Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/openload.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals, division
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..compat import (
   8     compat_chr,
   9     compat_ord,
  10 )
  11 from ..utils import (
  12     determine_ext,
  13     ExtractorError,
  14 )
  15 from ..jsinterp import (
  16     JSInterpreter,
  17     _NAME_RE
  18 )
  19
  20
  21 class OpenloadIE(InfoExtractor):
  22     _VALID_URL = r'https?://openload\.(?:co|io)/(?:f|embed)/(?P<id>[a-zA-Z0-9-_]+)'
  23
  24     _TESTS = [{
  25         'url': 'https://openload.co/f/kUEfGclsU9o',
  26         'md5': 'bf1c059b004ebc7a256f89408e65c36e',
  27         'info_dict': {
  28             'id': 'kUEfGclsU9o',
  29             'ext': 'mp4',
  30             'title': 'skyrim_no-audio_1080.mp4',
  31             'thumbnail': 're:^https?://.*\.jpg$',
  32         },
  33     }, {
  34         'url': 'https://openload.co/embed/rjC09fkPLYs',
  35         'info_dict': {
  36             'id': 'rjC09fkPLYs',
  37             'ext': 'mp4',
  38             'title': 'movie.mp4',
  39             'thumbnail': 're:^https?://.*\.jpg$',
  40             'subtitles': {
  41                 'en': [{
  42                     'ext': 'vtt',
  43                 }],
  44             },
  45         },
  46         'params': {
  47             'skip_download': True,  # test subtitles only
  48         },
  49     }, {
  50         'url': 'https://openload.co/embed/kUEfGclsU9o/skyrim_no-audio_1080.mp4',
  51         'only_matching': True,
  52     }, {
  53         'url': 'https://openload.io/f/ZAn6oz-VZGE/',
  54         'only_matching': True,
  55     }, {
  56         'url': 'https://openload.co/f/_-ztPaZtMhM/',
  57         'only_matching': True,
  58     }, {
  59         # unavailable via https://openload.co/f/Sxz5sADo82g/, different layout
  60         # for title and ext
  61         'url': 'https://openload.co/embed/Sxz5sADo82g/',
  62         'only_matching': True,
  63     }]
  64
  65     def openload_decode(self, txt):
  66         symbol_dict = {
  67             '(ﾟДﾟ) [ﾟΘﾟ]': '_',
  68             '(ﾟДﾟ) [ﾟωﾟﾉ]': 'a',
  69             '(ﾟДﾟ) [ﾟΘﾟﾉ]': 'b',
  70             '(ﾟДﾟ) [\'c\']': 'c',
  71             '(ﾟДﾟ) [ﾟｰﾟﾉ]': 'd',
  72             '(ﾟДﾟ) [ﾟДﾟﾉ]': 'e',
  73             '(ﾟДﾟ) [1]': 'f',
  74             '(ﾟДﾟ) [\'o\']': 'o',
  75             '(oﾟｰﾟo)': 'u',
  76             '(ﾟДﾟ) [\'c\']': 'c',
  77             '((ﾟｰﾟ) + (o^_^o))': '7',
  78             '((o^_^o) +(o^_^o) +(c^_^o))': '6',
  79             '((ﾟｰﾟ) + (ﾟΘﾟ))': '5',
  80             '(-~3)': '4',
  81             '(-~-~1)': '3',
  82             '(-~1)': '2',
  83             '(-~0)': '1',
  84             '((c^_^o)-(c^_^o))': '0',
  85         }
  86         delim = '(ﾟДﾟ)[ﾟεﾟ]+'
  87         end_token = '(ﾟДﾟ)[ﾟoﾟ]'
  88         symbols = '|'.join(map(re.escape, symbol_dict.keys()))
  89         txt = re.sub('(%s)\+\s?' % symbols, lambda m: symbol_dict[m.group(1)], txt)
  90         ret = ''
  91         for aacode in re.findall(r'{0}\+\s?{1}(.*?){0}'.format(re.escape(end_token), re.escape(delim)), txt):
  92             for aachar in aacode.split(delim):
  93                 if aachar.isdigit():
  94                     ret += compat_chr(int(aachar, 8))
  95                 else:
  96                     m = re.match(r'^u([\da-f]{4})$', aachar)
  97                     if m:
  98                         ret += compat_chr(int(m.group(1), 16))
  99                     else:
 100                         self.report_warning("Cannot decode: %s" % aachar)
 101         return ret
 102
 103     def _real_extract(self, url):
 104         video_id = self._match_id(url)
 105         webpage = self._download_webpage('https://openload.co/embed/%s/' % video_id, video_id)
 106
 107         if 'File not found' in webpage or 'deleted by the owner' in webpage:
 108             raise ExtractorError('File not found', expected=True)
 109
 110         # The following decryption algorithm is written by @yokrysty and
 111         # declared to be freely used in youtube-dl
 112         # See https://github.com/rg3/youtube-dl/issues/10408
 113         enc_data = self._html_search_regex(
 114             r'<span[^>]*>([^<]+)</span>\s*<span[^>]*>[^<]+</span>\s*<span[^>]+id="streamurl"',
 115             webpage, 'encrypted data')
 116
 117         enc_code = self._html_search_regex(r'<script[^>]+>(ﾟωﾟ[^<]+)</script>',
 118                                            webpage, 'encrypted code')
 119
 120         js_code = self.openload_decode(enc_code)
 121         jsi = JSInterpreter(js_code)
 122
 123         m_offset_fun = self._search_regex(r'slice\(0\s*-\s*(%s)\(\)' % _NAME_RE, js_code, 'javascript offset function')
 124         m_diff_fun = self._search_regex(r'charCodeAt\(0\)\s*\+\s*(%s)\(\)' % _NAME_RE, js_code, 'javascript diff function')
 125
 126         offset = jsi.call_function(m_offset_fun)
 127         diff = jsi.call_function(m_diff_fun)
 128
 129         video_url_chars = []
 130
 131         for idx, c in enumerate(enc_data):
 132             j = compat_ord(c)
 133             if j >= 33 and j <= 126:
 134                 j = ((j + 14) % 94) + 33
 135             if idx == len(enc_data) - offset:
 136                 j += diff
 137             video_url_chars += compat_chr(j)
 138
 139         video_url = 'https://openload.co/stream/%s?mime=true' % ''.join(video_url_chars)
 140
 141         title = self._og_search_title(webpage, default=None) or self._search_regex(
 142             r'<span[^>]+class=["\']title["\'][^>]*>([^<]+)', webpage,
 143             'title', default=None) or self._html_search_meta(
 144             'description', webpage, 'title', fatal=True)
 145
 146         entries = self._parse_html5_media_entries(url, webpage, video_id)
 147         subtitles = entries[0]['subtitles'] if entries else None
 148
 149         info_dict = {
 150             'id': video_id,
 151             'title': title,
 152             'thumbnail': self._og_search_thumbnail(webpage, default=None),
 153             'url': video_url,
 154             # Seems all videos have extensions in their titles
 155             'ext': determine_ext(title),
 156             'subtitles': subtitles,
 157         }
 158
 159         return info_dict