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     mimetype2ext,
   8     url_or_none,
   9 )
  10
  11
  12 class AparatIE(InfoExtractor):
  13     _VALID_URL = r'https?://(?:www\.)?aparat\.com/(?:v/|video/video/embed/videohash/)(?P<id>[a-zA-Z0-9]+)'
  14
  15     _TEST = {
  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             'age_limit': 0,
  23         },
  24         # 'skip': 'Extremely unreliable',
  25     }
  26
  27     def _real_extract(self, url):
  28         video_id = self._match_id(url)
  29
  30         # Note: There is an easier-to-parse configuration at
  31         # http://www.aparat.com/video/video/config/videohash/%video_id
  32         # but the URL in there does not work
  33         webpage = self._download_webpage(
  34             'http://www.aparat.com/video/video/embed/vt/frame/showvideo/yes/videohash/' + video_id,
  35             video_id)
  36
  37         title = self._search_regex(r'\s+title:\s*"([^"]+)"', webpage, 'title')
  38
  39         file_list = self._parse_json(
  40             self._search_regex(
  41                 r'fileList\s*=\s*JSON\.parse\(\'([^\']+)\'\)', webpage,
  42                 'file list'),
  43             video_id)
  44
  45         formats = []
  46         for item in file_list[0]:
  47             file_url = url_or_none(item.get('file'))
  48             if not file_url:
  49                 continue
  50             ext = mimetype2ext(item.get('type'))
  51             label = item.get('label')
  52             formats.append({
  53                 'url': file_url,
  54                 'ext': ext,
  55                 'format_id': label or ext,
  56                 'height': int_or_none(self._search_regex(
  57                     r'(\d+)[pP]', label or '', 'height', default=None)),
  58             })
  59         self._sort_formats(formats)
  60
  61         thumbnail = self._search_regex(
  62             r'image:\s*"([^"]+)"', webpage, 'thumbnail', fatal=False)
  63
  64         return {
  65             'id': video_id,
  66             'title': title,
  67             'thumbnail': thumbnail,
  68             'age_limit': self._family_friendly_search(webpage),
  69             'formats': formats,
  70         }