#!perl use strict; use warnings; my $infile = 'rangeFile.txt'; my $outfile = 'test.txt'; open IN, '<', $infile or die "Cannot Open InFile: $infile : $!"; open OUT, '>', $outfile or die "Cannot Open OutFile: $outfile : $!"; my %out=(); my @key=(); my $count_in = 0; my $count_out = 0; # input while (){ chomp; ++$count_in; my @in = split ',',$_; my $key = join ',',@in[0..2]; if ( ! defined $out{$key} ){ # initialise @{$out{$key}} = @in[3..4]; push @key,$key; # preserve order } else { # min if ($in[3] < $out{$key}[0]){ $out{$key}[0] = $in[3]; } # max if ($in[4] > $out{$key}[1]){ $out{$key}[1] = $in[4]; } } } # output for my $key (@key){ print OUT join ',',$key,@{$out{$key}},"\n"; ++$count_out; } close IN; close OUT; print " $count_in records read from $infile $count_out records written to $outfile\n"; __DATA__ IMB,060410,V1 ,371094378,371096338,1961 IMB,060410,V1 ,371096340,371096486,147 IMB,107951,V1 ,981157588,981164939,7352 IMB,107951,V1 ,981164941,981165606,666 IMB,107951,V1 ,981165608,981175100,9493 IMB,107951,V1 ,981175102,981176199,1098 IMB,060410|,folded ,307057959,307058193,235 IMB,060410|,selfmail ,307058194,307066458,8265 IMB,107951|,folded ,958090350,958091491,1142 IMB,107951|,selfmail ,958091492,958132856,41365 SEQ,,folded ,000000001,000001377,1377 SEQ,,selfmail ,000001378,000051007,49630