Passage

Passage [1] provides correct and complete results for SPARQL queries, despite time quotas enforced by public SPARQL endpoints. Contrarily to state-of-the-art approaches [2, 3], Passage works solely within SPARQL boundaries: a query execution returns correct (possibly partial) results along with a SPARQL continuation query; executing the continuation query provides missing results, itself with another SPARQL continuation query; until termination.

Installation

git clone https://github.com/Chat-Wane/passage.git
cd passage

mvn clean package

Usage

• passage-server.jar provides a server running passage/raw/blazegraph as a SPARQL endpoint, it requires a few changes to the configuration file. --help for more information.
• passage.jar provides a one time execution without server. --help for more information.
• raw.jar provides a one time execution of a online sample-based query execution. --help for more information.
• passage-comunica provides a client to automate the continuation loop.

Datasets

Passage relies on a Blazegraph storage. To ingest your RDF data, please follow their instructions. Alternatively, we provide a few datasets already ingested for trials:

• FedShop200 for federations of SPARQL endpoints;
• WatDiv for an e-commerce dataset;
• WDBench (Part A) (Part B) for a sample of Wikidata.

References

[1] T.H.T. Pham, G. Montoya, B. Nédelec, H. Skaf-Molli, and P. Molli. Passage: Ensuring Completeness and Responsiveness of Public SPARQL Endpoints with SPARQL Continuation Queries. WWW ‘25: Proceedings of the ACM on Web Conference 2025.

[2] T. Minier, H. Skaf-Molli and P. Molli. SaGe: Web Preemption for Public SPARQL Query services. In Proceedings of the World Wide Web Conference (2019).

[3] R. Verborgh, M. Vander Sande, O. Hartig, J. Van Herwegen, L. De Vocht, B. De Meester, G. Haesendonck and P. Colpaert. Triple Pattern Fragments: A Low-Cost Knowledge Graph Interface for the Web. In Journal of Web Semantics (2016).