Hello, I have a line matching exercise. I have identical lines prefixed with either a "s" or "d". The idea is to pair the "s" and "d" lines together. In the event the corresponding line is missing print "<NONE>" in its place. Please refer to the desired output. Input file: s00PGUJU0009G00000114GI295D s00PGUJU0009G00000114GI49HF s00PGUMU0009G00000114JPD4M5 s00PGUMU0009G00000114JPHNIC s00PGUMU0009G000001154IMMLA s00PGUMU0009G000001154IOR2K s00PGUMU0009G000001154IRCFN d00PGUJU0009G00000114GI295D d00PGUMU0009G00000114JPD4M5 d00PGUMU0009G00000114JPHNIC d00PGUMU0009G000001154IMMLA Desires output: s00PGUJU0009G00000114GI295D d00PGUJU0009G00000114GI295D s00PGUJU0009G00000114GI49HF <NONE> s00PGUMU0009G00000114JPD4M5 d00PGUMU0009G00000114JPD4M5 s00PGUMU0009G00000114JPHNIC d00PGUMU0009G00000114JPHNIC s00PGUMU0009G000001154IMMLA d00PGUMU0009G000001154IMMLA s00PGUMU0009G000001154IOR2K <NONE> s00PGUMU0009G000001154IRCFN <NONE>

I have a line matching exercise. That tells me that this is probably a homework assignment. We won't do someones homework, but if you post the awk command(s) that you've tried, someone will help guide you. ShaqDiesel showed a possible Perl solution, but it could be improved. There's no need to duplicate the data across 4 arrays. It could easily be done with a single hash. ============================================================ #!/usr/bin/perl use strict; use warnings; my %data; while( <DATA> ) { push @{$data{substr($_, 1)}}, $_; } for my $key ( sort keys %data ) { print $data{$key}->[0]; print $data{$key}->[1] ? $data{$key}->[1] : "<NONE>\n"; } __DATA__ s00PGUJU0009G00000114GI295D s00PGUJU0009G00000114GI49HF s00PGUMU0009G00000114JPD4M5 s00PGUMU0009G00000114JPHNIC s00PGUMU0009G000001154IMMLA s00PGUMU0009G000001154IOR2K s00PGUMU0009G000001154IRCFN d00PGUJU0009G00000114GI295D d00PGUMU0009G00000114JPD4M5 d00PGUMU0009G00000114JPHNIC d00PGUMU0009G000001154IMMLA

Do you have to use kornshell for it? Because in perl you can 1.create arrays @sLines, @dLines 2.copy the arrays: @sLinesCOPY=@sLines; @dLinesCOPY=@dLines; 3.put your data in two arrays so you can compare them later: while(INPUT_HANDLE) { if(/^s/) { push(@sLines, $_); } else if(/^d/) { push(@dLines, $_); } else { ; } } 4.compare the strings and pair them if possible: foreach(@sLinesCOPY) { $sElement=unshift(@sLines); $dElement=unshift(@dLines); if($sElement ~= m/^s(.*)/) { $sRemainder = $1; } if($dElement ~= m/^d(.*)/) { $dRemainder = $1 } print $sElement; if($sRemainder eq $dRemainder) { print $dElement; } else { print "<NONE>"; } }

you asked for awk awk ' /^s/{ s[substr($0,2)]++ } /^d/ { d[substr($0,2)]++} END { for(i in s) { print "s"i if ( i in d ){ print "d"i }else { print "None" } } } ' file

Ghostdog, Many thanks for the solution. This was exactly what I was looking for. The requirement was to troubleshoot a HACMP issue in particular for the grpsvcs subsystem which should have a "d" and "s" entry for each concurrent volume group. My solution was a little crude requiring manual cross checking: lssrc -ls grpsvcs| awk '$0 ~/^s00/ || $0 ~/^d00/ { print }'