No abstract

An iterator is proposed that allows to apply sequences of update operations in a set-oriented way with deterministic semantics. Because the mechanism is independent of a particular model, it can be used i n t h e relational and in object-oriented ones. Thus, the deterministic semantics of embedded SQL cursors, and of triggers that are a p p l i e d after (set-oriented) SQL updates can be c h e cked. Furthermore, the iterator can be used to apply object-oriented methods, which are usually update sequences de ned on a single object, also to sets in a deterministic way. It turns out that the criteria that guarantee determinism are a l s o u s e d in semantic or multi-level concurrency control.

One of the objectives of ooDBMSs is to use type-speci c methods for manipulating objects, in order to maintain the consistency of the database. This is, however, little help for the method implementor as far as the model-inherent constraints are concerned. We propose a set of generic update operations that maintain integrity constraints such a s t ypes, class memberships, subtype-, subclass-relationships, and class predicates. The operations can be used for implementing type-speci c update methods or directly by applications. We present an approach to consistently de ne update semantics for an object model including classes, views, and variables that is based on necessary and su cient predicates akin to de ned concepts in KL-ONE style languages.

No abstract

The language BCOOL is formally de ned using a denotational semantics approach. BCOOL is a functional object database language with a very exible, yet strong and statically checked, type system. Its main source of exibility is its support for object evolution, that is, dynamic type changes of existing objects. Originally, B C O O L w as used as a formal basis for a more traditional (relational algebra-style) database language, COOL. In this paper, though, BCOOL is presented on its own. The purpose being to compare with other functional languages and discuss the virtues and limitations that BCOOL and these functional languages have w.r.t. each other in terms of (i) the above-mentioned exibility in the type system, which w e consider essential for objects and (ii) the orthogonality of the language.