Professional consultancy and software development is available through my company Castle Amber.

Since most of my CGI programs are written in Perl most answers are illustrated with a piece of Perl code. See also Perl Answers.

• 1 General Information
• 2 Redirection
• 3 Logging

1 General Information

1.1 Which web sites have information on CGI?

• CGI Programming MetaFAQ

2 Redirection

2.1 How can I redirect a user time dependend?

You can use this script for example when you have a webcam on a server which is only available a few hours a day.

#!/usr/local/bin/perl -wT
#
# happyhour.cgi - redirects visitor time depended
# 0.01 (10-Jan-1999)
# (c) Castle Amber, 1999

use strict;
use CGI qw(:cgi);

my @time = localtime;

my $hour = $time[2];

my $location;

if ($hour >= 20 && $hour <= 21)
{
    $location = "http://www.yoursite.com/happyhour.html";
}
else
{
    $location = "http://www.yoursite.com/paymore.html";
}

print redirect($location);

For more information see Programming Perl page 185-186 (localtime), Perl Cookbook page 688-690 and perldoc CGI.

2.2 How can I redirect to a random URL?

#!/usr/local/bin/perl -wT
#
# random.cgi - redirects user to a random URL
# 0.01 (19-Jan-1998)
# (c) Castle Amber, 1999

use strict;
use CGI qw(:cgi);

my @URLS = qw(

	http://www.castleamber.com/
	http://www.perl.com/
	http://www.javasoft.com/
);

srand(time() ^ ($$ + ($$ << 15)));
print redirect($URLS[rand @URLS]);

<IMG SRC="/cgi-bin/random.cgi" ALT="Random Picture">

For more information see section 2.1, Programming Perl page 223 (srand), Perl Cookbook page 688-690 and perldoc CGI.

3 Logging

3.1 How do I keep track of the number of downloads?

For an example of a form to call this script see the Castle Amber AmberCam page.

#!/usr/local/bin/perl -wT
#
# dltrack.cgi - keeps track of downloads
# 0.01 (27-Jan-1998)
# (c) Castle Amber, 1999

use strict;
use Fcntl qw(:DEFAULT :flock);
use CGI;

$| = 1;

my $LOG = "/path/to/your/home/downloadlog.txt";

my $query = new CGI;

#--- try
eval
{
    #--- open the file and create if it doesn't exist
    sysopen(FILE, $LOG, O_WRONLY | O_CREAT, 0644)
	or die "can't open/create log: $!";
    #--- lock the file (exclusive, for writing)
    flock(FILE, LOCK_EX)
	or die "can't lock file: $!";
    #--- move to end of file
    seek(FILE, 0, 2)
	or die "seek failed: $!";
    
    #--- write out data
    print FILE	"name:\t",  $query->param('name'),  "\n", 
		"email:\t", $query->param('email'), "\n",  
		"--\n"
	or die "print failed: $!";
    close(FILE)
	or die "close failed: $!";
};
#--- catch
unless ($@)
{
    print $query->redirect('http://www.yoursite.com/file.zip');
}
else
{
    print $query->header(-TYPE => 'text/plain'), $@;
}

You can add more fields besides name and email by adding additional calls to the param function.

Notice that this script uses eval to simulate a try and catch construct. When an error occurs the message is returned to the browser as plain text. The message reports also the line in the script that caused the error. If you don't want this add an \n to the end of each die message.

If no error occured the visitor is redirected to the file to be downloaded.

See also 2 Redirection, Perl Cookbook page 245-247 (Locking a file).

© Copyright 1999 by John Bokma