From: Philipp Hagemeister <phihag@phihag.de>
Date: Sat, 13 Sep 2014 04:55:38 +0000 (+0200)
Subject: [spiegel] Add support for articles
X-Git-Url: http://git.oshgnacknak.de/?a=commitdiff_plain;h=89fb6a979765689a744e5e5bfbf3b397ed40b33a;p=youtube-dl

[spiegel] Add support for articles
---

diff --git a/youtube_dl/extractor/__init__.py b/youtube_dl/extractor/__init__.py
index 4423f78a5..f1a5b2608 100644
--- a/youtube_dl/extractor/__init__.py
+++ b/youtube_dl/extractor/__init__.py
@@ -325,7 +325,7 @@ from .southpark import (
 )
 from .space import SpaceIE
 from .spankwire import SpankwireIE
-from .spiegel import SpiegelIE
+from .spiegel import SpiegelIE, SpiegelArticleIE
 from .spiegeltv import SpiegeltvIE
 from .spike import SpikeIE
 from .sportdeutschland import SportDeutschlandIE
diff --git a/youtube_dl/extractor/spiegel.py b/youtube_dl/extractor/spiegel.py
index 340a38440..a3b89f751 100644
--- a/youtube_dl/extractor/spiegel.py
+++ b/youtube_dl/extractor/spiegel.py
@@ -4,6 +4,7 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
+from ..utils import compat_urlparse
 
 
 class SpiegelIE(InfoExtractor):
@@ -82,3 +83,34 @@ class SpiegelIE(InfoExtractor):
             'duration': duration,
             'formats': formats,
         }
+
+
+class SpiegelArticleIE(InfoExtractor):
+    _VALID_URL = 'https?://www\.spiegel\.de/(?!video/)[^?#]*?-(?P<id>[0-9]+)\.html'
+    IE_NAME = 'Spiegel:Article'
+    IE_DESC = 'Articles on spiegel.de'
+    _TEST = {
+        'url': 'http://www.spiegel.de/sport/sonst/badminton-wm-die-randsportart-soll-populaerer-werden-a-987092.html',
+        'info_dict': {
+            'id': '1516455',
+            'ext': 'mp4',
+            'title': 'Faszination Badminton: Nennt es bloß nicht Federball',
+            'description': 're:^Patrick Kämnitz gehört.{100,}',
+        },
+    }
+
+    def _real_extract(self, url):
+        m = re.match(self._VALID_URL, url)
+        video_id = m.group('id')
+
+        webpage = self._download_webpage(url, video_id)
+        video_link = self._search_regex(
+            r'<a href="([^"]+)" onclick="return spOpenVideo\(this,', webpage,
+            'video page URL')
+        video_url = compat_urlparse.urljoin(
+            self.http_scheme() + '//spiegel.de/', video_link)
+
+        return {
+            '_type': 'url',
+            'url': video_url,
+        }