How to Fix Chinese Character Garbling in JSP, Servlets, and Java Web Apps

Case 1: JSP page shows garbled Chinese

Problem: When a JSP page is accessed via a browser, Chinese characters are displayed as unreadable symbols.

Solution: Ensure the JSP file is saved with UTF‑8 encoding in the editor, then add the directive at the top of the page: <%@ pageEncoding="utf-8" %> This forces the JSP compiler to treat the source as UTF‑8 and resolves the garbling.

Case 2: Servlet response shows garbled Chinese

Problem: A servlet outputs Chinese text, but the browser displays it incorrectly.

Solution: Before writing any output, set the response’s content type with UTF‑8 charset: response.setContentType("text/html;charset=utf-8"); This tells the browser how to decode the bytes.

Case 3: POST form parameters are garbled

Problem: Parameters submitted via a form (POST) to a JSP or servlet appear as garbled Chinese when retrieved.

Solution: In the receiving JSP/servlet, set the request character encoding before reading parameters: request.setCharacterEncoding("utf-8"); Then retrieve parameters normally, e.g.: <%=request.getParameter("username")%> Using a UTF‑8 filter is also recommended for a more robust approach.

Case 4: URI (GET) parameters are garbled

Problem: When parameters are passed via the URL (GET), Chinese characters become garbled because the default encoding is ISO‑8859‑1.

Solution: Modify Tomcat’s server.xml connector to specify UTF‑8 for URI encoding. Change the connector line to include URIEncoding="utf-8":

<Connector port="8080" protocol="HTTP/1.1" maxThreads="150" connectionTimeout="200000" redirectPort="8443" URIEncoding="utf-8" />

This makes Tomcat decode query strings using UTF‑8.

Case 5: Properties files used for i18n are garbled

Problem: When creating .properties files for internationalization, Chinese characters appear as garbled text because the Java compiler expects ISO‑8859‑1.

Solution: Convert the properties file with the JDK’s native2ascii tool:

native2ascii -encoding utf-8 display.properties display_zh_CN.properties

This converts UTF‑8 Chinese characters to Unicode escape sequences that Java can read correctly.