Question

I am looking for a way to read an input file and print only select lines to an output file in Perl. The lines I want to print to the output file all begin with xxxx.xxxx.xxxx, where x is an alphanumeric character (the periods are periods, not wildcards). The lines do not all have the same ending, if that makes a difference. I’m thinking something like the following (the condition of the if statement is all that is really missing as far as I can tell).

open(IN, "<$csvfile");
my @LINES = <IN>;
close(IN);
open(OUT, ">$csvnewfile");
print OUT @LINES if ([line starts with xxxx.xxxx.xxxx]);
close(OUT);

Thanks in advance!

Answer 1

Here is a better way to loop through your lines. It avoids loading your whole input file into memory at once:

use strict;
use warnings;

open my $fhi, '<', $csvfile    or die "Can not open file $csvfile: $!";
open my $fho, '>', $csvnewfile or die "Can not open file $csvnewfile: $!";
while (<$fhi>) {
    print $fho $_ if m/^ \w{4} \. \w{4} \. \w{4} /x;
}
close $fho;
close $fhi;

Keep in mind that the \w character class also includes underscores. To avoid underscores:

print $fho $_ if m/^ [a-z\d]{4} \. [a-z\d]{4} \. [a-z\d]{4} /xi;