diff --git a/source_code/2-4-practice-baidu-baike.py b/source_code/2-4-practice-baidu-baike.py
index 4814e9f..1d14c23 100644
--- a/source_code/2-4-practice-baidu-baike.py
+++ b/source_code/2-4-practice-baidu-baike.py
@@ -2,13 +2,13 @@
 from urllib.request import urlopen
 import re
 import random
-
+from urllib.parse import quote
 
 base_url = "https://baike.baidu.com"
+
 his = ["/item/%E7%BD%91%E7%BB%9C%E7%88%AC%E8%99%AB/5162711"]
 
 for i in range(20):
-    # dealing with Chinese symbols
     url = base_url + his[-1]
 
     html = urlopen(url).read().decode('utf-8')
@@ -16,10 +16,16 @@
     print(i, soup.find('h1').get_text(), '    url: ', his[-1])
 
     # find valid urls
-    sub_urls = soup.find_all("a", {"target": "_blank", "href": re.compile("/item/(%.{2})+$")})
+    sub_urls = soup.find_all("a", {"target": "_blank", "href": re.compile("^(/item/)((%.{2})+|[\u4e00-\u9fa5]+)$")})
 
     if len(sub_urls) != 0:
-        his.append(random.sample(sub_urls, 1)[0]['href'])
+        temp=random.sample(sub_urls, 1)[0]['href']
+        if len(re.findall("([\u4e00-\u9fa5]+)",temp))==0:
+            his.append(temp)
+        else:
+            a,=re.findall("([\u4e00-\u9fa5]+)", temp)
+            b="/item/"+quote(a)
+            his.append(b)
     else:
         # no valid sub link found
         his.pop()