Re: PEAR DB abstraction layers

Aidan Lister wrote:

I didn't want to reply to pear-dev, as I think this topic should be kept on track as far as possible, but I have some questions about MDB2:

I hope you are ok with me replying into the original thread. I think alot of people have similar concerns then you.

Is it more complicated to use? I'm looking at all this set-type stuff and I'm thinking "ohh to hard". Is that optional? Is the API less complex than DB?

The type stuff is optional. The only difference when you dont want to use the types is that a few parameters will have to be set as "null".

Why not return a result object? I love the result object, why does MDB not use it? Is this the reason for the speed increase?

Yes performance was the reason. However I changed this in MDB2 already. There I return result objects. I did alot of performance tweaks and MDB2 is therefore still faster than MDB. And MDB is faster than DB. While I am at it: No MDB[2] doest not require you to parse XML ever. XML is just used optionally if you want to keep your SQL schema in an RDBMS independent form. regards, Lukas Smith smith@backendmedia.com _______________________________ BackendMedia www.backendmedia.com berlin@backendmedia.com Linn Zwoch Smith GbR Pariser Str. 44 D-10707 Berlin Tel +49 30 83 22 50 00 Fax +49 30 83 22 50 07