Services

This page lists available reconciliation services.

Servers

Beyond publicly available hosted services, some software exposes reconciliation endpoints locally.

csv-reconcile

csv-reconcile is a Python based project can be used to run a reconciliation service off any CSV file. It supports version 0.2 of the protocol.

reconcile-csv

reconcile-csv is a Clojure based project can be used to run a reconciliation service off any CSV file. It supports version 0.1 of the protocol.

OpenRefine’s RDF extension

The RDF extension can be used to wrap a SPARQL endpoint into a reconciliation service. It supports version 0.1 of the protocol.

datasette-reconcile

datasette-reconcile is a plugin for Datasette to expose a reconciliation endpoint from a table stored in a Datasette instance. It supports version 0.2 of the protocol.

Hosted services

A list of hosted services is maintained on Wikidata and can be viewed in our reconciliation test bench.

Other reconciliation services (including discontinued services) can be found on OpenRefine’s wiki.

How to add a service to the test bench

Requirements: Make sure your service is publicly accessible - local endpoints (such as those with localhost or 127.0.0.1 in their address) should not be added.

If your service fulfills these requirements, let us know about it by opening an issue or add the necessary information to Wikidata by yourself like in this example.

Steps:

Find out if there is already a Wikidata item about your service, your database or your identifier. You can use Wikidata’s search interface for that. If there is not, you can create a new one.
Add a statement on that item, using the API endpoint (P6269) property, adding the OpenRefine reconciliation service API (Q64490175) as protocol (P2700) in a qualifier.
Optionally add a link to the documentation for the endpoint by adding a described at URL (P973) or user manual link (P2078) qualifier.
Optionally add a link to the source code for the endpoint by adding a source code repository (P1324) qualifier.