在列中添加常用名称 - awk

Question

是否可以通过在第二列中添加名称来在第一列中打印唯一名称，如下所示？提前谢谢！

输入

tony  singapore
johnny  germany
johnny  singapore

输出

tony  singapore
johnny  germany;singapore

Answer 1

试试这个单行：

awk '{a[$1]=$1 in a?a[$1]";"$2:$2}END{for(x in a)print x, a[x]}' file

Answer 2

$ awk '{name2vals[$1] = name2vals[$1] sep[$1] $2; sep[$1] = ";"} END { for (name in name2vals) print name, name2vals[name]}' file
johnny germany;singapore
tony singapore

Answer 3

这是一个神秘的sed变体：

script.sed的内容

$ cat script.sed 
:a                        # Create a label called loop
$!N                       # If not last line, append the line to pattern space
s/^(([^ ]+ ).*)\n\2/\1;/  # If first column is same append second column to it separated by ;
ta                        # If the last substitution was successful loop back
P                         # Print up to the first \n of the current pattern space
D                         # Delete from current pattern space, up to the \n character

执行：

$ cat file
tony  singapore
johnny  germany
johnny  singapore

$ sed -rf script.sed file
tony  singapore
johnny  germany; singapore