Easily integrate the new version of Elasticsearch7.9 Chinese search to Laravel7 project

The following tutorial column of Laravel will introduce to you how to easily integrate the new version of Elasticsearch7.9 Chinese search into the Laravel7 project. I hope it will be helpful to friends in need!

                                               

Only five steps:

#1. Start the integrated ik Chinese word segmentation plug-in Elasticsearch7.9 Docker Image

##Course Recommendation→:"Elasticsearch Full Text Search Practical Combat" (Practical Video)

From the course

"Ten million-level data concurrency solution (theoretical practice)"

2, Laravel7 configuration Scout

3, configure Model model

4, Import data

5. Search

Demo address

www.ar414. com/search?query=php�%...

Search scope

Article content
• Title
• Tags

Result weight

Number of keywords appearing
1. Number of keywords appearing

Search page

High Highlight display
• Word segmentation display
• Result paging

##Foreword

Mainly because I just want to make a blog Search and organize it into articles by the way

Laravel Elasticsearch Many seniors have written tutorials and cases, but with the upgrade of Elasticsearch and laravel versions, many of the previous articles are not applicable to the new version. It is recommended that everyone use For any open source project, you should go through the documentation, mainly the version documentation currently in use, supplemented by tutorials

Elasticsearch 7.9

• Laravel 7
• elasticsearch-analysis-ik v7.9

Reference

ik Chinese word segmentation plug-in

• elasticsearch official document

Use integrated ik Chinese word segmentation plug-in Elasticsearch

Pull docker

$ docker pull ar414/elasticsearch-7.9-ik-plugin

Create log And the data storage directory

is mapped locally into the docker container to prevent data loss when docker restarts

$ mkdir -p /data/elasticsearch/data
$ mkdir -p /data/elasticsearch/log
$ chmod -R 777 /data/elasticsearch/data
$ chmod -R 777 /data/elasticsearch/log

Run

docker run -d -p 9200:9200 -p 9300:9300 -e "discovery.type=single-node" -v /data/elasticsearch/data:/var/lib/elasticsearch -v /data/elasticsearch/log:/var/log/elasticsearch ar414/elasticsearch-7.9-ik-plugin

Verification

$ curl http://localhost:9200{
  "name" : "01ac21393985",  "cluster_name" : "docker-cluster",  "cluster_uuid" : "h8L336qcRb2i1aydOv04Og",  "version" : {
    "number" : "7.9.0",    "build_flavor" : "default",    "build_type" : "docker",    "build_hash" : "a479a2a7fce0389512d6a9361301708b92dff667",    "build_date" : "2020-08-11T21:36:48.204330Z",    "build_snapshot" : false,    "lucene_version" : "8.6.0",    "minimum_wire_compatibility_version" : "6.8.0",    "minimum_index_compatibility_version" : "6.0.0-beta1"
  },  "tagline" : "You Know, for Search"}

Test Chinese word segmentation

curl -X POST "http://localhost:9200/_analyze?pretty" -H 'Content-Type: application/json' -d'
{
  "analyzer": "ik_max_word",
  "text":     "laravel天下无敌"
}
'{
  "tokens" : [
    {
      "token" : "laravel",      "start_offset" : 0,      "end_offset" : 7,      "type" : "ENGLISH",      "position" : 0    },    {
      "token" : "天下无敌",      "start_offset" : 7,      "end_offset" : 11,      "type" : "CN_WORD",      "position" : 1    },    {
      "token" : "天下",      "start_offset" : 7,      "end_offset" : 9,      "type" : "CN_WORD",      "position" : 2    },    {
      "token" : "无敌",      "start_offset" : 9,      "end_offset" : 11,      "type" : "CN_WORD",      "position" : 3    }
  ]}

Using Elasticsearch in Laravel project

Elasticsearch
The official SDK is provided. In the Laravel project, you can connect to Elasticsearch more elegantly. Laravel itself provides a Scout full-text search solution. We only need to change the default The Algolia driver is replaced with ElasticSearch driver.

Installation

laravel/scout

• Easily integrate the new version of Elasticsearch7.9 Chinese search to Laravel7 project

  $ composer require laravel/scout
  $ composer require Easily integrate the new version of Elasticsearch7.9 Chinese search to Laravel7 project

Configuration

Generate Scout configuration file (config/scout.php)

1. $ php artisan vendor:publish --provider="Laravel\Scout\ScoutServiceProvider"Copied File [\vendor\laravel\scout\config\scout.php] To [\config\scout.php]Publishing complete.

Specify Scout driver

First type: Specify (recommended)

SCOUT_DRIVER=Matchish\ScoutElasticSearch\Engines\ElasticSearchEngine

in the

.env
• fileSecond type: Specify in
config/scout.php
• Directly modify the default driver

  'driver' => env('SCOUT_DRIVER', 'algolia')改为'driver' => env('SCOUT_DRIVER', 'Matchish\ScoutElasticSearch\Engines\ElasticSearchEngine')

Specify the Elasticsearch service IP port

3. If you use docker deployment, use the IP of

   docker0

   , Linux checks through ifconfig

   Configure in

   .env

   <div class="code" style="position:relative; padding:0px; margin:0px;"><pre class="brush:php;toolbar:false">ELASTICSEARCH_HOST=172.17.0.1:9200</pre><div class="contentsignin">Copy after login</div></div>

Registration service

4. config/app.php
   <div class="code" style="position:relative; padding:0px; margin:0px;"><pre class="brush:php;toolbar:false">'providers' =&gt; [  // Other Service Providers  \Matchish\ScoutElasticSearch\ElasticSearchServiceProvider::class],</pre><div class="contentsignin">Copy after login</div></div>

Clear configuration cache

5. $ php artisan config:clear

Laravel has been connected to Elasticsearch

实际业务中使用

需求

通过博客右上角的搜索框可以搜索到与关键词相关的文章，从以下几点匹配

• 文章内容
• 文章标题
• 文章标签

涉及到2张 Mysql表 以及字段

• article
  • title
  • tags
• article_content
  • content

为文章配置 Elasticsearch 索引

1. 创建索引配置文件（config/elasticsearch.php）

   $ touch config/elasticsearch.php

2. elasticsearch.php 配置字段映射

   <?phpreturn  [
    &#39;indices&#39; => [
        'mappings' => [
            'blog-articles' => [
                "properties"=>  [
                    "content"=>  [
                        "type"=>  "text",
                        "analyzer"=>  "ik_max_word",
                        "search_analyzer"=>  "ik_smart"
                    ],
                    "tags"=>  [
                        "type"=>  "text",
                        "analyzer"=>  "ik_max_word",
                        "search_analyzer"=>  "ik_smart"
                    ],
                    "title"=>  [
                        "type"=>  "text",
                        "analyzer"=>  "ik_max_word",
                        "search_analyzer"=>  "ik_smart"
                    ]
                ]
            ]
        ]
    ],];

• analyzer：字段文本的分词器
  • search_analyzer：搜索词的分词器
  • 根据具体业务场景选择(颗粒小占用资源多,一般场景analyzer使用ik_max_word,search_analyzer使用ik_smart):
    • ik_max_word：ik中文分词插件提供,对文本进行最大数量分词
      laravel天下无敌 -> laravel，天下无敌,天下,无敌
    • ik_smart: ik中文分词插件提供,对文本进行最小数量分词
      laravel天下无敌 -> laravel，天下无敌

配置文章模型

建议先看一遍 Laravel Scout 使用文档

1. 引入Laravel Scout

    namespace App\Models\Blog;
   
    use Laravel\Scout\Searchable;
   
    class Article extends BlogBaseModel
    {
        use Searchable;
    }

2. 指定索引(刚刚配置文件中的elasticsearch.indices.mappings.blog-articles)

    /**
     * 指定索引
     * @return string
     */
    public function searchableAs()
    {
        return 'blog-articles';
    }

3. 设置导入索引的数据字段

    /**
     * 设置导入索引的数据字段
     * @return array
     */
    public function toSearchableArray()
    {
        return [
            'content' => ArticleContent::query()
                ->where('article_id',$this->id)
                ->value('content'),
            'tags'    => implode(',',$this->tags),
            'title'   => $this->title
        ];
    }

4. 指定 搜索索引中存储的唯一ID

    /**
     * 指定 搜索索引中存储的唯一ID
     * @return mixed
     */
    public function getScoutKey()
    {
        return $this->id;
    }
   
    /**
     * 指定 搜索索引中存储的唯一ID的键名
     * @return string
     */
    public function getScoutKeyName()
    {
        return 'id';
    }

数据导入

其实是将数据表中的数据通过Elasticsearch导入到Lucene
Elasticsearch 是 Lucene 的封装，提供了 REST API 的操作接口

• 一键自动导入: php artisan scout:import
• 导入指定模型: php artisan scout:import ${model}

$ php artisan scout:import "App\Models\Blog\Article"Importing [App\Models\Blog\Article]Switching to the new index
5/5 [⚬⚬⚬⚬⚬⚬⚬⚬⚬⚬⚬⚬⚬⚬⚬⚬⚬⚬⚬⚬⚬⚬⚬⚬⚬⚬⚬⚬] 100%[OK] All [App\Models\Blog\Article] records have been imported.

导入失败,常见原因:

• Unresolvable dependency resolving [Parameter #0 [  integer $retries ]] in class Elasticsearch\Transport
  • 解决: 修改配置后,没有清除配置缓存
• invalid_index_name_exception
  • 解决: searchableAs配置错误,为索引创建别名后,指定别名

检查索引是否正确

$ curl -XGET http://localhost:9200/blog-articles/_mapping?pretty{
  "blog-articles_1598362919" : {
    "mappings" : {
      "properties" : {
        "__class_name" : {
          "type" : "text",          "fields" : {
            "keyword" : {
              "type" : "keyword",              "ignore_above" : 256            }
          }
        },        "content" : {
          "type" : "text",          "analyzer" : "ik_max_word",          "search_analyzer" : "ik_smart"
        },        "tags" : {
          "type" : "text",          "analyzer" : "ik_max_word",          "search_analyzer" : "ik_smart"
        },        "title" : {
          "type" : "text",          "analyzer" : "ik_max_word",          "search_analyzer" : "ik_smart"
        }
      }
    }
  }}

测试

1. 创建一个测试命令行

   $ php artisan make:command ElasticTest

2. 代码

<?phpnamespace  App\Console\Commands;use App\Models\Blog\Article;use App\Models\Blog\ArticleContent;use Illuminate\Console\Command;use Illuminate\Support\Carbon;class ElasticTest extends Command{
    /**
     * The name and signature of the console command.
     *
     * @var string
     */
    protected $signature = &#39;elasticsearch {query}&#39;;

    /**
     * The console command description.
     *
     * @var string
     */
    protected $description = &#39;elasticsearch test&#39;;

    /**
     * Create a new command instance.
     *
     * @return void
     */
    public function __construct()
    {
        parent::__construct();
    }

    /**
     * Execute the console command.
     *
     * @return mixed
     */
    public function handle()
    {
        //
        $startTime = Carbon::now()->getPreciseTimestamp(3);
        $articles = Article::search($this->argument('query'))->get()->toArray();
        $userTime = Carbon::now()->getPreciseTimestamp(3) - $startTime;
        echo "耗时(毫秒)：{$userTime} \n";

        //content在另外一张表中，方便观察测试 这里输出
        if(!empty($articles)) {
            foreach($articles as &$article) {
                $article = ArticleContent::query()->where('article_id',$article['id'])->value('content');
            }
        }

        var_dump($articles);

    }}

3. 测试

   $ php artisan elasticsearch 周杰伦

4. 复杂查询
   例如：自定义高亮显示

   //ONGR\ElasticsearchDSL\Highlight\Highlight ArticleModel::search($query,function($client,$body) {
            $higlight = new Highlight();
            $higlight->addField('content',['type' => 'plain']);
            $higlight->addField('title');
            $higlight->addField('tags');
            $body->addHighlight($higlight);
            $body->setSource(['title','tags']);
            return $client->search(['index' => (new ArticleModel())->searchableAs(), 'body' => $body->toArray()]);
        })->raw();

复杂自定义查询回调中的$client和$body，可根据这两个包进行灵活操作

$client 官方 elasticsearch/elasticsearch package（https://packagist.org/packages/elasticsearch/elasticsearch）

$body ongr/elasticsearch-dsl package（https://packagist.org/packages/ongr/elasticsearch-dsl）

The above is the detailed content of Easily integrate the new version of Elasticsearch7.9 Chinese search to Laravel7 project. For more information, please follow other related articles on the PHP Chinese website!