Hi -

I am in the middle of some refactoring of legacy code and am not sure how to write the mappings for the new code.

Consider that now we have the following:

class A
class B

A has a String property U0
B has a String properties U1, U2

U0, U1, U2 are all the same concept, but are not an object at the moment.

Now, it has become necessary for A to have 0..n U values.


Therefore, I see the creation of an class U, that will have among other things a property that holds the current U values.

so:
class U {
....
int id;
String propertyToHoldUValue;
... more stuff.
}
and the table for U will have at the minimum

uId and uValue.

Now in

A and B I see having

class A{
.....
Set <U> zeroToNUs;
....
}
class B{
.....
Set <U> zeroToNUs;
....
}

Now how does map this? Should I use join tables? Should I use a discriminator? If so what would the mappings look like? I also see in the future other objects having 0..n U.

Thanks,
Bal

You could use <set> to map U objects in your A hbm xml file. Something like this in your A.hbm.xml




__________________________________________
Please do RATE if this post was helpful