Understand the structure of your RDF data at a glance using automatically built application profiles and spot differences between dataset structures.
An application profile, in this context, is the set of data shapes designed for a particular purpose acting as constraints on how the data are instantiated and so can be used to validate the data.
Fingerprinting is the action of generating, or rather, guessing, the application profile applied to a particular dataset. This is an inductive process of reconstructing the data shape for each class instantiated in the dataset.
RDF fingerprinter may be installed with pip as follows.
pip install rdf-fingerprinter
Note that Python version 3.8 or later is required.
The easiest way to build a fingerprint of a SPARQL endpoint is by calling the fingeprinting CLI command and write the report in an output folder.
fingerprint -e http://my.sparql.endpoint.com -o my/output/folderTo use the fingerprinter programmatically please refer to the API Reference.
You are more than welcome to help expand and mature this project. We adhere to Apache code of conduct, please follow it in all your interactions on the project.
When contributing to this repository, please first discuss the change you wish to make via issue, email, or any other method with the maintainers of this repository before making a change.
This project is licensed under Apache License 2.0.