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     _EMBED_REGEX = [r'<iframe .*?src="(?P<url>http://www\.aparat\.com/video/[^"]+)"']
  14
  15     _TESTS = [{
  16         'url': 'http://www.aparat.com/v/wP8On',
  17         'md5': '131aca2e14fe7c4dcb3c4877ba300c89',
  18         'info_dict': {
  19             'id': 'wP8On',
  20             'ext': 'mp4',
  21             'title': 'تیم گلکسی 11 - زومیت',
  22             'description': 'md5:096bdabcdcc4569f2b8a5e903a3b3028',
  23             'duration': 231,
  24             'timestamp': 1387394859,
  25             'upload_date': '20131218',
  26             'view_count': int,
  27         },
  28     }, {
  29         # multiple formats
  30         'url': 'https://www.aparat.com/v/8dflw/',
  31         'only_matching': True,
  32     }]
  33
  34     def _parse_options(self, webpage, video_id, fatal=True):
  35         return self._parse_json(self._search_regex(
  36             r'options\s*=\s*({.+?})\s*;', webpage, 'options', default='{}'), video_id)
  37
  38     def _real_extract(self, url):
  39         video_id = self._match_id(url)
  40
  41         # If available, provides more metadata
  42         webpage = self._download_webpage(url, video_id, fatal=False)
  43         options = self._parse_options(webpage, video_id, fatal=False)
  44
  45         if not options:
  46             webpage = self._download_webpage(
  47                 'http://www.aparat.com/video/video/embed/vt/frame/showvideo/yes/videohash/' + video_id,
  48                 video_id, 'Downloading embed webpage')
  49             options = self._parse_options(webpage, video_id)
  50
  51         formats = []
  52         for sources in (options.get('multiSRC') or []):
  53             for item in sources:
  54                 if not isinstance(item, dict):
  55                     continue
  56                 file_url = url_or_none(item.get('src'))
  57                 if not file_url:
  58                     continue
  59                 item_type = item.get('type')
  60                 if item_type == 'application/vnd.apple.mpegurl':
  61                     formats.extend(self._extract_m3u8_formats(
  62                         file_url, video_id, 'mp4',
  63                         entry_protocol='m3u8_native', m3u8_id='hls',
  64                         fatal=False))
  65                 else:
  66                     ext = mimetype2ext(item.get('type'))
  67                     label = item.get('label')
  68                     formats.append({
  69                         'url': file_url,
  70                         'ext': ext,
  71                         'format_id': 'http-%s' % (label or ext),
  72                         'height': int_or_none(self._search_regex(
  73                             r'(\d+)[pP]', label or '', 'height',
  74                             default=None)),
  75                     })
  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         })