[youtube] Fix extraction.
[youtube-dl] / youtube_dl / extractor / canalc2.py
index f6a1ff381a61af338fd7dec2419aaadfa1081c8b..407cc8084836cec7f82113eb71ff8e0314a8e932 100644 (file)
@@ -9,48 +9,65 @@ from ..utils import parse_duration
 
 class Canalc2IE(InfoExtractor):
     IE_NAME = 'canalc2.tv'
-    _VALID_URL = r'https?://(?:www\.)?canalc2\.tv/video/(?P<id>\d+)'
+    _VALID_URL = r'https?://(?:(?:www\.)?canalc2\.tv/video/|archives-canalc2\.u-strasbg\.fr/video\.asp\?.*\bidVideo=)(?P<id>\d+)'
 
-    _TEST = {
+    _TESTS = [{
         'url': 'http://www.canalc2.tv/video/12163',
         'md5': '060158428b650f896c542dfbb3d6487f',
         'info_dict': {
             'id': '12163',
-            'ext': 'flv',
+            'ext': 'mp4',
             'title': 'Terrasses du Numérique',
             'duration': 122,
         },
-        'params': {
-            'skip_download': True,  # Requires rtmpdump
-        }
-    }
+    }, {
+        'url': 'http://archives-canalc2.u-strasbg.fr/video.asp?idVideo=11427&voir=oui',
+        'only_matching': True,
+    }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
-        webpage = self._download_webpage(url, video_id)
-        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,
-            })
+
+        webpage = self._download_webpage(
+            'http://www.canalc2.tv/video/%s' % video_id, video_id)
 
         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))
+            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'])
 
-        return {
+        info.update({
             'id': video_id,
             'title': title,
-            'duration': duration,
-            'formats': formats,
-        }
+            'duration': parse_duration(self._search_regex(
+                r'id=["\']video_duree["\'][^>]*>([^<]+)',
+                webpage, 'duration', fatal=False)),
+        })
+        return info