#!/usr/bin/perl -w
use strict;

use Bio::SeqIO;
use Bio::Seq;

use Data::Dumper;
# To run this, pass in two filenames
my $file = $ARGV[0];
my $ofile = $ARGV[1];
my $ofile2 = $ARGV[2];

my $input = Bio::SeqIO->new(-format => 'genbank',
			    -file   => $file);
my $output = Bio::SeqIO->new(-format => 'fasta',
			     -file   => ">$ofile");
my $output_protein = Bio::SeqIO->new(-format => 'fasta',
				     -file   => ">$ofile2");
while( my $seq = $input->next_seq ) { 
  # get all the feature from the sequence object
 my @features = $seq->get_SeqFeatures;
 foreach my $f ( @features ) {
   # test if the feature TYPE is 'CDS'
   #print "feature object is $f\n";
   #print Dumper($f), "\n";
   if( $f->primary_tag eq 'CDS' ) {
     my @tags = $f->get_all_tags;
     print "the tags are @tags\n";
     my @namelist = $f->get_tag_values('gene'); 
     my $name = $namelist[0];
     my ($desc) = $f->get_tag_values('product');
     # process location information
     my $location = $f->location;
     print "location str is ", $location->to_FTstring(), "\n";
     my @sublocs = $location->each_Location; 
     my $cds_string;
     for my $l ( @sublocs ) {
       print $l->to_FTstring(), "\n";
       my $start = $l->start;
       my $end   = $l->end;
       my $strand= $l->strand;
       my $exon_seq = $seq->subseq($start,$end);
       print "exon_seq is $exon_seq\n";
       $cds_string .= $exon_seq;
     }
     my $cds_seq2 = $f->spliced_seq;
     my $cds_seq = Bio::Seq->new(-seq => $cds_string,
				 -id  => $name,
				 -desc=> $desc);     
     $output->write_seq($cds_seq);
     $output->write_seq($cds_seq2); 
     $output_protein->write_seq($cds_seq->translate);
     #print $f->start, "..", $f->end, " for feature ", $name, "\n";
   }
 } 
  
}