为什么bash命令不按列2对数据进行排序？

Question

给定数据文件

dogfood    dev        project    virtual-machine      z                  
dogfood    admin      z          dev-center           devcenter           
dogfood    admin      .          machine-definition   machinedefinition   
z          admin      .          network-setting      networksetting      
dogfood    z          .          project              project             
self-host    admin      .          vnet                 vnet                
dogfood    admin      project    z                    pool                
self-host  admin      vnet       subnet               default

$ cat data | sort -k2返回

z          admin      .          network-setting      networksetting      
dogfood    admin      .          machine-definition   machinedefinition   
self-host    admin      .          vnet                 vnet                
dogfood    admin      project    z                    pool                
dogfood    admin      z          dev-center           devcenter           
dogfood    dev        project    virtual-machine      z                  
dogfood    z          .          project              project             
self-host  admin      vnet       subnet               default

为什么z不在第2栏的底部？

如果我在其中一个“自主机”列后面再添加几个空格，它就会正确排序。我希望能够在列之间添加任意数量的空格，但这实际上会影响排序。

Answer 1

       -b, --ignore-leading-blanks
              ignore leading blanks

这样就行了！

cat data | sort -b -k2

Answer 2

文件似乎使用多个空格而不是制表符来分隔列。

试试这个：

awk 'BEGIN{OFS="\t"} $1=$1' file | sort -k2 | column -t

或,

sort -b -k2 file

任何一种指纹：

dogfood    admin  .        machine-definition  machinedefinition
z          admin  .        network-setting     networksetting
self-host  admin  .        vnet                vnet
dogfood    admin  project  z                   pool
self-host  admin  vnet     subnet              default
dogfood    admin  z        dev-center          devcenter
dogfood    dev    project  virtual-machine     z
dogfood    z      .        project             project