[snip] > a new connection, but in terms of op codes, a new connection is created.
no idea why your going on about op codes, and no a new connection is not made - there is merely switching between DBs on a single connection. it's the same as connecting with the mysql cmdline util and being able to browse any db by simply performing 'USE <DB>' queries ... those queries don't cause the mysql client to open new connections. [/snip] Exactly. One connection, no new overhead. A few years ago, and I think it was on this list, someone gave what I consider to be some sage advice; "allow the database to do all of the work that it can do before applying any PHP to it". If you use explicit database.table names in your queries it is placing the workload on the database engine rather than making a function call in PHP that essentially issues a USE <db> function call on the database. You're using 2 functions (and at least one additional line of code) to achieve what you could have achieved by stating the database name in the query. Like Jochem, I am unsure why you're going smackers over op code unless your systems are antiquated or under-performing. We quit counting clock cycles a long time ago. Is it because your application has a ton of users with lots of connections? If so you may want to look into connection pooling for the database. -- PHP General Mailing List (http://www.php.net/) To unsubscribe, visit: http://www.php.net/unsub.php
