📄 用java实现断点续传(http).htm
字号:
<BODY bgColor=#ffffff text=#000000>
<table>
<TBODY>
<TR>
<TD height=21>
<DIV align=center><B><FONT size=3>用Java实现断点续传(HTTP)
<BR><FONT size=2> </FONT></FONT></FONT>
<HR align=center color=#cccccc noShade SIZE=1>
</DIV></TD></TR>
<TR>
<TD class=line><FONT
color=#333300> 钟华 (zhong_hua@263.net)<BR>2001 年 5 月<BR><BR>(一)断点续传的原理<BR>其实断点续传的原理很简单，就是在Http的请求上和一般的下载有所不同而已。<BR>打个比方，浏览器请求服务器上的一个文时，所发出的请求如下：<BR>假设服务器域名为wwww.sjtu.edu.cn，文件名为down.zip。<BR>GET /down.zip HTTP/1.1<BR>Accept: image/gif, image/x-xbitmap, image/jpeg, image/pjpeg, application/vnd.ms-<BR>excel, application/msword, application/vnd.ms-powerpoint, */*<BR>Accept-Language: zh-cn<BR>Accept-Encoding: gzip, deflate<BR>User-Agent: Mozilla/4.0 (compatible; MSIE 5.01; Windows NT 5.0)<BR>Connection: Keep-Alive<BR><BR><BR>服务器收到请求后，按要求寻找请求的文件，提取文件的信息，然后返回给浏览器，返回信息如下：<BR><BR><BR>200<BR>Content-Length=106786028<BR>Accept-Ranges=bytes<BR>Date=Mon, 30 Apr 2001 12:56:11 GMT<BR>ETag=W/"02ca57e173c11:95b"<BR>Content-Type=application/octet-stream<BR>Server=Microsoft-IIS/5.0<BR>Last-Modified=Mon, 30 Apr 2001 12:56:11 GMT<BR><BR> <BR><BR>所谓断点续传，也就是要从文件已经下载的地方开始继续下载。所以在客户端浏览器传给<BR>Web服务器的时候要多加一条信息--从哪里开始。<BR>下面是用自己编的一个"浏览器"来传递请求信息给Web服务器，要求从2000070字节开始。<BR>GET /down.zip HTTP/1.0<BR>User-Agent: NetFox<BR>RANGE: bytes=2000070-<BR>Accept: text/html, image/gif, image/jpeg, *; q=.2, */*; q=.2<BR><BR><BR>仔细看一下就会发现多了一行RANGE: bytes=2000070-<BR>这一行的意思就是告诉服务器down.zip这个文件从2000070字节开始传，前面的字节不用传了。<BR>服务器收到这个请求以后，返回的信息如下：<BR>206<BR>Content-Length=106786028<BR>Content-Range=bytes 2000070-106786027/106786028<BR>Date=Mon, 30 Apr 2001 12:55:20 GMT<BR>ETag=W/"02ca57e173c11:95b"<BR>Content-Type=application/octet-stream<BR>Server=Microsoft-IIS/5.0<BR>Last-Modified=Mon, 30 Apr 2001 12:55:20 GMT<BR><BR><BR>和前面服务器返回的信息比较一下，就会发现增加了一行：<BR>Content-Range=bytes 2000070-106786027/106786028<BR>返回的代码也改为206了，而不再是200了。<BR><BR><BR>知道了以上原理，就可以进行断点续传的编程了。<BR><BR><BR>(二)Java实现断点续传的关键几点<BR><BR><BR>(1)用什么方法实现提交RANGE: bytes=2000070-。<BR>当然用最原始的Socket是肯定能完成的，不过那样太费事了，其实Java的net包中提供了这种功能。代码如下：<BR>URL url = new URL("http://www.sjtu.edu.cn/down.zip");<BR>HttpURLConnection httpConnection = (HttpURLConnection)url.openConnection<BR><BR> <BR><BR>();<BR>//设置User-Agent<BR>httpConnection.setRequestProperty("User-Agent","NetFox");<BR>//设置断点续传的开始位置<BR>httpConnection.setRequestProperty("RANGE","bytes=2000070");<BR>//获得输入流<BR>InputStream input = httpConnection.getInputStream();<BR><BR><BR>从输入流中取出的字节流就是down.zip文件从2000070开始的字节流。<BR>大家看，其实断点续传用Java实现起来还是很简单的吧。<BR>接下来要做的事就是怎么保存获得的流到文件中去了。<BR><BR><BR>保存文件采用的方法。<BR>我采用的是IO包中的RandAccessFile类。<BR>操作相当简单，假设从2000070处开始保存文件，代码如下：<BR>RandomAccess oSavedFile = new RandomAccessFile("down.zip","rw");<BR>long nPos = 2000070;<BR>//定位文件指针到nPos位置<BR>oSavedFile.seek(nPos);<BR>byte[] b = new byte[1024];<BR>int nRead;<BR>//从输入流中读入字节流，然后写到文件中<BR>while((nRead=input.read(b,0,1024)) > 0)<BR>{<BR>oSavedFile.write(b,0,nRead);<BR>}<BR><BR>怎么样，也很简单吧。<BR>接下来要做的就是整合成一个完整的程序了。包括一系列的线程控制等等。<BR><BR><BR><BR>(三)断点续传内核的实现<BR>主要用了6个类，包括一个测试类。<BR>SiteFileFetch.java负责整个文件的抓取，控制内部线程(FileSplitterFetch类)。<BR>FileSplitterFetch.java负责部分文件的抓取。<BR>FileAccess.java负责文件的存储。<BR>SiteInfoBean.java要抓取的文件的信息，如文件保存的目录，名字，抓取文件的URL等。<BR>Utility.java工具类，放一些简单的方法。<BR>TestMethod.java测试类。<BR><BR><BR>下面是源程序： <BR>/*<BR>**SiteFileFetch.java<BR>*/<BR>package NetFox;<BR>import java.io.*;<BR>import java.net.*;<BR><BR><BR>public class SiteFileFetch extends Thread {<BR><BR><BR>SiteInfoBean siteInfoBean = null; //文件信息Bean<BR>long[] nStartPos; //开始位置<BR>long[] nEndPos; //结束位置<BR>FileSplitterFetch[] fileSplitterFetch; //子线程对象<BR>long nFileLength; //文件长度<BR>boolean bFirst = true; //是否第一次取文件<BR>boolean bStop = false; //停止标志<BR>File tmpFile; //文件下载的临时信息<BR>DataOutputStream output; //输出到文件的输出流<BR><BR><BR>public SiteFileFetch(SiteInfoBean bean) throws IOException<BR>{<BR>siteInfoBean = bean;<BR>//tmpFile = File.createTempFile ("zhong","1111",new File(bean.getSFilePath()));<BR>tmpFile = new File(bean.getSFilePath()+File.separator + bean.getSFileName()+".info");<BR>if(tmpFile.exists ())<BR>{<BR>bFirst = false;<BR>read_nPos();<BR>}<BR>else<BR>{<BR>nStartPos = new long[bean.getNSplitter()];<BR>nEndPos = new long[bean.getNSplitter()];<BR>}<BR><BR><BR><BR>}<BR><BR><BR>public void run()<BR>{<BR>//获得文件长度<BR>//分割文件<BR>//实例FileSplitterFetch<BR>//启动FileSplitterFetch线程<BR>//等待子线程返回<BR>try{<BR>if(bFirst)<BR>{<BR>nFileLength = getFileSize();<BR>if(nFileLength == -1)<BR>{<BR>System.err.println("File Length is not known!");<BR>}<BR>else if(nFileLength == -2)<BR>{<BR>System.err.println("File is not access!");<BR>}<BR>else<BR>{<BR>for(int i=0;i new FileSplitterFetch( siteInfoBean.getSSiteURL(), <BR> siteInfoBean.getSFilePath() + File.separator + siteInfoBean.getSFileName(),<BR>nStartPos[i],nEndPos[i],i);<BR> Utility.log( " Thread " + i + " , nStartPos =" + nStartPos[i]+", nEndPos = " + nEndPos[i]);<BR>fileSplitterFetch[i].start();<BR>}<BR>// fileSplitterFetch[nPos.length-1] = new FileSplitterFetch(siteInfoBean.getSSiteURL(), <BR>siteInfoBean.getSFilePath() + File.separator + siteInfoBean.getSFileName() , nPos[nPos.length-1], nFileLength, nPos.length-1);<BR>// Utility.log("Thread " + (nPos.length-1 ) + " , nStartPos = " + nPos[nPos.length-1] + ",<BR>nEndPos = " + nFileLength);<BR>// fileSplitterFetch[nPos.length-1].start();<BR><BR><BR>//等待子线程结束<BR>//int count = 0;<BR>//是否结束while循环<BR>boolean breakWhile = false;<BR><BR><BR>while(!bStop)<BR>{<BR>write_nPos();<BR>Utility.sleep(500);<BR>breakWhile = true;<BR><BR><BR>for(int i=0;i4)<BR>// siteStop();<BR>}<BR><BR><BR>System.err.println("文件下载结束！");<BR>}<BR>catch(Exception e){e.printStackTrace ();}<BR>}<BR><BR><BR>//获得文件长度<BR>public long getFileSize()<BR>{<BR>int nFileLength = -1;<BR>try{<BR>URL url = new URL(siteInfoBean.getSSiteURL());<BR>HttpURLConnection httpConnection = (HttpURLConnection)url.openConnection ();<BR>httpConnection.setRequestProperty("User-Agent","NetFox");<BR><BR><BR>int responseCode=httpConnection.getResponseCode();<BR>if(responseCode>=400)<BR>{<BR>processErrorCode(responseCode);<BR>return -2; //-2 represent access is error<BR>}<BR><BR><BR>String sHeader;<BR><BR><BR>for(int i=1;;i++)<BR>{<BR>//DataInputStream in = new DataInputStream(httpConnection.getInputStream ());<BR>//Utility.log(in.readLine());<BR>sHeader=httpConnection.getHeaderFieldKey(i);<BR>if(sHeader!=null)<BR>{<BR>if(sHeader.equals("Content-Length"))<BR>{<BR>nFileLength = Integer.parseInt(httpConnection.getHeaderField(sHeader));<BR>break;<BR>}<BR>}<BR>else<BR>break;<BR>}<BR>}<BR>catch(IOException e){e.printStackTrace ();}<BR>catch(Exception e){e.printStackTrace ();}<BR><BR><BR>Utility.log(nFileLength);<BR><BR><BR>return nFileLength;<BR>}<BR><BR><BR>//保存下载信息（文件指针位置）<BR>private void write_nPos()<BR>{<BR>try{<BR>output = new DataOutputStream(new FileOutputStream(tmpFile));<BR>output.writeInt(nStartPos.length);<BR>for(int i=0;iDataInputStream(new FileInputStream(tmpFile));<BR>int nCount = input.readInt();<BR>nStartPos = new long[nCount];<BR>nEndPos = new long[nCount];<BR>for(int i=0;i 0 & & nStartPos < nEndPos && !bStop)<BR>{<BR>nStartPos += fileAccessI.write(b,0,nRead);<BR>//if(nThreadID == 1)<BR>// Utility.log("nStartPos = " + nStartPos + ", nEndPos = " + nEndPos);<BR>}<BR><BR><BR>Utility.log("Thread " + nThreadID + " is over!");<BR>bDownOver = true;<BR>//nPos = fileAccessI.write (b,0,nRead);<BR>}<BR>catch(Exception e){e.printStackTrace ();}<BR>}<BR>}<BR><BR><BR>//打印回应的头信息<BR>public void logResponseHead(HttpURLConnection con)<BR>{<BR>for(int i=1;;i++)<BR>{<BR>String header=con.getHeaderFieldKey(i);<BR>if(header!=null)<BR>//responseHeaders.put(header,httpConnection.getHeaderField(header));<BR>Utility.log(header+" : "+con.getHeaderField(header));<BR>else<BR>break;<BR>}<BR>}<BR><BR><BR>public void splitterStop()<BR>{<BR>bStop = true;<BR>}<BR><BR><BR>}<BR><BR><BR>/*<BR>**FileAccess.java<BR>*/<BR>package NetFox;<BR>import java.io.*;<BR><BR><BR>public class FileAccessI implements Serializable{<BR><BR><BR>RandomAccessFile oSavedFile;<BR>long nPos;<BR><BR><BR>public FileAccessI() throws IOException<BR>{<BR>this("",0);<BR>}<BR><BR><BR>public FileAccessI(String sName,long nPos) throws IOException<BR>{<BR>oSavedFile = new RandomAccessFile(sName,"rw");<BR>this.nPos = nPos;<BR>oSavedFile.seek(nPos);<BR>}<BR><BR><BR>public synchronized int write(byte[] b,int nStart,int nLen)<BR>{<BR>int n = -1;<BR>try{<BR>oSavedFile.write(b,nStart,nLen);<BR>n = nLen;<BR>}<BR>catch(IOException e)<BR>{<BR>e.printStackTrace ();<BR>}<BR><BR><BR>return n;<BR>}<BR><BR><BR>}<BR><BR><BR>/*<BR>**SiteInfoBean.java<BR>*/<BR>package NetFox;<BR><BR><BR>public class SiteInfoBean {<BR><BR><BR>private String sSiteURL; //Site's URL<BR>private String sFilePath; //Saved File's Path<BR>private String sFileName; //Saved File's Name<BR>private int nSplitter; //Count of Splited Downloading File<BR><BR><BR>public SiteInfoBean()<BR>{<BR>//default value of nSplitter is 5<BR>this("","","",5);<BR>}<BR><BR><BR>public SiteInfoBean(String sURL,String sPath,String sName,int nSpiltter)<BR>{<BR>sSiteURL= sURL;<BR>sFilePath = sPath;<BR>sFileName = sName;<BR>this.nSplitter = nSpiltter;<BR><BR><BR>}<BR><BR><BR>public String getSSiteURL()<BR>{<BR>return sSiteURL;<BR>}<BR><BR><BR>public void setSSiteURL(String value)<BR>{<BR>sSiteURL = value;<BR>}<BR><BR><BR>public String getSFilePath()<BR>{<BR>return sFilePath;<BR>}<BR><BR><BR>public void setSFilePath(String value)<BR>{<BR>sFilePath = value;<BR>}<BR><BR><BR>public String getSFileName()<BR>{<BR>return sFileName;<BR>}<BR><BR><BR>public void setSFileName(String value)<BR>{<BR>sFileName = value;<BR>}<BR><BR><BR>public int getNSplitter()<BR>{<BR>return nSplitter;<BR>}<BR><BR><BR>public void setNSplitter(int nCount)<BR>{<BR>nSplitter = nCount;<BR>}<BR>}<BR><BR><BR>/*<BR>**Utility.java<BR>*/<BR>package NetFox;<BR><BR><BR>public class Utility {<BR><BR><BR>public Utility()<BR>{<BR><BR><BR>}<BR><BR><BR>public static void sleep(int nSecond)<BR>{<BR>try{<BR>Thread.sleep(nSecond);<BR>}<BR>catch(Exception e)<BR>{<BR>e.printStackTrace ();<BR>}<BR>}<BR><BR><BR>public static void log(String sMsg)<BR>{<BR>System.err.println(sMsg);<BR>}<BR><BR><BR>public static void log(int sMsg)<BR>{<BR>System.err.println(sMsg);<BR>}<BR>}<BR><BR><BR>/*<BR>**TestMethod.java<BR>*/<BR>package NetFox;<BR><BR><BR>public class TestMethod {<BR><BR><BR>public TestMethod()<BR>{ ///xx/weblogic60b2_win.exe<BR>try{<BR>SiteInfoBean bean = new SiteInfoBean("http://localhost/xx/weblogic60b2_win.exe"," L:
emp","weblogic60b2_win.exe",5);<BR>//SiteInfoBean bean = new SiteInfoBean("http://localhost:8080/down.zip"," L:
emp","weblogic60b2_win.exe",5);<BR>SiteFileFetch fileFetch = new SiteFileFetch(bean);<BR>fileFetch.start();<BR>}<BR>catch(Exception e){e.printStackTrace ();}<BR><BR><BR>}<BR><BR><BR>public static void main(String[] args)<BR>{<BR>new TestMethod();<BR>}<BR>}<BR><BR> <BR>关于作者<BR>钟华，您可以通过电子邮件 zhong_hua@263.net 跟他联系。 <BR><BR> <BR></FONT></TD></TR>
<TR>
<TD height=5>
<HR align=center color=#cccccc noShade SIZE=1>
</TD></TR></TBODY></BODY></HTML>
⌨️ 快捷键说明
复制代码
Ctrl + C
搜索代码
Ctrl + F
全屏模式
F11
切换主题
Ctrl + Shift + D
显示快捷键
?
增大字号
Ctrl + =
减小字号
Ctrl + -