[KornLee]

#!/bin/bash
#scriptname:conver
#在BASH下简单实现十进制到二进制和十六进制的转换
cat<<info
1]  10=>2       2]  10=>16
info
read input?"please enter your choice:"
case $input in
1)      echo -n "please a dec number:";read i1
        echo "$i1==$(echo "obase=2;$i1"|bc)" ;;
2)      echo -n "please a dec number:";read i2
        echo "$i2==$(echo "obase=16;$i2"|bc)" ;;
*)      echo "wrong choice!"
        exit 1 ;;
esac

是不是很方便

[KornLee]

perl -e 'print time,"\n"'

[KornLee]

有些单词在一篇文章中经常会出现很多次,怎么统计这个单词的个数呢?!
如:文件kshfile,统计shell这个单词的个数,
$cat kshfile
ksh
The "Korn" shell, written by David Korn of AT&T Bell Labs (now Lucent). Written as a major upgrade to "sh", it is compatible with it, but has many more internal commands for the most frequently used functions. It also incorporates most of the same features from tcsh which enhance interactive use (command line history recall etc.). This shell is now available on most systems. It was slow
to gain acceptance because earlier versions were encumbered by AT&T licensing.
$cat kshfile|tr " " "\n"|grep -wc shell
2

[KornLee]

[javalee//home/javalee/myshell]cat tmp
aaaaaa
dddddd
kasdkfkk
djhasdjf
dfddf
kjsdfklkls
asdfjklkas
#显示文件tmp的奇数行的内容:
[javalee//home/javalee/myshell]sed -n '1,$p;n' tmp
aaaaaa
kasdkfkk
dfddf
asdfjklkas
#显示文件tmp的偶数行的内容:
[javalee//home/javalee/myshell]sed -n '1,$n;p' tmp
dddddd
djhasdjf
kjsdfklkls

[777]

#比较两个数的大小
#! /bin/bash
max=$1
while [ $2 ]; do
[ $max -lt $2 ] && max=$2
shift
done
echo $max

[KornLee]

多谢777的启发我也来一个!
比较多个数字的技巧:
#! /bin/bash
echo "max:$(for i in $(echo $*);do echo $i;done|sort -n|sed -n '$p')"

[777]

这是我抄的
比较两个数是这样：
#!/bin/bash
echo "first num:"
read a
echo "second num:"
read b
if [ $a -gt $b ]
then
max=$a;esle max=$b
fi
clear
echo $max

那怎么一个个地读多个数进一个变量

[KornLee]

用位置参数比较简单!
cat bj
#! /bin/bash
echo "max:$(for i in $(echo $*);do echo $i;done|sort -n|sed -n '$p')"

[javalee//home/javalee/myshell]bj 23 45 9 4 3 45 3 3 2 2 1 45
max:45

[KornLee]

例如文章:
$cat -n tmp
1 abcdefgh
2 123234234
3 sjdfk23423
1,行号倒序:
$cat -n tmp|tac #tac和cat很有趣吧~~
3 sjdfk23423
2 123234234
1 abcdefgh
2,每行倒读:
$cat tmp|rev
hgfedcba
432432321
32432kfdjs
3,全部倒过来:
$cat -n tmp|rev|tac
32432kfdjs 3
432432321 2
hgfedcba 1
4,用sed也可以解决:(要求对sed有足够的理解)
$cat -n tmp|sed '/\n/!G;s/\(.\)\(.*\n\)/&\2\1/;//D;s/.//'
hgfedcba 1
432432321 2
32432kfdjs 3
感谢:idkey兄!

[KornLee]

命令行下,利用perl的unpack函数,可以将汉字巧妙的变成十六进制和二进制,如:
[javalee//home/javalee/myshell]perl -le 'print unpack("B*","中");' #把汉字"中"转换成二进制
1101011011010000
[javalee//home/javalee/myshell]perl -le 'print unpack("H*","国");' #把汉字"国"转换成十六进制
b9fa

[idkey]

od 也可以，比如
$ echo '中国' | od -h
0000000 d0d6 fab9 000a
0000005

[ilmargaret]

To javalee:
我按照你的方法将脚本写成函数的形式并且在profile里指明了路径,然后在命令行里执行函数时不输出任何结果,也没有报错.但是将脚本改为原来普通的形式后重启系统,竟然在任何目录下都可以象执行ls命令一样成功地执行了脚本,也就是说不用定义函数也可以直接运行自己的脚本?

[KornLee]

我按照你的方法将脚本写成函数的形式并且在profile里指明了路径,然后在命令行里执行函数时不输出任何结果,也没有报错.但是将脚本改为原来普通的形式后重启系统,竟然在任何目录下都可以象执行ls命令一样成功地执行了脚本,也就是说不用定义函数也可以直接运行自己的脚本?

当然!但是直接运行脚本是因为在path中你已经为shell提供了你脚本所在的路径,而一些常用的小脚本,一函数的方法去定义的话,一般来说运行速度要比你每次从磁盘中读取文件要快得多而且占用资源很少!因为函数是在启动之后就已经驻留在内存啦.

[badb0y]

#!/bin/sh
if [ $# -ne 1 ]
then
echo "usage: $0 clean-ip"
exit 1
fi

if [ -e .bash_history ]
then
cat /dev/null > .bash_history
echo "history clean ok!"
else
echo "history not found!"
fi

if [ -e /var/log/lastlog ]
then
grep -v $1 /var/log/lastlog > /tmp/a ; mv /tmp/a /var/log/lastlog
echo "/var/log/lastlog clean ok!"
else
echo "/var/log/lastlog not found!"
fi

if [ -e /var/log/messages ]
then
grep -v $1 /var/log/messages > /tmp/a ; mv /tmp/a /var/log/messages
echo "/var/log/messages clean ok!"
else
echo "/var/log/messages not found!"
fi

if [ -e /var/log/wtmp ]
then
grep -v $1 /var/log/wtmp > /tmp/a ; mv /tmp/a /var/log/wtmp
echo "/var/log/wtmp clean ok!"
else
echo "/var/log/wtmp not found!"
fi

if [ -e /var/log/xferlog ]
then
grep -v $1 /var/log/xferlog > /tmp/a && mv /tmp/a /var/log/xferlog
echo "/var/log/xferlog clean ok!"
else
echo "/var/log/xferlog not found!"
fi

if [ -e /etc/utmp ]
then
grep -v $1 /etc/utmp > /tmp/a && mv /tmp/a /etc/utmp
echo "/etc/utmp clean ok!"
else
echo "/etc/utmp not found!"
fi

[KornLee]

to badb0y:
最好把说明,注释,用途之类的加上,以便兄弟们阅读!
谢谢~~