Hi all,

i've just discovered that mysql_connect() uses a hash that forces it to reuse a connection on a second call to mysql_connect() with the same arguments, instead of opening a new connection.
This was, at least for me, an unexpected behaviour: i was hoping to be able to use several database connections to the same mysql server (and same user) in order to be able to issue several LOCK TABLE or GET_LOCK() queries (mysql allows only one 'active' lock per connection)

[I know basic locking may be done on php server's system level, but it doesn't make sense in web-cluster environment, and using a database to centralize locks sounds interesting]

I also took a look into php's mysql api source (i'm not familiar with php sources), and adding an optional argument saying _not_ to reuse a connection or coding an alternative mysql_connect() from the existing one doesnt _seem_ to be a long or complicated task.

I _might_ do the change myself for my server, but as the sites i'm working on are unlikely to be hosted on my server, this is not very portable and would be probably useless.

Please tell me what you think about this issue, and if a change to mysql_connect() function is conceivable (in that case i'd be glad to do the change myself and send you the result :)

Thanks
-- vedad