my %sequence;
while (<>) {
  my @line = split;
  my $sequence = ++$sequence{$line[0]};
  print "@line $sequence\n";
}
[download]

Thank you for the reply. If the tags are in continuous order generating sequence number would be easy. For e.g. A 100 A 200 A 300. Each time A is encountered sequence number can be incremented. However if a series of tags are repeating, for e.g. A 100 B 100 A 200 B 300, then for the first occurance of A and B, the sequence number should be 1 and for the next it should be incremented. In other words whenever a tag is repeated in the input file, the sequence number should be incremented. The repeatition of the tags is not guarenteed. They may occur. Thank you, Gowtham

Yes, that's exactly what my code does, presuming "A 100" is on one line, and "B 100" is on the next. You aren't formatting it more clearly than that, so I'm not sure what you mean unless you say more.

-- Randal L. Schwartz, Perl hacker
Be sure to read my standard disclaimer if this is a reply.

use strict;
use warnings;
my %seq;
for (@data) {
  my ($tag) = split;
  no warnings 'uninitialized';
  my $seq = ++$seq{$tag} ;
  print "$_ -> seqno = $seq\n";
}
[download]

no warnings 'uninitialized';

What's that doing in there? Incrementing from undef to 1 is never a warnable offense.

-- Randal L. Schwartz, Perl hacker
Be sure to read my standard disclaimer if this is a reply.

I believe what you could use is a hash of arrays. Each hash key would be the tag (A, B, C, ...) the arrays would be the pushed values, with the array indicies being the sequence numbers you're after.

If I'm understanding you correctly, this issue was recently discussed in node464229. They do a fine job describing how to do this.

my %sequences;  # this will be a hash of arrays

while (<DATA>) {
    @fields = split;
    for ( my $i=0; $i<@fields; $i+=2 ) {
        push @{$sequences{$fields[$i]}}, $fields[$i+1];
    }
}

for my $tag ( sort keys %sequences ) {
    my $i = 0;
    for my $val ( @{$sequences{$tag}} ) {
        printf( "%s [ %d ] : %s\n", $tag, $i+1, $sequences{$tag}[$i++]
+ );
    }
}

__DATA__
A 100 B 200 C 400 A 150 C 250 D 550 B 350
A 200 B 300 C 500 A 600 B 700 C 800 D 900
[download]

Some of the stuff you said in a sub-reply to merlyn seemed not to make sense:

A $$ indicates the start of the record. The record is split on the tag. A hash is populated with key being the tag and value being the data. Some fields within a record repeat in a well defined manner.

I hope the snippet provided here makes sense -- note that it keeps all the values for each tag in an array that is stored as the value of the hash element for that tag. Then, we just use the array index (plus 1) to provide the sequence numbers that you want for each of the values.

use strict;
use warnings;

my %c;
while (<DATA>) {
    chomp; @_ = split / /, $_, 2;
    print "$_ -> seqno = " . ++$c{$_[0]} . "\n";
}

__DATA__
A 100
B 200
B 300
C 400
D 500
C 600
D 700
[download]