searx

commit: b5061b5f0fd2e3058af60038a47676ad26c671e6
parent: 909424e1e5fc3f7c3a7089d8a4ddae84e56db8dd
Author: Adam Tauber <asciimoo@gmail.com>
Date:   Tue,  9 Feb 2016 21:21:59 +0100

[fix] restrict wikipedia queries to the titles to avoid irrelevant results

Diffstat:
M
searx/engines/mediawiki.py
6
++++--

1 file changed, 4 insertions(+), 2 deletions(-)
diff --git a/searx/engines/mediawiki.py b/searx/engines/mediawiki.py
@@ -27,10 +27,10 @@ base_url = 'https://{language}.wikipedia.org/'
 search_postfix = 'w/api.php?action=query'\
     '&list=search'\
     '&{query}'\
-    '&srprop=timestamp'\
     '&format=json'\
     '&sroffset={offset}'\
-    '&srlimit={limit}'
+    '&srlimit={limit}'\
+    '&srwhat=nearmatch'  # search for a near match in the title
 
 
 # do search-request
@@ -74,6 +74,8 @@ def response(resp):
 
     # parse results
     for result in search_results['query']['search']:
+        if result.get('snippet', '').startswith('#REDIRECT'):
+            continue
         url = base_url.format(language=resp.search_params['language']) +\
             'wiki/' + quote(result['title'].replace(' ', '_').encode('utf-8'))