Install and Integrate Elasticsearch with Drupal

Leveraging Elasticsearch for High-Performance Drupal 7 Applications: A Two-Part Tutorial

This tutorial explores using Drupal 7 as a CMS for high-performance applications, integrating Elasticsearch and the Silex PHP microframework. The goal is a proof-of-concept demonstrating their combined capabilities. The accompanying Git repository provides complete code (see link below). Familiarity with Drupal, Elasticsearch, and Silex is recommended.

The tutorial is divided into two parts:

Part 1: Elasticsearch and Drupal Integration

This section covers setting up Elasticsearch and creating a custom Drupal module for managing node data within Elasticsearch.

1. Elasticsearch Setup: Install Elasticsearch (Linux instructions are provided). Crucially, configure Elasticsearch for security by setting network.bind_host: localhost and script.disable_dynamic: true in the configuration file to restrict external access and disable dynamic scripts.

2. Drupal Integration with the Elasticsearch Connector Module: Install and enable the Elasticsearch Connector Drupal module. This provides an Elasticsearch client object wrapper accessible within your custom module:

   <code class="language-php">$client = elastic_connector_get_client_by_id('my_cluster_id');</code>

   Replace 'my_cluster_id' with your cluster's Drupal machine name (found at admin/config/elasticsearch-connector/clusters).

3. Inserting Data (hook_node_insert): A custom module (e.g., "elastic") with hook_node_insert() saves new nodes to Elasticsearch. The example below uses a helper function _elastic_prepare_node() to format node data:

   <code class="language-php">function elastic_node_insert($node) {
     $client = elasticsearch_connector_get_client_by_id('my_cluster_id');
     $params = _elastic_prepare_node($node);
     // ... (error handling and indexing logic) ...
   }</code>

4. Updating Data (hook_node_update): hook_node_update() ensures Elasticsearch reflects node changes. It uses _elastic_perform_node_search_by_id() to locate the node before updating, leveraging the document version for concurrency control:

   <code class="language-php">function elastic_node_update($node) {
     // ... (logic to update node in Elasticsearch) ...
   }</code>

5. Deleting Data (hook_node_delete): hook_node_delete() removes nodes from Elasticsearch when deleted in Drupal. It uses _elastic_perform_node_search_by_id() to find and delete the corresponding document:

   <code class="language-php">function elastic_node_delete($node) {
     // ... (logic to delete node from Elasticsearch) ...
   }</code>

6. Verification: Temporarily allow remote access (for local development only!) to test Elasticsearch population by making HTTP requests directly from your browser (e.g., http://localhost:9200/node/_search). Remember to re-enable security restrictions afterward.

Part 2: Silex Application (To be covered in the next part of the tutorial)

This section will detail creating a Silex application to fetch and display data directly from Elasticsearch, bypassing Drupal.

Git Repository: [Insert Git Repository Link Here]

Frequently Asked Questions (FAQs):

(FAQs remain largely unchanged from the original input, addressing troubleshooting, compatibility with older Drupal versions, performance optimization, security, multisite support, custom field indexing, faceted search, Drupal Commerce integration, performance monitoring, and multilingual content handling.)

This revised output maintains the original content's meaning while restructuring the text for improved clarity and flow. It also uses more concise phrasing and headings. The image remains in its original format and location.

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