just to clarify, what I am saying is this.
when I have a single class that uses discriminator and no subclass, it generates a select statement like this select street1, street2, city, state_or_provide, postal_code from Address when I have subclass, it generates the following select statement select street1, street2, city, state_or_provide, postal_code, address_type_code from Address My point is the select statement should include "address_type_code" column in the column part of the select statement, not the where part. peter On Sep 26, 2:54 pm, "Jon Palmer" <[EMAIL PROTECTED]> wrote: > I disagree, the current implementation is exactly right. > > In a normal situation your table has only data that is mapped in NHibernate > (ie all the values of the discriminator column are mapped to some class). Now > if I perform a query for the base class I DON'T want NHibernate to put in an > spurious where clause which will be true for all rows in the table. > > I think its right that it works the way it does and you have to employ a work > around in your legacy data case. > > Jon > > -----Original Message----- > From: [email protected] [mailto:[EMAIL PROTECTED] On Behalf Of Peter > Lin > Sent: Friday, September 26, 2008 11:51 AM > To: nhusers > Subject: [nhusers] Re: Discriminator bug > > you could use where clause, filters, views or create a dummy subclass. > > my point is, it should include the discriminator column in the select > clause, so that users don't have to resort to these work arounds. > > having built several ORM in the past, that's how I would do it. If > anyone knows which file to look at, I'm happy to take a look and make > a patch for it. > > peter > > On Sep 26, 2:40 pm, "Tuna Toksöz" <[EMAIL PROTECTED]> wrote: > > I think this is where filters come into play? > > > On Fri, Sep 26, 2008 at 9:31 PM, Jon Palmer <[EMAIL PROTECTED]>wrote: > > > > Can't you just specify a Where clause in the definition of the single > > > class mapping. > > > > -----Original Message----- > > > From: [email protected] [mailto:[EMAIL PROTECTED] On > > > Behalf Of Peter Lin > > > Sent: Friday, September 26, 2008 11:29 AM > > > To: nhusers > > > Subject: [nhusers] Re: Discriminator bug > > > > the reason is simple. what if I'm reading data from an existing > > > database, which uses the concept of a discriminator with > > > address_type_code. what if the address table has entries with > > > different address_type_code values? > > > > If I only want to get a specific type of address from the database and > > > I don't have a subclass, NHibernate wouldn't return the correct > > > result. It would return all rows, instead of just the rows with the > > > specific address_type_code. > > > > I just looked at the release notes for NH 2.0 and it looks like this > > > scenario is supported. > > > > Added [ Table per subclass, using a discriminator ] Support to > > > Nhibernate > > > >http://ayende.com/Blog/archive/2008/03/31/NHibernate-2.0-Alpha-is-out.as > > > px > > > > unfortunately, I can't upgrade to NH 2.0. I understand NH might not > > > consider this a valid use case, but I definitely consider it a valid > > > use case. > > > > peter > > > > On Sep 26, 11:20 am, "Fabio Maulo" <[EMAIL PROTECTED]> wrote: > > > > Why you think that it is a bug ?If a class don't have a <subclass> > > > (mean you > > > > have only one value for discriminator and that mean don't have nothing > > > to > > > > discriminate), why NH must add the discriminator clause ? > > > > > 2008/9/26 Peter Lin <[EMAIL PROTECTED]> > > > -- > > Tuna Toksöz > > > Typos included to enhance the readers attention! --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "nhusers" 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/nhusers?hl=en -~----------~----~----~----~------~----~------~--~---
