## Transforming Databases with Recursive Data Structures

My PhD. thesis was completed in 1995 as the final requirement for the PhD. program
at the University of Pennsylvania Department of Computer and Information Science.

This thesis examines the problems of performing structural
transformations on databases involving complex data-structures and
object-identities, and proposes an approach to specifying and
implementing such transformations.

It starts by looking at various applications of such *database
transformations*, and at some of the more significant work in these
areas.
In particular it looks at work on transformations in the area of
*database integration*, which was one of the major motivating
areas for this work.
It also looks at various notions of correctness that have been
proposed for database transformations, and shows that the utility of
such notions is limited by the dependence of transformations on
certain implicit database constraints.
It draws attention to the limitations of existing work on
transformations,
and argues that there is a need for a more general formalism for
reasoning about database transformations and constraints.

It also argues that, in order to ensure that database
transformations are well-defined and meaningful, it is necessary to
understand the information capacity of the data-models being
transformed. To this end if provides a thorough analysis of the
information capacity of data-models supporting object identity, and
shows that this is dependent on the operations supported by a query
language for
comparing object identities.

A declarative language, *WOL*, based on Horn-clause logic, is introduced for
specifying database transformations and constraints. A
method of implementing transformations specified in this language is also proposed, by
manipulating their clauses into a *normal form* which can then be
translated into an underlying database programming language.

Finally a number of optimizations and techniques
necessary in order to build a practical implementation based on these
proposals are presented, together with a discussion of the results of some of the trials
that were carried out using a prototype of such a system.

See here
for the thesis.

My PhD thesis proposal, entitled *Types with Extents*
is also still available. See
here for
a summary and pointers.

Anthony Kosky (anthony@genelogic.com)