Re: New class in PEAR: Translation

On 2002-10-01 14:58, while getting another beer from the fridge I saw a little yellow card, on which Markus Wolff wrote:

Am Tue, 1 Oct 2002 14:01:55 +0200 schrieb "Lukas Smith" <smith@dybnet.de>:

actually what we use today is just associative arrays that are stored in a file. so we are doing pretty much that today as well (and it works nicely and performant too). We don’t have such a helper script however (eventhough I have been meaning to write one). The nice thing about having it in the db is that you can modify it easier and/or distribute it across a cluster ...

You could pass either a DSN, PEAR::DB or MDB connection object or an array to the constructor of such a translation class and it would then detect what kind of data it is and act accordingly. That way everyone would be happy :-) Regards, Markus Passing the DB connections is the plan for next version :)

But however passing the connections to e.g. flat files - honestly I do not (yet :) ) have an idea how to do that. But I will surely think about that and will have to make a closer look at MDB class (always used DB :) ) Voyteck