omaresmael / laravel-query-optimizer
An AI powered optimizer for laravel eloquent and query builder
Requires
- openai-php/laravel: ^0.4.0
Requires (Dev)
- laravel/pint: ^1.7
- nunomaduro/collision: ^6.1
- orchestra/testbench: ^7.22
- pestphp/pest: ^1.22
- phpunit/phpunit: ^9.6
- roave/security-advisories: dev-latest
README
Optimize laravel DB or eloquent read queries using openAI API, and get insights on how you can make your queries faster
Installation
First, install the package using composer
composer require omaresmael/laravel-query-optimizer --dev
Then, publish openAI configuration file
php artisan vendor:publish --provider="OpenAI\Laravel\ServiceProvider"
Lastly, add the openAPI key to your .env
file
OPENAI_API_KEY=sk-...
and that's it, you are ready to go
Usage
this package can help you optimize your eloquent or DB queries using openAI
User::query() //can be applied to `DB facade` as well (see tests) ->select('users.id', 'users.name') ->whereHas('roles', function ($query) { $query->where('name', 'author'); }) ->whereHas('posts', function ($query) { $query->where('title', 'Awesome post'); })->optimize()->get();
the package has the following methods
optimize()
This method is responsible for optimizing the query and return an instance of the Optimizer
class.
User::query() ->select('users.id', 'users.name') ->whereHas('roles', function ($query) { $query->where('name', 'author'); }) ->whereHas('posts', function ($query) { $query->where('title', 'Awesome post'); })->optimize(); //old query => select `id`, `name` from `users` where exists (select * from `roles` inner join `role_user` on `roles`.`id` = `role_user`.`role_id` where `users`.`id` = `role_user`.`user_id` and `type` = ?) and exists (select * from `posts` where `users`.`id` = `posts`.`user_id` and `title` = ?) //optimized query => SELECT `id`, `name` FROM `users` INNER JOIN `role_user` ON `users`.`id` = `role_user`.`user_id` INNER JOIN `roles` ON `roles`.`id` = `role_user`.`role_id` INNER JOIN `posts` ON `users`.`id` = `posts`.`user_id` WHERE `type` = ? AND `title` = ?
toSql()
this method will return the optimized query as a string
User::query() ->select('users.id', 'users.name') ->whereHas('roles', function ($query) { $query->where('type', 'author'); }) ->whereHas('posts', function ($query) { $query->where('title', 'Awesome post'); })->optimize()->toSql(); //output => SELECT `id`, `name` FROM `users` INNER JOIN `role_user` ON `users`.`id` = `role_user`.`user_id` INNER JOIN `roles` ON `roles`.`id` = `role_user`.`role_id` INNER JOIN `posts` ON `users`.`id` = `posts`.`user_id` WHERE `type` = ? AND `title` = ?
get()
this method will run the optimized query that is generated by optimize()
method and return the result
⚠️ make sure to know the query you are running before using this method, as it will run the optimized query and not the original query
User::query() ->select('users.id', 'users.name') ->whereHas('roles', function ($query) { $query->where('type', 'author'); }) ->whereHas('posts', function ($query) { $query->where('title', 'Awesome post'); })->optimize()->get()->toArray(); //output => ['id' => '1', 'name' => 'omar']
explain()
this method will return a key-value array that contains the optimized query, the reasoning behind performing such optimization, and suggestions to manually optimize the query even further
User::query() ->select('users.id', 'users.name') ->whereHas('roles', function ($query) { $query->where('type', 'author'); }) ->whereHas('posts', function ($query) { $query->where('title', 'Awesome post'); })->optimize()->explain();
the array format will be
[ 'optimizedQuery' => 'SELECT `id`, `name` FROM `users` INNER JOIN `role_user` ON `users`.`id` = `role_user`.`user_id` INNER JOIN `roles` ON `roles`.`id` = `role_user`.`role_id` INNER JOIN `posts` ON `users`.`id` = `posts`.`user_id` WHERE `type` = ? AND `title` = ?' // the optimized query 'reasoning' => 'This query optimizes the original query by using JOINs to reduce the number of subqueries and improve the performance of the query. By using JOINs, the query can access the data from multiple tables in a single query, instead of having to make multiple subqueries.' //the reasoning behind performing such optimization, 'suggestions' => 'It may be beneficial to add an index on the `type` and `title` columns to further improve the performance of the query.' //suggestions to manually optimize the query even further ]
Credits
License
The MIT License (MIT). Please see License File for more information.
Inspiration
This package is inspired by laravel-ask-database package
Thanks for the great work Marcel Pociot