Package Data | |
---|---|
Maintainer Username: | devemio |
Package Create Date: | 2016-03-10 |
Package Last Update: | 2021-03-28 |
Home Page: | https://devemio.github.io/elasticsearch-eloquent |
Language: | PHP |
License: | MIT |
Last Refreshed: | 2024-11-22 03:07:24 |
Package Statistics | |
---|---|
Total Downloads: | 13,118 |
Monthly Downloads: | 220 |
Daily Downloads: | 10 |
Total Stars: | 111 |
Total Watchers: | 6 |
Total Forks: | 14 |
Total Open Issues: | 0 |
This package allows you to interact with Elasticsearch as you interact with Eloquent models in Laravel.
Feel free to improve the project.
Via Composer
$ composer require isswp101/elasticsearch-eloquent
Warning! First of all you should create a base model and inherit from it their models.
use Elasticsearch\Client;
use Isswp101\Persimmon\DAL\ElasticsearchDAL;
use Isswp101\Persimmon\ElasticsearchModel as Model;
use Isswp101\Persimmon\Event\EventEmitter;
class ElasticsearchModel extends Model
{
public function __construct(array $attributes = [])
{
$dal = new ElasticsearchDAL($this, app(Client::class), app(EventEmitter::class));
parent::__construct($dal, $attributes);
}
public static function createInstance()
{
return new static();
}
}
In this example we use Laravel IoC Container to resolve Elasticsearch\Client
dependency as app(Client::class)
.
You must override static variables index
and type
to determine the document path.
class Product extends ElasticsearchModel
{
protected static $_index = 'test';
protected static $_type = 'test';
public $name;
public $price = 0;
}
Here name
and price
are fields which will be stored in Elasticsearch.
Warning! Don't use field names starting with underscore
$_*
, for example$_name
.
Use the static create()
method to create document in Elasticsearch:
$product = Product::create(['id' => 3, 'name' => 'Product 3', 'price' => 30]);
$product = new Product();
$product->id = 1;
$product->name = 'Product 1';
$product->price = 20;
$product->save();
Use save()
method to store model data in Elasticsearch. Let's see how this looks in Elasticsearch:
{
"_index": "test",
"_type": "test",
"_id": "1",
"_version": 1,
"found": true,
"_source": {
"name": "Product 1",
"price": 10,
"id": 1,
"user_id": null,
"created_at": "2016-06-03 08:11:08",
"updated_at": "2016-06-03 08:11:08"
}
}
Fields created_at
and updated_at
were created automatically. The user_id
field is persistent field to store user id.
$product = Product::find(1);
If you have big data in Elasticsearch you can specify certain fields to retrieve:
$product = Product::find(1, ['name']);
In this case the price
field equals 0
because it's populated as the default value that you specified in the model.
There are the following methods:
findOrFail()
returns ModelNotFoundException
exception if no result found.findOrNew()
returns a new model if no result found.There is a smart model cache when you use methods like find()
, findOrFail()
and so on.
$product = Product::find(1, ['name']); // will be retrieved from the elasticsearch
$product = Product::find(1, ['name']); // will be retrieved from the cache
$product = Product::find(1, ['price']); // elasticsearch
$product = Product::find(1, ['price']); // cache
$product = Product::find(1, ['name']); // cache
$product = Product::findOrFail(1); // elasticsearch
$product = Product::find(1); // cache
$product = Product::find(1, ['name']); // cache
$product = Product::find(1, ['price']); // cache
You can use partial update to update specific fields quickly.
$product = Product::find(1, ['name']);
$product->name = 'Product 3';
$product->save('name');
$product = Product::find(1);
$product->delete();
You can use the static method:
Product::destroy(1);
Out of the box you are provided with a simple implementation of events.
You can override the following methods to define events:
saving()
is called before saving, updating, creating the modelsaved()
is called after saving, updating, creating the modeldeleting()
is called before deleting the modeldeleted()
is called after deleting the modelFor example:
class Product extends ElasticsearchModel
{
public static $_index = 'test';
public static $_type = 'test';
public $name;
public $price = 0;
protected function saving()
{
if ($this->price <= 0) {
return false;
}
return true;
}
protected function deleting()
{
if (!$this->canDelete()) {
throw new LogicException('No permissions to delete the model');
}
return true;
}
}
There are helpers to search documents:
The first($query)
method returns the first document according to the query or null
.
$product = Product::first($query);
The firstOrFail($query)
method returns ModelNotFoundException
exception if first($query)
returns null
.
$product = Product::firstOrFail($query);
The search($query)
method returns documents (default 50 items) according to the query.
$products = Product::search($query);
The map($query, callable $callback)
method returns all documents (default 50 items per request) according to the query.
$total = Product::map([], function (Product $product) {
// ...
});
The all($query)
method returns all documents according to the query.
$products = Product::all($query);
If $query
is not passed the query will be as match_all
query.
use Isswp101\Persimmon\QueryBuilder\QueryBuilder;
$query = new QueryBuilder();
Simple usage:
$query = new QueryBuilder(['query' => ['match' => ['name' => 'Product']]]);
$products = Product::search($query);
The match
query:
$query = new QueryBuilder();
$query->match('name', 'Product');
$products = Product::search($query);
The range
query:
$query = new QueryBuilder();
$query->betweenOrEquals('price', 20, 30)->greaterThan('price', 15);
$products = Product::search($query);
Feel free to add your own filters.
The TermFilter
filter:
$query = new QueryBuilder();
$query->filter(new TermFilter('name', '2'));
$products = Product::search($query);
The IdsFilter
filter:
$query = new QueryBuilder();
$query->filter(new IdsFilter([1, 3]));
$products = Product::search($query);
The RangeOrExistFilter
filter:
$query = new QueryBuilder();
$query->filter(new RangeOrExistFilter('price', ['gte' => 20]));
$products = Product::search($query);
Feel free to add your own aggregations.
$query = new QueryBuilder();
$query->aggregation(new TermsAggregation('name'));
$products = Product::search($query);
$buckets = $products->getAggregation('name');
// Usage: $buckets[0]->getKey() and $buckets[0]->getCount()
The parent-child relationship is similar in nature to the nested model: both allow you to associate one entity with another. The difference is that, with nested objects, all entities live within the same document while, with parent-child, the parent and children are completely separate documents.
Let's create two models:
PurchaseOrder
has many PurchaseOrderLine
modelsPurchaseOrderLine
belongs to PurchaseOrder
modelclass PurchaseOrder extends ElasticsearchModel
{
protected static $_index = 'test_parent_child_rel';
protected static $_type = 'orders';
public $name;
public function lines()
{
return $this->hasMany(PurchaseOrderLine::class);
}
}
class PurchaseOrderLine extends ElasticsearchModel
{
protected static $_index = 'test_parent_child_rel';
protected static $_type = 'lines';
protected static $_parentType = 'orders';
public $name;
public function po()
{
return $this->belongsTo(PurchaseOrder::class);
}
}
To save()
models you can use the following code:
$po = new PurchaseOrder(['id' => 1, 'name' => 'PO1']);
$line = new PurchaseOrderLine(['id' => 1, 'name' => 'Line1']);
$po->save();
$po->lines()->save($line);
You can use the associate()
method to save models:
$po = new PurchaseOrder(['id' => 1, 'name' => 'PO1']);
$line = new PurchaseOrderLine(['id' => 1, 'name' => 'Line1']);
$po->save();
$line->po()->associate($po);
$line->save();
To get parent you can use the following code:
$line = PurchaseOrderLine::findWithParentId(1, 1);
$po = $line->po()->get();
To get children you can use the following code:
$po = PurchaseOrder::findOrFail(1);
$line = $po->lines()->find(1); // by id
$lines = $po->lines()->get(); // all children
The parent/child and nested features allow the return of documents that have matches in a different scope. In the parent/child case, parent document are returned based on matches in child documents or child document are returned based on matches in parent documents. In the nested case, documents are returned based on matches in nested inner objects.
You can get parent model using only one request with InnerHitsFilter
filter:
$query = new QueryBuilder();
$query->filter(new InnerHitsFilter(PurchaseOrderLine::getParentType()));
$line = PurchaseOrderLine::search($query)->first();
$po = $line->po()->get(); // will be retrieved from inner_hits cache
To debug all elasticsearch queries to search you can use own DALEmitter
class:
use Isswp101\Persimmon\DAL\DALEvents;
use Isswp101\Persimmon\Event\EventEmitter;
class DALEmitter extends EventEmitter
{
public function __construct()
{
$this->on(DALEvents::BEFORE_SEARCH, function (array $params) {
Log::debug('Elasticsearch query', $params);
});
}
}
And configure it in your service provider:
use Elasticsearch\Client;
use Isswp101\Persimmon\DAL\ElasticsearchDAL;
use Isswp101\Persimmon\ElasticsearchModel as Model;
use Isswp101\Persimmon\Test\Models\Events\DALEmitter;
class ElasticsearchModel extends Model
{
public function __construct(array $attributes = [])
{
$dal = new ElasticsearchDAL($this, app(Client::class), app(DALEmitter::class));
parent::__construct($dal, $attributes);
}
// ...
}
There are the following events:
DALEvents::BEFORE_SEARCH
is triggered before any search.DALEvents::AFTER_SEARCH
is triggered after any search.TO BE CONTINUED...
@TODO:
Please see CHANGELOG for more information what has changed recently.
$ composer test
Please see CONTRIBUTING and CONDUCT for details.
If you discover any security related issues, please email isswp101@gmail.com instead of using the issue tracker.
The MIT License (MIT). Please see License File for more information.