Correct.  I use a field access for most of my collections.. in fact I
don't know that I have a single collection I allow direct access to so
they all use field access.

On Oct 21, 8:42 pm, "Tim Barcz" <[EMAIL PROTECTED]> wrote:
> Then you just have to set the access attribute in you mapping file correct?
>
> Tim
>
> On Tue, Oct 21, 2008 at 8:59 PM, Shane C <[EMAIL PROTECTED]> wrote:
>
> > Well I personally never allow direct access to my collections so I
> > already have an AddPlayer method.  It's a simple enough matter to go
> > newPlayer.Game = this; and the two way relationship is intact.
>
> > On Oct 21, 7:06 pm, "Tim Barcz" <[EMAIL PROTECTED]> wrote:
> > > Thanks for the input, that's exactly what I'm bumping into now.  I
> > presume
> > > then you make it bidirectional to make your DBA happy?  If that is the
> > case
> > > what steps are you taking in your code to ensure everything is working
> > the
> > > way you want?
>
> > > Tim
>
> > > On Tue, Oct 21, 2008 at 8:00 PM, Shane C <[EMAIL PROTECTED]>
> > wrote:
>
> > > > Personally outside of an NHibernate environment I prefer uni because
> > > > their is less complexity BUT within NHibernate I have yet to make uni
> > > > relationships work WHEN there is a non-null constraint on the foreign
> > > > key.  The reason for this is that NHibernate ends up doing an insert,
> > > > and then a separate update to set the relevant ID.  For example the
> > > > following table setup will NOT work for me:
>
> > > > Game Table
> > > > GameId
> > > > Name
>
> > > > Player Table
> > > > PlayerId
> > > > GameId (non-null foreign key to GameId)
> > > > Name
>
> > > > NHibernate: INSERT INTO Game (GameId) VALUES (@p0); @p0 = '163840'
> > > > NHibernate: INSERT INTO Player (Id) VALUES (@p0); @p0 = '196608' <-
> > > > KABOOM! There is no GameId
>
> > > > Good luck finding a DBA that will agree to making GameId null-able
> > > > because of that "silly" tool you're using.
>
> > > > On Oct 21, 2:36 pm, "Symon Rottem" <[EMAIL PROTECTED]> wrote:
> > > > > I guess the primary issue is that if you do a bidirectional
> > association
> > > > you
> > > > > need to keep in mind that the other end of the association (from the
> > one
> > > > > that's being modified) is not automatically updated so you need to
> > add
> > > > the
> > > > > logic to manage this yourself.
>
> > > > > For example, take the ubiquitous example of an Order class with an
> > > > > OrderLines collection which contains instances of the OrderLine
> > class,
> > > > which
> > > > > in turn has has an Order property.
>
> > > > > What happens when you create a new OrderLine object?
>
> > > > > If you set the Order property of the OrderLine instance and you've
> > mapped
> > > > a
> > > > > bidirectional association in your mapping where the collection end is
> > > > mapped
> > > > > as inverse so it's not in control of the association the data in the
> > > > > database will be correct and when you load your Order or OrderLine
> > both
> > > > ends
> > > > > of the association are correct.
>
> > > > > The problem, however is before you save and reload the changes using
> > > > > NHibernate - the association in memory is not automatically handled
> > so
> > > > the
> > > > > Order instance's OrderLines collection will not contain the new
> > OrderLine
> > > > > unless you explicitly add it yourself in your code.
>
> > > > > That's the primary reason I think most people avoid bidirectional
> > > > > associations - if it's appropriate for you to use then use it, just
> > be
> > > > aware
> > > > > of the extra work involved that NHibernate won't do for you.
>
> > > > > Cheers,
>
> > > > > Symon.
>
> > > > > On Tue, Oct 21, 2008 at 5:58 PM, Tim Barcz <[EMAIL PROTECTED]>
> > wrote:
> > > > > > I'm hearing and reading different things....is unidirectional or
> > > > > > bidirectional the recommended way.
>
> > > > > > On certain scenarios I don't need to go unidirectional but then
> > have a
> > > > hard
> > > > > > time mapping but then see people say unidirectional is bad.
>
> > > > > > Which is it?  Uni or Bi?
>
> > > > > --
> > > > > Symon Rottemhttp://blog.symbiotic-development.com
--~--~---------~--~----~------------~-------~--~----~
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
-~----------~----~----~----~------~----~------~--~---

Reply via email to