We need to see runnable code that is equivalent to the code you're running. The code you've posted cannot ever yield an array with more than a single element:

c:\@Work\Perl\monks>perl -wMstrict -le
"my @list = qw(one two three four);
 ;;
 my @ra;
 for my $item (@list) {
   @ra = $item;
   printf 'print all in @ra: ';
   printf qq{$_ } for @ra;
   print 'done';
   }
"
print all in @ra: one done
print all in @ra: two done
print all in @ra: three done
print all in @ra: four done
[download]

Update: E.g., something using push would do the trick:

c:\@Work\Perl\monks>perl -wMstrict -le
"my @list = qw(one two three four);
 ;;
 my @ra;
 for my $item (@list) {
   push @ra, $item;
   printf 'print all in @ra: ';
   printf qq{$_ } for @ra;
   print 'done';
   }
"
print all in @ra: one done
print all in @ra: one two done
print all in @ra: one two three done
print all in @ra: one two three four done
[download]

Thanks for your reply. Actually that is actual code and output. The only thing missing is where i snatch the file name from command line.

 open IN, "<$ARGV[0]"
    or die "Could not open input file '$ARGV[0]' $!";
[download]

So i was reading into the push function but cannot seem to get it right. Not sure just where to place it. When i am loading the array via the substr function or printing it out?

Please see How do I post a question effectively? and How (Not) To Ask A Question.

---

Give a man a fish:  <%-{-{-{-<

My goal is to somehow stuff all neg values within one array and pos values within another...

I would think you want something like this:

my @pos_array;
my @neg_array;

while ( my $row = <IN> ) {
    if ( /(-\d[.\d]+)/ ) {
        push @neg_array, $1;
    }
    elsif ( /(\d[.\d]+)/ ) {
        push @pos_array $1;
    }
}
# now do stuff with those two arrays...
[download]

As for why your script is doing what it does, well, it's hard to say without seeing the actual input data. How about you post your code with data, like this, so we can see what's happening:

#!/usr/bin/perl

use strict;
use warnings;

while ( my $row = <DATA>) {
    print $row;
}

__DATA__
10.00
10.00
-30.00
[download]

One last point: substr() returns a scalar, not an array; your code is creating an array with a single element on each loop iteration, and that single element is a string of (up to) 17 characters long. The code doesn't make much sense, actually.

Why not just use the numeric value in the string:

$ perl -demo

  DB<1> $A = "123"

  DB<2> if ( $A < 0 ) { print 'negative' }else {print 'non-negative'}
non-negative
  DB<3> $A = "-123"

  DB<4>  if ( $A < 0 ) { print 'negative' }else {print 'non-negative'}
negative
[download]

As Occam said: Entia non sunt multiplicanda praeter necessitatem.

use strict;
use warnings;
*IN = *DATA;
my $row;
my $val;

while ( $row = <IN> ) {
    $val = substr( $row, 0, 1 );
    my @tranAmount = substr( $row, 68, 17 );

    foreach (@tranAmount) {
        print "$_\n";
    }
}
__DATA__
0                                                                   10
+.00000000000000
0                                                                   10
+.00000000000000
0                                                                   10
+.00000000000000
-                                                                   -3
+0.0000000000000
[download]

OUTPUT:

10.00000000000000
10.00000000000000
10.00000000000000
-30.0000000000000
[download]

UPDATE: Corrected formatting tag

use strict;
use warnings;
[download]

From command-line argument. It's a fixed text file, in this case it only contains 4 lines but it can be thousands. The amount is always on that position on every line. The only way i could think of snatching them was through substr.

open IN, "<$ARGV[0]"
    or die "Could not open input file '$ARGV[0]' $!";
[download]

Flat text file would be many lines of equal length with the amounts at that same position in every line.

If you could show us an actual data line(s), then most likely other techniques than substr() could be used. Split with an array slice is often a good approach. Just post your 4 lines within <code>...</code> blocks and we can tell exactly what is going on.