README

EsLoader is a lighweight PHP package for indexing data from multiple sources into Elasticsearch index.

Table of Contents

Features

• Simple and Easy 3 step integration.
• Supports CSV, XML and JSON files.
• Support for MongoDb and MySQL coming soon in next release.
• Indexes data quickly by leveraging the bulk indexing feature of Elasticsearch.
• Supports integration of AWS Elasticsearch Service.
• Fully configurable - Define your own index name, custom mapping and settings and even set the size of each batch request.

NOTE 1: This package is developed and tested on PHP 7.x and Elasticsearch 6.x.

NOTE 2: This package uses _doc for the _type meta field of Elasticsearch document. This setting is not configurable. To know more read : Removal of mapping types.

Installation

$ composer require neelkanthk/esloader

Usage

require __DIR__ . '/vendor/autoload.php';

use Neelkanthk\EsLoader\Core\EsLoader;

//1. Specify the path of file to be indexed.
$filePath = __DIR__ . "/data.csv";
//2. Load array of configurations.
$config = [
    "index" => "esloader",
    "doc_id_key" => NULL,
    "connection" => "local",
    "local" => [
        'host' => "localhost",
        'port' => "9200"
    ],
    "aws" => [
        'host' => "",
        'region' => "",
        'access_key' => "",
        'secret_key' => ""
    ],
    "mappings" => [
        "_doc" => [
            '_source' => [
                'enabled' => true
            ],
            "properties" => [
                "id" => ['type' => 'keyword'],
                "first_name" => ['type' => 'text'],
                "last_name" => ['type' => 'text'],
                "email" => ['type' => 'keyword'],
                "gender" => ['type' => 'keyword'],
                "points" => ['type' => 'integer']
            ]
        ],
    ],
    "settings" => [
        'number_of_shards' => 2,
        'number_of_replicas' => 0
    ],
    "batch_size" => 100
];
//3. Pass the file and configuration to the `EsLoader::load` method.
EsLoader::load($filePath, $config);

Configuration

<?php

/**
 * index : Name of the Elasticsearch index
 * doc_id_key : The field in your dataset which you want to keep as Es document id. NULL assigns a Es auto generated id
 * connection : Set it `local` if you have a self managed Es cluster. For AWS hosted Es set it to `aws`
 * local : Es configuration for your self managed Es
 * aws : Es configuration for your AWS managed Es
 * mappings : Define Es mappings as per your dataset
 * settings : Define Es settings as per your requirements
 * batch_size : The number of documents to index in a single bulk index request
 * 
 */
return [
    "index" => "esloader",
    "doc_id_key" => NULL,
    "connection" => "local",
    "local" => [
        'host' => "localhost",
        'port' => "9200"
    ],
    "aws" => [
        'host' => "",
        'region' => "",
        'access_key' => "",
        'secret_key' => ""
    ],
    "mappings" => [
        "_doc" => [
            '_source' => [
                'enabled' => true
            ],
            "properties" => [
                "id" => ['type' => 'keyword'],
                "first_name" => ['type' => 'text'],
                "last_name" => ['type' => 'text'],
                "email" => ['type' => 'keyword'],
                "gender" => ['type' => 'keyword'],
                "points" => ['type' => 'integer']
            ]
        ],
    ],
    "settings" => [
        'number_of_shards' => 2,
        'number_of_replicas' => 0
    ],
    "batch_size" => 100
];

Demo

The demo code is placed inside the testdata directory. The testdata contains sample csv, json and xml files having 1000 records each.

The configuration for the demo can be modified in config.php file.

To run the demo you need to have an elasticsearch cluster running. Once you have that you just need to run php example.php to run the demo.