[youtube] Fix extraction.
[youtube-dl] / youtube_dl / extractor / gamestar.py
1 # coding: utf-8
2 from __future__ import unicode_literals
3
4 import re
5
6 from .common import InfoExtractor
7 from ..utils import (
8     int_or_none,
9     remove_end,
10 )
11
12
13 class GameStarIE(InfoExtractor):
14     _VALID_URL = r'https?://(?:www\.)?game(?P<site>pro|star)\.de/videos/.*,(?P<id>[0-9]+)\.html'
15     _TESTS = [{
16         'url': 'http://www.gamestar.de/videos/trailer,3/hobbit-3-die-schlacht-der-fuenf-heere,76110.html',
17         'md5': 'ee782f1f8050448c95c5cacd63bc851c',
18         'info_dict': {
19             'id': '76110',
20             'ext': 'mp4',
21             'title': 'Hobbit 3: Die Schlacht der Fünf Heere - Teaser-Trailer zum dritten Teil',
22             'description': 'Der Teaser-Trailer zu Hobbit 3: Die Schlacht der Fünf Heere zeigt einige Szenen aus dem dritten Teil der Saga und kündigt den...',
23             'thumbnail': r're:^https?://.*\.jpg$',
24             'timestamp': 1406542380,
25             'upload_date': '20140728',
26             'duration': 17,
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         'url': 'http://www.gamestar.de/videos/top-10-indie-spiele-fuer-nintendo-switch-video-tolle-nindies-games-zum-download,95316.html',
33         'only_matching': True,
34     }]
35
36     def _real_extract(self, url):
37         mobj = re.match(self._VALID_URL, url)
38         site = mobj.group('site')
39         video_id = mobj.group('id')
40
41         webpage = self._download_webpage(url, video_id)
42
43         # TODO: there are multiple ld+json objects in the webpage,
44         # while _search_json_ld finds only the first one
45         json_ld = self._parse_json(self._search_regex(
46             r'(?s)<script[^>]+type=(["\'])application/ld\+json\1[^>]*>(?P<json_ld>[^<]+VideoObject[^<]+)</script>',
47             webpage, 'JSON-LD', group='json_ld'), video_id)
48         info_dict = self._json_ld(json_ld, video_id)
49         info_dict['title'] = remove_end(
50             info_dict['title'], ' - Game%s' % site.title())
51
52         view_count = int_or_none(json_ld.get('interactionCount'))
53         comment_count = int_or_none(self._html_search_regex(
54             r'<span>Kommentare</span>\s*<span[^>]+class=["\']count[^>]+>\s*\(\s*([0-9]+)',
55             webpage, 'comment count', fatal=False))
56
57         info_dict.update({
58             'id': video_id,
59             'url': 'http://gamestar.de/_misc/videos/portal/getVideoUrl.cfm?premium=0&videoId=' + video_id,
60             'ext': 'mp4',
61             'view_count': view_count,
62             'comment_count': comment_count
63         })
64
65         return info_dict