scispace - formally typeset
Search or ask a question

Showing papers by "Alessandro D'Atri published in 1983"


Proceedings ArticleDOI
01 May 1983
TL;DR: A formal treatment of unambiguous contexts is presented and it is shown that they provide a suitable basis for discussing the universal relation assumption, the relationship uniqueness assumption, maximal objects, and other related concepts.
Abstract: This paper concerns query answering in relational databases. We assume a universe U of attributes and a set of values associated with each attribute. A database scheme is a given collection R = {1, 2, ..., n} of subsets of U, called relation schemes. A query in R is any subset of U. We call "context" any joinable subset of R. An unambiguous context is one in which a query receives the same answer independently of the subcontext used for the computation. A formal treatment of unambiguous contexts is presented and it is shown that they provide a suitable basis for discussing the universal relation assumption, the relationship uniqueness assumption, maximal objects, and other related concepts.

12 citations


Journal ArticleDOI
TL;DR: This paper investigates, as refînement restrictions, boundings in renamings of objects and prove several results concerning meaningful properties in the relational theory o f data.
Abstract: — The design of a data base schema can be organized through a set of refînement steps, that gradually introducé progressive details in the description of reality. In this approach, suitable restrictions to refînement rules allow to guarantee both conceptual and logical properties. Following this approach the formalism of hypergraphs and hypergraph grammars can be applied to investigate various aspects of data base schema design, in the framework ofthe relational model of data. In this paper we investigate, as refînement restrictions, boundings in renamings of objects and prove several results concerning meaningful properties in the relational theory o f data. Résumé. — La conception du schéma d'une base de données peut se structurer en étapes introduisant progressivement les détails de la description du monde réel. Dans ce travail, nous proposons des contraintes sur les étapes pour garantir certaines propriétés au niveau conceptuel et logique. Nous utilisons des hypergraphes et des grammaires d'hyper graphes pour étudier plusieurs aspects de la conception du schéma dune base, dans le cadre du modèle relationnel. Nous considérons des contraintes liées au changement de noms dobjects, et nous présentons des résultats liés à des propriétés importantes de la théorie relationnelle.

4 citations