Title of a Perl CGI script

[!const]

I wrote a small Perl script that can download an existing file on the server. Everything works fine except that the file download has systematically filename perl (filedl.pl). Someone knows how to ensure that the file name to the proposed registration of the original file, not the perl script? I tried to add the head tag with title, but as can be expected, the mime type is automatically in text and binary file content is displayed in the browser, instead of being proposed to recording.

Here is an extract from the script code:

# Create the CGI 
my $cgi = new CGI;
my $file = $cgi -> param ("file"); 
# Type MIME
my $mimetype = &get_mime_type($file);
# HTML Headers 
print $cgi->header('Content-type: $mimetype; charset=utf-8');
open(FILE, "<$file" ) or die "Cannot open $file";
my @filedata = <FILE>;
close (FILE) or die "Cannot close $file";
print @filedata

[Lemog]

To do this, usually I send the following HTTP header:

Content-Disposition: inline; filename="%s"; size=%d
Content-Length: %d
Content-Type: application/binary

<contents of file>

Where you replace %d with size in bytes of the file and %s the file name.

But in general it gets stuck with unicode characters >127. If you have this kind of character, will probably have to go through RewriteRule.

[Ashok.M]

in the header, add this:

print $cgi->header("Content-Disposition: attachment; filename=nameofyourfile" );