yt_dlp/extractor/vidzi.py

   1 import re
   2
   3 from .common import InfoExtractor
   4 from ..utils import (
   5     decode_packed_codes,
   6     js_to_json,
   7     NO_DEFAULT,
   8     PACKED_CODES_RE,
   9 )
  10
  11
  12 class VidziIE(InfoExtractor):
  13     _VALID_URL = r'https?://(?:www\.)?vidzi\.(?:tv|cc|si|nu)/(?:embed-)?(?P<id>[0-9a-zA-Z]+)'
  14     _TESTS = [{
  15         'url': 'http://vidzi.tv/cghql9yq6emu.html',
  16         'md5': '4f16c71ca0c8c8635ab6932b5f3f1660',
  17         'info_dict': {
  18             'id': 'cghql9yq6emu',
  19             'ext': 'mp4',
  20             'title': 'youtube-dl test video  1\\\\2\'3/4<5\\\\6ä7↭',
  21         },
  22         'params': {
  23             # m3u8 download
  24             'skip_download': True,
  25         },
  26     }, {
  27         'url': 'http://vidzi.tv/embed-4z2yb0rzphe9-600x338.html',
  28         'only_matching': True,
  29     }, {
  30         'url': 'http://vidzi.cc/cghql9yq6emu.html',
  31         'only_matching': True,
  32     }, {
  33         'url': 'https://vidzi.si/rph9gztxj1et.html',
  34         'only_matching': True,
  35     }, {
  36         'url': 'http://vidzi.nu/cghql9yq6emu.html',
  37         'only_matching': True,
  38     }]
  39
  40     def _real_extract(self, url):
  41         video_id = self._match_id(url)
  42
  43         webpage = self._download_webpage(
  44             'http://vidzi.tv/%s' % video_id, video_id)
  45         title = self._html_search_regex(
  46             r'(?s)<h2 class="video-title">(.*?)</h2>', webpage, 'title')
  47
  48         codes = [webpage]
  49         codes.extend([
  50             decode_packed_codes(mobj.group(0)).replace('\\\'', '\'')
  51             for mobj in re.finditer(PACKED_CODES_RE, webpage)])
  52         for num, code in enumerate(codes, 1):
  53             jwplayer_data = self._parse_json(
  54                 self._search_regex(
  55                     r'setup\(([^)]+)\)', code, 'jwplayer data',
  56                     default=NO_DEFAULT if num == len(codes) else '{}'),
  57                 video_id, transform_source=lambda s: js_to_json(
  58                     re.sub(r'\s*\+\s*window\[.+?\]', '', s)))
  59             if jwplayer_data:
  60                 break
  61
  62         info_dict = self._parse_jwplayer_data(jwplayer_data, video_id, require_title=False)
  63         info_dict['title'] = title
  64
  65         return info_dict