]> Raphaël G. Git Repositories - youtubedl/blobdiff - youtube_dl/extractor/ina.py
Prepare to upload
[youtubedl] / youtube_dl / extractor / ina.py
index c19b956592e4b6099bb0d6d42b8c1f9bb3aea5cd..9544ff9d469c52a932cc3a8fed0dafbed9f4ae83 100644 (file)
@@ -1,31 +1,36 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
 import re
 
 from .common import InfoExtractor
 
 
 class InaIE(InfoExtractor):
-    """Information Extractor for Ina.fr"""
-    _VALID_URL = r'(?:http://)?(?:www\.)?ina\.fr/video/(?P<id>I[0-9]+)/.*'
-
-    def _real_extract(self,url):
+    _VALID_URL = r'https?://(?:www\.)?ina\.fr/video/(?P<id>I?[A-Z0-9]+)'
+    _TEST = {
+        'url': 'http://www.ina.fr/video/I12055569/francois-hollande-je-crois-que-c-est-clair-video.html',
+        'md5': 'a667021bf2b41f8dc6049479d9bb38a3',
+        'info_dict': {
+            'id': 'I12055569',
+            'ext': 'mp4',
+            'title': 'François Hollande "Je crois que c\'est clair"',
+        }
+    }
+
+    def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
 
         video_id = mobj.group('id')
-        mrss_url='http://player.ina.fr/notices/%s.mrss' % video_id
-        video_extension = 'mp4'
-        webpage = self._download_webpage(mrss_url, video_id)
+        mrss_url = 'http://player.ina.fr/notices/%s.mrss' % video_id
+        info_doc = self._download_xml(mrss_url, video_id)
 
         self.report_extraction(video_id)
 
-        video_url = self._html_search_regex(r'<media:player url="(?P<mp4url>http://mp4.ina.fr/[^"]+\.mp4)',
-            webpage, u'video URL')
-
-        video_title = self._search_regex(r'<title><!\[CDATA\[(?P<titre>.*?)]]></title>',
-            webpage, u'title')
+        video_url = info_doc.find('.//{http://search.yahoo.com/mrss/}player').attrib['url']
 
-        return [{
-            'id':       video_id,
-            'url':      video_url,
-            'ext':      video_extension,
-            'title':    video_title,
-        }]
+        return {
+            'id': video_id,
+            'url': video_url,
+            'title': info_doc.find('.//title').text,
+        }