Java Chinese garbled learning and the garbled characters in spring @ResponseBody-garbled __java in spring @ResponseBody

If you find garbled characters in the content returned by spring's @responsbody, you need to solve them in the following ways.

1. Ensure that spring character Encoding Filter is configured in Web.xml:

  XML Code    <!-- Servlet Encoding Start -->        <filter>           <filter-name>Set  character encoding</filter-name>           < filter-class>org.springframework.web.filter.characterencodingfilter</filter-class>            <init-param>                <param-name>encoding</param-name>                <param-value>UTF-8</param-value>            </init-param>            <init-param>                <param-name>forceencoding</param-name>                <param-value>true</param-value>            </init-param>       </filter>       <filter-mapping>           < filter-name>set character encoding</filter-name>            <url-pattern>/*</url-pattern>       </ filter-mapping>       <!-- Servlet Encoding End -->   

2. If the return value type under the @responsebody annotation method is String, the <mvc:annotation-driven/> is configured in the spring configuration file before the result is returned to the user; Then spring defaults to invoke a call Stringhttpmessageconverter

Class for the output of the content. Why would there be garbled?

Workaround 1: View the source code of the Stringhttpmessageconverter, the default encoding in the Stringhttpmessageconverter class is iso-8859-1 encoding, the encoding is the Western European character set code, Obviously will not support Chinese. PS: It feels like a spring bug, obviously it should be UTF-8, I've reported this bug on the Spring official web site.

Java code public static final Charset Default_charset = Charset.forname ("iso-8859-1");      Private final Charset Defaultcharset; Private final list<charset> availablecharsets;

And as you can see from the code above, the 3 variables associated with a character are final, meaning that we cannot dynamically change the 3 values above through a set or a constructor injection.

Further analysis of the source can be seen, stringhttpmessageconverter inheritance and abstracthttpmessageconverter<string>, analysis of the abstract class, the relevant operation character encoding method can be overridden, So we can customize a class to inherit Stringhttpmessageconverter, and then rewrite the related method, the code is as follows:

  Java code   /**   *    */   package  com.chuanliu.platform.activity.basic.converter;            import  java.io.ioexception;   import java.nio.charset.charset;   import java.util.Arrays;    import java.util.list;      import  org.springframework.http.httpoutputmessage;   import org.springframework.http.mediatype;    import org.springframework.http.converter.stringhttpmessageconverter;   import  org.springframework.util.streamutils;     /**   *  for handling Chinese garbled problems:  spring  bug -   *    * In StringHttpMessageConverter, the  Default char set is iso-8859-1 (Western European character set)    *    *  @author   Josh wang (Sheng)    *   *  @email   josh_wang23@hotmail.com   */   Public class utf8stringhttpmessageconverter  extends StringHttpMessageConverter {               private static final mediatype utf8 = new mediatype ("Text ", " plain ",                Charset.forname ("UTF-8"));