Remove row and create new array - perl

Question

I am reading a CSV file
A B C D
A 1, 2, 3, 4
B 5, 1, 7, 8
C 9, 4, 1, 2
D 2, 7, 8, 1
The idea is to compute matrix correlation
How can I remove a row and create a new array?
I tried this
my #new_row = split(/\s+/, $header_line);
This is my first Perl program
my #row = /#desired_row/ && $_;
current output
A 1,2,3,4
What I am trying now
my #newarray = ( );
#newarray = grep ($_ > 2, #row);
print "#newarray\n";
Result I am trying to get
A 3, 4

Answer 1

Your question is confusing. My interpretation is that you are trying to read a correlation matrix into a Perl data structure. The fact the matrix has unit diagonal seems to support that interpretation. On the other hand, the matrix is not symmetric, so that's even more confusing.
I am assuming you are confused about how to remove the variable labels, and just getting the numbers into a matrix. In Perl, you can represent a matrix either as an array of references to anonymous arrays (dense matrix) or as a hash whose keys are row or column indexes and values are references to anonymous arrays (sparse matrix). The choice of whether to store column vectors or row vectors can affect how well you can work with the data, but, once again, there is not enough information in your question to deduce what would be most appropriate in your situation.
The code below shows the most basic way of reading that matrix into a Perl data structure. In your code, you would open the data file and assign a filehandle, e.g. $fh, and read from that rather than the __DATA__ section of your script.
#!/usr/bin/env perl
use strict;
use warnings;
my #vars = split ' ', <DATA>;
my #correl;
while (my $line = <DATA>) {
push #correl, [ $line =~ /([0-9]+)/g ];
}
print join("\t", #vars), "\n";
print join("\t", #$_), "\n" for #correl;
__DATA__
A B C D
A 1, 2, 3, 4
B 5, 1, 7, 8
C 9, 4, 1, 2
D 2, 7, 8, 1
Output:
A B C D
1 2 3 4
5 1 7 8
9 4 1 2
2 7 8 1