[fix] extract_text: use html.tostring instead html_to_text. Fix #711

2016-12-31 13:56:09 +01:00 · 2016-12-31 13:56:09 +01:00 · 90e1db3e5c
commit 90e1db3e5c
parent eaa0fb8102
1 changed files with 3 additions and 1 deletions
--- a/searx/engines/xpath.py
+++ b/searx/engines/xpath.py
@ -42,7 +42,9 @@ def extract_text(xpath_results):
        return ''.join(xpath_results)
    else:
        # it's a element
-        return html_to_text(xpath_results.text_content()).strip()
+        text = html.tostring(xpath_results, encoding='unicode', method='text', with_tail=False)
+        text = text.strip().replace('\n', ' ')
+        return ' '.join(text.split())


 def extract_url(xpath_results, search_url):