jsonld-to-elasticsearch

A command-line tool that loads line-separated JSON-LD documents into Elasticsearch.

Usage

Compile using Leiningen:

lein uberjar

Run from the command-line:

java -jar jsonld_to_elasticsearch.jar --help

You will need to provide the tool with several parameters. Provide configuration as an EDN file that contains the following keys:

• endpoint: A HTTP URL of the Elasticsearch REST API
• index: Name of the Elasticsearch index to use.
• batch-size (optional, default = 5000): Number of documents to index in one batch.

Besides the configuration, you will need to provide an Elasticsearch mapping for the documents you want to index. The mapping must contain exactly one mapping type, which will be used as the type of the indexed documents.

The input JSON-LD documents can be read from a file or by default from the standard input. The tool expects the JSON-LD provided as Newline Delimited JSON, with each document on a single line. @context and @id attributes are removed from each document, while @id is used as the Elasticsearch _id field. You can prepare JSON-LD documents in the expected format using sparql-to-jsonld.

License

Copyright © 2016 Jindřich Mynarz

Distributed under the Eclipse Public License version 1.0.