一种结构化的编译型 PROLOG数据库及其管理策略

PROLOG数据库存放着构成PROLOG程序的子句以及表示成子句的全局数据,它的组织与管理是实用PROLOG系统的关键,这一点在提高执行效率的编译型PROLOG系统中尤为如此。本文以YH-SIM——一个扩充了的WAM模型为基础,提出一种新颖的编译型PROLOG数据库结构形式,其中·过程中的子句索引与子句彼此分离; ·子句索引块分成多个专门的子索引块; ·一个子句按两种形式存放,子句代码和子句项(源子句形式)。二者分离且共享同一索引,因此代码库和项库的管理是一体化的。这些结构化的特征,有力地支持PROLOG数据库的操作及管理(子句的插入删除,以及空间分配与回收)。文中还介绍了一种简明有效的管理方法。

A Structured Organization for the Compiler-based PROLOG Database and Its Management

In PROLOG database stored are clauses comprising PROLOG procedures andother long-term data structures represented as clauses. The implementationtechniques for the manipulation of clauses are very important for PROLOG'Sbeing a powerful AI language. It is even more the case when considering thestorage of compiled clauses. With the supports of YH-SIM architecture, wepropose a new structured organization for compiler-based PROLOG database,and present a simple and efficient management strategy that a procedure is theunit for claiming, and a clause the unit for dooming. Also different schemesfor clause representation in the database are investigated briefly.