There's an excellent javascript site with hundreds of examples to do that kind of thing at http://javascript.internet.com.

If instead you want to do form validation after the user has submitted then HTML::FormValidator as suggested by rob_au is a good bet.

Alternately you can do something like the following:

use strict;
use CGI;

my $cgi=CGI->new;

print $cgi->header;

my @un_filled;
foreach ($cgi->param)
{
    unless($cgi->param($_))
    {
        push @un_filled, $_;
    }
}
if(@un_filled)
{
    print "You've forgotten to fill in the following
        fields: @un_filled";
    exit;
}
# do stuff...
[download]

Good luck.

Update: thanks to TomK32 for reminding me. You never want to rely on Javascript to have validated your forms. It's nice as a quick reminder to the user that a field should be filled in, but nothing more than that.

Ideally your script should handle proper validation as a matter of course. Checking that all the fields you expected have been passed back, that the fields contain valid entries (numbers for years, letters for names etc). Javascript is really just a convenience thing.

Doing taint checking before using these values is also recommended. Look at perldoc perlsec.

Using Javascript (JS) for checking a field is no good idea.
• The user might not have JS activated
• You can go around the JS
• Servers are more powerful than they were 1999
Update: OK, for slow connections client-side checking is a good idea, as long as you keep the server-side checking. But hey, why do a job twice?

Imho JS checks should be added to make it more user friendly. Especially for users that have a dialup or slow connection. If JS is not supported by the browser the user experience just degrades a little.
In any case, the server has to do the checking too. This is not only because the browser may not support JS, but it is possible for a malicious person to take your html page and submit the doctored version.

Thank you monks! Works now! It wuz very simple once I understood I could call the param without agruments. I'm new to Perl, and I'm embarassed to show you my bloated script.

I would also direct you to heed the excellent advice given to you by demerphq here in reply to your previous question with regard to how to get the most out of your questions asked here on PM.

Ooohhh, Rob no beer function well without!

Here's another way. Add a comment - <--error_message--> - to the form page where you want the error message to display, then use something like this:

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

# create CGI object
my $q = new CGI;

# create list of fields to check
my @non_empty_form_fields = qw(name email address tel);

# store errors here
my @err = ();

# check fields exist
for (@non_empty_form_fields) {
  unless ( $q->param($_) ) {
    push @err, "You must complete the '$_' field";
  }
}

# if error, represent
if (@err) {
  # read in form
  open (FORM,'/path/to/form.html') || die $!;
  my $form = join '', (<FORM>);
  close(FORM);

  # create error message
  my $err_message = $q->h3('Error').
                    $q->p('I could not process the form:').
                    $q->ul(
                        $q->li([@err])
                    ).
                    $q->p('Check your input and resubmit.');

  # put message in page
  $form =~ s/<!--error_message-->/$err_message/is;

  # show page
  print $q->header,
        $form;
  exit(0);
}

# otherwise input ok, so do whatever...
[download]

If you are using relative URLs to call page images, you may also want to add a <BASE HREF...> tag to the page (or get the script to add it).

HTH

cLive ;-)

#!/usr/bin/perl
use CGI qw/:standard/;
print header;

$form_field = param("form_field"); # get the value from form
if ($form_field eq ""){            # if value is nothing
&print_error;                      # run subroutine
}else{                             # else
print "$form_field\n";             # do your stuff
}

sub print_error{                   # Error subroutine
print "there was an error\n";
}
[download]