Why is Setting a SQL Server Bit to "false" in a stored procedure definition valid?

Question

Bits are often used in SQL Server to represent true and false however you cannot literally assign true and false to them it must be 1 or 0.

I recently stumbled across the following code:

CREATE PROCEDURE TestProcedure
@parameter bit = false
AS
BEGIN
...
END

Why is it valid? Especially when setting a bit to false in the body of the stored procedure is not.

If you query the parameter with a select it is indeed set to 0 which represents false.

Answer 1

SQL Server has in built functionality to convert string values 'true' to 1 and 'false' to 0:

Run this code and see:

DECLARE @someval BIT = 'true'

SELECT @someval

SET @someval = 'false'

SELECT @someval

Reference:

bit (Transact-SQL)

The string values TRUE and FALSE can be converted to bit values: TRUE is converted to 1 and FALSE is converted to 0.

As @hvd states in comments, the reason the declaration works without quotes in your sample is based on quotes being optional in parameter assignments of a stored procedure declaration.

Reference:

Parameter Assignment to unquoted string

In a stored procedure definition, a parameter can be assigned a default string value that is an 'unquoted' string.

...this is very widely used legacy behavior that needs to be maintained for compatibility reasons. The same behavior can be seen when passing parameters in an exec statement.