From 7b4ec5c5e9a89fc1bc3b3fc8dfad26450530a2da Mon Sep 17 00:00:00 2001
From: asciimoo <asciimoo@gmail.com>
Date: Fri, 10 Jan 2014 23:38:08 +0100
Subject: [fix] highlighting only html

---
 searx/utils.py | 26 ++++++++++++++++++++++++++
 1 file changed, 26 insertions(+)

(limited to 'searx/utils.py')

diff --git a/searx/utils.py b/searx/utils.py
index 670499805..53300181f 100644
--- a/searx/utils.py
+++ b/searx/utils.py
@@ -3,6 +3,32 @@ from HTMLParser import HTMLParser
 import csv
 import codecs
 import cStringIO
+import re
+
+def highlight_content(content, query):
+
+    if not content:
+        return None
+    # ignoring html contents
+    # TODO better html content detection
+    if content.find('<') != -1:
+        return content
+
+    query = query.decode('utf-8')
+    if content.lower().find(query.lower()) > -1:
+        query_regex = u'({0})'.format(re.escape(query))
+        content = re.sub(query_regex, '<b>\\1</b>', content, flags=re.I | re.U)
+    else:
+        regex_parts = []
+        for chunk in query.split():
+            if len(chunk) == 1:
+                regex_parts.append(u'\W+{0}\W+'.format(re.escape(chunk)))
+            else:
+                regex_parts.append(u'{0}'.format(re.escape(chunk)))
+        query_regex = u'({0})'.format('|'.join(regex_parts))
+        content = re.sub(query_regex, '<b>\\1</b>', content, flags=re.I | re.U)
+
+    return content
 
 class HTMLTextExtractor(HTMLParser):
     def __init__(self):
-- 
cgit v1.2.3