本文共 1800 字，大约阅读时间需要 6 分钟。

引言

最近需要对一个MarkDown表格进行排序，关键字段是在第一个的两个|中间，如下

|8| | |Medium| |5| | |Medium| |17| | |Medium| |538| | |Medium| |1122| | |Easy| |40| | |Medium| |3| | |Medium| |1| | |Easy| |925| | |Easy| |7| | |Easy| |113| | |Medium| |2| | |Medium| |9| | |Easy| |6| | |Medium| |1370| | |Easy| |78| | |Medium| |208| | |Medium| |1024| | |Medium| |529| | |Medium| |39| | |Medium| |15| | |Medium| |1291| | |Medium| |11| | |Medium| |865| | |Medium| |16| | |Medium| |89| | |Medium| |12| | |Medium| |14| | |Easy| |402| | |Medium| |856| | |Medium| |19| | |Medium| |290| | , |Easy| |389| | |Easy|

解决方法

想到用shell脚本通过awk来进行排序，代码如下：

for i in `awk '{split($1,a,"[\|\|]");print NR,a[2]}' test.txt | sort -n -k2 | awk '{print $1}'`; do awk 'NR=='$i'{print}' test.txt; done

思路就是：

• 先通过split($1,a,"[\|\|]"分割出test.txt第一列中的关键字段
• 然后通过print NR,a[2]记录每一列的列号和关键字段
• 然后对关键字段进行排序sort -nr -k2
• 其次在外层使用for i infor循环将每一个列号输入到 'NR=='$i'
• 最后根据'NR=='$i'排序好的列号打印出test.txt每一行数据

结果如下：

|1| | |Easy| |2| | |Medium| |3| | |Medium| |5| | |Medium| |6| | |Medium| |7| | |Easy| |8| | |Medium| |9| | |Easy| |11| | |Medium| |12| | |Medium| |14| | |Easy| |15| | |Medium| |16| | |Medium| |17| | |Medium| |19| | |Medium| |39| | |Medium| |40| | |Medium| |78| | |Medium| |89| | |Medium| |113| | |Medium| |208| | |Medium| |290| | , |Easy| |389| | |Easy| |402| | |Medium| |529| | |Medium| |538| | |Medium| |856| | |Medium| |865| | |Medium| |925| | |Easy| |1024| | |Medium| |1122| | |Easy| |1291| | |Medium| |1370| | |Easy|

效果如下：

#	Title	Solution	Difficulty
1			Easy
2			Medium
3			Medium
5			Medium
6			Medium
7			Easy
8			Medium
9			Easy
11			Medium
12			Medium
14			Easy
15			Medium
16			Medium
17			Medium
19			Medium
39			Medium
40			Medium
78			Medium
89			Medium
113			Medium
208			Medium
290		,	Easy
389			Easy
402			Medium
529			Medium
538			Medium
856			Medium
865			Medium
925			Easy
1024			Medium
1122			Easy
1291			Medium
1370			Easy

转载地址：http://axrr.baihongyu.com/