linux shell 命令怎么遍历目录

先设定实验环境：

# 造 5 个 目录，每个目录下，造 3 个 文件和两个子目录如下：

cd $HOME/tmp

for i in d1 d2 d3 d4 d5

do

mkdir -p $i

touch $i/1.txt $i/2.txt $i/3.txt

mkdir -p $i/tmp1 $i/tmp2

done

# 检验测试环境：

$ ls -lR d1

total 0

-rw-r--r-- 1 wenlee comm 0 Dec 22 10:35 1.txt

-rw-r--r-- 1 wenlee comm 0 Dec 22 10:35 2.txt

-rw-r--r-- 1 wenlee comm 0 Dec 22 10:35 3.txt

drwxr-sr-x 2 wenlee comm256 Dec 22 10:35 tmp1/

drwxr-sr-x 2 wenlee comm256 Dec 22 10:35 tmp2/

# 利用下列脚本来实现你要做的：

cd $HOME/tmp

for i in */1.txt

do

echo "Found $i, save $i and remove everything else under $(dirname $i)/"

save_this_file=$(basename $i)

curr_dir=$(dirname $i)

# 把这个1.txt暂时存到/tmp里面去,为了避免已经有同样的档案名称在/tmp，加上$$ (i.e. PID)

mv $i /tmp/${save_this_file}.$$

rm -rf $curr_dir

mkdir -p $curr_dir

mv /tmp/${save_this_file}.$$ $curr_dir

done

# 屏幕执行输出如下：

Found d1/1.txt, save d1/1.txt and remove everything else under d1/

Found d2/1.txt, save d2/1.txt and remove everything else under d2/

Found d3/1.txt, save d3/1.txt and remove everything else under d3/

Found d4/1.txt, save d4/1.txt and remove everything else under d4/

Found d5/1.txt, save d5/1.txt and remove everything else under d5/

# 复验实验环境：

$ ls -l d?/*

-rw-r--r-- 1 wenlee comm 0 Dec 22 10:35 d1/1.txt

-rw-r--r-- 1 wenlee comm 0 Dec 22 10:35 d2/1.txt

-rw-r--r-- 1 wenlee comm 0 Dec 22 10:35 d3/1.txt

-rw-r--r-- 1 wenlee comm 0 Dec 22 10:35 d4/1.txt

-rw-r--r-- 1 wenlee comm 0 Dec 22 10:35 d5/1.txt

OK?

thanks!

为了避免目录列举消耗时间过长,请指定一个目录来模拟,命令行参数:代表路径的字符串.

如果认可代码,请加分50,谢谢

----

import javax.swing.*

import javax.swing.tree.*

import java.awt.*

import java.io.*

final public class FileTree extends JFrame {

public FileTree(File dir) throws HeadlessException {

super("File Tree")

JTree tree

add(new JScrollPane(tree =new JTree(buildTreeModel(dir))))

tree.setCellRenderer(new FileTreeRenderer())

setSize(400,600)

setVisible(true)

}

private TreeModel buildTreeModel(File dir){

DefaultMutableTreeNode root = new DefaultMutableTreeNode(dir)

walkthrough(dir,root)

return new DefaultTreeModel(root)

}

private static void walkthrough(File f,DefaultMutableTreeNode node){

for (File fle : f.listFiles()) {

DefaultMutableTreeNode n = new DefaultMutableTreeNode(fle)

node.add(n)

if (fle.isDirectory()){

walkthrough(fle, n)

}

}

}

private class FileTreeRenderer extends DefaultTreeCellRenderer {

public Component getTreeCellRendererComponent(JTree tree, Object value, boolean sel, boolean expanded, boolean leaf, int row, boolean hasFocus) {

JLabel cmp = (JLabel)super.getTreeCellRendererComponent(tree, value, sel, expanded, leaf, row, hasFocus)

if (value instanceof DefaultMutableTreeNode) {

DefaultMutableTreeNode n = (DefaultMutableTreeNode)value

Object obj = n.getUserObject()

if (obj instanceof File) {

File f = (File)obj

cmp.setText(f.getName())

cmp.setForeground(f.isDirectory()?Color.BLUE:Color.BLACK)

}

}

return cmp

}

}

public static void main(String[] args) {

new FileTree(new File(args[0]))

}

}

写一个函数,函数的参数是目录路径字符串

函数内使用 ls -s dir_path , 然后for 遍历循环

如果是目录则继续调用自身

如果是文件则答应文件名

从执行优化的角度来讲,可以把判断目录还是文件的代码放在循环外层.

好久没写shell了 ,我这也没环境测试 , 只能给个思路,函数的具体写法自己找一下资料吧.

另外,find命令可以直接完成你要做的事.

---