ', '')
  89             m = re.match(r'^\d+', aachar)
  90             if m:
  91                 ret += compat_chr(int(m.group(0), 8))
  92             else:
  93                 m = re.match(r'^u([\da-f]+)', aachar)
  94                 if m:
  95                     ret += compat_chr(int(m.group(1), 16))
  96         return cls.openload_level2(ret)
  97
  98     def _real_extract(self, url):
  99         video_id = self._match_id(url)
 100         webpage = self._download_webpage(url, video_id)
 101
 102         if 'File not found' in webpage:
 103             raise ExtractorError('File not found', expected=True)
 104
 105         code = self._search_regex(
 106             r'</video>\s*</div>\s*<script[^>]+>[^>]+</script>\s*<script[^>]+>([^<]+)</script>',
 107             webpage, 'JS code')
 108
 109         decoded = self.openload_decode(code)
 110
 111         video_url = self._search_regex(
 112             r'return\s+"(https?://[^"]+)"', decoded, 'video URL')
 113
 114         title = self._og_search_title(webpage, default=None) or self._search_regex(
 115             r'<span[^>]+class=["\']title["\'][^>]*>([^<]+)', webpage,
 116             'title', default=None) or self._html_search_meta(
 117             'description', webpage, 'title', fatal=True)
 118
 119         ext = mimetype2ext(self._search_regex(
 120             r'window\.vt\s*=\s*(["\'])(?P<mimetype>.+?)\1', decoded,
 121             'mimetype', default=None, group='mimetype')) or determine_ext(
 122             video_url, 'mp4')
 123
 124         return {
 125             'id': video_id,
 126             'title': title,
 127             'ext': ext,
 128             'thumbnail': self._og_search_thumbnail(webpage, default=None),
 129             'url': video_url,
 130         }