youtube_dl/extractor/vidzi.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     decode_packed_codes,
   9     js_to_json,
  10     NO_DEFAULT,
  11     PACKED_CODES_RE,
  12 )
  13
  14
  15 class VidziIE(InfoExtractor):
  16     _VALID_URL = r'https?://(?:www\.)?vidzi\.(?:tv|cc|si|nu)/(?:embed-)?(?P<id>[0-9a-zA-Z]+)'
  17     _TESTS = [{
  18         'url': 'http://vidzi.tv/cghql9yq6emu.html',
  19         'md5': '4f16c71ca0c8c8635ab6932b5f3f1660',
  20         'info_dict': {
  21             'id': 'cghql9yq6emu',
  22             'ext': 'mp4',
  23             'title': 'youtube-dl test video  1\\\\2\'3/4<5\\\\6ä7↭',
  24         },
  25         'params': {
  26             # m3u8 download
  27             'skip_download': True,
  28         },
  29     }, {
  30         'url': 'http://vidzi.tv/embed-4z2yb0rzphe9-600x338.html',
  31         'only_matching': True,
  32     }, {
  33         'url': 'http://vidzi.cc/cghql9yq6emu.html',
  34         'only_matching': True,
  35     }, {
  36         'url': 'https://vidzi.si/rph9gztxj1et.html',
  37         'only_matching': True,
  38     }, {
  39         'url': 'http://vidzi.nu/cghql9yq6emu.html',
  40         'only_matching': True,
  41     }]
  42
  43     def _real_extract(self, url):
  44         video_id = self._match_id(url)
  45
  46         webpage = self._download_webpage(
  47             'http://vidzi.tv/%s' % video_id, video_id)
  48         title = self._html_search_regex(
  49             r'(?s)<h2 class="video-title">(.*?)</h2>', webpage, 'title')
  50
  51         codes = [webpage]
  52         codes.extend([
  53             decode_packed_codes(mobj.group(0)).replace('\\\'', '\'')
  54             for mobj in re.finditer(PACKED_CODES_RE, webpage)])
  55         for num, code in enumerate(codes, 1):
  56             jwplayer_data = self._parse_json(
  57                 self._search_regex(
  58                     r'setup\(([^)]+)\)', code, 'jwplayer data',
  59                     default=NO_DEFAULT if num == len(codes) else '{}'),
  60                 video_id, transform_source=lambda s: js_to_json(
  61                     re.sub(r'\s*\+\s*window\[.+?\]', '', s)))
  62             if jwplayer_data:
  63                 break
  64
  65         info_dict = self._parse_jwplayer_data(jwplayer_data, video_id, require_title=False)
  66         info_dict['title'] = title
  67
  68         return info_dict