A new kind of data model has recently emerged in which the database is not constrained by a conventional schema. Systems like ACeDB, which has become very popular with biologists, and the recent Tsimmis proposal for data integration organize data in tree-like structures whose components can be used equally well to represent sets and tuples. Such structures allow great flexibility in data representation What query language is appropriate for such structures? Here we propose a simple language UnQL for querying data organized as a rooted, edge-labeled graph. In this model, relational data may be represented as fixed-depth trees, and on such trees UnQL is equivalent to the relational algebra. The novelty of UnQL consists in its programming constructs for arbitrarily deep data and for cyclic structures. While strictly more powerful than query languages with path expressions like XSQL, UnQL can still be efficiently evaluated. We describe new optimization techniques for the deep or "vertical" dimension of UnQL queries. Furthermore, we show that known optimization techniques for operators on flat relations apply to the "horizontal" dimension of UnQL.

One source of partial information in databases is the need to combine information from several databases. Even if each database is complete for some "world", the combined databases will not be, and answers to queries against such combined databases can only be approximated. In this paper we describe various situations in which a precise answer cannot be obtained for a query asked against multiple databases. Based on an analysis of these situations, we propose a classification of constructs that can be used to model approximations. A major goal is to obtain universality properties for these models of approximations. Universality properties suggest syntax for languages with approximations based on the operations which are naturally associated with them. We prove universality properties for most of the approximation constructs. Then we use them to design languages built around datatypes given by the approximation constructs. A straightforward approach results in langauges that have a numb...

We present a new principle for the development of database query languages that the primitive operations should be organized around types. Viewing a relational database as consisting of sets of records, this principle dictates that we should investigate separately operations for records and sets. There are two immediate advantages of this approach, which is partly inspired by basic ideas from category theory. First, it provides a language for structures in which record and set types may be freely combined: nested relations or complex objects. Second, the fundamental operations for sets are closely related to those for other "collection types" such as bags or lists, and this suggests how database languages may be uniformly extended to these new types. The most general operation on sets, that of structural recursion, is one in which not all programs are welldefined. In looking for limited forms of this operation that always give rise to well-defined operations, we find a number of close ...

Abstract not found

Data of interest to biomedical researchers associated with the Human Genome Project (HGP) is stored all over the world in a number of different electronic data formats and accessible through a varietyof interfaces and retrieval languages. These data sources include conventional relational databases with SQL interfaces, formatted text files on top of which indexing is provided for efficient retrieval (ASN.1-Entrez), and binary files that can be interpreted textually or graphically via special purpose interfaces (ACeDB). Researchers within the HGP wanttocombine data from these different data sources, add value through sophisticated data analysis techniques (such as the biosequence comparison software BLAST and FASTA), and view it using special purpose scientific visualization tools. However, currently there are no commercial tools for enabling such an integrated digital library, and a fundamental barrier to developing such tools appears to be one of language design and optimization: The data f...

Scientific data of importance to biologists in the Human Genome Project resides not only in conventional databases, but in structured files maintained in a number of different formats (e.g. ASN.1 and ACE) as well as sequence analysis packages (e.g. BLAST and FASTA). These formats and packages contain a number of data types not found in conventional databases, such as lists and variants, and may be deeply nested. We present in this paper techniques for querying and transforming such data, and illustrate their use in a prototype system developed in conjunction with the Human Genome Center for Chromosome 22. We also describe optimizations performed by the system, a crucial issue for bulk data. 1 Introduction The goal of the Human Genome Project (HGP) is to sequence the 24 distinct chromosomes comprising the human genome. Much of the information associated with the HGP resides not in conventional databases, but in files that have been formatted according to a variety of conventions. These...

We define a standard of effectiveness for a database calculus relative to a query language. Effectiveness judges suitability to serve as a processing framework for the query language, and comprises aspects of coverage, manipulability and efficient evaluation. We present the monoid calculus, and argue its effectiveness for object-oriented query languages, exemplified by OQL of ODMG-93. The monoid calculus readily captures such features as multiple collection types, aggregations, arbitrary composition of type constructors and nested query expressions. We also show how to extend the monoid calculus to deal with vectors and arrays in more expressive ways than current query languages do, and illustrate how it can handle identity and updates. 1 Introduction 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...

This paper argues that the research on semistructured data is receiving a new set of challenges with the advent of XML (Extensible Mark-up Language [Bos97, Con98]). This is a new standard approved by the World Wide Web Consortium that many believe will become the de facto data exchange format for the Web. XML supports the electronic exchange of machine-readable data (while HTML is designed primarily for human-readable documents). XML data shares many features of semistructured data: its structure can be irregular, is not always known ahead of time, and may change frequently and without notice. On the other hand it is easy to convert data from any source into XML which will make it attractive for organizations to "publish" their information sources in XML, and thus make them available to other XML applications on the Web. For XML applications to reach their full potential however, we need to build the right tools to process data in this new format. Existing Web tools (browsers, search engines) are oriented toward document operations . For XML we need database operations , like data extraction, data integration, data translation, data storage. The research done so far on semistructured data may offer some solutions to the database problems posed by XML. For example the recently proposed query language for XML, called XML-QL [DFF

Bags, i.e. sets with duplicates, are often used to implement relations in database systems. In this paper, we study the expressive power of algebras for manipulating bags. The algebra we present is a simple extension of the nested relation algebra. Our aim is to investigate how the use of bags in the language extends its expressive power, and increases its complexity. We consider two main issues, namely (i) the impact of the depth of bag nesting on the expressive power, and (ii) the complexity and the expressive power induced by the algebraic operations. We show that the bag algebra is more expressive than the nested relation algebra (at all levels of nesting), and that the difference may be subtle. We establish a hierarchy based on the structure of algebra expressions. This hierarchy is shown to be highly related to the properties of the powerset operator. Invited to a special issue of the Journal of Computer and System Sciences selected from ACM Princ. of Database Systems,...

Strong normalization results are obtained for a general language for collection types. An induced normal form for sets and bags is then used to show that the class of functions whose input has height (that is, the maximal depth of nestings of sets/bags/lists in the complex object) at most i and output has height at most o definable in a nested relational query language without powerset operator is independent of the height of intermediate expressions used. Our proof holds regardless of whether the language is used for querying sets, bags, or lists, even in the presence of variant types. Moreover, the normal forms are useful in a general approach to query optimization. Paredaens and Van Gucht proved a similar result for the special case when i = o = 1. Their result is complemented by Hull and Su who demonstrated the failure of independence when powerset operator is present and i = o = 1. The theorem of Hull and Su was generalized to all i and o by Grumbach and Vianu. Our result genera...