# convert output of DOS DIR command to RDF conforming to 
# http://www.semanticdesktop.org/ontologies/2007/03/22/nfo ontology
# run with
# dir  | perl dir2nfordf.pl
# 2009-11-06 Bob DuCharme no warrantee expressed or implied

$header =  "
<rdf:RDF xmlns:rdf='http://www.w3.org/1999/02/22-rdf-syntax-ns#' 
         xmlns:nfo='http://www.semanticdesktop.org/ontologies/2007/03/22/nfo#'
         xmlns:dc='http://purl.org/dc/elements/1.1/' 
         xml:base='http:www.snee.com/ns/base/files#' 
         xmlns:rdfs='http://www.w3.org/2000/01/rdf-schema#'>
  <rdf:Description rdf:about='http://www.w3.org/2000/01/rdf-schema#comment'>
    <rdfs:domain rdf:resource='http://www.semanticdesktop.org/ontologies/2007/03/22/nfo#FileDataObject'/>
  </rdf:Description>";

print $header;

while (<>) {
    chop($_);
    if (/(\d\d\/\d\d\/\d\d\d\d)\s+(\d\d):(\d\d) ([A|P]M)\s+((\d+,)?\d+) (.*)/) {

        $date = $1;
        $hour = $2;
        $fileSize = $5;
        $filename = $7;

        # Convert date to ISO 8609 format
        $hour += 12 if ($4 == "PM");

        @dateFields = split(/\//,$date);
        $dateTime = @dateFields[2] . "-" . @dateFields[0] . "-" . @dateFields[1] . "T" . $hour . ":" . $2 . ":00";

        $fileSize =~ s/\,//g; # remove commas

        print "  <nfo:FileDataObject nfo:fileCreated='$dateTime'\n";
        print "          nfo:fileName='$filename' nfo:fileSize='$fileSize' rdfs:comment=''/>\n";
    }
}

print "<\/rdf:RDF>\n";