X-Git-Url: http://git.cielonegro.org/gitweb.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Farte.py;h=b528a9ec50ca6c2dac1a52fe66de2cd66194dd23;hb=9e8753911cf37a4f8e26ce442e8938a7a52f3dad;hp=257dc1f61f05b2701350004d817f0624cc2d32fc;hpb=893f8832b52926847353f2b678e313687806a775;p=youtube-dl.git

diff --git a/youtube_dl/extractor/arte.py b/youtube_dl/extractor/arte.py
index 257dc1f61..b528a9ec5 100644
--- a/youtube_dl/extractor/arte.py
+++ b/youtube_dl/extractor/arte.py
@@ -2,7 +2,6 @@
 from __future__ import unicode_literals
 
 import re
-import json
 
 from .common import InfoExtractor
 from ..utils import (
@@ -25,6 +24,10 @@ class ArteTvIE(InfoExtractor):
     IE_NAME = 'arte.tv'
 
     def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        lang = mobj.group('lang')
+        video_id = mobj.group('id')
+
         ref_xml_url = url.replace('/videos/', '/do_delegate/videos/')
         ref_xml_url = ref_xml_url.replace('.html', ',view,asPlayerXml.xml')
         ref_xml_doc = self._download_xml(
@@ -37,8 +40,9 @@ class ArteTvIE(InfoExtractor):
         formats = [{
             'forma_id': q.attrib['quality'],
             'url': q.text,
+            'ext': 'flv',
             'quality': 2 if q.attrib['quality'] == 'hd' else 1,
-        } for q in config.findall('.//quality')]
+        } for q in config.findall('./urls/url')]
         self._sort_formats(formats)
 
         title = config.find('.//name').text
@@ -47,8 +51,7 @@ class ArteTvIE(InfoExtractor):
             'id': video_id,
             'title': title,
             'thumbnail': thumbnail,
-            'url': video_url,
-            'ext': 'flv',
+            'formats': formats,
         }
 
 
@@ -71,7 +74,8 @@ class ArteTVPlus7IE(InfoExtractor):
         return self._extract_from_webpage(webpage, video_id, lang)
 
     def _extract_from_webpage(self, webpage, video_id, lang):
-        json_url = self._html_search_regex(r'arte_vp_url="(.*?)"', webpage, 'json url')
+        json_url = self._html_search_regex(
+            r'arte_vp_url="(.*?)"', webpage, 'json vp url')
         return self._extract_from_json_url(json_url, video_id, lang)
 
     def _extract_from_json_url(self, json_url, video_id, lang):
@@ -117,14 +121,17 @@ class ArteTVPlus7IE(InfoExtractor):
                 return ['HQ', 'MQ', 'EQ', 'SQ'].index(f['quality'])
         else:
             def sort_key(f):
+                versionCode = f.get('versionCode')
+                if versionCode is None:
+                    versionCode = ''
                 return (
                     # Sort first by quality
-                    int(f.get('height',-1)),
-                    int(f.get('bitrate',-1)),
+                    int(f.get('height', -1)),
+                    int(f.get('bitrate', -1)),
                     # The original version with subtitles has lower relevance
-                    re.match(r'VO-ST(F|A)', f.get('versionCode', '')) is None,
+                    re.match(r'VO-ST(F|A)', versionCode) is None,
                     # The version with sourds/mal subtitles has also lower relevance
-                    re.match(r'VO?(F|A)-STM\1', f.get('versionCode', '')) is None,
+                    re.match(r'VO?(F|A)-STM\1', versionCode) is None,
                     # Prefer http downloads over m3u8
                     0 if f['url'].endswith('m3u8') else 1,
                 )