I don't think it should be that difficult. Here's a snippet from one of my scripts that I use for serving up files but it shouldn't take too much to extract what you need from it.

#!/usr/bin/perl -wT

use strict;
use CGI;
$CGI::HEADERS_ONCE = 1;

use constant BUF_SIZE => 4096;

# Unbuffer the output stream:
$| = 1;

# Set the environment for Taint mode:
BEGIN {
   $ENV{PATH} = "/bin:/usr/bin";
   delete @ENV{qw (IFS CDPATH ENV BASH_ENV)};
}

my $q            = new CGI;
my $source       = &get_file_name($q);

# IE is funky, doesn't seem to understand 
# "-type => 'text/octet-stream'", 
# wants a .whatever filename
my $save_as_name = &get_saveas_name($source);

if (open(IN, $source)) {
   print $q->header(
      -type       => 'text/octet-stream',
      -attachment => $save_as_name
      );

   my $buffer;
   binmode STDOUT;
   while (read(IN, $buffer, BUF_SIZE)) {
      print $buffer;
   }
   close IN;
} else {
   print &get_error_page($q, $save_as_name, $!);
}

...
[download]

Thanks. I try use your code but I must translate your code to embperl before.

Programing in Embperl and all questions are related to Embperl.
Excuse my bad English !!!