这篇文章主要介绍了Shell中创建序列和数组(list、array)的方法,
一、seq方法生成:
// seq [OPTION]... LAST
// seq [OPTION]... FIRST LAST
// seq [OPTION]... FIRST INCREMENT LAST
[chengmo@centos5 ~]$ seq 1000 //起始默认是 1,间隔默认也是1
[chengmo@centos5 ~]$ seq 2 1000 //间隔默认是1
[chengmo@centos5 ~]$ seq 1 3 10 //从1开始,到10 间隔为3 结果是:1 4 7 10
[chengmo@centos5 ~]$ seq -s '#' 1 3 10 //默认间隔是“空格” 如果想换成其它的可以带参数:-s
1#4#7#10
[chengmo@centos5 ~]$ seq -s '#' 30 | sed -e 's/[0-9]*//g' //生成连续的#
#############################
[chengmo@centos5 shell]$ aNumList=$(seq 100);
[chengmo@centos5 shell]$ echo $aNumList
1 2 3 4 5 6 7 8 9 10 11 12 13 14 ..... 100
aNumList得到是字符串,不同之处以:空格分隔开。在linux里面,可以把它看作是list. 可以通过for…in 循环读取。
[chengmo@centos5 shell]$ for i in $aNumList; do echo $i; done ;
1
2
3
...
100
如果需要生成array,只需要将$(seq 100) 再加个”()”即可。
[chengmo@centos5 ~]$ aNumList=($(seq 100))
[chengmo@centos5 ~]$ echo $aNumList
1
[chengmo@centos5 ~]$ echo ${aNumList[0]}
1
[chengmo@centos5 ~]$ echo ${#aNumList[@]}
100
[chengmo@centos5 ~]$ echo ${aNumList[@]}
1 2 3 4 5 6 7 8 9 10 11 12 13 14....100
aNumList是长度为100的数组。
二、通过内部{begin..end}生成
这种方法生成seq非常方便。通过内部运算符完成。
[chengmo@centos5 ~]$ echo {1..10}
1 2 3 4 5 6 7 8 9 10
[chengmo@centos5 ~]$ for a in {1..10}; do echo $a; done ;
1
2
3
4
5
6
7
8
9
10
三、性能比较
[chengmo@centos5 ~]$ time echo {1..100}
1 2 3 4 5 6 7 8 9 10 11 12 13 14 ..... 100
real 0m0.000s
user 0m0.001s
sys 0m0.000s
[chengmo@centos5 ~]$ time echo $( seq 100)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 ..... 100
real 0m0.003s
user 0m0.002s
sys 0m0.001s
从上面可以看到,{begin..end}速度比seq调用快了不少了。 以后调用时候可以考虑通过内部操作符完成。
题目:请用linux shell 写一段脚本,实现从1..1000中所有偶数的和值。
方法一:
通过while 循环得到需要的结果:
start=1;
total=0;
while [ $start -le 1000 ];do
[[ $(($start%2)) == 0 ]]&&total=$(($total+$start));
start=$(($start+1));
done;
echo $total;
[chengmo@centos5 ~]$ start=1;total=0;while [ $start -le 1000 ];do [[ $(($start%2)) == 0 ]]&&total=$(($total+$start)); start=$(($start+1));done;echo $total;
250500
以上运行结果是:249500,在linux shell 中,”;”作为命令行分隔符。如果大家对于$(()) 运算符号不是很理解,可以查看:linux shell 实现 四则运算(整数及浮点) 简单方法 ,如果对于:[[]] [] 符号,可以参考另外一篇文章linux shell 逻辑运算符、逻辑表达式详解。
方法二:
通过 for 循环得到结果:
start=0;
total=0;
for i in $(seq $start 2 1000); do
total=$(($total+$i));
done;
echo $total;
[chengmo@centos5 ~]$ start=0;total=0;for i in $(seq $start 2 1000); do total=$(($total+$i));done;echo $total;
250500
上面语句已经代码方面明显优于方法一,而且性能方面表现也很好。下面比较就可以发现:
比较性能:
[chengmo@centos5 ~]$ time (start=0;total=0;for i in $(seq $start 2 1000); do total=$(($total+$i));done;echo $total;)
250500
real 0m0.016s
user 0m0.012s
sys 0m0.003s
[chengmo@centos5 ~]$ time (start=1;total=0;while [ $start -le 1000 ];do [[ $(($start%2)) == 0 ]]&&total=$(($total+$start)); start=$(($start+1));done;echo $total;)
250500
real 0m0.073s
user 0m0.069s
sys 0m0.004s
方法一耗时 是方法二的 6倍!