SCULPT: A Schema Language for Tabular Data on the Web

TL;DR

SCULPT is a formal schema language designed for defining and constraining the structure of tabular web data, enabling efficient validation and streaming evaluation, with potential extensions for richer semantics and transformations.

Contribution

The paper introduces SCULPT, a novel schema language with formal semantics and efficient algorithms for tabular data validation and streaming, including region selection expressions.

Findings

Linear time evaluation algorithm for SCULPT

Support for weak and strong streaming evaluation

Extensions for types and complex content

Abstract

Inspired by the recent working effort towards a recommendation by the World Wide Web Consortium (W3C) for tabular data and metadata on the Web, we present in this paper a concept for a schema language for tabular web data called SCULPT. The language consists of rules constraining and defining the structure of regions in the table. These regions are defined through the novel formalism of region selection expressions. We present a formal model for SCULPT and obtain a linear time combined complexity evaluation algorithm. In addition, we consider weak and strong streaming evaluation for SCULPT and present a fragment for each of these streaming variants. Finally, we discuss several extensions of SCULPT including alternative semantics, types, complex content, and explore region selection expressions as a basis for a transformation language.