searx

commit: 5695ec58df14615e154072e97afc4bd6ed505812
parent: 474f6a861d1be5a874deefd5d60dd45b40db0a03
Author: asciimoo <asciimoo@gmail.com>
Date:   Tue,  4 Feb 2014 19:42:32 +0100

[fix] whitespace duplication fix in nonhtml outputs

Diffstat:
M
searx/webapp.py
4
+++-

1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/searx/webapp.py b/searx/webapp.py
@@ -190,7 +190,9 @@ def index():
         else:
             if 'content' in result:
                 result['content'] = html_to_text(result['content']).strip()
-            result['title'] = html_to_text(result['title']).strip()
+            # removing html content and whitespace duplications
+            result['title'] = ' '.join(html_to_text(result['title'])\
+                                       .strip().split())
         if len(result['url']) > 74:
             url_parts = result['url'][:35], result['url'][-35:]
             result['pretty_url'] = '{0}[...]{1}'.format(*url_parts)