15. Naming Conventions

In many cases MPI names for C functions are of the form Class_action_subset. This convention originated with MPI-1. Since MPI-2 an attempt has been made to standardize the names of MPI functions according to the following rules. The C++ bindings in particular follow these rules (see Section C++ Binding Issues on page C++ Binding Issues ).

1. In C, all routines associated with a particular type of MPI object should be of the form Class_action_subset or, if no subset exists, of the form Class_action. In Fortran, all routines associated with a particular type of MPI object should be of the form CLASS_ACTION_SUBSET or, if no subset exists, of the form CLASS_ACTION. For C and Fortran we use the C++ terminology to define the Class. In C++, the routine is a method on Class and is named MPI::Class::Action_subset. If the routine is associated with a certain class, but does not make sense as an object method, it is a static member function of the class.


2. If the routine is not associated with a class, the name should be of the form Action_subset in C and ACTION_SUBSET in Fortran, and in C++ should be scoped in the MPI namespace, MPI::Action_subset.


3. The names of certain actions have been standardized. In particular, Create creates a new object, Get retrieves information about an object, Set sets this information, Delete deletes information, Is asks whether or not an object has a certain property.

MPI identifiers are limited to 30 characters (31 with the profiling interface). This is done to avoid exceeding the limit on some compilation systems.