[nba] improve(fixes #7068)

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

diff --git a/youtube_dl/extractor/canalc2.py b/youtube_dl/extractor/canalc2.py
index 648af2e18ba62afdc8a5558cd74c6e1ad7256a82..f6a1ff381a61af338fd7dec2419aaadfa1081c8b 100644 (file)
--- a/youtube_dl/extractor/canalc2.py
+++ b/youtube_dl/extractor/canalc2.py
@@ -4,19 +4,21 @@ 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'https?://(www\.)?canalc2\.tv/video/(?P<id>\d+)'
+    _VALID_URL = r'https?://(?:www\.)?canalc2\.tv/video/(?P<id>\d+)'
 
     _TEST = {
         'url': 'http://www.canalc2.tv/video/12163',
         'md5': '060158428b650f896c542dfbb3d6487f',
         'info_dict': {
             'id': '12163',
-            'ext': 'mp4',
-            'title': 'Terrasses du Numérique'
+            'ext': 'flv',
+            'title': 'Terrasses du Numérique',
+            'duration': 122,
         },
         'params': {
             'skip_download': True,  # Requires rtmpdump
@@ -27,13 +29,14 @@ class Canalc2IE(InfoExtractor):
         video_id = self._match_id(url)
         webpage = self._download_webpage(url, video_id)
         video_url = self._search_regex(
-            r'jwplayer\("Player"\).setup\({[^}]*file: "([^"]+)"',
-            webpage, 'video_url')
+            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,
@@ -41,9 +44,13 @@ class Canalc2IE(InfoExtractor):
 
         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,
-            'formats': formats,
             'title': title,
+            'duration': duration,
+            'formats': formats,
         }