Easy Web API Development with SPARQL Transformer
Intervenant⋅e⋅s
Résumé
How often, as a web developer, do you struggle with the JSON output of a SPARQL endpoint?
In a document-based world as the one of Web APIs, the triple-based output of SPARQL endpoints can be a barrier for developers who want to integrate Linked Data in their applications. As the query results represent all the valid solutions of aquery, it is possible that two bindings describe the same real-world object and differ only by a single field, appearing almost as duplicates.
In order to have a suitable structure for manipulation in any web framework, repetitive manual tasks are required, including skipping irrelevant metadata, reducing and parsing the RDF types, merging the rows referring to the same object, and mapping to a destination structure.
All these tasks are automatised in SPARQL Transformer, which relies on a single JSON object for defining which data should be extracted from the endpoint (query) and which shape should they assume (template). The library automatically merges the results on the base of identifiers, giving to the returned JSON the classic tree-based structure.
SPARQL Transformer is integrated into the grlc framework, which can build a web API from a set of queries stored in a GitHub repository, in order to create new bridges between the Web of Data and the Web of applications.
SPARQL Transformer is currently used in different projects and application. Some real-world use cases will be presented to demonstrate how easy it is to develop your Semantic Web application with SPARQL Transformer.
Références
Lisena P., Meroño-Peñuela A., Kuhn T. & Troncy R. Easy Web API Development with SPARQL Transformer. In 18th International Semantic Web Conference (ISWC), Auckland, New Zealand, October 26-30, 2019.
Lisena P. & Troncy R. Transforming the JSON Output of SPARQL Queries for Linked Data Clients. In The Web Conference 2018, Developer Track, Lyon, France, April 23-27, 2018.
Auteurs/Autrices
Pasquale Lisena is finishing his PhD (foreseen defense on 11th October) in the Data Science department at EURECOM, working on music representation and recommendation under the supervision of Raphaël Troncy. He got his Master in Media Engineering at Politecnico di Torino, and he has a previous work experience in the field of Web Interaction. From 2016 to 2018, he was part of the DOREMUS project, being in charge of the knowledge base population and recommendation tasks, contributing papers to conferences in the field like ISWC, EKAW and TheWebConf.