[traileraddict] Fix extraction

youtube_dl/extractor/traileraddict.py

@@ -1,3 +1,5 @@
+from __future__ import unicode_literals
+
 import re
 
 from .common import InfoExtractor
@@ -6,12 +8,13 @@ from .common import InfoExtractor
 class TrailerAddictIE(InfoExtractor):
     _VALID_URL = r'(?:http://)?(?:www\.)?traileraddict\.com/(?:trailer|clip)/(?P<movie>.+?)/(?P<trailer_name>.+)'
     _TEST = {
-        u'url': u'http://www.traileraddict.com/trailer/prince-avalanche/trailer',
+        'url': 'http://www.traileraddict.com/trailer/prince-avalanche/trailer',
-        u'file': u'76184.mp4',
+        'md5': '41365557f3c8c397d091da510e73ceb4',
-        u'md5': u'57e39dbcf4142ceb8e1f242ff423fd71',
+        'info_dict': {
-        u'info_dict': {
+            'id': '76184',
-            u"title": u"Prince Avalanche Trailer",
+            'ext': 'mp4',
-            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."
+            '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.',
         }
     }
 
@@ -22,9 +25,15 @@ class TrailerAddictIE(InfoExtractor):
 
         title = self._search_regex(r'<title>(.+?)</title>',
                 webpage, 'video title').replace(' - Trailer Addict','')
-        view_count = self._search_regex(r'Views: (.+?)<br />',
+        view_count_str = self._search_regex(
-                webpage, 'Views Count')
+            r'<span class="views_n">([0-9,.]+)</span>',
-        video_id = self._og_search_property('video', webpage, 'Video id').split('=')[1]
+            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):
@@ -39,14 +48,16 @@ class TrailerAddictIE(InfoExtractor):
                 info_webpage, 'Download url').replace('%3F','?')
         thumbnail_url = self._search_regex(r'&image=(.+?)&',
                 info_webpage, 'thumbnail url')
-        ext = final_url.split('.')[-1].split('?')[0]
+
-
+        description = self._html_search_regex(
-        return [{
+            r'(?s)<div class="synopsis">.*?<div class="movie_label_info"[^>]*>(.*?)</div>',
-            'id'          : video_id,
+            webpage, 'description', fatal=False)
-            'url'         : final_url,
+
-            'ext'         : ext,
+        return {
-            'title'       : title,
+            'id': video_id,
-            'thumbnail'   : thumbnail_url,
+            'url': final_url,
-            'description' : self._og_search_description(webpage),
+            'title': title,
-            'view_count'  : view_count,
+            'thumbnail': thumbnail_url,
-        }]
+            'description': description,
+            'view_count': view_count,
+        }