Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/aparat.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 from .common import InfoExtractor
   5 from ..utils import (
   6     int_or_none,
   7     merge_dicts,
   8     mimetype2ext,
   9     url_or_none,
  10 )
  11
  12
  13 class AparatIE(InfoExtractor):
  14     _VALID_URL = r'https?://(?:www\.)?aparat\.com/(?:v/|video/video/embed/videohash/)(?P<id>[a-zA-Z0-9]+)'
  15
  16     _TESTS = [{
  17         'url': 'http://www.aparat.com/v/wP8On',
  18         'md5': '131aca2e14fe7c4dcb3c4877ba300c89',
  19         'info_dict': {
  20             'id': 'wP8On',
  21             'ext': 'mp4',
  22             'title': 'تیم گلکسی 11 - زومیت',
  23             'description': 'md5:096bdabcdcc4569f2b8a5e903a3b3028',
  24             'duration': 231,
  25             'timestamp': 1387394859,
  26             'upload_date': '20131218',
  27             'view_count': int,
  28         },
  29     }, {
  30         # multiple formats
  31         'url': 'https://www.aparat.com/v/8dflw/',
  32         'only_matching': True,
  33     }]
  34
  35     def _real_extract(self, url):
  36         video_id = self._match_id(url)
  37
  38         # Provides more metadata
  39         webpage = self._download_webpage(url, video_id, fatal=False)
  40
  41         if not webpage:
  42             # Note: There is an easier-to-parse configuration at
  43             # http://www.aparat.com/video/video/config/videohash/%video_id
  44             # but the URL in there does not work
  45             webpage = self._download_webpage(
  46                 'http://www.aparat.com/video/video/embed/vt/frame/showvideo/yes/videohash/' + video_id,
  47                 video_id)
  48
  49         options = self._parse_json(
  50             self._search_regex(
  51                 r'options\s*=\s*JSON\.parse\(\s*(["\'])(?P<value>(?:(?!\1).)+)\1\s*\)',
  52                 webpage, 'options', group='value'),
  53             video_id)
  54
  55         player = options['plugins']['sabaPlayerPlugin']
  56
  57         formats = []
  58         for sources in player['multiSRC']:
  59             for item in sources:
  60                 if not isinstance(item, dict):
  61                     continue
  62                 file_url = url_or_none(item.get('src'))
  63                 if not file_url:
  64                     continue
  65                 item_type = item.get('type')
  66                 if item_type == 'application/vnd.apple.mpegurl':
  67                     formats.extend(self._extract_m3u8_formats(
  68                         file_url, video_id, 'mp4',
  69                         entry_protocol='m3u8_native', m3u8_id='hls',
  70                         fatal=False))
  71                 else:
  72                     ext = mimetype2ext(item.get('type'))
  73                     label = item.get('label')
  74                     formats.append({
  75                         'url': file_url,
  76                         'ext': ext,
  77                         'format_id': 'http-%s' % (label or ext),
  78                         'height': int_or_none(self._search_regex(
  79                             r'(\d+)[pP]', label or '', 'height',
  80                             default=None)),
  81                     })
  82         self._sort_formats(
  83             formats, field_preference=('height', 'width', 'tbr', 'format_id'))
  84
  85         info = self._search_json_ld(webpage, video_id, default={})
  86
  87         if not info.get('title'):
  88             info['title'] = player['title']
  89
  90         return merge_dicts(info, {
  91             'id': video_id,
  92             'thumbnail': url_or_none(options.get('poster')),
  93             'duration': int_or_none(player.get('duration')),
  94             'formats': formats,
  95         })