I'm trying to sort the following file:
我试着对以下文件进行排序:
a 2
b 1
a 10
I need to get:
我需要:
a 2
a 10
b 1
I know about the -kPOS[opts] option, and try to use it:
我知道-kPOS[opts]选项，并尝试使用它:
sort -k1 -k2n file
but this command gives me only:
但这个命令只给了我:
a 10
a 2
b 1
So it sorts by the first column, but no by the second. Running just sort -k2n file sorts by the second column.
按第一列排序，但不按第二列排序。运行的只是按第二列排序-k2n文件。
b 1
a 2
a 10
How could I sort it by two columns?
我怎么能把它分成两列呢?
Edit:
sort (GNU coreutils) 5.93
(GNU coreutils)5.93
3 个解决方案
#1
24
You have to terminate the primary key, otherwise, sort uses all the fields starting from the given one:
必须终止主键，否则，sort将使用从给定键开始的所有字段:
sort -k1,1 -k2n
#2
7
It is almost correct. Try this:
这几乎是正确的。试试这个:
sort -k1,1 -k2,2n
#3
5
If you have GNU sort sort then you can do a version sort:
如果你有GNU排序，你可以做一个版本排序:
$ sort -V file
a 2
a 10
b 1
Option:
选择:
-V, --version-sort natural sort of (version) numbers within text
The nice thing about version sorting is it will work regardless of columns:
版本排序的好处是，不管列是什么，它都可以工作:
$ cat file
a2
b1
a10
$ sort -V file
a2
a10
b1
#1
24
You have to terminate the primary key, otherwise, sort uses all the fields starting from the given one:
必须终止主键，否则，sort将使用从给定键开始的所有字段:
sort -k1,1 -k2n
#2
7
It is almost correct. Try this:
这几乎是正确的。试试这个:
sort -k1,1 -k2,2n
#3
5
If you have GNU sort sort then you can do a version sort:
如果你有GNU排序，你可以做一个版本排序:
$ sort -V file
a 2
a 10
b 1
Option:
选择:
-V, --version-sort natural sort of (version) numbers within text
The nice thing about version sorting is it will work regardless of columns:
版本排序的好处是，不管列是什么，它都可以工作:
$ cat file
a2
b1
a10
$ sort -V file
a2
a10
b1