Jon Stevens wrote:
> .sql -> MySQLDatabaseMapBuilder -> generated .java
>
> As you can see, this is much fewer steps, which is much easier and cleaner
> to me since you don't have to waste your time trying to make a square peg
> (xml) fit into a round hole (.sql).
As far as i can see, the current way sql -> MapBuilder is short but is
not very clean. SQL is a command language and _not_ a description
language. So, using sql files to describe a database structure may
hardly be appropriate. Actually, your .sql generator tool has probably
an internal format for describing the database structure, and outputs
.sql files that create this stucture into a specific sgbd (say MySQL).
Of course, as long as you do not have access to this descriptor you have
to play with the sql files which is easier for you :-) You are probably
not alone and that is why the current schema has to stay there. It is
simple and efficient.
Anyway, what I propose is to define a XML file that describes a database
structure. This file may then be used to generate the .sql scripts for
creating the database (in the same way that your db-tool does) and to
generate the turbine peers .java classes. Of course, one way for doing
this may be
description.xml -> script.sql
script.sql -> MyMapBuilder
I am not dealing with object persistance but only with relational
database description. So, I think that OPaL is not exactly doing the
same job. But I do not really know OPaL.
One more time, I want to underscore that I am quite a newbie into
Turbine and I will not be "angry" if you flame me because I am totally
wrong :-)
Regis
------------------------------------------------------------
To subscribe: [EMAIL PROTECTED]
To unsubscribe: [EMAIL PROTECTED]
Search: <http://www.mail-archive.com/turbine%40list.working-dogs.com/>
Problems?: [EMAIL PROTECTED]
