belyys7/elasticsearch-its-easy

Simple search with elasticsearch

1.1.1 2023-05-04 16:48 UTC

Requires

• php: ^7.1
• elasticsearch/elasticsearch: 7.9.0
• saikiran/geohash: 1.0.1

MIT

• Sergey Belyy

• elasticsearch

README

Elasticsearch it`s easy SDK for working with Elasticsearch like with constructor

Documentation

The documentation for the Elasticsearch REST API can be found here.

Installation

The preferred way to install this extension is through composer.

Either run

composer require idapgroup/elasticsearch-its-easy

or add

{
  "require": {
    "idapgroup/elasticsearch-its-easy": "^1.0.0"
  }
}

to the requirement section of your composer.json file.

Quickstart

Prepare the data you want to store in elasticsearch

$data = [
    [
        'user' => [
            'id' => 100,
            'email' => '[e-mail]',
            'name' => 'Stepan',
            'age' => 21,
            'birthday' => '2001-06-15',
        ],
        'work' => [
            'position' => [
                'id' => 25,
                'name' => 'php developer',
            ],
            'skills' => [
                [
                    'id' => 36,
                    'name' => 'php'
                ],
                [
                    'id' => 40,
                    'name' => 'mysql'
                ],
                [
                    'id' => 56,
                    'name' => 'js'
                ],
            ],
            'salary' => 4000
        ],
        'location' => [
            'lat' => 50.445077,
            'lon' => 30.521215
        ],
    ],
    [
        'user' => [
            'id' => 101,
            'email' => '[e-mail]',
            'name' => 'Luigi',
            'age' => 29,
            'birthday' => '2005-03-20',
        ],
        'work' => [
            'position' => [
                'id' => 12,
                'name' => 'js developer',
            ],
            'skills' => [
                [
                    'id' => 56,
                    'name' => 'js'
                ],
                [
                    'id' => 70,
                    'name' => 'mongodb'
                ],
                [
                    'id' => 1,
                    'name' => 'html'
                ],
                [
                    'id' => 2,
                    'name' => 'css'
                ],
            ],
            'salary' => 2700
        ],
        'location' => [
            'lat' => 47.454589,
            'lon' => 32.915673
        ],
    ],
];

Create your class to be expanded by basic search

use IdapGroup\ElasticsearchItsEasy\ModelSearchBase;

class StaffModelSearch extends ModelSearchBase
{   
    public function setRules() : void
    {
        $this->rules = [
            self::GROUP_MUST => [
                self::RULE_EQUAL => [
                    'userId' => 'user.id',
                ],
            ],
            self::GROUP_SHOULD => [
                self::RULE_LIKE => [
                    'userEmail' => 'user.email',
                    'userName' => 'user.name',
                ],
                self::RULE_IN => [
                    'workSkillsId' => 'work.skills.id',
                ],
            ],
            self::GROUP_FILTER => [
                self::RULE_EQUAL => [
                    'workPositionId' => 'work.position.id'
                ],
                self::RULE_RANGE_NUMBER => [
                    'userAge' => 'user.age',
                    'workSalary' => 'work.salary',
                ],
                self::RULE_RANGE_DATE => [
                    'birthday' => 'user.birthday',
                ],
            ],
            self::GROUP_LOCATION => [
                'location' => self::SORT_DESC,
            ],
        ];
    }
   
    public function setSort() : void
    {
        $this->sort = [
            'user.id' => self::SORT_DESC,
        ];
    }

}

Create model based on elasticsearch configuration

$staffModelSearch = new StaffModelSearch('es01', '9200', 'staff_search');

Indexing Documents in Elasticsearch

$staffModelSearch->reCreateIndex();

foreach ($data as $item) {
    $staffModelSearch->addDocument($item, 'user.id', $item['user']['id']);
}

Example #1: search as list with pagination

Specify search keys and their values

$params = [
    'userId' => 100,
    'workPositionId' => 25,
    'userEmail' => '[e-mail]',
    'userName' => 'Stepan',
    'workSkillsId' => [36, 40],
    'userAge' => ['min' => 18, 'max' => 65],
    'workSalary' => ['min' => 500, 'max' => 5000],
    'location' => [
        'point' => [
            'lat' => 48.454589,
            'lon' => 33.915673,
            'distance' => 100000
        ],
        'rectangle' => [
            'topLeftLat' => 55.710929,
            'topLeftLng' => 14.090451,
            'bottomRightLat' => 41.830140,
            'bottomRightLng' => 41.802791
        ],
    ],
    'page' => 1,
    'limit' => 20
];

Set data output limits if required and search

//$staffModelSearch->enableFixLimitResult(50);
$result = $staffModelSearch->searchList($params);

The result of the response will be in the format

[
    'result' => [
        [
            'user' => [
                'id' => 100,
                'email' => '[e-mail]',
                'name' => 'Stepan',
                'age' => 21
            ],
            'work' => [
                'position' => [
                    'id' => 25,
                    'name' => 'php developer',
                ],
                'skills' => [
                    [
                        'id' => 36,
                        'name' => 'php'
                    ],
                    [
                        'id' => 40,
                        'name' => 'mysql'
                    ],
                    [
                        'id' => 56,
                        'name' => 'js'
                    ],
                ],
                'salary' => 4000
            ],
            'location' => [
                'lat' => 50.445077,
                'lon' => 30.521215
            ],
        ],
        //... etc.
    ],
    'pagination' => [
        'totalCount' => (int),
        'pageCount' => (int),
        'currentPage' => (int)
    ]
]

Example #2: search for map

Specify search keys and their values

$params = [
    'userId' => 100,
    'workPositionId' => 25,
    'userEmail' => '[e-mail]',
    'userName' => 'Stepan',
    'workSkillsId' => [36, 40],
    'userAge' => ['min' => 18, 'max' => 65],
    'workSalary' => ['min' => 500, 'max' => 5000],
    'location' => [
        'point' => [
            'lat' => 48.454589,
            'lon' => 33.915673,
            'distance' => 100000
        ],
        'rectangle' => [
            'topLeftLat' => 55.710929,
            'topLeftLng' => 14.090451,
            'bottomRightLat' => 41.830140,
            'bottomRightLng' => 41.802791
        ],       
    ],
];

Execute search

$result = $staffModelSearch->searchMap($params);

The result of the response will be in the format

[
    [
        'user' => [
            'id' => 100,
            'email' => '[e-mail]',
            'name' => 'Stepan',
            'age' => 21
        ],
        'work' => [
            'position' => [
                'id' => 25,
                'name' => 'php developer',
            ],
            'skills' => [
                [
                    'id' => 36,
                    'name' => 'php'
                ],
                [
                    'id' => 40,
                    'name' => 'mysql'
                ],
                [
                    'id' => 56,
                    'name' => 'js'
                ],
            ],
            'salary' => 4000
        ],
        'location' => [
            'lat' => 50.445077,
            'lon' => 30.521215
        ],
    ],
    //... etc.
]

Additional settings

Custom clustering (useful when markers on the map overlap each other and the maximum zoom does not solve the problem)

$params = [
    //...
    'location' => [
        'point' => [
            'lat' => 48.454589,
            'lon' => 33.915673,
            'distance' => 100000
        ],
        'rectangle' => [
            'topLeftLat' => 55.710929,
            'topLeftLng' => 14.090451,
            'bottomRightLat' => 41.830140,
            'bottomRightLng' => 41.802791
        ],
        'clustering' => true,
        'zoom' => 1,
    ],
    //...
];

Response structure example

[
    // Cluster 1
    [
        [
            // data
        ],
        [
            // data
        ],
        //...
    ],
    // Cluster 2
    [
        [
            // data
        ],
        [
            // data
        ],
        //...
    ],
    //... etc.
]

Overwrite rules

// Describe the rules in the associative array as required by the ES documentation
$overWriteRules = [
    'must' => [
        [
            'term' => [
                'user.name.keyword' => 'Stepan',
            ],
        ],
        [
            'term' => [
                'user.age' => 21,
            ],
        ],
        //...
    ],
    //...
];

// Use a Method to Set Your Rules
$staffModelSearch->setOverWriteRules($overWriteRules);