Hello soulbleach, and welcome to the Monastery!

Note that the regex in this line:

my @strings = $line =~ m/(RC[0-9][0-9][0-9][0-9])/;
[download]

contains only a single capture, so if the match succeeds @strings will contain only one element. Did you intend more than one capture per $line? If so, you need to add a /g modifier to the regex:

my @strings = $line =~ m/(RC[0-9]{4})/g;
[download]

Update: See “Global matching” in perlretut. (Also shortened regex: see “Quantifiers” in Regular Expressions.)

But without sample input data and desired output, it’s hard to know what you’re trying to do.

Anyway, hope that helps,

The similarity in the two programs should suggest 'subroutine' to you. Call it once for each file, have it pass back the array of unique lines you are already creating. Use a similar method to your %seen to populate your final hash with the values from the file2 array, removing entries that appear in the file1 array.

It would be nice to see an example of what the input file looks like.

UPDATE - suggested subroutine

my @array1 = unique($filename1);

sub unique {
  my ($file) = shift;
  my (%seen);
  open my $fh, '<', $file or die "Could not open $file\n";
  while(<$fh>) {
    chomp;
    $_ =~ m/(RC=\d{4})/;
    $seen{$_} = 1;
  }
  close $fh;
  return keys %seen;
}
[download]

Don't know exactly if this is what you were after, but I'm sitting at a hotel before I cross the border and thought I'd slap something together quickly. There are many ways to do this; this was just a quick idea.

It reads in two files (see below), each with unique RC#### and some overlap. It keeps all the individual lines (that have unique RC numbers), and if file2 has one that overlaps, it overwrites the line entry from file1.

Hopefully this is kind of what you were looking for. If not, please clarify a little better. Also note that in file2, it uses the '=' character in the RC number, but I removed that before any comparisons.

#!/usr/bin/perl

use 5.12.0;

my ( $file1, $file2 ) = qw( 1.txt 2.txt );

my %seen;

for my $file ( $file1, $file2 ){
    open my $fh, '<', $file
      or die "Can't open file $file: $!";

    while ( my $line = <$fh> ){
        chomp;
        next if $line !~ /RC=?\d{4}\s+/;
        $line =~ s/RC=/RC/;
        $line =~ /(RC\d{4})\s+/;
        my $string = $1 if $1;
        $seen{ $string } = $line;
    }

    close $fh;
}

for my $key ( keys %seen ){
    say $seen{ $key };
}
[download]

Output:

This is line RC0003 in file 2.
This is line RC0001 in file 2.
This is line RC0002 in file 1
This is line RC0004 in file 2.
[download]

file1:

This is line RC0001 in file 1
This is line RC0002 in file 1
This is line RC0003 in file 1
[download]

file2:

This is line RC=0001 in file 2.
This is line RC=0003 in file 2.
This is line RC=0004 in file 2.
[download]

-stevieb

Thanks very much this is exactly what I was needing. You guys all rock, I'm a Perl convert.

I don't know if you are permitted to use modules (i fear not).

But if so if you are - and if i where in your shoes i would consider using IO::All and then apply a set operation using Set::Scalar to check for uniqueness.

N.B.: Assuming that i guessed your specs right.

As some of my predecessors already pointed out, it would be helpful if you provide some more details to avoid unnecessary reverse engineering ;-)

Update: tried to fix bad English.

Regards, Karl