RegExp reading only first unique occurance

[biobrain]

I have a data like this

SEQRES 1 A 1522 U U U G U U G G A G A G U
SEQRES 2 A 1522 U U G A U C C U G G C U C
SEQRES 3 B 1522 A G G G U G A A C G C U G
SEQRES 4 B 1522 G C G G C G U G C C U A A
SEQRES 5 B 1522 G A C A U G C A A G U C G
SEQRES 6 B 1522 U G C G G G C C G C G G G
SEQRES 7 C 1522 G U U U U A C U C C G U G
SEQRES 8 D 1522 G U C A G C G G C G G A C
SEQRES 9 F 1522 G G G U G A G U A A C G C
SEQRES 10 F 1522 G U G G G U G A C C U A C
SEQRES 11 F 1522 C C G G A A G A G G G G G

I am interested only to print , A, B, C, D, F from the third column only once irrespective of how many time these appear in each column.

Here is my code

I have written a code
open (SP, "myfile");
        my $flag=1;
        my $chainid;
        my $chainid2;
        my $chainidB;
	while(<SP>){
             if ($flag==1){
             	if ( $_=~/SEQRES\s+\S+\s(\S)/){
             	$flag++;
            	#print  $1 ;
                $chainid=$1;
                print "this is chain 1 $chainid\n";
             }
             }else{
             	if ( $_=~/SEQRES\s+\S+\s(\S)/){
		#print  $1 ;
                $chainid2=$1;
                print " this is chain 2 $chainid2\n";}
           }
           
           }

please help to get the desired results

 

[stevexff]

Can you give a sample of the expected output?

Steve

[small]"Every program can be reduced by one instruction, and every program has at least one bug. Therefore, any program can be reduced to one instruction which doesn't work." (Object:erlDesignPatterns)[/small]

 

[biobrain]

The expected result should be like that

this is chain A

this is chain B

this is chain C

this is chain D

this is Chain F

but I am getting out put like this

this is chain A
this is chain A

this is chain B
this is chain B
this is chain B
this is chain B

this is chain C

this is chain D

this is Chain F
this is Chain F
this is Chain F

actually i do not want to have these repetitions in my results

 

[travs69]

my %duplicate;
 while(<SP>){
  if ( $_=~/SEQRES\s+\S+\s(\S)/){
   next if exists $duplicate{$1};
   print "this is chain 1 $1\n";
   $duplicate{$1} = 1;
 }
}

not tested..

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Travis - Those who say it cannot be done are usually interrupted by someone else doing it; Give the wrong symptoms, get the wrong solutions;

 

[stevexff]

use strict;
use warnings;

my %uniq;

while (<DATA>) {
   my $id = (split)[2];
   print unless exists $uniq{$id};
   $uniq{$id}++;
}


__DATA__
SEQRES   1 A 1522    U   U   U   G   U   U   G   G   A   G   A   G   U          
SEQRES   2 A 1522    U   U   G   A   U   C   C   U   G   G   C   U   C          
SEQRES   3 B 1522    A   G   G   G   U   G   A   A   C   G   C   U   G          
SEQRES   4 B 1522    G   C   G   G   C   G   U   G   C   C   U   A   A          
SEQRES   5 B 1522    G   A   C   A   U   G   C   A   A   G   U   C   G          
SEQRES   6 B 1522    U   G   C   G   G   G   C   C   G   C   G   G   G          
SEQRES   7 C 1522    G   U   U   U   U   A   C   U   C   C   G   U   G          
SEQRES   8 D 1522    G   U   C   A   G   C   G   G   C   G   G   A   C          
SEQRES   9 F 1522    G   G   G   U   G   A   G   U   A   A   C   G   C          
SEQRES  10 F 1522    G   U   G   G   G   U   G   A   C   C   U   A   C          
SEQRES  11 F 1522    C   C   G   G   A   A   G   A   G   G   G   G   G

which is pretty much the same as Travis' but without the regex...

Steve

[small]"Every program can be reduced by one instruction, and every program has at least one bug. Therefore, any program can be reduced to one instruction which doesn't work." (Object:erlDesignPatterns)[/small]

 

[KevinADC]

split() is a regexp. If those are fixed-width fields unpack() will be the most efficient way to get the third column.

------------------------------------------
- Kevin, perl coder unexceptional!

 

[stevexff]

split() is a regexp

All right officer, I'll come quietly, you've got me bang to rights.


What I should have said was "without the unnecessarily complicated regex"

Steve

[small]"Every program can be reduced by one instruction, and every program has at least one bug. Therefore, any program can be reduced to one instruction which doesn't work." (Object:erlDesignPatterns)[/small]