Use canonical URL in Vcs-Git.

[youtubedl] / youtube_dl / extractor / traileraddict.py

diff --git a/youtube_dl/extractor/traileraddict.py b/youtube_dl/extractor/traileraddict.py
index 9dd26c1637f58e5a823a6cf75ba8fd4ee42f0750..747370d12d7fc8fd1c66b1ac101db0ba01c963e5 100644 (file)
--- a/youtube_dl/extractor/traileraddict.py
+++ b/youtube_dl/extractor/traileraddict.py
@@ -1,49 +1,64 @@
+from __future__ import unicode_literals
+
 import re
 
 from .common import InfoExtractor
 
 
 class TrailerAddictIE(InfoExtractor):
-    _VALID_URL = r'(?:http://)?(?:www\.)?traileraddict\.com/trailer/([^/]+)/(?:trailer|feature-trailer)'
+    _WORKING = False
+    _VALID_URL = r'(?:https?://)?(?:www\.)?traileraddict\.com/(?:trailer|clip)/(?P<movie>.+?)/(?P<trailer_name>.+)'
     _TEST = {
-        u'url': u'http://www.traileraddict.com/trailer/prince-avalanche/trailer',
-        u'file': u'76184.mp4',
-        u'md5': u'41365557f3c8c397d091da510e73ceb4',
-        u'info_dict': {
-            u"title": u"Prince Avalanche Trailer",
-            u"description": u"Trailer for Prince Avalanche.Two highway road workers spend the summer of 1988 away from their city lives. The isolated landscape becomes a place of misadventure as the men find themselves at odds with each other and the women they left behind."
+        'url': 'http://www.traileraddict.com/trailer/prince-avalanche/trailer',
+        'md5': '41365557f3c8c397d091da510e73ceb4',
+        'info_dict': {
+            'id': '76184',
+            'ext': 'mp4',
+            'title': 'Prince Avalanche Trailer',
+            'description': 'Trailer for Prince Avalanche.\n\nTwo highway road workers spend the summer of 1988 away from their city lives. The isolated landscape becomes a place of misadventure as the men find themselves at odds with each other and the women they left behind.',
         }
     }
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group(1)
-        webpage = self._download_webpage(url, video_id)
-        
+        name = mobj.group('movie') + '/' + mobj.group('trailer_name')
+        webpage = self._download_webpage(url, name)
+
         title = self._search_regex(r'<title>(.+?)</title>',
-                webpage, 'video title').replace(' - Trailer Addict','')
-        view_count = self._search_regex(r'Views: (.+?)<br />',
-                webpage, 'Views Count')
-        description = self._search_regex(r'<meta property="og:description" content="(.+?)" />',
-                webpage, 'video description')
-        video_id = self._search_regex(r'<meta property="og:video" content="(.+?)" />',
-                webpage, 'Video id').split('=')[1]
-        
-        info_url = "http://www.traileraddict.com/fvar.php?tid=%s" %(str(video_id))
-        info_webpage = self._download_webpage(info_url, video_id , "Downloading the info webpage")
-        
+                                   webpage, 'video title').replace(' - Trailer Addict', '')
+        view_count_str = self._search_regex(
+            r'<span class="views_n">([0-9,.]+)</span>',
+            webpage, 'view count', fatal=False)
+        view_count = (
+            None if view_count_str is None
+            else int(view_count_str.replace(',', '')))
+        video_id = self._search_regex(
+            r'<param\s+name="movie"\s+value="/emb/([0-9]+)"\s*/>',
+            webpage, 'video id')
+
+        # Presence of (no)watchplus function indicates HD quality is available
+        if re.search(r'function (no)?watchplus()', webpage):
+            fvar = 'fvarhd'
+        else:
+            fvar = 'fvar'
+
+        info_url = 'http://www.traileraddict.com/%s.php?tid=%s' % (fvar, str(video_id))
+        info_webpage = self._download_webpage(info_url, video_id, 'Downloading the info webpage')
+
         final_url = self._search_regex(r'&fileurl=(.+)',
-                info_webpage, 'Download url').replace('%3F','?')
+                                       info_webpage, 'Download url').replace('%3F', '?')
         thumbnail_url = self._search_regex(r'&image=(.+?)&',
-                info_webpage, 'thumbnail url')
-        ext = final_url.split('.')[-1].split('?')[0]
-        
-        return [{
-            'id'          : video_id,
-            'url'         : final_url,
-            'ext'         : ext,
-            'title'       : title,
-            'thumbnail'   : thumbnail_url,
-            'description' : description,
-            'view_count'  : view_count,
-        }]
+                                           info_webpage, 'thumbnail url')
+
+        description = self._html_search_regex(
+            r'(?s)<div class="synopsis">.*?<div class="movie_label_info"[^>]*>(.*?)</div>',
+            webpage, 'description', fatal=False)
+
+        return {
+            'id': video_id,
+            'url': final_url,
+            'title': title,
+            'thumbnail': thumbnail_url,
+            'description': description,
+            'view_count': view_count,
+        }