文件上传在web应用中非常普遍，要在jsp环境中实现文件上传功能是非常容易的，因为网上有许多用java开发的文件上传组件，本文以commons-fileupload组件为例，为jsp应用添加文件上传功能。
common-fileupload组件是apache的一个开源项目之一，可以从http://jakarta.apache.org/commons/fileupload/下载。
用该组件可实现一次上传一个或多个文件，并可限制文件大小。
下载后解压zip包，将commons-fileupload-1.0.jar复制到tomcat的webapps你的webappWEB-INFlib下，目录不存在请自建目录。
新建一个servlet: Upload.java用于文件上传：
import java.io.*; import java.util.*; import javax.servlet.*; import javax.servlet.http.*; import org.apache.commons.fileupload.*; public class Upload extends HttpServlet {private String uploadPath = "C:upload"; // 上传文件的目录private String tempPath = "C:uploadtmp"; // 临时文件目录public void doPost(HttpServletRequest request,HttpServletResponse response)throws IOException, ServletException{}}在doPost()方法中，当servlet收到浏览器发出的Post请求后，实现文件上传。以下是示例代码：
public void doPost(HttpServletRequest request,HttpServletResponse response)throws IOException, ServletException{try {DiskFileUpload fu = new DiskFileUpload(); // 设置最大文件尺寸，这里是4MBfu.setSizeMax(4194304); // 设置缓冲区大小，这里是4kbfu.setSizeThreshold(4096); // 设置临时目录：fu.setRepositoryPath(tempPath); // 得到所有的文件：List fileItems = fu.parseRequest(request); Iterator i = fileItems.iterator(); // 依次处理每一个文件：while(i.hasNext()) {FileItem fi = (FileItem)i.next(); // 获得文件名，这个文件名包括路径：String fileName = fi.getName(); // 在这里可以记录用户和文件信息// ...// 写入文件，暂定文件名为a.txt，可以从fileName中提取文件名：fi.write(new File(uploadPath + "a.txt")); }}catch(Exception e) {// 可以跳转出错页面}}如果要在配置文件中读取指定的上传文件夹，可以在init()方法中执行：
public void init() throws ServletException {uploadPath = ....tempPath = ....// 文件夹不存在就自动创建：if(!new File(uploadPath).isDirectory())new File(uploadPath).mkdirs(); if(!new File(tempPath).isDirectory())new File(tempPath).mkdirs(); }编译该servlet，注意要指定classpath，确保包含commons-upload-1.0.jar和tomcatcommonlibservlet-api.jar。
配置servlet，用记事本打开tomcatwebapps你的webappWEB-INFweb.xml，没有的话新建一个。
典型配置如下：
〈?xml version="1.0" encoding="ISO-8859-1"?〉〈!DOCTYPE web-appPUBLIC "-//Sun Microsystems, Inc.//DTD Web Application 2.3//EN""http://java.sun.com/dtd/web-app_2_3.dtd"〉〈web-app〉〈servlet〉〈servlet-name〉Upload〈/servlet-name〉〈servlet-class〉Upload〈/servlet-class〉〈/servlet〉〈servlet-mapping〉〈servlet-name〉Upload〈/servlet-name〉〈url-pattern〉/fileupload〈/url-pattern〉〈/servlet-mapping〉〈/web-app〉
配置好servlet后，启动tomcat，写一个简单的html测试：
〈form action="fileupload" method="post"enctype="multipart/form-data" name="form1"〉〈input type="file" name="file"〉〈input type="submit" name="Submit" value="upload"〉〈/form〉
注意action="fileupload"其中fileupload是配置servlet时指定的url-pattern。
下面是某个大虾的代码：
这个Upload比smartUpload好用多了.完全是我一个个byte调试出来的,不象smartUpload的bug具多.
调用方法:
Upload up = new Upload(); up.init(request); /**此处可以调用setSaveDir(String saveDir); 设置保存路径调用setMaxFileSize(long size)设置上传文件的最大字节.调用setTagFileName(String)设置上传后文件的名字(只对第一个文件有效)*/up. uploadFile();
然后String[] names = up.getFileName(); 得到上传的文件名,文件绝对路径应该是
保存的目录saveDir+"/"+names[i];
可以通过up.getParameter("field"); 得到上传的文本或up.getParameterValues("filed")
得到同名字段如多个checkBox的值.
其它的自己试试.
源码如下所示:____________________________________________________________
package com.inmsg.beans; import java.io.*; import java.util.*; import javax.servlet.*; import javax.servlet.http.*; public class Upload {private String saveDir = "."; //要保存文件的路径private String contentType = ""; //文档类型private String charset = ""; //字符集private ArrayList tmpFileName = new ArrayList(); //临时存放文件名的数据结构private Hashtable parameter = new Hashtable(); //存放参数名和值的数据结构private ServletContext context; //程序上下文,用于初始化private HttpServletRequest request; //用于传入请求对象的实例private String boundary = ""; //内存数据的分隔符private int len = 0; //每次从内在中实际读到的字节长度private String queryString; private int count; //上载的文件总数private String[] fileName; //上载的文件名数组private long maxFileSize = 1024 * 1024 * 10; //最大文件上载字节; private String tagFileName = ""; public final void init(HttpServletRequest request) throws ServletException {this.request = request; boundary = request.getContentType().substring(30); //得到内存中数据分界符queryString = request.getQueryString(); }public String getParameter(String s) { //用于得到指定字段的参数值,重写request.getParameter(String s)if (parameter.isEmpty()) {return null; }return (String) parameter.get(s); }public String[] getParameterValues(String s) { //用于得到指定同名字段的参数数组,重写request.getParameterValues(String s)ArrayList al = new ArrayList(); if (parameter.isEmpty()) {return null; }Enumeration e = parameter.keys(); while (e.hasMoreElements()) {String key = (String) e.nextElement(); if ( -1 != key.indexOf(s + "||||||||||") || key.equals(s)) {al.add(parameter.get(key)); }}if (al.size() == 0) {return null; }String[] value = new String[al.size()]; for (int i = 0; i 〈 value.length; i++) {value[i] = (String) al.get(i); }return value; }public String getQueryString() {return queryString; }public int getCount() {return count; }public String[] getFileName() {return fileName; }public void setMaxFileSize(long size) {maxFileSize = size; }public void setTagFileName(String filename) {tagFileName = filename; }public void setSaveDir(String saveDir) { //设置上载文件要保存的路径this.saveDir = saveDir; File testdir = new File(saveDir); //为了保证目录存在,如果没有则新建该目录if (!testdir.exists()) {testdir.mkdirs(); }}public void setCharset(String charset) { //设置字符集this.charset = charset; }public boolean uploadFile() throws ServletException, IOException { //用户调用的上载方法setCharset(request.getCharacterEncoding()); return uploadFile(request.getInputStream()); }private boolean uploadFile(ServletInputStream servletinputstream) throws //取得央存数据的主方法ServletException, IOException {String line = null; byte[] buffer = new byte[256]; while ( (line = readLine(buffer, servletinputstream, charset)) != null) {if (line.startsWith("Content-Disposition: form-data; ")) {int i = line.indexOf("filename="); if (i 〉= 0) { //如果一段分界符内的描述中有filename=,说明是文件的编码内容String fName = getFileName(line); if (fName.equals("")) {continue; }if (count == 0 && tagFileName.length() != 0) {String ext = fName.substring( (fName.lastIndexOf(".") + 1)); fName = tagFileName + "." + ext; }tmpFileName.add(fName); count++; while ( (line = readLine(buffer, servletinputstream, charset)) != null) {if (line.length() 〈= 2) {break; }}File f = new File(saveDir, fName); FileOutputStream dos = new FileOutputStream(f); long size = 0l; while ( (line = readLine(buffer, servletinputstream, null)) != null) {if (line.indexOf(boundary) != -1) {break; }size += len; if (size 〉 maxFileSize) {throw new IOException("文件超过" + maxFileSize + "字节!"); }dos.write(buffer, 0, len); }dos.close(); }else { //否则是字段编码的内容String key = getKey(line); String value = ""; while ( (line = readLine(buffer, servletinputstream, charset)) != null) {if (line.length() 〈= 2) {break; }}while ( (line = readLine(buffer, servletinputstream, charset)) != null) {if (line.indexOf(boundary) != -1) {break; }value += line; }put(key, value.trim(), parameter); }}}if (queryString != null) {String[] each = split(queryString, "&"); for (int k = 0; k 〈 each.length; k++) {String[] nv = split(each[k], "="); if (nv.length == 2) {put(nv[0], nv[1], parameter); }}}fileName = new String[tmpFileName.size()]; for (int k = 0; k 〈 fileName.length; k++) {fileName[k] = (String) tmpFileName.get(k); //把ArrayList中临时文件名倒入数据中供用户调用}if (fileName.length == 0) {return false; //如果fileName数据为空说明没有上载任何文件}return true; }private void put(String key, String value, Hashtable ht) {if (!ht.containsKey(key)) {ht.put(key, value); }else { //如果已经有了同名的KEY,就要把当前的key更名,同时要注意不能构成和KEY同名try {Thread.currentThread().sleep(1); //为了不在同一ms中产生两个相同的key}catch (Exception e) {}key += "||||||||||" + System.currentTimeMillis(); ht.put(key, value); }}/*调用ServletInputstream.readLine(byte[] b,int offset,length)方法,该方法是从ServletInputstream流中读一行到指定的byte数组,为了保证能够容纳一行,该byte[]b不应该小于256,重写的readLine中,调用了一个成员变量len为实际读到的字节数(有的行不满256),则在文件内容写入时应该从byte数组中写入这个len长度的字节而不是整个byte[]的长度,但重写的这个方法返回的是String以便分析实际内容,不能返回len,所以把len设为成员变量,在每次读操作时把实际长度赋给它.也就是说在处理到文件的内容时数据既要以String形式返回以便分析开始和结束标记,又要同时以byte[]的形式写到文件输出流中.*/private String readLine(byte[] Linebyte,ServletInputStream servletinputstream, String charset) {try {len = servletinputstream.readLine(Linebyte, 0, Linebyte.length); if (len == -1) {return null; }if (charset == null) {return new String(Linebyte, 0, len); }else {return new String(Linebyte, 0, len, charset); }}catch (Exception _ex) {return null; }}private String getFileName(String line) { //从描述字符串中分离出文件名if (line == null) {return ""; }int i = line.indexOf("filename="); line = line.substring(i + 9).trim(); i = line.lastIndexOf(""); if (i 〈 0 || i 〉= line.length() - 1) {i = line.lastIndexOf("/"); if (line.equals("""")) {return ""; }if (i 〈 0 || i 〉= line.length() - 1) {return line; }}return line.substring(i + 1, line.length() - 1); }private String getKey(String line) { //从描述字符串中分离出字段名if (line == null) {return ""; }int i = line.indexOf("name="); line = line.substring(i + 5).trim(); return line.substring(1, line.length() - 1); }public static String[] split(String strOb, String mark) {if (strOb == null) {return null; }StringTokenizer st = new StringTokenizer(strOb, mark); ArrayList tmp = new ArrayList(); while (st.hasMoreTokens()) {tmp.add(st.nextToken()); }String[] strArr = new String[tmp.size()]; for (int i = 0; i 〈 tmp.size(); i++) {strArr[i] = (String) tmp.get(i); }return strArr; }}下载其实非常简单，只要如下处理，就不会发生问题。public void downLoad(String filePath,HttpServletResponse response,boolean isOnLine)throws Exception{File f = new File(filePath); if(!f.exists()){response.sendError(404,"File not found!"); return; }BufferedInputStream br = new BufferedInputStream(new FileInputStream(f)); byte[] buf = new byte[1024]; int len = 0; response.reset(); //非常重要if(isOnLine){ //在线打开方式URL u = new URL("file:///"+filePath); response.setContentType(u.openConnection().getContentType()); response.setHeader("Content-Disposition", "inline; filename="+f.getName()); //文件名应该编码成UTF-8}else{ //纯下载方式response.setContentType("application/x-msdownload"); response.setHeader("Content-Disposition", "attachment; filename=" + f.getName()); }OutputStream out = response.getOutputStream(); while((len = br.read(buf)) 〉0)out.write(buf,0,len); br.close(); out.close(); }