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 )
   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
  14     _TEST = {
  15         'url': 'http://www.aparat.com/v/wP8On',
  16         'md5': '131aca2e14fe7c4dcb3c4877ba300c89',
  17         'info_dict': {
  18             'id': 'wP8On',
  19             'ext': 'mp4',
  20             'title': 'تیم گلکسی 11 - زومیت',
  21             'age_limit': 0,
  22         },
  23         # 'skip': 'Extremely unreliable',
  24     }
  25
  26     def _real_extract(self, url):
  27         video_id = self._match_id(url)
  28
  29         # Note: There is an easier-to-parse configuration at
  30         # http://www.aparat.com/video/video/config/videohash/%video_id
  31         # but the URL in there does not work
  32         webpage = self._download_webpage(
  33             'http://www.aparat.com/video/video/embed/vt/frame/showvideo/yes/videohash/' + video_id,
  34             video_id)
  35
  36         title = self._search_regex(r'\s+title:\s*"([^"]+)"', webpage, 'title')
  37
  38         file_list = self._parse_json(
  39             self._search_regex(
  40                 r'fileList\s*=\s*JSON\.parse\(\'([^\']+)\'\)', webpage,
  41                 'file list'),
  42             video_id)
  43
  44         formats = []
  45         for item in file_list[0]:
  46             file_url = item.get('file')
  47             if not file_url:
  48                 continue
  49             ext = mimetype2ext(item.get('type'))
  50             label = item.get('label')
  51             formats.append({
  52                 'url': file_url,
  53                 'ext': ext,
  54                 'format_id': label or ext,
  55                 'height': int_or_none(self._search_regex(
  56                     r'(\d+)[pP]', label or '', 'height', default=None)),
  57             })
  58         self._sort_formats(formats)
  59
  60         thumbnail = self._search_regex(
  61             r'image:\s*"([^"]+)"', webpage, 'thumbnail', fatal=False)
  62
  63         return {
  64             'id': video_id,
  65             'title': title,
  66             'thumbnail': thumbnail,
  67             'age_limit': self._family_friendly_search(webpage),
  68             'formats': formats,
  69         }