_ Git - youtube-dl/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\.)?game(?:pro|star)\.de/videos/.*,(?P<id>[0-9]+)\.html'
  13     _TESTS = [
  14         {
  15             'url': 'http://www.gamestar.de/videos/trailer,3/hobbit-3-die-schlacht-der-fuenf-heere,76110.html',
  16             'md5': 'ee782f1f8050448c95c5cacd63bc851c',
  17             'info_dict': {
  18                 'id': '76110',
  19                 'ext': 'mp4',
  20                 'title': 'Hobbit 3: Die Schlacht der Fünf Heere - Teaser-Trailer zum dritten Teil',
  21                 'description': 'Der Teaser-Trailer zu Hobbit 3: Die Schlacht der Fünf Heere zeigt einige Szenen aus dem dritten Teil der Saga und kündigt den...',
  22                 'thumbnail': r're:^https?://.*\.jpg$',
  23                 'timestamp': 1406542380,
  24                 'upload_date': '20140728',
  25                 'duration': 17,
  26             }
  27         },
  28         {
  29             'url': 'http://www.gamepro.de/videos/top-10-indie-spiele-fuer-nintendo-switch-video-tolle-nindies-games-zum-download,95316.html',
  30             'only_matching': True,
  31         },
  32     ]
  33
  34     def _real_extract(self, url):
  35         video_id = self._match_id(url)
  36         webpage = self._download_webpage(url, video_id)
  37
  38         url = 'http://gamestar.de/_misc/videos/portal/getVideoUrl.cfm?premium=0&videoId=' + video_id
  39
  40         # TODO: there are multiple ld+json objects in the webpage,
  41         # while _search_json_ld finds only the first one
  42         json_ld = self._parse_json(self._search_regex(
  43             r'(?s)<script[^>]+type=(["\'])application/ld\+json\1[^>]*>(?P<json_ld>[^<]+VideoObject[^<]+)</script>',
  44             webpage, 'JSON-LD', group='json_ld'), video_id)
  45         info_dict = self._json_ld(json_ld, video_id)
  46         info_dict['title'] = remove_end(info_dict['title'], ' - GameStar')
  47         info_dict['title'] = remove_end(info_dict['title'], ' - GamePro')
  48
  49         view_count = int_or_none(json_ld.get('interactionCount'))
  50         comment_count = int_or_none(self._html_search_regex(
  51             r'<span>Kommentare</span><span class="count">\(([0-9]+)\)</span>',
  52             webpage, 'comment_count', fatal=False))
  53
  54         info_dict.update({
  55             'id': video_id,
  56             'url': url,
  57             'ext': 'mp4',
  58             'view_count': view_count,
  59             'comment_count': comment_count
  60         })
  61
  62         return info_dict