isis-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Stephen Cameron <>
Subject Re: Datanucleus issues
Date Tue, 01 Sep 2015 12:40:01 GMT
Further to this, for the time being the answer seems to be to not use
abstract parent classes. This will mean that have tables in my database
schema (that correspond to concrete parent classes) that aren't needed, not
a big deal I guess, but just annoying that I've spent much of today trying
to find solutions for issues associated with something that seemed
perfectly reasonable.

The Datanucleus documentation provides a solution for this inheritance
scenario, which is to make the abstract parent class table part of its
concrete child class tables, by using:

@Inheritance(strategy = InheritanceStrategy.SUBCLASS_TABLE)

Here is a quote: "DataNucleus supports persistence of classes in the tables
of subclasses where this is required. This is typically used where you have
an abstract base class and it doesn't make sense having a separate table
for that class."

I couldn't get this to work in my refactored code, that had something to do
with collections, I'll provide a simple test case to show that later on,
but  now I cannot use abstract classes as the properties of the abstract
parent class, with no collections, aren't visible anyway in the viewer

I know this sounds like a winge, sorry for that.

How to get past this situation? There are good reasons to use Datanucleus I
know, but its supposed to make things easier surely. Maybe we need another
kitchen-sink like demo just to show Datanucleus 'recipes' working inside

Tomorrow is another day.


On Tue, Sep 1, 2015 at 8:39 PM, Stephen Cameron <>

> Hi,
> I've been trying today to find solutions to what seem to be Datanucleus
> issues, but without much success. So rather than solve the problems created
> in my refactoring of my application, which has been too hard frankly, I'm
> trying to start afresh with some simple test cases, show these work, then
> add more complexity till I get where I want to be.
> So, I am starting this approach and I immediately have an issue, but not a
> Datanucleus one, I find that the value properties of an abstract parent
> class don't appear in the Wicket viewer, whereas they do if the parent
> class is concrete. Is this correct and if so what is the reason?
> Thanks.

  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message