> -----Original Message-----
> From: Jason van Zyl [mailto:[EMAIL PROTECTED]]
> Sent: Thursday, September 27, 2001 12:39 PM
> To: [EMAIL PROTECTED]
> Subject: Re: Torque and <filesets>
> 
> 
> On 9/27/01 3:27 PM, "Fedor Karpelevitch" 
> <[EMAIL PROTECTED]>
> wrote:
> 
> > I think, a better way to handle these would be to add support for
> > "including" one schema into another. maybe some <include> element or
> > something... FOr example a turbine app would just "include" 
> turbine schema
> > into it's schema. that would remove the need for nasty 
> hacks when you have
> > to create alias for turbine_user in your own schema...
> > Any ideas on what would be most kosher way to do this?
> 
> I personally don't like this method as you can't really take 
> advantage of an
> <uptodate> task for rebuilds.
> 
> I think separate schemas are better for maintenance. Even 
> with the changes
> I'm proposing you could still do that if you wanted but I 
> wouldn't recommend
> it. And I certainly would be against making that the method 
> that is employed
> in the TDK. If you create subsequent schemas you must edit 
> and fart around
> with the main schema that includes all other schemas. By 
> using a fileset you
> can process 1 or 10 in the same fashion.

We are trying to solve slightly different problems, I guess.
I am talking about a case when some schema is an integral part of a bigger
schema. An example is turbine schema which is a part of a turbine app's
schema. In this case, I guess, it would be valuable to be able to "include"
turbine schema into app schema, so they are considered to be the same db.
This would not let you provide a wrong FK from your table to TURBINE_USER
and such and would not require you to do any cut-and-paste which is never
good. This also reflects dependencies correctly - app schema depends on
turbine schema. So when you do an uptodate check you would be able to (and
should to be correct) also walk down the deps - id app schema is updated it
needs to be regenerated, if turbine schema is updated - both need to be
regenerated.
What you are trying to do is to process several independent schemas in one
shot. It may have it's valid applications, but it's a different case...

fedor

> 
> >> -----Original Message-----
> >> From: Jason van Zyl [mailto:[EMAIL PROTECTED]]
> >> Sent: Thursday, September 27, 2001 12:20 PM
> >> To: [EMAIL PROTECTED]
> >> Subject: Torque and <filesets>
> >> 
> >> 
> >> Hi,
> >> 
> >> I am trying to provide an easy way of dealing with multiple
> >> databases in the
> >> TDK and I would like to add the ability to the torque task of
> >> dealing with
> >> a set of xml torque schemas.
> >> 
> >> The tasks would work with their current syntax accepting 
> xmlFile as a
> >> parameters, but if that attribute is not present than a
> >> <fileset> of xml
> >> torque schemas will be processed.
> >> 
> >> For now I am interested in changing the following tasks:
> >> 
> >> TorqueCreateDatabase
> >> TorqueObjectModelTask
> >> TorqueSQLTask
> >> 
> >> I would also like to change the TorqueSQLTask to produce a
> >> small table that
> >> maps the SQL file to the database it should be inserted into.
> >> I would than
> >> make another task using the <sql> task as a base that would
> >> take this little
> >> table and slot each SQL file into the appropriate database.
> >> 
> >> I am thinking that a project may potentially have several xml
> >> schemas, as
> >> tambora does, and that it might be better to keep them
> >> separate so that an
> >> <uptodate> task can be used to automate rebuilds in the
> >> future. In tambora
> >> we like to keep the turbine schema in one database and the
> >> tambora schema in
> >> another. But the changes the tasks won't affect current
> >> operation: if you
> >> use one schema than that's fine.
> >> 
> >> I'm trying to find an easy way to maintain the data model
> >> given the use of
> >> one database or many. Again these changes won't affect
> >> current use, and I
> >> primarily want these changes so I can simplify the sample app
> >> initialization
> >> and deployment in the TDK.
> >> 
> >> I'm off to bed now :-)
> >> 
> >> -- 
> >> 
> >> jvz.
> >> 
> >> Jason van Zyl
> >> 
> >> http://tambora.zenplex.org
> >> http://jakarta.apache.org/turbine
> >> http://jakarta.apache.org/velocity
> >> http://jakarta.apache.org/alexandria
> >> http://jakarta.apache.org/commons
> >> 
> >> 
> >> 
> >> 
> ---------------------------------------------------------------------
> >> To unsubscribe, e-mail: [EMAIL PROTECTED]
> >> For additional commands, e-mail: 
> [EMAIL PROTECTED]
> >> 
> > 
> > 
> ---------------------------------------------------------------------
> > To unsubscribe, e-mail: [EMAIL PROTECTED]
> > For additional commands, e-mail: [EMAIL PROTECTED]
> 
> -- 
> 
> jvz.
> 
> Jason van Zyl
> 
> http://tambora.zenplex.org
> http://jakarta.apache.org/turbine
> http://jakarta.apache.org/velocity
> http://jakarta.apache.org/alexandria
> http://jakarta.apache.org/commons
> 
> 
> 
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: [EMAIL PROTECTED]
> For additional commands, e-mail: [EMAIL PROTECTED]
> 

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

Reply via email to