当参数有时可以为NULL时，如何参数化SQL查询？

Question

With pyodbc I can parametrize the query like this;

value = "testval"

query = \
    """
    SELECT *
    FROM TestTable
    WHERE Column = ?;
    """

cursor.execute(query, value)

But the problem is that if the value is None, the query should look like this;

value = None

query = \
    """
    SELECT *
    FROM TestTable
    WHERE Column IS NULL;
    """

cursor.execute(query)

So how should the query look like when the value can either be None or a string;

value = get_value()  # can return a string or None

query = \
    """
    SELECT *
    FROM TestTable
    WHERE Column ???????????
    """

cursor.execute(query, value)

Answer 1

The solution is to use the ISO/ANSI standard NULL -safe comparison:

WHERE Column IS NOT DISTINCT FROM ?

Not all databases support this, so you can also use:

WHERE Column = ? OR (Column IS NULL AND ? IS NULL)

If you are reluctant to pass the parameter twice, you can include it in the FROM clause:

. . .
FROM . . . CROSS JOIN
     (SELECT ? as compColumn) params
WHERE (Column = params.compColumn0 or (Column IS NULL and params.compColumn IS NULL)