问题
After a bit of searching and reading the documentation, it's clear that you can write user defined functions in SQL Server that are marked as either deterministic or nondeterministic depending on which built-infunctions are used within the body.
RAND() is listed under the nondeterministic functions (see msdn article). So why can't I use it in a function?
回答1:
Because it has side effects.
Constructs with side effects are not allowed in a function. The side effect that it has is to change some internal state that keeps track of the last rand()
value issued.
I think you can get around it by including it in a View definition then selecting from the View.
回答2:
Using a View might work for you.
From Returning Random Numbers from a select statement
CREATE VIEW vRandNumber
AS
SELECT RAND() as RandNumber
The view is necessary because, as you already found out, a UDF cannot use the rand() function because that would make the function non-determistic. You can trick the UDF to accept a random number by using a View.
CREATE FUNCTION RandNumber()
RETURNS float
AS
BEGIN
RETURN (SELECT RandNumber FROM vRandNumber)
END
Finally, you can use this function in any SELECT to now return a random number between 0 and 1 per row:
SELECT dbo.RandNumber(), *
FROM Northwind..Customers
回答3:
I found this solution that doesn't create a view:
Basically:
Instead of
SET @R = Rand()
Use
SET @R = ABS(CHECKSUM(PWDENCRYPT(N''))) / 2147483647.0
In my case I wanted a number between 1 and 10:
ROUND(((10 - 1 -1) * ( ABS(CHECKSUM(PWDENCRYPT(N''))) / 2147483647.0) + 1), 0))
ROUND(((@max - @lower -1) * ( ABS(CHECKSUM(PWDENCRYPT(N''))) / 2147483647.0) + @lower), 0))
If you want a full explanation: Using (Or Simulating) Rand() In A T-Sql User-Defined Function
来源:https://stackoverflow.com/questions/3531450/creating-nondeterministic-functions-in-sql-server-using-rand