SQL Server Column From Another Table With Static Value In View - sql-server

Question

Is there a way to have a column from another table with value which is always the same inside a View> Example:
SELECT *,
(SELECT value FROM tblStudentPrefixes WHERE PrefixName = 'SeniorPrefix')
AS StudentPrefix
FROM tblStudents
Will the above nested query get executed fro each row? Is there a way to execute it once and use for all rows.
Please note, I'm specifically talking about a View, not a Stored Procedure. I know this can be done in a Stored Procedure.

Answer 1

This actually depends on your table set up. Unless prefixName is constrained to be unique you could come across errors, where the subquery returns more than one row. If it is not constrained to be unique, but happens to be unique for SeniorPrefix then your query will be executed 1000 times. To demonstrate I have used the following DDL:
CREATE TABLE #tblStudents (ID INT IDENTITY(1, 1), Filler CHAR(100));
INSERT #tblStudents (Filler)
SELECT TOP 10000 NULL
FROM sys.all_objects a, sys.all_objects b;
CREATE TABLE #tblStudentPrefixes (Value VARCHAR(10), PrefixName VARCHAR(20));
INSERT #tblStudentPrefixes (Value, PrefixName) VALUES ('A Value', 'SeniorPrefix');
Running your query gives the following IO output:
Table '#tblStudentPrefixes'. Scan count 10000, logical reads 10000
Table '#tblStudents'. Scan count 1, logical reads 142
The key being the 1000 logical reads on tblStudentPrefixes. The other problem with it not being constrained to be unique is that if you have duplicates your query will fail with the error:
Subquery returned more than 1 value. This is not permitted when the subquery follows =, !=, <, <= , >, >= or when the subquery is used as an expression.
If you can't constrain PrefixName to be unique, then you can stop it executing for each row and avoid the errors by using TOP:
SELECT *,
(SELECT TOP 1 value FROM #tblStudentPrefixes WHERE PrefixName = 'SeniorPrefix' ORDER BY Value)
AS StudentPrefix
FROM #tblStudents
The IO now becomes:
Table '#tblStudentPrefixes'. Scan count 1, logical reads 1
Table '#tblStudents'. Scan count 1, logical reads 142
However, I would still recommend switching to a CROSS JOIN here:
SELECT s.*, p.Value AS StudentPrefix
FROM #tblStudents AS s
CROSS JOIN
( SELECT TOP 1 value
FROM #tblStudentPrefixes
WHERE PrefixName = 'SeniorPrefix'
ORDER BY Value
) AS p;
Inspection of the execution plans shows that a sub-select using a table spool which is very unnecessary for a single value:
So in summary, it depends on your table set up whether it will execute for each row, but regardless you are giving the optimiser a better chance if you switch to a cross join.
EDIT
In light of the fact that you need to return rows from tblstudent when there is no match for SeniorPrefix in tblStudentPrefixes, and that PrefixName is not currenty constrianed to be unique then the best solution is:
SELECT *,
(SELECT MAX(value) FROM #tblStudentPrefixes WHERE PrefixName = 'SeniorPrefix')
AS StudentPrefix
FROM #tblStudents;
If you do constrain it to be unique, then the following 3 queries produce (essentially) the same plan and the same results, it is simply personal preference:
SELECT *,
(SELECT value FROM #tblStudentPrefixes WHERE PrefixName = 'SeniorPrefix')
AS StudentPrefix
FROM #tblStudents;
SELECT s.*, p.Value AS StudentPrefix
FROM #tblStudents AS s
LEFT JOIN #tblStudentPrefixes AS p
ON p.PrefixName = 'SeniorPrefix';
SELECT s.*, p.Value AS StudentPrefix
FROM #tblStudents AS s
OUTER APPLY
( SELECT Value
FROM #tblStudentPrefixes
WHERE PrefixName = 'SeniorPrefix'
) AS p;

Answer 2

I hope I understand your question right, but try this
SELECT *
FROM tblStudents
Outer Apply
(
SELECT value
FROM tblStudentPrefixes
WHERE PrefixName = 'SeniorPrefix'
) as tble

Answer 3

This is OK. Subquery would be executed for every row on every row (which could provide bad performance).
You could try also:
SELECT tblStudents.*,StudentPrefix.value
FROM tblStudents,
(SELECT value
FROM tblStudentPrefixes
WHERE PrefixName = 'SeniorPrefix')StudentPrefix