Browse Prior Art Database

Web Enabled SPARQL and SQL Query Generation Based on Common JSON Model

IP.com Disclosure Number: IPCOM000239663D
Publication Date: 2014-Nov-24
Document File: 5 page(s) / 118K

Publishing Venue

The IP.com Prior Art Database

Abstract

Disclosed is a query generation model, based on the JavaScript Object Notation (JSON) model, which supports visual creation of a query for enterprise data. This allows users to seamlessly switch between data sources, without having to know the complex query languages.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 42% of the total text.

Page 01 of 5

Web Enabled SPARQL and SQL Query Generation Based on Common JSON Model

An enterprise business needs to be able to report on the data it collects . This data can come from many different sources and be stored in different ways using different technologies. A relational database might store some of the data in tables accessible by a structured query language (SQL). A linked data strategy might be used with a triple store repository, which can be queried using SPARQL Protocol and Resource Description Framework (SPARQL). Each of these strategies comprises a different data source that requires reporting.

It is preferable for application users trying to create queries to retrieve data from those data stores to not be affected by the implementation specifics. Existing solutions require users to use different tools to create queries against relational database and triple store. In addition, users are sometimes required to know the complex textual query language, as the visual query builder is either not available or does not support all needed aspects of query creation. Some approaches allow query creation in one technology (i.e. SPARQL) and then use an intermediary driver to translate the SPARQL query to SQL before sending it for execution to the relational database. The first approach is inconvenient and second results in SQL queries that do not perform

well.

A method is needed that allows users to seamlessly switch between data sources, without having to know the complex query languages.

The novel solution addresses the query generation part of those problems and supports a web -based query builder user interface.

Figure1: Improving query creation across different data sources

1


Page 02 of 5

The solution supports visual creation of the query. Visual query presentation is converted to the JavaScript Object Notation (JSON) model, which is passed to the server for query generation and storing. The same visual presentation can be used for various kinds of data sources, as the JSON model does not contain any low level details about the data; it just identifies the resources that are involved in the query. This identification is possible as the information about the resources in each data store is loaded into internal meta models such that all have the same Application Programming Interface (API). As the user selects the resource in the user interface, the identifier of that resource is recorded in the JSON model. That identifier later helps the user find the information about that resource in the meta model, the information that is needed for query generation. (Figure 2)

Figure 2: High-level design

Information about the data structure in data stores is loaded into a common in -memory presentation, the meta model. The meta model of a data store consists of meta types defined for that data store and meta properties defined on those meta types . Specifically, for relational database, this means that the database schema information about the t...