[youtube] Fix extraction.

[youtube-dl] / youtube_dl / extractor / canalc2.py

diff --git a/youtube_dl/extractor/canalc2.py b/youtube_dl/extractor/canalc2.py
index 50832217a8ad5fe52b4cfa62731171b6dfaadcd6..407cc8084836cec7f82113eb71ff8e0314a8e932 100644 (file)
--- a/youtube_dl/extractor/canalc2.py
+++ b/youtube_dl/extractor/canalc2.py
@@ -1,35 +1,73 @@
 # coding: utf-8
+from __future__ import unicode_literals
+
 import re
 
 from .common import InfoExtractor
+from ..utils import parse_duration
 
 
 class Canalc2IE(InfoExtractor):
-    _IE_NAME = 'canalc2.tv'
-    _VALID_URL = r'http://.*?\.canalc2\.tv/video\.asp\?idVideo=(\d+)&voir=oui'
-
-    _TEST = {
-        u'url': u'http://www.canalc2.tv/video.asp?idVideo=12163&voir=oui',
-        u'file': u'12163.mp4',
-        u'md5': u'060158428b650f896c542dfbb3d6487f',
-        u'info_dict': {
-            u'title': u'Terrasses du Numérique'
-        }
-    }
+    IE_NAME = 'canalc2.tv'
+    _VALID_URL = r'https?://(?:(?:www\.)?canalc2\.tv/video/|archives-canalc2\.u-strasbg\.fr/video\.asp\?.*\bidVideo=)(?P<id>\d+)'
+
+    _TESTS = [{
+        'url': 'http://www.canalc2.tv/video/12163',
+        'md5': '060158428b650f896c542dfbb3d6487f',
+        'info_dict': {
+            'id': '12163',
+            'ext': 'mp4',
+            'title': 'Terrasses du Numérique',
+            'duration': 122,
+        },
+    }, {
+        'url': 'http://archives-canalc2.u-strasbg.fr/video.asp?idVideo=11427&voir=oui',
+        'only_matching': True,
+    }]
 
     def _real_extract(self, url):
-        video_id = re.match(self._VALID_URL, url).group(1)
-        webpage = self._download_webpage(url, video_id)
-        file_name = self._search_regex(
-            r"so\.addVariable\('file','(.*?)'\);",
-            webpage, 'file name')
-        video_url = 'http://vod-flash.u-strasbg.fr:8080/' + file_name
+        video_id = self._match_id(url)
+
+        webpage = self._download_webpage(
+            'http://www.canalc2.tv/video/%s' % video_id, video_id)
 
         title = self._html_search_regex(
-            r'class="evenement8">(.*?)</a>', webpage, u'title')
-        
-        return {'id': video_id,
-                'ext': 'mp4',
-                'url': video_url,
-                'title': title,
-                }
+            r'(?s)class="[^"]*col_description[^"]*">.*?<h3>(.+?)</h3>',
+            webpage, 'title')
+
+        formats = []
+        for _, video_url in re.findall(r'file\s*=\s*(["\'])(.+?)\1', webpage):
+            if video_url.startswith('rtmp://'):
+                rtmp = re.search(
+                    r'^(?P<url>rtmp://[^/]+/(?P<app>.+/))(?P<play_path>mp4:.+)$', video_url)
+                formats.append({
+                    'url': rtmp.group('url'),
+                    'format_id': 'rtmp',
+                    'ext': 'flv',
+                    'app': rtmp.group('app'),
+                    'play_path': rtmp.group('play_path'),
+                    'page_url': url,
+                })
+            else:
+                formats.append({
+                    'url': video_url,
+                    'format_id': 'http',
+                })
+
+        if formats:
+            info = {
+                'formats': formats,
+            }
+        else:
+            info = self._parse_html5_media_entries(url, webpage, url)[0]
+
+        self._sort_formats(info['formats'])
+
+        info.update({
+            'id': video_id,
+            'title': title,
+            'duration': parse_duration(self._search_regex(
+                r'id=["\']video_duree["\'][^>]*>([^<]+)',
+                webpage, 'duration', fatal=False)),
+        })
+        return info