Re: Why PHPLIB::DB sucks

On Fri, 6 Jun 2003, Bertrand Mansion wrote: > >> Isn't that the way they do it in PEAR::DB too ? > > > > I think it uses native sequences for postgres. > > IIRC, they also use a separate sequence table for each table in MySQL, but > we all know these tables can have autoincremented fields. If the sequence > table doesn't exist, it even seems to create one. I guess this adds > overhead. Yes, but the original mail was about Postgres, and PEAR::DB uses native sequences with Postgres: -cut- test-> \d person Table "public.person" Column | Type | Modifiers -----------+-----------------------+----------- ... test-> \d person_seq; Sequence "public.person_seq" Column | Type ---------------+--------- ... -cut- -- Mika Tuupola http://www.appelsiini.net/~tuupola/