X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fcanalc2.py;h=f6a1ff381a61af338fd7dec2419aaadfa1081c8b;hb=b25f7533975d7463cc772e9d106802ef29252796;hp=d0e2ed536c3551f5bdb7f1bd82b2ee956a554f21;hpb=cd0abcc0bb4c218fd02850a139b626d252e22599;p=youtube-dl

diff --git a/youtube_dl/extractor/canalc2.py b/youtube_dl/extractor/canalc2.py
index d0e2ed536..f6a1ff381 100644
--- a/youtube_dl/extractor/canalc2.py
+++ b/youtube_dl/extractor/canalc2.py
@@ -1,37 +1,56 @@
 # coding: utf-8
-"""Extractor for canalc2.tv"""
+from __future__ import unicode_literals
+
 import re
-import lxml.html
 
 from .common import InfoExtractor
+from ..utils import parse_duration
+
 
 class Canalc2IE(InfoExtractor):
-    """Extractor for canalc2.tv"""
-    _VALID_URL = r'http://.*?\.canalc2\.tv/video\.asp\?idVideo=(\d+)&voir=oui'
+    IE_NAME = 'canalc2.tv'
+    _VALID_URL = r'https?://(?:www\.)?canalc2\.tv/video/(?P<id>\d+)'
 
     _TEST = {
-        u'url': u'http://www.canalc2.tv/video.asp?idVideo=12163&voir=oui',
-        u'file': u'12163.mp4',
-        u'md5': u'c00fa80517373764ff5c0b5eb5a58780',
-        u'info_dict': {
-            u'title': u'Terrasses du Numérique'
+        'url': 'http://www.canalc2.tv/video/12163',
+        'md5': '060158428b650f896c542dfbb3d6487f',
+        'info_dict': {
+            'id': '12163',
+            'ext': 'flv',
+            'title': 'Terrasses du Numérique',
+            'duration': 122,
+        },
+        'params': {
+            'skip_download': True,  # Requires rtmpdump
         }
     }
 
     def _real_extract(self, url):
-        video_id = re.match(self._VALID_URL, url).group(1)
+        video_id = self._match_id(url)
         webpage = self._download_webpage(url, video_id)
-        file_name = re.search(r"so\.addVariable\('file','(.*?)'\);",
-            webpage).group(1)
-        
-        video_url = 'http://vod-flash.u-strasbg.fr:8080/' + file_name
-        
-        html   = lxml.html.fromstring(webpage)
-        
-        title = html.cssselect('.evenement8')[0].text_content()
-        
-        return {'id': video_id,
-                'ext' : 'mp4',
-                'url' : video_url,
-                'title' : title
-                }
+        video_url = self._search_regex(
+            r'jwplayer\((["\'])Player\1\)\.setup\({[^}]*file\s*:\s*(["\'])(?P<file>.+?)\2',
+            webpage, 'video_url', group='file')
+        formats = [{'url': video_url}]
+        if video_url.startswith('rtmp://'):
+            rtmp = re.search(r'^(?P<url>rtmp://[^/]+/(?P<app>.+/))(?P<play_path>mp4:.+)$', video_url)
+            formats[0].update({
+                'url': rtmp.group('url'),
+                'ext': 'flv',
+                'app': rtmp.group('app'),
+                'play_path': rtmp.group('play_path'),
+                'page_url': url,
+            })
+
+        title = self._html_search_regex(
+            r'(?s)class="[^"]*col_description[^"]*">.*?<h3>(.*?)</h3>', webpage, 'title')
+        duration = parse_duration(self._search_regex(
+            r'id=["\']video_duree["\'][^>]*>([^<]+)',
+            webpage, 'duration', fatal=False))
+
+        return {
+            'id': video_id,
+            'title': title,
+            'duration': duration,
+            'formats': formats,
+        }