On Fri, 2002-08-23 at 06:44, Rana Bhattacharyya wrote:
> Hi,
>   For one of my project I am using torque (well
> almost). After going through the torque sources, I
> decided to use a subset of torque. I have copied some
> parts of the code, modified, tested. You guys are
> doing a really great job. 
> 
> But it seems the codebase is not stable yet.
> Here are my observations and what I did with sources:
> 1. I have removed all village stuff.

Village is a useful but thin layer over jdbc.  But as torque removes the
need to use either directly, I have considered removing the dependency,
but it has always seemed like too much work with too little benefit. 
Did you keep the PreparedStatement's used by village?

> 
> 2. I got rid of org\apache\torque\om\* classes.
> 

Not sure what this buys you.  So you are saying you do not wish to have
an OID class that uniquely identifies your objects.  I think it is good
to have this abstraction, so that you can have methods that take an
ObjectKey to create the object.  

It might be a better api, if the properties which use the column names
returned the column type, but getPrimaryKey() should return an
ObjectKey.


> 7. From programming point of view DB should return
> IdGenerator. But the Idgenerator id, object etc. are
> really confusing.

Can you explain this further?

john mcnally




--
To unsubscribe, e-mail:   <mailto:[EMAIL PROTECTED]>
For additional commands, e-mail: <mailto:[EMAIL PROTECTED]>

Reply via email to