From: "Craig R. McClanahan" <[EMAIL PROTECTED]>
> On Fri, 18 Oct 2002 [EMAIL PROTECTED] wrote:
> Interesting ... I was looking at the same class for the same reason
> recently.  My thought was to split the JDBC-->XML functionality into two
> pieces:  JDBC-->Internal graph of objects starting with
> o.a.c.s.model.Database, and Object Graph->XML (which already exists via
> DatabaseWriter).  That way, a tool or app that wanted to introspect a set
> of database metadata, and then use it directly, could do so without having
> to do it in two steps.

Agreed. Thats always been the intention. Then a model-diff could be
implemented for determining the DDL required to perform a schema upgrade, to
compare the database changes required when upgrading software or to test if
a real database has the required schema that a codebase expects etc.


> One question I had when looking at this, though ... the data model
> represented in the XML document format doesn't totally match that
> represented in the o.a.c.s.model package classes.  Shouldn't it?

It should. I'm sure we can fix it fairly easily.

The model needs to be extended a bit right now for indices and maybe extend
the relationship support a little. I've been chatting to some of the
middlegen folks lately

http://boss.bekk.no/boss/middlegen/

who might use commons-sql as the database model on which their UI and
Ant/XDoclet tools will generate JDO, CMP and other code from. A few minor
extensions to commons-sql model will be required to facilitate this (like
supporting relationship cardinality etc).

James
-------
http://radio.weblogs.com/0112098/

__________________________________________________
Do You Yahoo!?
Everything you'll ever need on one web page
from News and Sport to Email and Music Charts
http://uk.my.yahoo.com

--
To unsubscribe, e-mail:   <mailto:commons-dev-unsubscribe@;jakarta.apache.org>
For additional commands, e-mail: <mailto:commons-dev-help@;jakarta.apache.org>

Reply via email to