[gamestar] Fix metadata extraction (closes #10479)

This commit is contained in:
Yen Chi Hsuan 2016-09-07 23:07:50 +08:00
parent 92c9c2a88b
commit 84b91dd4e3
No known key found for this signature in database
GPG key ID: 3FDDD575826C5C30
2 changed files with 18 additions and 34 deletions

View file

@ -1,6 +1,7 @@
version <unreleased> version <unreleased>
Extractors Extractors
* [gamestar] Fix metadata extraction (#10479)
+ [bilibili] Support episodes (#10190) + [bilibili] Support episodes (#10190)
+ [tvnoe] New extractor (#10524) + [tvnoe] New extractor (#10524)

View file

@ -1,14 +1,10 @@
# coding: utf-8 # coding: utf-8
from __future__ import unicode_literals from __future__ import unicode_literals
import re
from .common import InfoExtractor from .common import InfoExtractor
from ..utils import ( from ..utils import (
int_or_none, int_or_none,
parse_duration, remove_end,
str_to_int,
unified_strdate,
) )
@ -21,8 +17,9 @@ class GameStarIE(InfoExtractor):
'id': '76110', 'id': '76110',
'ext': 'mp4', 'ext': 'mp4',
'title': 'Hobbit 3: Die Schlacht der Fünf Heere - Teaser-Trailer zum dritten Teil', 'title': 'Hobbit 3: Die Schlacht der Fünf Heere - Teaser-Trailer zum dritten Teil',
'description': 'Der Teaser-Trailer zu Hobbit 3: Die Schlacht der Fünf Heere zeigt einige Szenen aus dem dritten Teil der Saga und kündigt den vollständigen Trailer an.', 'description': 'Der Teaser-Trailer zu Hobbit 3: Die Schlacht der Fünf Heere zeigt einige Szenen aus dem dritten Teil der Saga und kündigt den...',
'thumbnail': 'http://images.gamestar.de/images/idgwpgsgp/bdb/2494525/600x.jpg', 'thumbnail': 're:^https?://.*\.jpg$',
'timestamp': 1406542020,
'upload_date': '20140728', 'upload_date': '20140728',
'duration': 17 'duration': 17
} }
@ -32,41 +29,27 @@ def _real_extract(self, url):
video_id = self._match_id(url) video_id = self._match_id(url)
webpage = self._download_webpage(url, video_id) webpage = self._download_webpage(url, video_id)
og_title = self._og_search_title(webpage)
title = re.sub(r'\s*- Video (bei|-) GameStar\.de$', '', og_title)
url = 'http://gamestar.de/_misc/videos/portal/getVideoUrl.cfm?premium=0&videoId=' + video_id url = 'http://gamestar.de/_misc/videos/portal/getVideoUrl.cfm?premium=0&videoId=' + video_id
description = self._og_search_description(webpage).strip() # TODO: there are multiple ld+json objects in the webpage,
# while _search_json_ld finds only the first one
thumbnail = self._proto_relative_url( json_ld = self._parse_json(self._search_regex(
self._og_search_thumbnail(webpage), scheme='http:') r'(?s)<script[^>]+type=(["\'])application/ld\+json\1[^>]*>(?P<json_ld>[^<]+VideoObject[^<]+)</script>',
webpage, 'JSON-LD', group='json_ld'), video_id)
upload_date = unified_strdate(self._html_search_regex( info_dict = self._json_ld(json_ld, video_id)
r'<span style="float:left;font-size:11px;">Datum: ([0-9]+\.[0-9]+\.[0-9]+)&nbsp;&nbsp;', info_dict['title'] = remove_end(info_dict['title'], ' - GameStar')
webpage, 'upload_date', fatal=False))
duration = parse_duration(self._html_search_regex(
r'&nbsp;&nbsp;Länge: ([0-9]+:[0-9]+)</span>', webpage, 'duration',
fatal=False))
view_count = str_to_int(self._html_search_regex(
r'&nbsp;&nbsp;Zuschauer: ([0-9\.]+)&nbsp;&nbsp;', webpage,
'view_count', fatal=False))
view_count = json_ld.get('interactionCount')
comment_count = int_or_none(self._html_search_regex( comment_count = int_or_none(self._html_search_regex(
r'>Kommentieren \(([0-9]+)\)</a>', webpage, 'comment_count', r'([0-9]+) Kommentare</span>', webpage, 'comment_count',
fatal=False)) fatal=False))
return { info_dict.update({
'id': video_id, 'id': video_id,
'title': title,
'url': url, 'url': url,
'ext': 'mp4', 'ext': 'mp4',
'thumbnail': thumbnail,
'description': description,
'upload_date': upload_date,
'duration': duration,
'view_count': view_count, 'view_count': view_count,
'comment_count': comment_count 'comment_count': comment_count
} })
return info_dict