HP 3000 Manuals HP 3000 Manuals
Function Parameters Some of the functions require one or more parameters. Function parameters can be identifiers, arithmetic expressions, or literals. See the description of each function for specific information about its parameters. If a function requires parameters and you do not supply any, or if the parameters supplied do not comply with all restrictions for parameters to that function, the value returned by the function is undefined. Using ALL as a Table Subscript Some functions allow a variable number of arguments, for example MAX, MEAN, and SUM. You can pass all the elements of a table to one of these functions by specifying the table as a parameter with ALL as the table subscript. See the examples under these functions. See also "Referencing Table Items with Subscripting" in Chapter 4 . In multi-dimensional tables, ALL may occur in one or more subscripts. Precision of Numeric Functions Some of the numeric functions convert the parameters you pass to intermediate floating point values to calculate the function result. The precision of these functions is limited to 15 significant digits. Also, fractional values may have rounding errors even if the total size of the argument is less than or equal to 15 digits. For example, using the SUM function on a table of dollars and cents values, such as PIC S9(9)V99, might not produce the correct answer. Use of the ROUNDED phrase is recommended. See the HP COBOL II/XL Programmer's Guide for more information on performance and the precision of functions.