which is better, please?

[Anon]

For a mySQL lookup, I have always used:
mysql_select_db(\'my_db\', mysql_connect( \'localhost\', \'my_username\', \'password\'))

But, when I had programming work done for me, they used:

define(\"DB_HOST\", \"localhost\");
define(\"DB_USER\", \"my_username\");
define(\"DB_PASS\", \"password\");
define(\"DB_NAME\", \"my_db\");

$DBH = \"\";

if( !$DBH = mysql_connect(DB_HOST, DB_USER, DB_PASS) ) {
echo \"<h2>Cannot connect to database server!</h2>\";
echo \"Mysql error: \", mysql_error();
die;
}
if( !mysql_select_db(DB_NAME, $DBH) ) {
echo \"<h2>Cannot access database!</h2>\";
echo \"Mysql error: \", mysql_error();
die;
}
?>

What on EARTH did they do all that for?

Does php connect to mySQL FASTER than my way? Or is it actually a bit slower. What kind of programming did they use, since it looks different than regular php???

Thank you!

[mrmufin]

"Regular PHP" has a quite a few different faces.

"They" did it that way (I surmise)... because "that way" allows for your database connection parameters to be stored in a single, external file (preferably outside of the servers ~/htdocs path).

Suppose you move the application from the current server to another server, using a different database name, etc... your way means editing every single instance where a database connection is established. "Their way" means editing a single set of configuration constants.

Further, they exit on "can't connect" and "can't select" errors. If what follows depends upon the results of a query, there's no use generating the rest of the page if a database connection or selection can't be established.

Sometimes bad things happen in web server and database server world. They way they did it allows for errors to happen without dumping a bunch of ugly PHP error messages to the screen.