Here is a more detailed answer. A case when a criteria for a
subclass / superclass is based on a presence of a given relationship
is better addressed with vertical inheritance - something we'll be
adding in 3.0. Single table inheritance mapping requires a criteria
based on one or more colum
On Aug 6, 2008, at 2:54 PM, Dave Dombrosky wrote:
Do I have to add another column just to hold the type like the
documentation example shows? Not the best solution in this case, but
I guess it can be done if need be.
That would certainly work.
Andrus
Oh boy. What do you recommend in the meantime?
Do I have to add another column just to hold the type like the
documentation example shows? Not the best solution in this case, but
I guess it can be done if need be.
-Dave
On Wed, Aug 6, 2008 at 2:47 PM, Andrus Adamchik <[EMAIL PROTECTED]> wrote
Ok, I see. This indeed is a limitation of the inheritance algorithm.
We can't evaluate the inheritance qualifier expression in-memory in
this case. We can probably improve on that in the future, but there is
no easy fix.
Andrus
On Aug 6, 2008, at 2:29 PM, Dave Dombrosky wrote:
Yes it is
Yes it is just a 1:1 mapping on the relationship.
Here is the stack trace:
org.apache.cayenne.exp.ExpressionException: [v.3.0-SNAPSHOT Aug 01
2008 19:55:10] Error evaluating expression 'db:some_relationship !=
null'
at
org.apache.cayenne.exp.parser.SimpleNode.evaluate(SimpleNode.java:331
Hi Chris,
Since I haven't got nowhere near the cause yet, I can't suggest a good
name either. What's important is that you add steps to reproduce it.
Call it anything you feel like.
Thanks,
Andrus
On Aug 6, 2008, at 1:51 PM, Chris Gamache wrote:
Just a squeak from the wheel here.
I'd cr
Just a squeak from the wheel here.
I'd create a bug report, but I'm not sure how to describe it...
Join on ExtendedType invalidates ExtendedType when retrieving rows in parent
table using SelectQuery ??
That's one idea I have. Please advise.
On Mon, Aug 4, 2008 at 5:55 PM, Chris Gamache <[EMAIL P
Could you post a bigger part of the stack trace (at least everything
that has org.apache.cayenne in it)?
From what I can see, this is related to in-memory evaluation of the
qualifier. If that's a to-one relationship based on FK, intuitively
this should work, but it probably hits a limitatio
On Aug 6, 2008, at 11:26 AM, Andrus Adamchik wrote:
Not quite sure why the root is null.
Actually I have a few suspicions why this is happenning, and this may
be a bug. Care to log a bug report in Jira?
https://issues.apache.org/cayenne/
In the meantime you can use a workaround:
Hi Jens,
Not quite sure why the root is null. So is named query mapped via the
Modeler or do you add it in the code?
As an aside (this may have nothing to do with your problem), all
mapped queries are shared across DataContexts, so you may have a race
condition when you attempt to modify
On Aug 5, 2008, at 9:55 PM, Lachlan Deck wrote:
Are you +1 on providing these with accessors?
Actually no, as there is lots of private state there. I want to hide
as much as possible. I think we are exposing too much stuff at this
level already. If you have a specific adapter customizatio
On 06/08/2008, at 11:32 AM, Andrus Adamchik wrote:
On Aug 5, 2008, at 9:24 PM, Lachlan Deck wrote:
Was just thinking that there didn't seem to be any hook available
to createSqlString() (which calls the above method) to ensure
suppressingDistinct is true for this case.
Sure. I am +1 on su
Hi,
i'm using Cayenne 2.0.4 with Oracle 10 and JDK 1.5.
Please have a short look at this code:
... dataCtx, myID are set...
SQLTemplate rawQuery =
(SQLTemplate)dataCtx.getEntityResolver().lookupQuery("myQuery");
Map params = new HashMap();
params.put("processID", myID);
rawQuery.setParameters
13 matches
Mail list logo
