Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/gamestar.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     remove_end,
   8 )
   9
  10
  11 class GameStarIE(InfoExtractor):
  12     _VALID_URL = r'https?://(?:www\.)?gamestar\.de/videos/.*,(?P<id>[0-9]+)\.html'
  13     _TEST = {
  14         'url': 'http://www.gamestar.de/videos/trailer,3/hobbit-3-die-schlacht-der-fuenf-heere,76110.html',
  15         'md5': '96974ecbb7fd8d0d20fca5a00810cea7',
  16         'info_dict': {
  17             'id': '76110',
  18             'ext': 'mp4',
  19             'title': 'Hobbit 3: Die Schlacht der Fünf Heere - Teaser-Trailer zum dritten Teil',
  20             'description': 'Der Teaser-Trailer zu Hobbit 3: Die Schlacht der Fünf Heere zeigt einige Szenen aus dem dritten Teil der Saga und kündigt den...',
  21             'thumbnail': r're:^https?://.*\.jpg$',
  22             'timestamp': 1406542020,
  23             'upload_date': '20140728',
  24             'duration': 17
  25         }
  26     }
  27
  28     def _real_extract(self, url):
  29         video_id = self._match_id(url)
  30         webpage = self._download_webpage(url, video_id)
  31
  32         url = 'http://gamestar.de/_misc/videos/portal/getVideoUrl.cfm?premium=0&videoId=' + video_id
  33
  34         # TODO: there are multiple ld+json objects in the webpage,
  35         # while _search_json_ld finds only the first one
  36         json_ld = self._parse_json(self._search_regex(
  37             r'(?s)<script[^>]+type=(["\'])application/ld\+json\1[^>]*>(?P<json_ld>[^<]+VideoObject[^<]+)</script>',
  38             webpage, 'JSON-LD', group='json_ld'), video_id)
  39         info_dict = self._json_ld(json_ld, video_id)
  40         info_dict['title'] = remove_end(info_dict['title'], ' - GameStar')
  41
  42         view_count = json_ld.get('interactionCount')
  43         comment_count = int_or_none(self._html_search_regex(
  44             r'([0-9]+) Kommentare</span>', webpage, 'comment_count',
  45             fatal=False))
  46
  47         info_dict.update({
  48             'id': video_id,
  49             'url': url,
  50             'ext': 'mp4',
  51             'view_count': view_count,
  52             'comment_count': comment_count
  53         })
  54
  55         return info_dict