On 2/9/06, Alexandru Popescu <[EMAIL PROTECTED]> wrote:
> #: Christophe Lombart changed the world a bit at a time by saying (astral 
> date: 2/9/2006 10:21 PM) :#

> Considering that the class-descriptor element is quite simple at this moment 
> and that in Java we
> have only single class inheritance, I would rather see "extends" as an 
> attribute in class-descriptor.
>

ok

> Another problem that I see at this moment is: when defining a class mapping 
> and considering
> inheritance too, we have to deal with 2 hierarchies: Java hierarchy and also 
> node type hierarchy. I
> think it is not possible to define a Java hierarchy that is persisted on 
> different/unreleated node
> types. What I am trying to say is: if B extends A, B is of type node:b and A 
> is node:a than node:b
> must inherit from node:a.
>
Yes, the node type manager (under construction by Oliver) can manage this .

> At the first glance the part about the query should work. Still, I don't see 
> how you will be able to
> decide after fetching the nodes which object to instantiate: ContentImpl or 
> FolderImpl? How can you
> decide it? Considering that both classes can be persisted to the same node 
> type, than I cannot see a
> way to differentiate them. Reading a node property set and deciding upon this 
> if a ContentImpl or
> FolderImpl must be initialize is not gonna work.
>
>

No, both classes are not persisted to the same node type.
Following my previous example,
Folder is mapped to the node type : graffito:folder
Content is mapped to the node type : graffito:content

both node types inherit from graffito:cmsobject.

Of course, this solution doesn't works with  nt:unstructured.

>
> > 2/ Interface  : I don't understand why mixin node types should be use.
> > Java interface have not some attributes. So, I don't see how it can be
> > set into the JCR node type hierarchy.
> > I think we can also reuse the extent-class element but more
> > modification in the current code is required. Still thinking about
> > that.
> >
>
> I am not sure about this point :-S.
>
> ./alex
> --
> .w( the_mindstorm )p.
>
> >
> >
> > --
> > Best regards,
> >
> > Christophe
> >
>
>


--
Best regards,

Christophe

Reply via email to