Good practice for using "NOT IN" in a LINQ query - asp.net-mvc-3

Question

I have converted the following SQL query to LINQ with the exception of the "NOT IN" subquery.
What would be the most effective way to implement this using LINQ? Should I use a join instead?
If anybody is able to provide an example or some guidance, I'd appreciate it.
New LINQ Query:
return from objOpenCalls in db.OpenItemss
from objTasks in db.Tasks
.Where(t => (t.Task_ID == objOpenCalls.Parent_Task_ID))
where ((objTasks.Item_ID > 0) && (objTasks.Type_ID > 0) && (objTasks.OwnerTypeItem_ID == user) && (objOpenCalls.CallEnd < DateTime.Now))
orderby objOpenCalls.CallStart descending
select new CallMiniViewModel
{
ID = objOpenCalls.ID,
CallStart = objOpenCalls.CallStart,
Name = objTasks.Task_Title
};
Old SQL Query:
SELECT TOP (100) ta.ID, t.Task_Title, ta.CallStart
FROM OpenItems AS ta INNER JOIN
Tasks AS t ON ta.Parent_Task_ID = t.Task_ID
WHERE
(t.Item_ID > 0) AND (t.[Type_ID] > 0) AND (ta.CallStart > DATEADD(m, -6, GETDATE()))
AND (ta.ID NOT IN (SELECT CallId FROM CallFeedback)) AND (t.OwnerTypeItem_ID = #Username) AND (ta.CallEnd < GETDATE())
ORDER BY ta.CallStart DESC

Answer 1

There are a couple of ways of doing the not in. Below is just a quick sample put in LinqPad as a test.
class MyClass {
public int Id {get;set;}
}
void Main()
{
int[] myItems = new[] { 1, 2, 3, 4, 5, 6 };
IEnumerable<MyClass> classes = new []{
new MyClass { Id = 3 },
new MyClass { Id = 6 },
new MyClass { Id = 8 }
};
var results = from cl in classes
where !myItems.Contains( cl.Id )
select cl;
foreach(var result in results) {
Console.WriteLine( "Class {0}", result.Id);
}
var results2 = from cl in classes
where (
from i in myItems
where i == cl.Id
select i ).Count( ) == 0
select cl;
foreach(var result in results2) {
Console.WriteLine( "Class {0}", result.Id);
}
}
I normally play with code first in LinqPad as it helps me understand any problems, and it will (if you're working with SQL) then show you what SQL the query will generate and you can fine tune a little. Sometimes it takes a little time to get your stuff able to be run in there, but it's worth it with the more complex queries.