Seemingly equivalent queries in LINQ and SQL returns different results [duplicate]

Answer 1

Try to take a look on query that you run after LINQ execution in IntelliTrace window while you debugging this place. And then compare to your SQL, I think they are different. If you don't have IntelliTrace, try Express Profiler. Just set name of SQL Server, click Trace and spy it. I hope it will help!

Answer 2

This is a long shot but would this LINQ expression change anything?

(from uno in context.Unos
 join duo in context.Duos on uno.CommonId equals duo.CommonId
 where uno.Amount > 0 && duo.Type == 2
 select new {Uno = uno, Duo = duo}
).ToList()

I like this form better because it looks like the real SQL request. Somehow, the Join statement in your LINQ bothers me...

Answer 3

This is something that often happens when views are mapped in Entity Framework. Unlike regular database tables, views often don't have clearly unique key values. When EF encounters rows with identical primary key values, it just duplicates the row that belongs to the key it already knows. In joins, this may cause EF to produce more child records because this first-known row may have more children than the real database row.

The fix is to make sure that the view has a field (or combination of fields) that uniquely identifies a row. And, of course, in EF this should be mapped as a (possibly composite) primary key.

I don't know why EF displays this behavior. I think it should be possible for it to throw an exception that it materializes duplicate entities. This "feature" always causes a lot of confusion.