yt_dlp/extractor/aparat.py

   1 from .common import InfoExtractor
   2 from ..utils import (
   3     get_element_by_id,
   4     int_or_none,
   5     merge_dicts,
   6     mimetype2ext,
   7     url_or_none,
   8 )
   9
  10
  11 class AparatIE(InfoExtractor):
  12     _VALID_URL = r'https?://(?:www\.)?aparat\.com/(?:v/|video/video/embed/videohash/)(?P<id>[a-zA-Z0-9]+)'
  13
  14     _TESTS = [{
  15         'url': 'http://www.aparat.com/v/wP8On',
  16         'md5': '131aca2e14fe7c4dcb3c4877ba300c89',
  17         'info_dict': {
  18             'id': 'wP8On',
  19             'ext': 'mp4',
  20             'title': 'تیم گلکسی 11 - زومیت',
  21             'description': 'md5:096bdabcdcc4569f2b8a5e903a3b3028',
  22             'duration': 231,
  23             'timestamp': 1387394859,
  24             'upload_date': '20131218',
  25             'view_count': int,
  26         },
  27     }, {
  28         # multiple formats
  29         'url': 'https://www.aparat.com/v/8dflw/',
  30         'only_matching': True,
  31     }]
  32
  33     def _parse_options(self, webpage, video_id, fatal=True):
  34         return self._parse_json(self._search_regex(
  35             r'options\s*=\s*({.+?})\s*;', webpage, 'options', default='{}'), video_id)
  36
  37     def _real_extract(self, url):
  38         video_id = self._match_id(url)
  39
  40         # If available, provides more metadata
  41         webpage = self._download_webpage(url, video_id, fatal=False)
  42         options = self._parse_options(webpage, video_id, fatal=False)
  43
  44         if not options:
  45             webpage = self._download_webpage(
  46                 'http://www.aparat.com/video/video/embed/vt/frame/showvideo/yes/videohash/' + video_id,
  47                 video_id, 'Downloading embed webpage')
  48             options = self._parse_options(webpage, video_id)
  49
  50         formats = []
  51         for sources in (options.get('multiSRC') or []):
  52             for item in sources:
  53                 if not isinstance(item, dict):
  54                     continue
  55                 file_url = url_or_none(item.get('src'))
  56                 if not file_url:
  57                     continue
  58                 item_type = item.get('type')
  59                 if item_type == 'application/vnd.apple.mpegurl':
  60                     formats.extend(self._extract_m3u8_formats(
  61                         file_url, video_id, 'mp4',
  62                         entry_protocol='m3u8_native', m3u8_id='hls',
  63                         fatal=False))
  64                 else:
  65                     ext = mimetype2ext(item.get('type'))
  66                     label = item.get('label')
  67                     formats.append({
  68                         'url': file_url,
  69                         'ext': ext,
  70                         'format_id': 'http-%s' % (label or ext),
  71                         'height': int_or_none(self._search_regex(
  72                             r'(\d+)[pP]', label or '', 'height',
  73                             default=None)),
  74                     })
  75         self._sort_formats(formats)
  76
  77         info = self._search_json_ld(webpage, video_id, default={})
  78
  79         if not info.get('title'):
  80             info['title'] = get_element_by_id('videoTitle', webpage) or \
  81                 self._html_search_meta(['og:title', 'twitter:title', 'DC.Title', 'title'], webpage, fatal=True)
  82
  83         return merge_dicts(info, {
  84             'id': video_id,
  85             'thumbnail': url_or_none(options.get('poster')),
  86             'duration': int_or_none(options.get('duration')),
  87             'formats': formats,
  88         })