USE OF ADAPTER DESIGN PATTERN FOR THE SCREENING OF DIFFERENCES IN API DATABASE

Actual ERP systems use for data storage mainly relational databases, but in specific situations, particularly for modeling complex relationships can also use the object databases. These information systems are usually developed using object-oriented programming languages like C++, Java, C #, and it would seem natural that for the data storage they also use the object-oriented databases instead of the relational databases, thus dispensing with the need for mapping objects into relations. But this is not true especially because of two reasons. The first reason is historical and the resulting enormous widespread of the relational databases, which were here before the object-oriented databases. The second reason for minority use of the object-oriented databases is the minimum compliance of standards by the object DBMS (Database Management System) producers, such as SQL standard in relational databases, resulting in differences in the API (Application Programming Interface) of object-oriented DBMS and the impossibility to change the object-oriented DBMS in an application without having to rewrite the code of an entire application which is built on that object-oriented database. The following paper attempts to resolve this issue.