If you want to get unique records, the idiomatic way in Perl is by using a hash.

I don't know what you want to accomplish with your ($b, $b, $b, $b, $base_email, $b, $b, $b) = split but I assume that you are using split to get the fields into the scalars you want to process.
Here is a simple test case, that is using DATA as input and STDOUT as output.
There is a duplicate in my sample DATA, and it is skipped by the hash mechanism.

#!/usr/bin/perl -w
use strict;

my @regfile = <DATA>;

my %uniques = ();

foreach my $i (@regfile) {
    my ($email, $name, $date, $format) = split (/\|/, $i);
    $uniques{$email.$name.$date.$format}++;
    print "$email\|$name\|$date\|$format" 
        unless $uniques{$email.$name.$date.$format} > 1;
}

__DATA__
dummy@fake.com|John|10-10-2000|XZXXXXXXXXX
dummy2@fake.com|Joe|11-10-2000|XXXXXXXXXXX
alone@dunno.com|Jim|10-09-2000|YYYYYYYYYYYYY
dummy2@fake.com|Joe|11-10-2000|XXXXXXXXXXX
[download]

Output:

dummy@fake.com|John|10-10-2000|XZXXXXXXXXX
dummy2@fake.com|Joe|11-10-2000|XXXXXXXXXXX
alone@dunno.com|Jim|10-09-2000|YYYYYYYYYYYYY
[download]

The key to the hash should be the concatenation of all pieces of information that will make your record unique.

Before I forget. Did you remember to use strict and  -w ?

Hope it was helpful.

 _  _ _  _  
(_|| | |(_|><
 _|
[download]