I've just committed the changes I spoke of. There's now no restrictions on
the types you can use for discriminator, so although you can specify it as
an int, you can still pass in a string value; this will allow you to pass in
"null" or "not null" for these special cases. In addition I've added a
little enum class to make it read a bit nicer.
DiscriminateSubClassesOnColumn("type")
.SubClass<Child>(DiscriminatorValue.Null, sc =>
{
})
.SubClass<Child2>(DiscriminatorValue.NotNull, sc =>
{
});
On Sun, May 3, 2009 at 4:38 PM, jcavaliere <[email protected]>wrote:
>
> Ok, now I feel stupid!
>
> It’s "not null” as opposed to "not-null” when I'm using a
> discriminator value.
>
> Exporting the xml helped me a lot, I should have done that first off.
> And yes, it did work with specifying a string as the type, even if it
> is a numeric data type, so that's great! I'll keep an eye out for
> future updates where that will not be necessary, so that it doesn’t
> confuse the team.
>
> Thanks again!
> -Jason
>
> On May 3, 10:08 am, James Gregory <[email protected]> wrote:
> > Well, FNH just generates XML under the hood so there shouldn't be a
> > distinction between a keyword and the string literal.
> > There's a way to export the xml that FNH is passing to NH, it might be
> worth
> > having a look at that and comparing it to what you'd expect.
> >
> > If you're using the Fluently.Configure style configuration, there's an
> > ExportTo method on the FluentMappings property
> > (m.FluentMappings.AddFromAssemblyOf<X>().ExportTo("Path")). If you're
> using
> > the PersistenceModel directly there's a WriteMappingsTo call you can use.
> >
> > On Sun, May 3, 2009 at 4:03 PM, jcavaliere <[email protected]
> >wrote:
> >
> >
> >
> >
> >
> > > Version 2.1
> >
> > > I did try changing the type to a string, but is seemed to be looking
> > > for the literal of "null" and "not-null” rather than the keyword
> > > interpretation that the xml supports.
> >
> > > Take your time and to it right - and if it takes a couple weeks I'm
> > > all happy. I see fluent NHibernate as the long term solution and, if
> > > I have to, I'll put an extra column in the tables as an interim
> > > solution until it's implemented.
> >
> > > Best regards,
> > > -Jason
> >
> > > On May 3, 9:55 am, James Gregory <[email protected]> wrote:
> > > > Can't you change then type in the Discriminate call to be string?
> > > > I'm actually in the middle of some changes to this area (not yet
> > > committed),
> > > > one of which is to remove the strictness of the types for the
> > > discriminator.
> > > > Interestingly, I was not aware of the null/not-null behavior, and it
> > > seems
> > > > to be very poorly documented. What version of NH are you using?
> >
> > > > On Sun, May 3, 2009 at 3:49 PM, jcavaliere <
> [email protected]
> > > >wrote:
> >
> > > > > Great - I'll make that change. It still doesn’t compile though,
> > > > > because I'm testing the value of an int property on my domain with
> a
> > > > > string value. Does this all lead to disappointment for me in the
> > > > > short term?
> >
> > > > > Thanks again for the super quick response – I’m really happy w/ the
> > > > > framework overall!
> > > > > -Jason
> >
> > > > > On May 3, 9:36 am, James Gregory <[email protected]> wrote:
> > > > > > You should be chaining your subclass calls, like so:
> > > > > > DiscriminateSubClassesOnColumn<int>("CustomerID")
> > > > > > .SubClass<NewSalesPerson>("null", m => { })
> > > > > > .SubClass<SalesPerson>("not-null", m => { });
> >
> > > > > > On Sun, May 3, 2009 at 2:49 PM, jcavaliere <
> > > [email protected]
> > > > > >wrote:
> >
> > > > > > > Maybe I'm not following you - so I have an interface mapped to
> my
> > > db,
> > > > > > > ISalesPerson, and I want to elevate privileges based on whether
> > > they
> > > > > > > have made a sale or not, based on an ID those points to the
> > > customer
> > > > > > > table.
> >
> > > > > > > So if CustomerID is null I map to a NewSalesPerson Class, if it
> is
> > > not
> > > > > > > null, I map to SalesPerson subclass. Her'es the map I have so
> far.
> > > > > > > This works w/ the XML files, but I can't figure out how to make
> it
> > > > > > > work with the fluent format. I'm sad too, because I really
> like
> > > the
> > > > > > > testability that goes with this setup.
> >
> > > > > > > Thanks for the help!
> > > > > > > -Jason
> >
> > > > > > > public class SalesPersonMap : ClassMap<ISalesPerson>
> > > > > > > {
> > > > > > > public SalesPersonMap()
> > > > > > > {
> > > > > > > Id(x => x.SalesPersonID);
> > > > > > > Map(x => x.CustomerID);
> > > > > > > Map(x => x.Address);
> > > > > > > Map(x => x.City);
> > > > > > > Map(x => x.Country);
> > > > > > > Map(x => x.Email);
> > > > > > > Map(x => x.Fax);
> > > > > > > Map(x => x.IsActive);
> > > > > > > Map(x => x.Phone);
> > > > > > > Map(x => x.State);
> > > > > > > Map(x => x.Zip);
> > > > > > > Map(x => x.IsVisible);
> > > > > > > WithTable("SalesPerson");
> >
> > > > > > > DiscriminateSubClassesOnColumn<int>("CustomerID
> > > > > > > ").SubClass<NewSalesPerson>("null", m => { });
> > > > > > > DiscriminateSubClassesOnColumn<int>("CustomerID
> > > > > > > ").SubClass<SalesPerson>("not-null", m => { });
> >
> > > > > > > }
> >
> > > > > > > }
> >
> > > > > > > On May 3, 5:44 am, James Gregory <[email protected]>
> wrote:
> > > > > > > > There's an overload on the SubClass method that takes a
> > > discriminator
> > > > > > > value,
> > > > > > > > otherwise it defaults to the class name.
> >
> > > > > > > > On Sat, May 2, 2009 at 5:40 PM, jcavaliere <
> > > > > [email protected]
> > > > > > > >wrote:- Hide quoted text -
> >
> > > > > > - Show quoted text -- Hide quoted text -
> >
> > > > - Show quoted text -- Hide quoted text -
> >
> > - Show quoted text -
> >
>
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"Fluent NHibernate" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to
[email protected]
For more options, visit this group at
http://groups.google.com/group/fluent-nhibernate?hl=en
-~----------~----~----~----~------~----~------~--~---
