An Effective Framework for Processing OODB Languages

关系数据模型的一个备受吹捧的优点是存在一个形式演算和代数来建模数据库查询。在实践中，这些形式主义未能对商业查询语言中存在的许多特征（例如，SQL）：分组、聚合、重复值和排序顺序。差距只会随着人们转向面向对象的查询语言而扩大，例如ODMG-93的OQL，它必须处理多个批量（集合）类型、类型构造函数的任意嵌套、方法和嵌入式查询表达式。这些特性也出现在关系查询语言标准提案中，比如SQL 3。这项工作解决了现代数据库语言的高效查询处理。它试图通过原型实现来证明，如果新的语言功能在查询演算中得到直接支持，则可以更有效地处理这些功能。研究框架是基于这样的演算，称为幺半群理解演算。我们的形式化模型不是从关系代数开始进行扩展，而是自然地表示数据库语言的新特性。该研究的成功将为商用数据库系统的查询处理器的实现打下良好的基础，无论是纯面向对象模型的查询处理器还是关系对象的查询处理器

A much-touted advantage of the relational data model is the existence of a formal calculus and algebra to model database queries. In practice, these formalisms fail to model many of the features present in commercial query languages (e.g., SQL): grouping, aggregation, duplicate values, and sort orders. The gap only widens as one moves to object- oriented query languages, such as OQL of ODMG-93, which must deal with multiple bulk (collection) types, arbitrary nesting of type constructors, methods and embedded query expressions. Such features also appear in relational query language standards proposals, such as SQL3. This work addresses efficient query processing for modern database languages. It seeks to demonstrate through prototype implementation that new language features can be handled more effectively if these features have direct support in a query calculus. The research framework is based on such a calculus, called the monoid comprehension calculus. Instead of starting from relational algebra and making extensions, our formal model represents naturally the new emerging features of database lang uages. A successful outcome for the research will lead to a better basis for implementing the query processors of commercial database systems, both of pure object-oriented models and relational-object