From 90e1db3e5c628ae73800d51b53032064aadbfb44 Mon Sep 17 00:00:00 2001 From: Alexandre Flament Date: Sat, 31 Dec 2016 13:56:09 +0100 Subject: [PATCH] [fix] extract_text: use html.tostring instead html_to_text. Fix #711 --- searx/engines/xpath.py | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/searx/engines/xpath.py b/searx/engines/xpath.py index e5c0c5bea..18943bba4 100644 --- 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):