Fluent NHibernate AutoMapping Override ignoring column name

Question

Two tables:

CREATE TABLE [dbo].[Error](
[ErrorId] [int] IDENTITY(1,1) NOT NULL,
[ResponseId] [int] NOT NULL,
<Other fields>

 CONSTRAINT [PK_Error] PRIMARY KEY CLUSTERED 
(
    [ErrorId] ASC
)

CREATE TABLE [dbo].[Response](
[ResponseId] [int] IDENTITY(1,1) NOT NULL,
<other fields>
 CONSTRAINT [PK_Response] PRIMARY KEY CLUSTERED 
(
    [ResponseId] ASC
)

And the classes

public partial class ErrorType
{
    public virtual long Id { get; set; }
    public virtual hr_information_type Response { get; set; }
<other fields>
}

public class hr_information_type 
{
    public virtual long Id { get; set; }
    public virtual ErrorType[] Errors { get; set;}
<other fields>
}

I'm using Auto mapping, and overriding it thusly:

public class hr_information_typeMap : IAutoMappingOverride<hr_information_type>
{
    public void Override(AutoMapping<hr_information_type> mapping)
    {
        mapping.Table("Response");
        mapping.Id(x => x.Id).Column("ResponseId");
        mapping.HasMany(many => many.Errors).AsArray(a => a.Response, x => x.Column("ResponseId"));         
    }
}

public class ErrorTypeMap : IAutoMappingOverride<ErrorType>
{
    public void Override(AutoMapping<ErrorType> mapping)
    {
        mapping.Table("Error");
        mapping.Id(id => id.Id).Column("ErrorId");
        mapping.References(r => r.Response, "ResponseId");
    }
}

The problem I have is, when I call Session.Load(id), I get the following error

"could not initialize a collection: " with an Inner Exception "Invalid column name 'hr_information_type_id'"

because it generates the SQL as

SELECT 
errors0_.hr_information_type_id as hr7_1_, 
errors0_.ErrorId as ErrorId1_, 
errors0_.ResponseId as ResponseId1_, 
errors0_.ErrorId as ErrorId26_0_,
errors0_.ResponseId as ResponseId26_0_ 
FROM Error errors0_ 
WHERE errors0_.hr_information_type_id=?

I have no idea why it's looking for a column I never said was there, nor why it's looking for any columns twice.

What am I doing wrong? I have very similar code in other projects that are not using Auto Mapping, so is this not the correct way to do the override?

score 0 · Answer 1 · answered Apr 14 '14 at 19:30

mapping.HasMany(many => many.Errors).AsArray(a => a.Response, part => part.Column("ErrorId")).KeyColumn("ResponseId").Cascade.AllDeleteOrphan();

It needs an Index and a Key. I looked at the HBM files it generated ( using

.Mappings(m =>
{
    m.FluentMappings.AddFromAssembly(assembly).ExportTo(@"C:\nh.out");
    m.AutoMappings.Add(am).ExportTo(@"C:\nh.out");
}

)

and saw that. The Index is the part.Column("ErrorId") piece, which is the way it indexes into the collection (even though it's just an array). This has gotten me farther, but not 100% sure its the complete answer.

Fluent NHibernate AutoMapping Override ignoring column name

1 Answers1