unimodel-elasticsearch

Unimodel library for ElasticSearch.

Installation

$ npm install --save unimodel-elasticsearch

Basic Usage

In this section, we will walk through basic usage for the library.

Initiate the default connection to elasticsearch:

let es = require('unimodel-elasticsearch');
es.connect('http://localhost:9200', {
  'warehouse_*': { // index settings for any index matching warehouse_*
    shards: 16,
    replicas: 4
  }
});

Create an ElasticsearchModel:

let Animal = es.createModel(
  'Animal', // type name
  { // common-schema specification
    animalId: { type: String, index: true, id: true },
    name: { type: String, index: true }
  },
  'warehouse_animals' // index to store this type in
);

Register the ElasticsearchModel with the default model registry:

es.model('Animal', Animal);

Use the model registry for CRUD operations on the model:

let Animal = es.model('Animal');
let animal = Animal.create({ animalId: 'dog-charles-barkley', name: 'Charles Barkley' });
animal.save().then(() => {/* after save! */});

Components

For more information on each of these components, see the generated docs.

ElasticsearchConnection

Connection is a wrapper around an Elasticsearch Client, which ensures a connection is established before allowing operations against the Elasticsearch server.

ElasticsearchIndexManager

The Index Manager is responsible for storing the index configuration, and instantiating new Indexes when they are requrested by a Model.
An ElasticsearchIndexManager should not be directly instantiated, but instead should be created along with a connection.

ElasticsearchIndex

An ElasticsearchIndex is responsible for creating, updating, and storing information about an ElasticSearch Index.
An ElasticsearchIndex should nto be directly instantiated, but should instead be created by an ElasticsearchIndexManager.

ElasticsearchModel

An ElasticserachModel is analogous to a Type in ElasticSearch's langauge. It is responsible for registering a mapping with the ElasticsearchIndex, and creating/performing bulk operations on ElasticsearchDocuments.

ElasticsearchDocument

An ElasticsearchDocument directly cooresponds to a Document in ElasticSearch. It is responsible for saving and removing itself from ElasticSearch.

Testing

You must have a running elasticsearch instance running on http://localhost:9200 to run tests.
Once you have that, simply run:

$ npm test