Integration of PHP and Elasticsearch

Integration of PHP and Elasticsearch

With the development of big data and data mining, search engines have become an indispensable tool in our lives. Elasticsearch is a fast, open, and scalable search and analysis engine that can easily perform full-text retrieval, data analysis, and real-time data storage and query. So how to integrate with Elasticsearch using PHP?

1. Install Elasticsearch

First, we need to install Elasticsearch. You can go to the Elasticsearch official website to download the installation package of the corresponding version, and then unzip it to the location you want. You can see the elasticsearch.bat (Windows)/elasticsearch command (Linux) in the bin directory of Elasticsearch.

Execute the elasticsearch.bat/ command to start Elasticsearch. If everything goes well, you now have an Elasticsearch node started. Use the http://localhost:9200/ address to access the RESTful API provided by Elasticsearch.

2. Install and configure PHP's Elasticsearch client

We need to download and install PHP's Elasticsearch client, such as Elasticsearch-PHP or the official Elasticsearch library php-elasticsearch (Elasticsearch 7 version needs to be installed) ). These libraries are all very useful and have complete documentation and examples. Let’s take the Elasticsearch-PHP client as an example:

1. Installation

You can use composer to install the Elasticsearch-PHP client. Switch to the root directory of your php project and execute the following command:

composer require elasticsearch/elasticsearch

2. Use

to introduce autoloader, and then instantiate the client connected to Elasticsearch End object:

require 'vendor/autoload.php';
$client = ElasticsearchClientBuilder::create()->build();

Here we build in PHP An Elasticsearch client. Next, you can perform CRUD operations on Elasticsearch data.

3. Elasticsearch operation

1. Create index

Index is one of the most important concepts in Elasticsearch. We need to create different indexes for different business needs. An index named my_index can be created using the following code:

$params = [

'index' => 'my_index',
'body' => [
    'settings' => [
        'number_of_shards' => 3,
        'number_of_replicas' => 2
    ]
]

];
$response = $client->indices()->create ($params);

In the above code, we specify the name of the index to be built, and the settings attributes (number of shards and replicas) corresponding to the index. In actual use, it is recommended to configure more detailed settings attributes according to business needs.

2. Insert documents

Use bulk method to insert multiple documents:

$params = [

'body' => [
    ['index' => ['_id' => 1]],
    ['name' => 'product1', 'price' => 10.0, 'description' => 'description of product1'],
    ['index' => ['_id' => 2]],
    ['name' => 'product2', 'price' => 20.0, 'description' => 'description of product2'],
    ['index' => ['_id' => 3]],
    ['name' => 'product3', 'price' => 30.0, 'description' => 'description of product3'],
    ['index' => ['_id' => 4]],
    ['name' => 'product4', 'price' => 40.0, 'description' => 'description of product4'],
],
'index' => 'my_index',
'type' => 'my_type'

];
$response = $ client->bulk($params);

3. Query documents

Use the search method to query documents:

$params = [

'index' => 'my_index',
'type' => 'my_type',
'body' => [
    'query' => [
        'match' => [
            'name' => 'product1'
        ]
    ]
]

] ;
$response = $client->search($params);

4. Delete index

To delete an index, you can use the delete method:

$params = [

'index' => 'my_index'

];
$response = $client->indices()->delete($params);

Summary

Through the above code, You can easily create an Elasticsearch index, insert documents, query documents, delete indexes and other operations, which will undoubtedly improve your search experience. The integration of Elasticsearch and PHP is a very practical and powerful tool.

The above is the detailed content of Integration of PHP and Elasticsearch. For more information, please follow other related articles on the PHP Chinese website!