v2.9.7

Bugfix

• Hybrid relations eloquent builder to return native Laravel builder for non-ES models
• Escaped '/' for query builder
• Escaped values that start with a dash

Full Changelog: v2.9.6...v2.9.7

v2.8.6

Bugfix

• Hybrid relations eloquent builder to return native Laravel builder for non-ES models
• Escaped '/' for query builder
• Escaped values that start with a dash

Full Changelog: v2.8.5...v2.8.6

v2.7.6

Bugfix

• Hybrid relations eloquent builder to return native Laravel builder for non ES models
• Escaped '/' for query builder
• Escaped values that starts with a dash

Full Changelog: v2.7.5...v2.7.6

v2.10.4

Upgrades

• Re-index method now has explicit access to the connection's prefix

Full Changelog: v2.10.3...v2.10.4

v2.9.6

Upgrades

• Re-index method now has explicit access to the connection's prefix

Full Changelog: v2.9.5...v2.9.6

v2.8.5

Upgrades

• Re-index method now has explicit access to the connection's prefix

Full Changelog: v2.8.4...v2.8.5

v2.7.5

Upgrades

• Re-index method now has explicit access to the connection's prefix

Full Changelog: v2.7.4...v2.7.5

v2.10.3

Upgrades

• Updated docs to include re-indexing example
• Updated docs with TOC

New Feature

Distinct and GroupBy

This section covers the implementation of distinct() and groupBy() methods in the Elasticsearch Eloquent model. These methods are interchangeable and uses term aggregation under the hood.

This tends to be a core use case for Elasticsearch, for example, to get all the unique user_ids of the users who have been logged in the last 30 days:

Basic Usage

• Distinct:

// Unique user_ids of users logged in the last 30 days
$users = UserLog::where('created_at', '>=', Carbon::now()->subDays(30))->distinct()->get('user_id');
//or:
$users = UserLog::where('created_at', '>=', Carbon::now()->subDays(30))->select('user_id')->distinct()->get();

• GroupBy:

// Equivalent to the above distinct query
$users = UserLog::where('created_at', '>=', Carbon::now()->subDays(30))->groupBy('user_id')->get();

Working with Collections

• The results from these queries are returned as collections, allowing use of standard collection methods.
• Example of loading related user data:

$users = UserLog::where('created_at', '>=', Carbon::now()->subDays(30))->distinct()->get('user_id');
return $users->load('user');

Multiple Fields Aggregation

• You can pass multiple fields to perform term aggregation.
• Example:

$users = UserLog::where('created_at', '>=', Carbon::now()->subDays(30))->distinct()->get(['user_id', 'log_title']);
/** returns: 
{
    "user_id": "1",
    "log_title": "LOGGED_IN"
},
{
    "user_id": "2",
    "log_title": "LOGGED_IN"
},
{
    "user_id": "2",
    "log_title": "LOGGED_OUT"
},
 **/

Ordering by Aggregation Count

• Results can be sorted based on the count of the aggregated field.
• Example of ordering by the most logged users:

$users = UserLog::where('created_at', '>=', Carbon::now()->subDays(30))->distinct()->orderBy('_count')->get('user_id');

• Or you can order by the distinct field, example:

$users = UserLog::where('created_at', '>=', Carbon::now()->subDays(30))->distinct()->orderBy('user_id')->get('user_id');

Returning Count with Distinct Results

• To include the count of distinct values in the results, use distinct(true):

$users = UserLog::where('created_at', '>=', Carbon::now()->subDays(30))->distinct(true)->orderBy('_count')->get(['user_id']);
/** returns: 
{
    "user_id": "5",
    "user_id_count": 65
},
{
    "user_id": "1",
    "user_id_count": 61
},
{
    "user_id": "9",
    "user_id_count": 54
},
 **/

Pagination Support

• The distinct() and groupBy() methods support pagination.
• Example:

$users = UserLog::where('log_title', 'LOGGED_IN')->select('user_id')->distinct()->orderBy('_count')->paginate(20);
//or
$users = UserLog::where('log_title', 'LOGGED_IN')->groupBy('user_id')->orderBy('_count')->paginate(20);

Full Changelog: v2.10.2...v2.10.3

v2.9.5

New Feature

Distinct and GroupBy

This section covers the implementation of distinct() and groupBy() methods in the Elasticsearch Eloquent model. These methods are interchangeable and uses term aggregation under the hood.

This tends to be a core use case for Elasticsearch, for example, to get all the unique user_ids of the users who have been logged in the last 30 days:

Basic Usage

• Distinct:

// Unique user_ids of users logged in the last 30 days
$users = UserLog::where('created_at', '>=', Carbon::now()->subDays(30))->distinct()->get('user_id');
//or:
$users = UserLog::where('created_at', '>=', Carbon::now()->subDays(30))->select('user_id')->distinct()->get();

• GroupBy:

// Equivalent to the above distinct query
$users = UserLog::where('created_at', '>=', Carbon::now()->subDays(30))->groupBy('user_id')->get();

Working with Collections

• The results from these queries are returned as collections, allowing use of standard collection methods.
• Example of loading related user data:

$users = UserLog::where('created_at', '>=', Carbon::now()->subDays(30))->distinct()->get('user_id');
return $users->load('user');

Multiple Fields Aggregation

• You can pass multiple fields to perform term aggregation.
• Example:

$users = UserLog::where('created_at', '>=', Carbon::now()->subDays(30))->distinct()->get(['user_id', 'log_title']);
/** returns: 
{
    "user_id": "1",
    "log_title": "LOGGED_IN"
},
{
    "user_id": "2",
    "log_title": "LOGGED_IN"
},
{
    "user_id": "2",
    "log_title": "LOGGED_OUT"
},
 **/

Ordering by Aggregation Count

• Results can be sorted based on the count of the aggregated field.
• Example of ordering by the most logged users:

$users = UserLog::where('created_at', '>=', Carbon::now()->subDays(30))->distinct()->orderBy('_count')->get('user_id');

• Or you can order by the distinct field, example:

$users = UserLog::where('created_at', '>=', Carbon::now()->subDays(30))->distinct()->orderBy('user_id')->get('user_id');

Returning Count with Distinct Results

• To include the count of distinct values in the results, use distinct(true):

$users = UserLog::where('created_at', '>=', Carbon::now()->subDays(30))->distinct(true)->orderBy('_count')->get(['user_id']);
/** returns: 
{
    "user_id": "5",
    "user_id_count": 65
},
{
    "user_id": "1",
    "user_id_count": 61
},
{
    "user_id": "9",
    "user_id_count": 54
},
 **/

Pagination Support

• The distinct() and groupBy() methods support pagination.
• Example:

$users = UserLog::where('log_title', 'LOGGED_IN')->select('user_id')->distinct()->orderBy('_count')->paginate(20);
//or
$users = UserLog::where('log_title', 'LOGGED_IN')->groupBy('user_id')->orderBy('_count')->paginate(20);

Full Changelog: v2.9.4...v2.9.5

v2.8.4

New Feature

Distinct and GroupBy

This section covers the implementation of distinct() and groupBy() methods in the Elasticsearch Eloquent model. These methods are interchangeable and uses term aggregation under the hood.

This tends to be a core use case for Elasticsearch, for example, to get all the unique user_ids of the users who have been logged in the last 30 days:

Basic Usage

• Distinct:

// Unique user_ids of users logged in the last 30 days
$users = UserLog::where('created_at', '>=', Carbon::now()->subDays(30))->distinct()->get('user_id');
//or:
$users = UserLog::where('created_at', '>=', Carbon::now()->subDays(30))->select('user_id')->distinct()->get();

• GroupBy:

// Equivalent to the above distinct query
$users = UserLog::where('created_at', '>=', Carbon::now()->subDays(30))->groupBy('user_id')->get();

Working with Collections

• The results from these queries are returned as collections, allowing use of standard collection methods.
• Example of loading related user data:

$users = UserLog::where('created_at', '>=', Carbon::now()->subDays(30))->distinct()->get('user_id');
return $users->load('user');

Multiple Fields Aggregation

• You can pass multiple fields to perform term aggregation.
• Example:

$users = UserLog::where('created_at', '>=', Carbon::now()->subDays(30))->distinct()->get(['user_id', 'log_title']);
/** returns: 
{
    "user_id": "1",
    "log_title": "LOGGED_IN"
},
{
    "user_id": "2",
    "log_title": "LOGGED_IN"
},
{
    "user_id": "2",
    "log_title": "LOGGED_OUT"
},
 **/

Ordering by Aggregation Count

• Results can be sorted based on the count of the aggregated field.
• Example of ordering by the most logged users:

$users = UserLog::where('created_at', '>=', Carbon::now()->subDays(30))->distinct()->orderBy('_count')->get('user_id');

• Or you can order by the distinct field, example:

$users = UserLog::where('created_at', '>=', Carbon::now()->subDays(30))->distinct()->orderBy('user_id')->get('user_id');

Returning Count with Distinct Results

• To include the count of distinct values in the results, use distinct(true):

$users = UserLog::where('created_at', '>=', Carbon::now()->subDays(30))->distinct(true)->orderBy('_count')->get(['user_id']);
/** returns: 
{
    "user_id": "5",
    "user_id_count": 65
},
{
    "user_id": "1",
    "user_id_count": 61
},
{
    "user_id": "9",
    "user_id_count": 54
},
 **/

Pagination Support

• The distinct() and groupBy() methods support pagination.
• Example:

$users = UserLog::where('log_title', 'LOGGED_IN')->select('user_id')->distinct()->orderBy('_count')->paginate(20);
//or
$users = UserLog::where('log_title', 'LOGGED_IN')->groupBy('user_id')->orderBy('_count')->paginate(20);

Full Changelog: v2.8.3...v2.8.4