Getting rip of multiple rows based on column1

Shell Programming and Scripting
1

Hi,

I want to get rid of multiple rows (duplicate, triplicate etc..) for only column 1.

e.g.

iu  2 
iu  1
iu  3
k   4
jk  3
nm 4
nm 2

output

k 4
jk 3

thanks

2 
cat << EOF |
iu  2
iu  1
iu  3
k   4
jk  3
nm 4
nm 2
EOF
nawk '{
  aa_cnt[$1]++;
  aa_val[$1] = $0;
  }
  END{
  for ( var in aa_cnt ){
    if ( aa_cnt[var] == 1 ){
      print aa_val[var];
      }
    }
  }'
3

Alternatively, with Perl:

$
$ cat f0
iu  2
iu  1
iu  3
k   4
jk  3
nm 4
nm 2
$
$ ##
$ perl -F'\s+' -lane 'if ($x ne $F[0] && $x ne ""){print $y if $c==1; $c=0}
>                     $x=$F[0]; $y=$_; $c++;
>                     END {print $y if $c==1}' f0
k   4
jk  3
$
$

tyler_durden

4 
uniq -w2 -u file
5

-w is a non-standard option of the uniq command.

6 
wget -O- -q http://www.unix.com/misc.php?do=cfrules | gawk 'BEGIN{IGNORECASE}!/non-standard/{print "ok";exit}'
7

Please spare me your cryptic nonsensical replies.

The Open Group Base Specifications Issue 6

8 
wget -O- -q http://www.unix.com/302381237-post1.html | gawk '
{
    if ( $0 ~ /I cannot use non standard uniq options/ ){
       command = "pkill ichigo"
    }else{
       command = "pkill pretend_to_know_it_all"
    }
}
END{
    system(cmd)
}'