Data Abstraction in a Language of Multilevel Computations Based on Pattern Matching
Igor B. Shchenkov
Abstract. It is well-known that data abstraction badly coexists with pattern matching. Pattern matching is applicable in the domain with "transparent" structure of symbolical data while the traditional data abstraction supposes hiding data from the user. In this paper, it is shown how it is possible to provide data abstraction at reception of all completeness of convenience of it by simple means within the framework of the system of symbolic manipulations based on pattern matching. At that the problem is not put to hide abstract data from the programmer completely.
In our case data abstraction is based on possibility of obtaining of values of functions, being inversed to functions-constructors, that in turn is based on possibility of dynamic computation of patterns. At that multilevel computations are widely used, providing, in particular, used way of pattern matching.
The examples of data abstraction given in this paper are based on the subsystem of algebraic computations that has been built in the system of symbolic manipulations.