Restrictions on Overloading

There are several restrictions on how you can overload programs. When the PL/SQL engine compiles and runs your program, it has to be able to distinguish between the different overloaded versions of a program; after all, it can't run two different modules at the same time. So when you compile your code, PL/SQL will reject any improperly overloaded modules. It cannot distinguish between the modules by their names, because by definition they are the same in all overloaded programs. Instead, PL/SQL uses the parameter lists of these sibling programs to determine which one to execute and/or the types of the programs (procedure versus function). As a result, the following restrictions apply to overloaded programs.

The datatype family of at least one of the parameters of overloaded programs must differ

INTEGER, REAL, DECIMAL, FLOAT, etc., are NUMBER subtypes. CHAR, VARCHAR2, and LONG are character subtypes. If the parameters differ only by datatype within the supertype or family of datatypes, PL/SQL does not have enough information to determine the appropriate program to execute.

Overloaded programs with parameter lists that differ only by name must be called using named notation

If you don't use the name of the argument, how can the compiler distinguish between calls to two overloaded programs? Please note, however, that it is always risky to use named notation as an enforcement paradigm. One should avoid situations where named notation yields different semantic meaning from positional notation.

The parameter list of overloaded programs must differ by more than parameter mode

Even if a parameter in one version is IN and that same parameter in another version is IN OUT, PL/SQL cannot tell the difference at the point in which the program is called.

Overloaded functions must differ by more than their return type (the datatype specified in the RETURN clause of the function)

At the time that the overloaded function is called, the compiler doesn't know what type of data that function will return. The compiler therefore cannot determine which version of the function to use if all the parameters are the same.

All of the overloaded programs must be defined within the same PL/SQL scope or block (anonymous block, standalone procedure or function, or package)

You cannot define one version in one block (scope level) and define another version in a different block. You cannot overload two standalone programs; one simply replaces the other.