erlasticsearch

Elasticsearch Erlang Client, OTP 19, rebar3

Overview

This module allows to perform different operations against a single elasticsearch cluster, one of the most interesting features is the bulk indexing that allows to push a number of documents to the cluster, also there is an scroll function that allows to scroll over a set of results from the cluster, that helps to get a big quantity of documents using an iterator, and process them as they are retrieved until the result set is done.

If environment_var atom is used as host, the module will look in the os:getenv(ENV_ELASTICSEARCH_HOST) for the hostname, if var is not yet defined it will default to localhost.

Configuration

%%-*- mode: erlang -*-
[
    {elasticsearch, [
      {host, "localhost"}, %% or environment_var atom
      {port, 9200},
      {ssl, false},
      {max_sessions, 10},
      {shards, 2},
      {replicas, 1},
      {max_pipeline_size, 1},
      {http_options, [{timeout, 1500000}]},
      {basic_auth, {"fake-user", "fake-pass"}}
    ]}
].
% vim:ft=erlang

Ownership

This project is supported by John Diaz, email me for any issues email

Please see project definition documentation

Run: make shell

Eshell V8.2  (abort with ^G)
(erl@local)1> elasticsearch:create_index("index-name").
(erl@local)1> elasticsearch:index_exists("index-name").

WARNING: This will create and delete a test-index Test

Eshell V8.2  (abort with ^G)
(erl@local)1> elasticsearch:test().

Function	Parameters	Description
create_index/1	"index-name"	Checks if the index exists, if not, creates it
create_index/2	"index-name", {Shards, Replicas}	Creates an index with settings
path_exists/1	"path"	checks a specific path in Elasticsearch "index-name/mapping-name/docId"
bulk/2	{Index, Type}, term()	uses bulk api to push documents into elasticsearch, see Bulk Api

An erlang client for elasticsearch's REST interface