yt_dlp/extractor/aparat.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 from .common import InfoExtractor
   5 from ..utils import (
   6     get_element_by_id,
   7     int_or_none,
   8     merge_dicts,
   9     mimetype2ext,
  10     url_or_none,
  11 )
  12
  13
  14 class AparatIE(InfoExtractor):
  15     _VALID_URL = r'https?://(?:www\.)?aparat\.com/(?:v/|video/video/embed/videohash/)(?P<id>[a-zA-Z0-9]+)'
  16
  17     _TESTS = [{
  18         'url': 'http://www.aparat.com/v/wP8On',
  19         'md5': '131aca2e14fe7c4dcb3c4877ba300c89',
  20         'info_dict': {
  21             'id': 'wP8On',
  22             'ext': 'mp4',
  23             'title': 'تیم گلکسی 11 - زومیت',
  24             'description': 'md5:096bdabcdcc4569f2b8a5e903a3b3028',
  25             'duration': 231,
  26             'timestamp': 1387394859,
  27             'upload_date': '20131218',
  28             'view_count': int,
  29         },
  30     }, {
  31         # multiple formats
  32         'url': 'https://www.aparat.com/v/8dflw/',
  33         'only_matching': True,
  34     }]
  35
  36     def _parse_options(self, webpage, video_id, fatal=True):
  37         return self._parse_json(self._search_regex(
  38             r'options\s*=\s*({.+?})\s*;', webpage, 'options', default='{}'), video_id)
  39
  40     def _real_extract(self, url):
  41         video_id = self._match_id(url)
  42
  43         # If available, provides more metadata
  44         webpage = self._download_webpage(url, video_id, fatal=False)
  45         options = self._parse_options(webpage, video_id, fatal=False)
  46
  47         if not options:
  48             webpage = self._download_webpage(
  49                 'http://www.aparat.com/video/video/embed/vt/frame/showvideo/yes/videohash/' + video_id,
  50                 video_id, 'Downloading embed webpage')
  51             options = self._parse_options(webpage, video_id)
  52
  53         formats = []
  54         for sources in (options.get('multiSRC') or []):
  55             for item in sources:
  56                 if not isinstance(item, dict):
  57                     continue
  58                 file_url = url_or_none(item.get('src'))
  59                 if not file_url:
  60                     continue
  61                 item_type = item.get('type')
  62                 if item_type == 'application/vnd.apple.mpegurl':
  63                     formats.extend(self._extract_m3u8_formats(
  64                         file_url, video_id, 'mp4',
  65                         entry_protocol='m3u8_native', m3u8_id='hls',
  66                         fatal=False))
  67                 else:
  68                     ext = mimetype2ext(item.get('type'))
  69                     label = item.get('label')
  70                     formats.append({
  71                         'url': file_url,
  72                         'ext': ext,
  73                         'format_id': 'http-%s' % (label or ext),
  74                         'height': int_or_none(self._search_regex(
  75                             r'(\d+)[pP]', label or '', 'height',
  76                             default=None)),
  77                     })
  78         self._sort_formats(formats)
  79
  80         info = self._search_json_ld(webpage, video_id, default={})
  81
  82         if not info.get('title'):
  83             info['title'] = get_element_by_id('videoTitle', webpage) or \
  84                 self._html_search_meta(['og:title', 'twitter:title', 'DC.Title', 'title'], webpage, fatal=True)
  85
  86         return merge_dicts(info, {
  87             'id': video_id,
  88             'thumbnail': url_or_none(options.get('poster')),
  89             'duration': int_or_none(options.get('duration')),
  90             'formats': formats,
  91         })