Schematron-based Semantic Constraints Specification Framework and Validation Rules Engine for JSON

Abstract

Business data is a critical asset for organizations. Data quality issues can creep in at multiple stages during the creation and processing of the business data. Syntax and semantic validation of business data at inception and during processing are of utmost importance to mitigate financial and other types of risks. Currently, XML is a predominant format for business data. XML has a mature syntax and semantic constraint specification standards and validation toolset. JSON is an emerging format for business data. It has a relatively mature syntax specification standard and a variety of validation tools but it lacks a common semantic/co-constraints specification standard and reusable validation toolset. This research developed a system to overcome these serious limitations. The system consists of a) an ISO/IEC 19757– 3 Schematron compliant framework to specify the semantic constraints in JSON itself and b) a JavaScript-based reusable and extensible validation component. The semantic constraints specification framework can be leveraged to specify arbitrarily complex constraints whereas the validation tool can be used as a standalone component or can be embedded in other data processing enterprise systems as a module. Together this system can be used as a test bed for further research in the area of semantic validation for JSON data. This paper will describe in detail the first component while briefly touching upon the second component.