Google Reader

压缩解压性能探究

by

at 2010-12-08 09:21:21

original http://www.javaeye.com/topic/836327

     最近正在做一个项目，最终需要通过JAVA将一部分内容打包通过FTP上传到另外一台服务器上，当时压缩方法是这么做的：（先帖一个Test）

public class Test {
    public static void main(String[] args) {
        try {
            Date date1 = new Date();
            ZipOutputStream out = new ZipOutputStream(new FileOutputStream("test.zip"));
            File inputFile = new File("d://test");
            new Test().zip(out, inputFile, "");
            out.close();
            Date date2 = new Date();
            System.out.println("耗时："+(date2.getTime()-date1.getTime()));
        } catch (Exception e) {
            e.printStackTrace();
        }


}

private void zip(ZipOutputStream out, File f, String base) throws Exception {
    if (f.isDirectory()) {
        File[] fl = f.listFiles();
        out.putNextEntry(new ZipEntry(base + "/"));
        base = base.length() == 0 ? "" : base + "/";
        for (int i = 0; i < fl.length; i++) {
            zip(out, fl[i], base + fl[i].getName());
        }
    } else {
        out.putNextEntry(new ZipEntry(base));
        FileInputStream in = new FileInputStream(f);
        int b;
        while ((b = in.read()) != -1) {
            out.write(b);
        }
        in.close();
    }
}


}

     使用这种压缩方式效率极其低下，为了使效率提高，我先后发现了两种方法，第一种方法是利用BufferedInputStream和BufferedOutputStream对文件流进行包装，具体代码如下：

private void zip(ZipOutputStream out, File f, String base) throws Exception {
        if (f.isDirectory()) {
            File[] fl = f.listFiles();
            out.putNextEntry(new ZipEntry(base + "/"));
            base = base.length() == 0 ? "" : base + "/";
            for (int i = 0; i < fl.length; i++) {
                zip(out, fl[i], base + fl[i].getName());
            }
        } else {
            out.putNextEntry(new ZipEntry(base));
            BufferedOutputStream bou = new BufferedOutputStream(out);
            BufferedInputStream in = new BufferedInputStream(new FileInputStream(f));
            int b;
            while ((b = in.read()) != -1) {
                bou.write(b);
            }
            in.close();
            bou.flush();
        }
        }

     虽然第二种方式使用之后速度得到明显提升，但第三种方式较第二种还快了几倍，那就是在读写的时候增加每次读写量，具体代码如下：

private void zip(ZipOutputStream out, File f, String base) throws Exception {
        if (f.isDirectory()) {
            File[] fl = f.listFiles();
            out.putNextEntry(new ZipEntry(base + "/"));
            base = base.length() == 0 ? "" : base + "/";
            for (int i = 0; i < fl.length; i++) {
                zip(out, fl[i], base + fl[i].getName());
            }
        } else {
              out.putNextEntry(new ZipEntry(base));
              FileInputStream in = new FileInputStream(f);
              byte[] bs = new byte[10240];
              int b;
              while ((b = in.read(bs)) != -1) {
                  out.write(bs, 0, b);
              }
              in.close();
            }


}

     目前就我测试的结果看，第三种方式是最快的，不知道各位博友有没有更快的方式，欢迎大家一起过来探讨。

      <br><br>
      作者: <a href="http://jason--chen.javaeye.com">Jason__Chen</a> 
      <br>
      声明: 本文系JavaEye网站发布的原创文章，未经作者书面许可，严禁任何网站转载本文，否则必将追究法律责任！
      <br><br>
      <span style="color:red">
        <a href="http://www.javaeye.com/topic/836327" style="color:red">已有 <strong>0</strong> 人发表回复，猛击-&gt;&gt;<strong>这里</strong>&lt;&lt;-参与讨论</a>
      </span>
      <br><br><br>

JavaEye推荐

• —软件人才免语言低担保 赴美带薪读研！—