FactTemplates are an experimental feature and only supports primitives
and strings as fields. It was based around using arrays for the models,
this made the implementation easy but tooling and integration with other
libraries being very hard. For 5.0 we using "type declarations" which
basically are generated pojos, which causes a nightmare for custom
classloader stuff, but is much more tooling and 3rd party library
integration friendly. These type declarations support primitives and
classes for their fields.
Mark
Stephen Kestle wrote:
I'm trying to make some fact templates in the form of
template "Address"
String line1
String line2
end
template "Person"
String name
int age
Address address
end
It complains that it is "Unable to resolve class "Address""
1. Is there any way around this - am I just getting syntax wrong?
2. Is this being fixed in 5.0 or > 4.0.7 (which I have)
3. Am I able to modify the class loader to grab the FactTemplate
class types and infer the correct object? e.g. is this something
that's ridiculously hard, and so hasn't already been included?
4. Am I going about this entirely the wrong way?
1. Are nested types at all useful in the rules?
2. Is it possible to even have a rule like:
p: Person( age == 5, Address(line1 == "My address))
? (given a java object schema)
3. Should I be using variables and other rules to assert facts?
4. Off topic: Is rules composition achieved by the assertion of
facts for other rules to use?
What I want to do:
I am dealing with (xml) schemas and ontologies, and need to be able to
create Fact Templates from these definitions. While I can do this
easily enough in java side, I believe I need to create the templates
to match my java objects so that the rules can be compiled.
Thanks heaps
Stephen
_______________________________________________
rules-users mailing list
rules-users(a)lists.jboss.org
https://lists.jboss.org/mailman/listinfo/rules-users