Is there a way to limit the match as /^$one$/ would work for example?

Most likely, yes. However, you've not shown what the expected values are for the data that you're matching against, nor what qualifies as a match, nor how you tell when two & three are empty.

Personally, I use something where an indentifier may be made up of 4 parts ... (provider, source, instrument, detector), and they're identified by strings of the format "PROV.SOURCE.INST.DET", each of the individual identifiers is alphanumeric:

my $pattern = qr( '^'.
    ( join '\.', map { defined($_) ? $_ : '[^.]*' }
        @query{ qw( provider source instrument detector ) }
    ). '$' );
my @matches = grep { $_ =~ m/$pattern/ } @data;
[download]

Note -- I'm using '[^.]*' as opposed to '[^.]+', as not all of the identifiers have all 4 items populated, and would be empty for those slots.

I do not recommend using regular expressions to parse a CSV. What happens if part of the input includes a comma? In the Massive Overkill department, I would suggest using DBD::CSV:

   use DBI;
   $dbh = DBI->connect("DBI:CSV:")
       or die "Cannot connect: " . $DBI::errstr;
   $dbh->{'csv_tables'}->{'data'} = { 'file' => 'data.csv'};

   # Find the matching rows in the table
   $sth = $dbh->prepare('SELECT * FROM data ' .
                        'WHERE column1 = ? AND column2 = ? ' .
                        'AND column7 = ? AND column5 = ?', 
                        $one, $two, $three, $four)
        or die "Cannot prepare: " . $dbh->errstr();
   $sth->execute() or die "Cannot execute: " . $sth->errstr();

   my($id, $name);
   $sth->bind_columns(undef, \$id, \$name);
   while ($sth->fetch) 
   {
      # Print the name and id of each matching row
      printf("Found result row: id = %s, name = %s\n",
             defined($id) ? $id : "NULL",
             defined($name) ? $name : "NULL");
   }
   $sth->finish();
   $dbh->disconnect();
[download]

Thanks will have a look at the DBD::CSV module. I would strip out any comma's but here the values are in drop downs so I can specify these.

What is @data?
@data is the contents of the csv file in an array.
What are $one, $two, $three, $four?
This is the incoming data form the four drop downs in a html page
What are the requirements for @result?
A simple specific match against the incoming data and the CSV data. If drop downs $one and $four have been selected I want to display the reults.The data is mostly numerical

Testing against constent incoming data (for example $one $two $three $four) is fine as I can match the incoming data against where I want to look for it in the CSV @data and I know there is always something incoming.
But I don't know how to do this for say $one and $two or $two and $four etc which is why I was looking at grep.

CSV file is comma delemited file , you could loop through the file and do what you want , for example :

# open CSV file 
open(FH,"myfile.csv") or die $!;
# Loop through file content 
while (my $line = <FH>) {
    # split each line so ( $data[0] == first col , $data[1] == seconds
+ col ... so on )
    my @data = split(/,/$line);

    # now you can test if something matched or not 
    if ( ... ) {
        # Do Something here

        # if you want to end the loop after the first match , we just 
+use last : 
        last;
    }
}
# close File handle 
close(FH);
[download]

HTH