The correct syntax to invoke a user-defined function (UDF) in a SQL query is to use the function's name followed by parentheses containing the arguments. In this case, the function price requires two arguments, which are provided by the customerspend and customerunits columns. The AS clause is then used to alias the output of the function to the new column name customerprice. This syntax directly applies the defined function row by row to the specified columns in the customersummary table.

A: This syntax is invalid. PRICE is not a SQL operator; functions must be called with parentheses () enclosing their arguments.

B: This attempts to select a column named price, not call the function price with the required column arguments.

C: The function() wrapper is not a valid SQL construct. The UDF is called directly by its name without this extra keyword.

D: This is redundant. The UDF price is already defined to return a DOUBLE, so casting the result again is unnecessary.

1. Databricks SQL Language Reference

CREATE FUNCTION (SQL): The documentation provides examples of creating and then invoking a SQL UDF. The invocation syntax is shown as functionname(arguments). For instance

after creating a function

it is called within a SELECT statement like SELECT myudf(c1) FROM .... This supports the structure price(customerspend

customerunits).

Source: Databricks Documentation > SQL reference > Data definition language > CREATE FUNCTION (SQL).

2. Databricks SQL Language Reference

SELECT: The documentation for the SELECT statement specifies that the selectclause can contain expressions

which include function calls. The standard syntax for a function call is functionname([ ALL | DISTINCT ] expression [

...]). This confirms that price(customerspend

customerunits) is the correct expression.

Source: Databricks Documentation > SQL reference > Data manipulation language > SELECT.