[youtube] Fix extraction.

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

diff --git a/youtube_dl/extractor/ctvnews.py b/youtube_dl/extractor/ctvnews.py
index e14b30085a397a43c9b0c0999b584bdd084bf45c..03f8cefb77beb1185e14c2f435616d8f25f70f83 100644 (file)
--- a/youtube_dl/extractor/ctvnews.py
+++ b/youtube_dl/extractor/ctvnews.py
@@ -4,16 +4,17 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
+from ..utils import orderedSet
 
 
 class CTVNewsIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?ctvnews\.ca/(?:video\?(?:clip|playlist|bin)Id=|.*?)(?P<id>[0-9.]+)'
+    _VALID_URL = r'https?://(?:.+?\.)?ctvnews\.ca/(?:video\?(?:clip|playlist|bin)Id=|.*?)(?P<id>[0-9.]+)'
     _TESTS = [{
         'url': 'http://www.ctvnews.ca/video?clipId=901995',
-        'md5': '10deb320dc0ccb8d01d34d12fc2ea672',
+        'md5': '9b8624ba66351a23e0b6e1391971f9af',
         'info_dict': {
             'id': '901995',
-            'ext': 'mp4',
+            'ext': 'flv',
             'title': 'Extended: \'That person cannot be me\' Johnson says',
             'description': 'md5:958dd3b4f5bbbf0ed4d045c790d89285',
             'timestamp': 1467286284,
@@ -27,18 +28,21 @@ class CTVNewsIE(InfoExtractor):
         },
         'playlist_mincount': 19,
     }, {
-        'url': 'http://www.ctvnews.ca/video?binId=1.810401',
+        'url': 'http://www.ctvnews.ca/video?binId=1.2876780',
         'info_dict':
         {
-            'id': '1.810401',
+            'id': '1.2876780',
         },
-        'playlist_mincount': 91,
+        'playlist_mincount': 100,
     }, {
         'url': 'http://www.ctvnews.ca/1.810401',
         'only_matching': True,
     }, {
         'url': 'http://www.ctvnews.ca/canadiens-send-p-k-subban-to-nashville-in-blockbuster-trade-1.2967231',
         'only_matching': True,
+    }, {
+        'url': 'http://vancouverisland.ctvnews.ca/video?clipId=761241',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
@@ -57,8 +61,8 @@ class CTVNewsIE(InfoExtractor):
         else:
             webpage = self._download_webpage('http://www.ctvnews.ca/%s' % page_id, page_id, query={
                 'ot': 'example.AjaxPageLayout.ot',
-                'maxItemsPerPage': 20,
+                'maxItemsPerPage': 1000000,
             })
-            entries = [ninecninemedia_url_result(clip_id) for clip_id in set(
+            entries = [ninecninemedia_url_result(clip_id) for clip_id in orderedSet(
                 re.findall(r'clip\.id\s*=\s*(\d+);', webpage))]
             return self.playlist_result(entries, page_id)