I'm still a lowly Novice here, so be gentle oh great ones :)

While I'm not sure I can get you exactly what you're after, I can provide you some of the general logic I use to parse HL7 files. I know there's better ways to do it, but this is the project I learned Perl with, and so it's very rudimentary.

I work primarily with HL7 lab results, but the logic is pretty general for any HL7 record.

First off, I call the procedure with a file as a parameter, and assign it to the variable $b... Then call the following subroutine logic:

 sub parse_lab_test {
    @segs = split('\r', $b);  # \r is the record terminator
    foreach $seg (@segs) {
        $segtype = substr($seg, 0, 3);
        if ($segtype eq "MSH") {
           msh_fields($seg);
        } elsif ($segtype eq "PID") {
           pid_fields($seg);
        } elsif ($segtype eq "PV1") {
           pv1_fields($seg);

          ... etc ...

    } else {
           &segment_error;
        } 
    }
}
[download]

Then, for example, if you're working on the MSH segment, you'd have the following subroutine:

sub msh_fields {
    ($mshseg) =  @_;
    @vals      = split('\|', $mshseg);
    $docsource = $vals[3];
    $trandate  = $vals[6];
    $doctype2  = $vals[7];
    $rectype   = $vals[8];   
    $unique_hl7_id  = $vals[9];
    if ($rectype ne 'ORU^R01'){    # Discrete Results only
          next MAIN;     # Label that ends the script
    } 
}
[download]

Then, you just define all the HL7 segments you're interested in. In your case, you'd have $segtype = 'FT1', then call ft1_fields.

Anyways, that's how I do it... It's not the smallest or most elegant way to do it, but it's pretty easy to maintain :)

Trek