[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [GNUe-dev] Some questions about Introspection
From: |
Jason Cater |
Subject: |
Re: [GNUe-dev] Some questions about Introspection |
Date: |
Tue, 8 Jun 2004 15:42:40 -0500 |
User-agent: |
KMail/1.6.2 |
Reinhard,
> We had this idea of providing a function that accepts a "should-be"
> description of the db schema, and changes the current state into the
> should-be state.
>
> That is, this function would first compare the wanted schema with the
> existing schema and then decide which tables have to be added, which
> tables have to be extended etc.
Ok, that makes a little more sense. But even then, I see this as the Creation
class using the introspection instance to get the data it needs. I personally
think it'd be better to keep these separated, as there's still little overlap
Aside from Creation asking Introspection for the description of a table, how
would they interact?
> Of course, we could also put this logic on top of the Creation and
> Introspection modules, as a separate layer.
The original plans for gnue-schema included support for this. I feel that the
process of writing the schema modifications to the database and the process
of creating the schema definitions should probably be kept separate. Or at
least have the writing process be optional; i.e., have different methods in
the class for creating the sql versus actually doing the updates.
I'm not 100% set on all the above and could be convinced otherwise if you
really feel differently, but this is just my gut feeling.
-- Jason