問題分析

不過在實(shí)際使用中, 還是發(fā)現(xiàn)按照最基本的方式調(diào)用 HttpClient 時(shí), 并不支持 UTF-8 編碼, 在網(wǎng)絡(luò)上找過一些文章, 也不得要領(lǐng), 于是查看了 commons-httpClient3.0.1 的一些代碼, 首先在 PostMethod 中找到了 generateRequestEntity() 方法:

/**      * Generates a request entity from the post parameters, if present.   Calls      * {@link EntityEnclosingMethod#generateRequestBody()} if parameters have not been set.      *      * @since 3.0      */     protected RequestEntity generateRequestEntity() {         if (!this.params.isEmpty()) {             // Use a ByteArrayRequestEntity instead of a StringRequestEntity.             // This is to avoid potential encoding issues.   Form url encoded strings             // are ASCII by definition but the content type may not be.   Treating the content             // as bytes allows us to keep the current charset without worrying about how             // this charset will effect the encoding of the form url encoded string.             String content = EncodingUtil.formUrlEncode(getParameters(), getRequestCharSet());             ByteArrayRequestEntity entity = new ByteArrayRequestEntity(                 EncodingUtil.getAsciiBytes(content),                 FORM_URL_ENCODED_CONTENT_TYPE             );             return entity;         } else {             return super.generateRequestEntity();         }     }

原來使用 NameValuePair 加入的 HTTP 請(qǐng)求的參數(shù)最終都會(huì)轉(zhuǎn)化為 RequestEntity 提交到 HTTP 服務(wù)器, 接著在 PostMethod 的父類 EntityEnclosingMethod 中找到了如下的代碼:

/**      * Returns the request's charset.   The charset is parsed from the request entity's      * content type, unless the content type header has been set manually.      *      * @see RequestEntity#getContentType()      *      * @since 3.0      */     public String getRequestCharSet() {         if (getRequestHeader("Content-Type") == null) {             // check the content type from request entity             // We can't call getRequestEntity() since it will probably call             // this method.             if (this.requestEntity != null) {                 return getContentCharSet(                     new Header("Content-Type", requestEntity.getContentType()));             } else {                 return super.getRequestCharSet();             }         } else {             return super.getRequestCharSet();         }     }

解決方案

從上面兩段代碼可以看出是 HttpClient 是如何依據(jù) "Content-Type" 獲得請(qǐng)求的編碼(字符集), 而這個(gè)編碼又是如何應(yīng)用到提交內(nèi)容的編碼過程中去的. 按照這個(gè)原來, 其實(shí)我們只需要重載 getRequestCharSet() 方法, 返回我們需要的編碼(字符集)名稱, 就可以解決 UTF-8 或者其它非默認(rèn)編碼提交 POST 請(qǐng)求時(shí)的亂碼問題了.

測(cè)試

首先在 Tomcat 的 ROOT WebApp 下部署一個(gè)頁面 test.jsp, 作為測(cè)試頁面, 主要代碼片段如下:

<%@ page contentType="text/html;charset=UTF-8"%> <%@ page session="false" %> <% request.setCharacterEncoding("UTF-8"); String val = request.getParameter("TEXT"); System.out.println(">>>> The result is " + val); %>

接著寫一個(gè)測(cè)試類, 主要代碼如下:

public static void main(String[] args) throws Exception, IOException {         String url = "http://localhost:8080/test.jsp";         PostMethod postMethod = new UTF8PostMethod(url);         //填入各個(gè)表單域的值         NameValuePair[] data = {                 new NameValuePair("TEXT", "中文"),         };         //將表單的值放入postMethod中         postMethod.setRequestBody(data);         //執(zhí)行postMethod         HttpClient httpClient= new HttpClient();         httpClient.executeMethod(postMethod);     }          //Inner class for UTF-8 support     public static class UTF8PostMethod extends PostMethod{         public UTF8PostMethod(String url){             super(url);         }         @Override         public String getRequestCharSet() {             //return super.getRequestCharSet();             return "UTF-8";         }     }

運(yùn)行這個(gè)測(cè)試程序, 在 Tomcat 的后臺(tái)輸出中可以正確打印出 ">>>> The result is 中文" .