[youtube] Fix extraction.

[youtube-dl] / youtube_dl / extractor / franceinter.py

diff --git a/youtube_dl/extractor/franceinter.py b/youtube_dl/extractor/franceinter.py
index 47e06f5a1ada8a976a7988b438e0fd06115cc799..05806895c0a3300dd9db8e68efb15fcb70bb8006 100644 (file)
--- a/youtube_dl/extractor/franceinter.py
+++ b/youtube_dl/extractor/franceinter.py
@@ -1,40 +1,56 @@
-import re
+# coding: utf-8
+from __future__ import unicode_literals
 
 from .common import InfoExtractor
+from ..utils import month_by_name
+
+
 class FranceInterIE(InfoExtractor):
-    
-    _VALID_URL=r'http://www.franceinter.fr/player/reecouter\?play=(?P<id>[0-9]{6})'
-    IE_NAME='FranceInter'
-    _TEST={
-           u'url':u'http://www.franceinter.fr/player/reecouter?play=793962',
-           u'file':u'793962.mp3'
-           
-           
-           }
-           
-    #Easier to use python string matching than regex for a simple match
-    def get_download_url(self,webpage):
-        
-        start=webpage.index('&urlAOD=')+8
-        end=webpage.index('&startTime')
-        return u'http://www.franceinter.fr/%s'%webpage[start:end]
-        
-    def get_title(self,webpage):
-        start=webpage.index('<span class="title diffusion">')+30
-        end=webpage.index('</span> dans')
-        
-        return webpage[start:end]
-    def _real_extract(self,url):
-
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
-        
-        webpage=self._download_webpage(url,video_id)
-        
-        title=self.get_title(webpage)
-        
-        video_url=self.get_download_url(webpage)
-        
-        return{'id': video_id,u'url': video_url,u'title': title}
-        
-        
+    _VALID_URL = r'https?://(?:www\.)?franceinter\.fr/emissions/(?P<id>[^?#]+)'
+
+    _TEST = {
+        'url': 'https://www.franceinter.fr/emissions/affaires-sensibles/affaires-sensibles-07-septembre-2016',
+        'md5': '9e54d7bdb6fdc02a841007f8a975c094',
+        'info_dict': {
+            'id': 'affaires-sensibles/affaires-sensibles-07-septembre-2016',
+            'ext': 'mp3',
+            'title': 'Affaire Cahuzac : le contentieux du compte en Suisse',
+            'description': 'md5:401969c5d318c061f86bda1fa359292b',
+            'upload_date': '20160907',
+        },
+    }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, video_id)
+
+        video_url = self._search_regex(
+            r'(?s)<div[^>]+class=["\']page-diffusion["\'][^>]*>.*?<button[^>]+data-url=(["\'])(?P<url>(?:(?!\1).)+)\1',
+            webpage, 'video url', group='url')
+
+        title = self._og_search_title(webpage)
+        description = self._og_search_description(webpage)
+
+        upload_date_str = self._search_regex(
+            r'class=["\']\s*cover-emission-period\s*["\'][^>]*>[^<]+\s+(\d{1,2}\s+[^\s]+\s+\d{4})<',
+            webpage, 'upload date', fatal=False)
+        if upload_date_str:
+            upload_date_list = upload_date_str.split()
+            upload_date_list.reverse()
+            upload_date_list[1] = '%02d' % (month_by_name(upload_date_list[1], lang='fr') or 0)
+            upload_date_list[2] = '%02d' % int(upload_date_list[2])
+            upload_date = ''.join(upload_date_list)
+        else:
+            upload_date = None
+
+        return {
+            'id': video_id,
+            'title': title,
+            'description': description,
+            'upload_date': upload_date,
+            'formats': [{
+                'url': video_url,
+                'vcodec': 'none',
+            }],
+        }