Random Function behavior in SELECT query in SQL Server - sql

Question

I have a written a random function dbo.UDF_Q_RandomNumber() that generates a floating type random number between 0 and 1.
DECLARE #upper = 10
DECLARE #lower = 1
SELECT
ROUND(CAST((#lower + (#upper - #lower) * dbo.UDF_Q_RandomNumber()) AS INT), 0)
The above code generates a random number between 1 and 10.
Now I created a temporary table #tempTable with 10 rows in it with columns Id and Number.
Id Number
1 1
2 2
3 3
4 4
5 5
6 6
7 7
8 8
9 9
10 10
SQL query:
CREATE TABLE #tempTable(Id INT, Number INT)
INSERT INTO #tempTable VALUES (1,1)
INSERT INTO #tempTable VALUES (2,2)
INSERT INTO #tempTable VALUES (3,3)
INSERT INTO #tempTable VALUES (4,4)
INSERT INTO #tempTable VALUES (5,5)
INSERT INTO #tempTable VALUES (6,6)
INSERT INTO #tempTable VALUES (7,7)
INSERT INTO #tempTable VALUES (8,8)
INSERT INTO #tempTable VALUES (9,9)
INSERT INTO #tempTable VALUES (10,10)
DECLARE #maxCount INT;
SELECT #maxCount= COUNT(1) FROM #tempTable
SELECT * FROM #tempTable
SELECT Number
FROM #tempTable
WHERE Id = ROUND(CAST((1+(#maxCount-1)*dbo.UDF_Q_RandomNumber())AS INT),0)
DROP TABLE #tempTable
Here the query
SELECT Number
FROM #tempTable
WHERE Id = ROUND(CAST((1+(#maxCount-1)*dbo.UDF_Q_RandomNumber()) AS INT), 0)
Sometimes it returns 2 rows and sometimes null which should not come as Id selected is between 1 and 10 (the rows in temptable) and every Id has value too.
Please help .

Answer 1

Without seeing the function, it's hard to say, but I suspect your function is non-deterministic, so it gets interpreted for each row of the table.
If you use rand() it only returns one row
SELECT Number FROM #tempTable WHERE Id= ROUND(CAST((1+(#maxCount-1)*rand())AS INT),0)
Alternatively, if you just want a random #n rows...
select top (#n) * from #tempTable order by newid()