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     ExtractorError,
   7     HEADRequest,
   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         embed_url = 'http://www.aparat.com/video/video/embed/vt/frame/showvideo/yes/videohash/' + video_id
  33         webpage = self._download_webpage(embed_url, video_id)
  34
  35         file_list = self._parse_json(self._search_regex(
  36             r'fileList\s*=\s*JSON\.parse\(\'([^\']+)\'\)', webpage, 'file list'), video_id)
  37         for i, item in enumerate(file_list[0]):
  38             video_url = item['file']
  39             req = HEADRequest(video_url)
  40             res = self._request_webpage(
  41                 req, video_id, note='Testing video URL %d' % i, errnote=False)
  42             if res:
  43                 break
  44         else:
  45             raise ExtractorError('No working video URLs found')
  46
  47         title = self._search_regex(r'\s+title:\s*"([^"]+)"', webpage, 'title')
  48         thumbnail = self._search_regex(
  49             r'image:\s*"([^"]+)"', webpage, 'thumbnail', fatal=False)
  50
  51         return {
  52             'id': video_id,
  53             'title': title,
  54             'url': video_url,
  55             'ext': 'mp4',
  56             'thumbnail': thumbnail,
  57             'age_limit': self._family_friendly_search(webpage),
  58         }