Simple, by using a dispatch table:

my %handlers = (
    table1 => \&_upload_table1,
    table2 => \&_upload_table2,
);

$tablename=cgi->param('tablename');

my $handler = $handlers{$tablename};
$handler->(args);
[download]

#!/usr/bin/perl

use strict;
use warnings;

my %dispatch = (
  table1 => \&_upload_table1,
  table2 => \&_upload_table1,
);
 
my $tablename = cgi->param('tablename');

die qq{bad param\n} unless exists $dispatch{$tablename};

$dispatch{$tablename}->();
 
sub _upload_table1 {

}

sub _upload_table2 {
}
[download]

Thankyou, but I was more interested to find out the way to build the dispatch hash in a dynamic way using an array of table names.

I had a feeling you'd say that. :-)

You would be generating code based on user input (data from outside). Most monks will argue quite strenuously that this is a bad idea. Really.

You will be tempted to turn off strict and warnings, you will probably have to turn off taint mode too.

Your code should decide which sub to run, not the the user. Have a look at the Tutorials on web programing (particularly Security in CGI Programming) and, please, take another look at dispatch tables.

update: added link

#!/usr/bin/perl
use strict;
use warnings;

sub sa { print "sub sa\n"; }
sub sb { print "sub sb\n"; }

my $name = 'a';

# using the symbol table
$main::{'s' . $name}->();

$name = 'b';

# using symbolic references:
no strict 'refs';
&{"s$name"}();
[download]

Be aware that this might open some security holes depending on your application. You have been warned!

you can use variables as sub names

Or method names:

my $method = "s" . $name;
$object->$method(@args);
[download]

--
Online Fortune Cookie Search