Question

We have a stored procedure that has a select statement in it:

select convert(int, c.ID) as ID,
       convert(nvarchar(255), c.name) as name,
       convert(varchar(32), a.state) as state
from   customer c join address a on
       c.addressid = a.ID
where  c.name like @custNameSpec

This produces two records when executed in a T-SQL window:

ID  Name    State
1   Robert  PA
2   Rob     VA

When executed in Entity Framework 4 as a function import, it returns two records, but the first record is duplicated:

ID  Name    State
1   Robert  PA
1   Robert  PA

We deleted the function import and the imported function, recreated it, etc. We also added those SQL convert() statements above to guarantee Entity Framework understands the data types coming back from the server.

What could we do to fix it? What causes duplicates like that?

Our tests include:

var myresult3 = myUOW.DC.GetAdDir(todaysdate: null, store_nbr: 14, 
                                  adtype: null).ToList();
var myresult4 = DB.GetAdDir(todaysdate: null, store_nbr: 14, 
                            adtype: null).ToList();

Both return the same incorrect result. The SQL profiler shows this call:

exec [dbo].[GetCust] @todaysdate=NULL,@custNameSpec='Rob',@adtype=NULL

EDIT:

Apparently, the business rules changed. The POCO generated from Entity Framework had a primary key improperly set, so it returned the correct quantity of fields, but “removed” duplicates by making all duplicates the same (based on the POCO primary key fields.)

There was MergeOption referenced in other remotely related questions that may explain why this happens.

Answer 1

When you create a unique restriction on an Entity, like a primary key, and you have duplicates coming from the data, Entity Framework will repeat the first duplicate record for all fields for each duplicate found.

In other words, Entity Framework returns a mangled mess if your underlying data doesn’t reflect the primary key correctly.