Uploading progress statistics to a website automatically

I've seen a few questions related to storing & accessing simulation results on the mailing list, so I thought I would share how I implemented this on my own simulations.

This may be useful for anyone who wants to run simulations for long periods (days) on remote computers, and have statistics reported and stored in an online database for easy access. This method can also work behind most firewalls.

I've successfully used curl to have my (heavily) modified version of the breve Walker demo send average fitness, max fitness & variance results per generation to a website, where it is stored in a database using a few PHP lines. I can then check the progress at any time just by visiting another PHP page that lists all the results stored so far.

Here is what I use in breve to send the data to my website (this is just one line of code):
(self execute command "curl -F \"currentGenerationIndex=$currentGenerationIndex\" -F \"bestFitness=$bestFitness\" -F \"avgFitness=$avgFitness\" -F \"varFitness=$varFitness\" http://www.somewebsite.com.ar/simbug.php").

simbug.php is the webpage that receives the data (it can be named anything you want, I named it to match my simulation's name). It looks like this:

<?php

function connectDB() {
        $con = pg_connect("dbname=yourdatabasename user=yourusername password=yourpassword host=yourhost");
        if (!$con) {
                print "this isn't working :(";
                exit;
        }
        return $con;
}

$connection = connectDB();
$currentGenerationIndex = $_POST['currentGenerationIndex'];
$bestFitness = $_POST['bestFitness'];
$avgFitness = $_POST['avgFitness'];
$varFitness = $_POST['varFitness'];

if ($myresult = pg_exec($connection, "insert into bugsim (currentGenerationIndex, bestFitness, avgFitness, varFitness) values ('$currentGenerationIndex', '$bestFitness', '$avgFitness', '$varFitness')"))
        print "OK";
else
        print "ERROR"; 

?>

Then, I have another php page that just lists the contents of the table using a select * and a for loop (very similar to the code above)

Note: this code assumes you are using PostgreSQL and that you have a table called bugsim with four columns:
- currentGenerationIndex
- bestFitness
- avgFitness
- varFitness

Ignacio -