Reputation: 2573
Consider the following tables:
Client(Id bigint, Name varchar(50))
Employee(Id bigint, Name varchar(50), ClientId bigint)
that are mapped like this:
<class name="Client" table="`Client`">
<id name="Id" column="`Id`" type="long">
<generator class="native" />
</id>
<property name="Name" column="`Name`" />
<bag name="Employees" cascade="all" inverse="true" >
<key column="`ClientId`" />
<one-to-many class="Employee" />
</bag>
</class>
<class name="Employee" table="`Employee`">
<id name="Id" column="`Id`" type="long">
<generator class="native" />
</id>
<property name="Name" column="`Name`" />
<many-to-one name="Client" cascade="all" column="`ClientId`" />
</class>
If I get a Client, I also get a collection of Employees where the Employee.Client = Client.Id. Great.
Now consider this:
Client(Id bigint, Name varchar(50), AlternativeId int)
Employee(Id bigint, Name varchar(50), ClientId bigint, AlternativeClientId int)
I want to return a Client with a collection of Employees where Employee.AlternativeClientId = Client.AlternativeId.
I would assume the key node would now read:
<key column="`AlternativeClientId`"/>
But beyond that I'm stumped. There are filters that can be applied to collections, but the set of Employees in the second version may not be a subset of the Employees in the first version, so I don't think that is the way forward. I tried but that seemed to be a dead end. IS there some way of specifying a query, but not only on Employees that have the ClientId = Client.Id?
(For the 'whys': It's to do with how different systems have different views of the data.)
Upvotes: 0
Views: 958
Reputation: 1112
I'm pretty new to nhibernate but this sounds like a problem I had due to working with a brownfield database. The property-ref attribute allows a column other than the primary key to be referenced. So in your case it should be:
<key column="AlternativeClientId" property-ref="AlternativeId"/>
Upvotes: 2