youtube_dlc/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 _real_extract(self, url):
  37         video_id = self._match_id(url)
  38
  39         # Provides more metadata
  40         webpage = self._download_webpage(url, video_id, fatal=False)
  41
  42         if not webpage:
  43             webpage = self._download_webpage(
  44                 'http://www.aparat.com/video/video/embed/vt/frame/showvideo/yes/videohash/' + video_id,
  45                 video_id)
  46
  47         options = self._parse_json(self._search_regex(
  48             r'options\s*=\s*({.+?})\s*;', webpage, 'options'), 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         })