Laravel GraphQL

Use Facebook's GraphQL with Laravel 6.0+. It is based on the PHP port of GraphQL reference implementation. You can find more information about GraphQL in the GraphQL Introduction on the React blog or you can read the GraphQL specifications.

Allows creating queries and mutations as request endpoints
Supports multiple schemas
- per schema queries/mutations/types
- per schema HTTP middlewares
Custom GraphQL resolver middleware (not HTTP middleware) can be defined for each query/mutation

When using the SelectFields class for Eloquent support, additional features are available:

Queries return types, which can have custom privacy settings.
The queried fields will have the option to be retrieved dynamically from the database.

It offers following features and improvements over the original package by Folklore:

Per-operation authorization
Per-field callback defining its visibility (e.g. hiding from unauthenticated users)
SelectFields abstraction available in resolve(), allowing for advanced eager loading and thus dealing with n+1 problems
Pagination support
Server-side support for query batching
Support for file uploads

Installation

Dependencies:

Installation:

Require the package via Composer:

composer require rebing/graphql-laravel

Laravel

Publish the configuration file:

$ php artisan vendor:publish --provider="Rebing\GraphQL\GraphQLServiceProvider"

Review the configuration file:

config/graphql.php

The default GraphiQL view makes use of the global csrf_token() helper function.

Usage

Laravel GraphQL

Concepts

Before diving head first into code, it's good to familiarize yourself with the concepts surrounding GraphQL. If you've already experience with GraphQL, feel free to skip this part.

"schema"
A GraphQL schema defines all the queries, mutations and types associated with it.
"queries" and "mutations"
The "methods" you call in your GraphQL request (think about your REST endpoint)
"types"
Besides the primitive scalars like int and string, custom "shapes" can be defined and returned via custom types. They can map to your database models or basically any data you want to return.
"resolver"
Any time data is returned, it is "resolved". Usually in query/mutations this specified the primary way to retrieve your data (e.g. using SelectFields or dataloaders)

Typically, all queries/mutations/types are defined using the $attributes property and the args() / fields() methods as well as the resolve() method.

args/fields again return a configuration array for each field they supported. Those fields usually support these shapes

the "key" is the name of the field
type (required): a GraphQL specifier for the type supported here

Optional keys are:

description: made available when introspecting the GraphQL schema
resolve: override the default field resolver
deprecationReason: document why something is deprecated

Data loading

The act of loading/retrieving your data is called "resolving" in GraphQL. GraphQL itself does not define the "how" and leaves it up to the implementor.

In the context of Laravel it's natural to assume the primary source of data will be Eloquent. This library therefore provides a convenient helper called SelectFields which tries its best to eager load relations and to avoid n+1 problems.

Be aware that this is not the only way and it's also common to use concepts called "dataloaders". They usually take advantage of "deferred" executions of resolved fields, as explained in graphql-php solving n+1 problem.

The gist is that you can use any kind of data source you like (Eloquent, static data, ElasticSearch results, caching, etc.) in your resolvers but you've to be mindful of the execution model to avoid repetitive fetches and perform smart pre-fetching of your data.

GraphiQL

GraphiQL is lightweight "GraphQL IDE" in your browser. It takes advantage of the GraphQL type system and allows autocompletion of all queries/mutations/types and fields.

GraphiQL in the meantime evolved in terms of features and complexity, thus for convenience an older version is directly included with this library.

As enabled by the default configuration, it's available under the /graphiql route.

If you are using multiple schemas, you can access them via /graphiql/<schema name>.

Schemas

Schemas are required for defining GraphQL endpoints. You can define multiple schemas and assign different HTTP middleware to them, in addition to the global middleware. For example:

'schema' => 'default',

'schemas' => [
    'default' => [
        'query' => [
            ExampleQuery::class,
            // It's possible to specify a name/alias with the key
            // but this is discouraged as it prevents things
            // like improving performance with e.g. `lazyload_types=true`
            // It's recommended to specifcy just the class here and
            // rely on the `'name'` attribute in the query / type.
            'someQuery' => AnotherExampleQuery::class,
        ],
        'mutation' => [
            ExampleMutation::class,
        ],
        'types' => [
        
        ],
    ],
    'user' => [
        'query' => [
            App\GraphQL\Queries\ProfileQuery::class
        ],
        'mutation' => [

        ],
        'types' => [
        
        ],
        'middleware' => ['auth'],
    ],
],

Together with the configuration, in a way the schema defines also the route by which it is accessible. Per the default configuration of prefix = graphql, the default schema is accessible via /graphql.

Schema classes

You may alternatively define the configuration of a schema in a class that implements ConfigConvertible.

In your config, you can reference the name of the class, rather than an array.

'schemas' => [
    'default' => DefaultSchema::class
]

namespace App\GraphQL\Schemas;

use Rebing\GraphQL\Support\Contracts\ConfigConvertible;

class DefaultSchema implements ConfigConvertible
{
    public function toConfig(): array
    {
        return [
            'query' => [
                ExampleQuery::class,
            ],
            'mutation' => [
                ExampleMutation::class,
            ],
            'types' => [
            
            ],
        ];
    }
}

Creating a query

First you usually create a type you want to return from the query. The Eloquent 'model' is only required if specifying relations.

Note: The selectable key is required, if it's a non-database field or not a relation

namespace App\GraphQL\Types;

use App\User;
use GraphQL\Type\Definition\Type;
use Rebing\GraphQL\Support\Type as GraphQLType;

class UserType extends GraphQLType
{
    protected $attributes = [
        'name'          => 'User',
        'description'   => 'A user',
        // Note: only necessary if you use `SelectFields`
        'model'         => User::class,
    ];

    public function fields(): array
    {
        return [
            'id' => [
                'type' => Type::nonNull(Type::string()),
                'description' => 'The id of the user',
                // Use 'alias', if the database column is different from the type name.
                // This is supported for discrete values as well as relations.
                // - you can also use `DB::raw()` to solve more complex issues
                // - or a callback returning the value (string or `DB::raw()` result)
                'alias' => 'user_id',
            ],
            'email' => [
                'type' => Type::string(),
                'description' => 'The email of user',
                'resolve' => function($root, $args) {
                    // If you want to resolve the field yourself,
                    // it can be done here
                    return strtolower($root->email);
                }
            ],
            // Uses the 'getIsMeAttribute' function on our custom User model
            'isMe' => [
                'type' => Type::boolean(),
                'description' => 'True, if the queried user is the current user',
                'selectable' => false, // Does not try to query this from the database
            ]
        ];
    }

    // You can also resolve a field by declaring a method in the class
    // with the following format resolve[FIELD_NAME]Field()
    protected function resolveEmailField($root, $args)
    {
        return strtolower($root->email);
    }
}

The best practice is to start with your schema in config/graphql.php and add types directly to your schema (e.g. default):

'schemas' => [
    'default' => [
        // ...
        
        'types' => [
            App\GraphQL\Types\UserType::class,
        ],

Alternatively you can:

add the type on the "global" level, e.g. directly in the root config:
```
'types' => [
    App\GraphQL\Types\UserType::class,
],
```
Adding them on the global level allows to share them between different schemas but be aware this might make it harder to undertand which types/fields are used where.
or add the type with the GraphQL Facade, in a service provider for example.
```
GraphQL::addType(\App\GraphQL\Types\UserType::class);
```

As with queries/mutations, you can use an alias name (though again this prevents it from taking advantage of lazy type loading):

'schemas' => [
    'default' => [
        // ...
        
        'types' => [
            'Useralias' => App\GraphQL\Types\UserType::class,
        ],

Then you need to define a query that returns this type (or a list). You can also specify arguments that you can use in the resolve method.

namespace App\GraphQL\Queries;

use Closure;
use App\User;
use Rebing\GraphQL\Support\Facades\GraphQL;
use GraphQL\Type\Definition\ResolveInfo;
use GraphQL\Type\Definition\Type;
use Rebing\GraphQL\Support\Query;

class UsersQuery extends Query
{
    protected $attributes = [
        'name' => 'users',
    ];

    public function type(): Type
    {
        return Type::nonNull(Type::listOf(Type::nonNull(GraphQL::type('User'))));
    }

    public function args(): array
    {
        return [
            'id' => [
                'name' => 'id', 
                'type' => Type::string(),
            ],
            'email' => [
                'name' => 'email', 
                'type' => Type::string(),
            ]
        ];
    }

    public function resolve($root, $args, $context, ResolveInfo $resolveInfo, Closure $getSelectFields)
    {
        if (isset($args['id'])) {
            return User::where('id' , $args['id'])->get();
        }

        if (isset($args['email'])) {
            return User::where('email', $args['email'])->get();
        }

        return User::all();
    }
}

Add the query to the config/graphql.php configuration file

'schemas' => [
    'default' => [
        'query' => [
            App\GraphQL\Queries\UsersQuery::class
        ],
        // ...
    ]
]

And that's it. You should be able to query GraphQL with a request to the url /graphql (or anything you choose in your config). Try a GET request with the following query input

query FetchUsers {
    users {
        id
        email
    }
}

For example, if you use homestead:

http://homestead.app/graphql?query=query+FetchUsers{users{id,email}}

Creating a mutation

A mutation is like any other query. It accepts arguments and returns an object of a certain type. Mutations are meant to be used for operations modifying (mutating) the state on the server (which queries are not supposed to perform).

This is conventional abstraction, technically you can do anything you want in a query resolve, including mutating state.

For example, a mutation to update the password of a user. First you need to define the Mutation:

namespace App\GraphQL\Mutations;

use Closure;
use App\User;
use GraphQL;
use GraphQL\Type\Definition\Type;
use GraphQL\Type\Definition\ResolveInfo;
use Rebing\GraphQL\Support\Mutation;

class UpdateUserPasswordMutation extends Mutation
{
    protected $attributes = [
        'name' => 'updateUserPassword'
    ];

    public function type(): Type
    {
        return Type::nonNull(GraphQL::type('User'));
    }

    public function args(): array
    {
        return [
            'id' => ['
                name' => 'id', 
                'type' => Type::nonNull(Type::string()),
            ],
            'password' => [
                'name' => 'password', 
                'type' => Type::nonNull(Type::string()),
            ]
        ];
    }

    public function resolve($root, $args, $context, ResolveInfo $resolveInfo, Closure $getSelectFields)
    {
        $user = User::find($args['id']);
        if(!$user) {
            return null;
        }

        $user->password = bcrypt($args['password']);
        $user->save();

        return $user;
    }
}

As you can see in the resolve() method, you use the arguments to update your model and return it.

You should then add the mutation to the config/graphql.php configuration file:

'schemas' => [
    'default' => [
        'mutation' => [
            App\GraphQL\Mutations\UpdateUserPasswordMutation::class,
        ],
        // ...
    ]
]

You can then use the following query on your endpoint to do the mutation:

mutation users {
    updateUserPassword(id: "1", password: "newpassword") {
        id
        email
    }
}

if you use homestead:

http://homestead.app/graphql?query=mutation+users{updateUserPassword(id: "1", password: "newpassword"){id,email}}

Adding validation to a mutation

It is possible to add validation rules to arguments of any query or mutation. It uses the Laravel Validator to perform validation against the $args.

You can define rules on each argument:

class UpdateUserEmailMutation extends Mutation
{
    //...

    public function args(): array
    {
        return [
            'id' => [
                'name' => 'id',
                'type' => Type::string(),
                'rules' => ['required']
            ],
            'email' => [
                'name' => 'email',
                'type' => Type::string(),
                'rules' => ['required', 'email']
            ]
        ];
    }

    //...
}

Alternatively, uou can add a method to define the validation rules that apply, giving you more freedom on composing the rules (e.g. multi-field validations).

namespace App\GraphQL\Mutations;

use Closure;
use App\User;
use GraphQL;
use GraphQL\Type\Definition\ResolveInfo;
use GraphQL\Type\Definition\Type;
use Rebing\GraphQL\Support\Mutation;

class UpdateUserEmailMutation extends Mutation
{
    protected $attributes = [
        'name' => 'updateUserEmail'
    ];

    public function type(): Type
    {
        return GraphQL::type('User');
    }

    public function args(): array
    {
        return [
            'id' => [
                'name' => 'id', 
                'type' => Type::string(),
            ],
            'email' => [
                'name' => 'email', 
                'type' => Type::string(),
            ]
        ];
    }

    protected function rules(array $args = []): array
    {
        return [
            'id' => ['required'],
            'email' => ['required', 'email'],
            'password' => function($inputArguments, $mutationArguments) {
                if ($inputArguments['id'] !== 1337) {
                    return ['required'];
                }
                return [];
            }
        ];
    }

    public function resolve($root, $args, $context, ResolveInfo $resolveInfo, Closure $getSelectFields)
    {
        $user = User::find($args['id']);
        if (!$user) {
            return null;
        }

        $user->email = $args['email'];
        $user->save();

        return $user;
    }
}

Note: the following is a non-standard extension to the GraphQL error response!

When you execute a mutation, it will return any validation errors that occur. Since the GraphQL specification defines a certain format for errors, the validation errors are added to the error object as an extra validation attribute. To find the validation error, you should check for the error with a message equals to 'validation', then the validation attribute will contain the normal errors messages returned by the Laravel Validator:

{
    "data": {
        "updateUserEmail": null
    },
    "errors": [
        {
            "message": "validation",
            "locations": [
                {
                    "line": 1,
                    "column": 20
                }
            ],
            "validation": {
                "email": [
                    "The email is invalid."
                ]
            }
        }
    ]
}

The validation errors returned can be customised by overriding the validationErrorMessages method. This method should return an array of custom validation messages in the same way documented by Laravel's validation. For example, to check an email argument doesn't conflict with any existing data, you could perform the following:

Note: the keys should be in field_name.validator_type format so you can return specific errors per validation type.

public function validationErrorMessages(array $args = []): array
{
    return [
        'name.required' => 'Please enter your full name',
        'name.string' => 'Your name must be a valid string',
        'email.required' => 'Please enter your email address',
        'email.email' => 'Please enter a valid email address',
        'email.exists' => 'Sorry, this email address is already in use',
    ];
}

File uploads

This library provides a middleware compliant with the spec at https://github.com/jaydenseric/graphql-multipart-request-spec .

You have to add the \Rebing\GraphQL\Support\UploadType first to your config/graphql schema types definition (either global or in your schema):

'types' => [
    \Rebing\GraphQL\Support\UploadType::class,
],

It is important that you send the request as multipart/form-data:

WARNING: when you are uploading files, Laravel will use FormRequest - it means that middlewares which are changing request, will not have any effect.

namespace App\GraphQL\Mutations;

use Closure;
use GraphQL;
use GraphQL\Type\Definition\ResolveInfo;
use GraphQL\Type\Definition\Type;
use Rebing\GraphQL\Support\Mutation;

class UserProfilePhotoMutation extends Mutation
{
    protected $attributes = [
        'name' => 'userProfilePhoto',
    ];

    public function type(): Type
    {
        return GraphQL::type('User');
    }

    public function args(): array
    {
        return [
            'profilePicture' => [
                'name' => 'profilePicture',
                'type' => GraphQL::type('Upload'),
                'rules' => ['required', 'image', 'max:1500'],
            ],
        ];
    }

    public function resolve($root, $args, $context, ResolveInfo $resolveInfo, Closure $getSelectFields)
    {
        $file = $args['profilePicture'];

        // Do something with file here...
    }
}

Note: You can test your file upload implementation using Altair as explained here.

Vue.js and Axios example

<template>
  <div class="input-group">
    <div class="custom-file">
      <input type="file" class="custom-file-input" id="uploadFile" ref="uploadFile" @change="handleUploadChange">
      <label class="custom-file-label" for="uploadFile">
        Drop Files Here to upload
      </label>
    </div>
    <div class="input-group-append">
      <button class="btn btn-outline-success" type="button" @click="upload">Upload</button>
    </div>
  </div>
</template>

<script>
  export default {
    name: 'FileUploadExample',
    data() {
      return {
        file: null,
      };
    },
    methods: {
      handleUploadChange() {
        this.file = this.$refs.uploadFile.files[0];
      },
      async upload() {
        if (!this.file) {
          return;
        }
        // Creating form data object
        let bodyFormData = new FormData();
        bodyFormData.set('operations', JSON.stringify({
                   // Mutation string
            'query': `mutation uploadSingleFile($file: Upload!) {
                        upload_single_file  (attachment: $file)
                      }`,
            'variables': {"attachment": this.file}
        }));
        bodyFormData.set('operationName', null);
        bodyFormData.set('map', JSON.stringify({"file":["variables.file"]}));
        bodyFormData.append('file', this.file);

        // Post the request to GraphQL controller
        let res = await axios.post('/graphql', bodyFormData, {
          headers: {
            "Content-Type": "multipart/form-data"
          }
        });

        if (res.data.status.code == 200) {
          // On success file upload
          this.file = null;
        }
      }
    }
  }
</script>

<style scoped>
</style>

jQuery or vanilla javascript

<input type="file" id="fileUpload">

// Get the file from input element
// In jQuery:
let file = $('#fileUpload').prop('files')[0];
// Vanilla JS:
let file = document.getElementById("fileUpload").files[0];

// Create a FormData object
let bodyFormData = new FormData();
bodyFormData.set('operations', JSON.stringify({
         // Mutation string
  'query': `mutation uploadSingleFile($file: Upload!) {
              upload_single_file  (attachment: $file)
            }`,
  'variables': {"attachment": this.file}
}));
bodyFormData.set('operationName', null);
bodyFormData.set('map', JSON.stringify({"file":["variables.file"]}));
bodyFormData.append('file', this.file);

// Post the request to GraphQL controller via Axios, jQuery.ajax, or vanilla XMLHttpRequest
let res = await axios.post('/graphql', bodyFormData, {
  headers: {
    "Content-Type": "multipart/form-data"
  }
});

Resolve method

The resolve method is used in both queries and mutations, and it's here that responses are created.

The first three parameters to the resolve method are hard-coded:

The $root object this resolve method belongs to (can be null)
The arguments passed as array $args (can be an empty array)
The query specific GraphQL context, can be customized by overriding \Rebing\GraphQL\GraphQLController::queryContext

Arguments here after will be attempted to be injected, similar to how controller methods works in Laravel.

You can typehint any class that you will need an instance of.

There are two hardcoded classes which depend on the local data for the query:

GraphQL\Type\Definition\ResolveInfo has information useful for field resolution process.
Rebing\GraphQL\Support\SelectFields allows eager loading of related Eloquent models, see Eager loading relationships.

Example:

namespace App\GraphQL\Queries;

use Closure;
use App\User;
use GraphQL;
use GraphQL\Type\Definition\Type;
use GraphQL\Type\Definition\ResolveInfo;
use Rebing\GraphQL\Support\SelectFields;
use Rebing\GraphQL\Support\Query;
use SomeClassNamespace\SomeClassThatDoLogging;

class UsersQuery extends Query
{
    protected $attributes = [
        'name' => 'users',
    ];

    public function type(): Type
    {
        return Type::listOf(GraphQL::type('User'));
    }

    public function args(): array
    {
        return [
            'id' => [
                'name' => 'id', 
                'type' => Type::string(),
            ]
        ];
    }

    public function resolve($root, $args, $context, ResolveInfo $info, SelectFields $fields, SomeClassThatDoLogging $logging)
    {
        $logging->log('fetched user');

        $select = $fields->getSelect();
        $with = $fields->getRelations();

        $users = User::select($select)->with($with);

        return $users->get();
    }
}

Resolver middleware

These are GraphQL specific resolver middlewares and are only conceptually related to Laravels "HTTP middleware". The main difference:

Laravels HTTP middleware:
- works on the schema / route level
- is compatible with any regular Laravel HTTP middleware
- is the same for all queries/mutations in a schema
Resolver middleware
- Works similar in concept
- But applies on the query/mutation level, i.e. can be different for every query/mutation
- Is technically not compatible with HTTP middleware
- Takes different arguments

Defining middleware

To create a new middleware, use the make:graphql:middleware Artisan command

php artisan make:graphql:middleware ResolvePage

This command will place a new ResolvePage class within your app/GraphQL/Middleware directory. In this middleware, we will set the Paginator current page to the argument we accept via our PaginationType:

namespace App\GraphQL\Middleware;

use Closure;
use GraphQL\Type\Definition\ResolveInfo;
use Illuminate\Pagination\Paginator;
use Rebing\GraphQL\Support\Middleware;

class ResolvePage extends Middleware
{
    public function handle($root, $args, $context, ResolveInfo $info, Closure $next)
    {
        Paginator::currentPageResolver(function () use ($args) {
            return $args['pagination']['page'] ?? 1;
        });

        return $next($root, $args, $context, $info);
    }
}

Registering middleware

If you would like to assign middleware to specific queries/mutations, list the middleware class in the $middleware property of your query class.

namespace App\GraphQL\Queries;

use App\GraphQL\Middleware;
use Rebing\GraphQL\Support\Query;
use Rebing\GraphQL\Support\Query;

class UsersQuery extends Query
{
    protected $middleware = [
        Middleware\Logstash::class,
        Middleware\ResolvePage::class,
    ];
}

If you want a middleware to run during every GraphQL query/mutation to your application, list the middleware class in the $middleware property of your base query class.

namespace App\GraphQL\Queries;

use App\GraphQL\Middleware;
use Rebing\GraphQL\Support\Query as BaseQuery;

abstract class Query extends BaseQuery
{
    protected $middleware = [
        Middleware\Logstash::class,
        Middleware\ResolvePage::class,
    ];
}

Alternatively, you can override getMiddleware to supply your own logic:

    protected function getMiddleware(): array
    {
        return array_merge([...], $this->middleware);
    }

Terminable middleware

Sometimes a middleware may need to do some work after the response has been sent to the browser. If you define a terminate method on your middleware and your web server is using FastCGI, the terminate method will automatically be called after the response is sent to the browser:

namespace App\GraphQL\Middleware;

use Countable;
use GraphQL\Language\Printer;
use GraphQL\Type\Definition\ResolveInfo;
use Illuminate\Contracts\Pagination\LengthAwarePaginator;
use Illuminate\Pagination\AbstractPaginator;
use Illuminate\Support\Arr;
use Illuminate\Support\Facades\Log;
use Illuminate\Support\Facades\Route;
use Rebing\GraphQL\Support\Middleware;

class Logstash extends Middleware
{
    public function terminate($root, $args, $context, ResolveInfo $info, $result): void
    {
        Log::channel('logstash')->info('', (
            collect([
                'query' => $info->fieldName,
                'operation' => $info->operation->name->value ?? null,
                'type' => $info->operation->operation,
                'fields' => array_keys(Arr::dot($info->getFieldSelection($depth = PHP_INT_MAX))),
                'schema' => Arr::first(Route::current()->parameters()) ?? config('graphql.default_schema'),
                'vars' => $this->formatVariableDefinitions($info->operation->variableDefinitions),
            ])
                ->when($result instanceof Countable, function ($metadata) use ($result) {
                    return $metadata->put('count', $result->count());
                })
                ->when($result instanceof AbstractPaginator, function ($metadata) use ($result) {
                    return $metadata->put('per_page', $result->perPage());
                })
                ->when($result instanceof LengthAwarePaginator, function ($metadata) use ($result) {
                    return $metadata->put('total', $result->total());
                })
                ->merge($this->formatArguments($args))
                ->toArray()
        ));
    }

    private function formatArguments(array $args): array
    {
        return collect(Arr::sanitize($args))
            ->mapWithKeys(function ($value, $key) {
                return ["\${$key}" => $value];
            })
            ->toArray();
    }

    private function formatVariableDefinitions(?iterable $variableDefinitions = []): array
    {
        return collect($variableDefinitions)
            ->map(function ($def) {
                return Printer::doPrint($def);
            })
            ->toArray();
    }
}

The terminate method receives both the resolver arguments and the query result.

Once you have defined a terminable middleware, you should add it to the list of middleware in your queries and mutations.

Authorization

For authorization similar to Laravel's Request (or middleware) functionality, we can override the authorize() function in a Query or Mutation. An example of Laravel's 'auth' middleware:

namespace App\GraphQL\Queries;

use Auth;
use Closure;
use GraphQL\Type\Definition\ResolveInfo;

class UsersQuery extends Query
{
    public function authorize($root, array $args, $ctx, ResolveInfo $resolveInfo = null, Closure $getSelectFields = null): bool
    {
        // true, if logged in
        return ! Auth::guest();
    }

    // ...
}

Or we can make use of arguments passed via the GraphQL query:

namespace App\GraphQL\Queries;

use Auth;
use Closure;
use GraphQL\Type\Definition\ResolveInfo;

class UsersQuery extends Query
{
    public function authorize($root, array $args, $ctx, ResolveInfo $resolveInfo = null, Closure $getSelectFields = null): bool
    {
        if (isset($args['id'])) {
            return Auth::id() == $args['id'];
        }

        return true;
    }

    // ...
}

You can also provide a custom error message when the authorization fails (defaults to Unauthorized):

namespace App\GraphQL\Queries;

use Auth;
use Closure;
use GraphQL\Type\Definition\ResolveInfo;

class UsersQuery extends Query
{
    public function authorize($root, array $args, $ctx, ResolveInfo $resolveInfo = null, Closure $getSelectFields = null): bool
    {
        if (isset($args['id'])) {
            return Auth::id() == $args['id'];
        }

        return true;
    }

    public function getAuthorizationMessage(): string
    {
        return 'You are not authorized to perform this action';
    }

    // ...
}

Privacy

Note: this only applies when making use of the SelectFields class to query Eloquent models!

You can set custom privacy attributes for every Type's Field. If a field is not allowed, null will be returned. For example, if you want the user's email to only be accessible to themselves:

class UserType extends GraphQLType
{
    // ...

    public function fields(): array
    {
        return [
            'id' => [
                'type'          => Type::nonNull(Type::string()),
                'description'   => 'The id of the user'
            ],
            'email' => [
                'type'          => Type::string(),
                'description'   => 'The email of user',
                'privacy'       => function(array $args, $ctx): bool {
                    return $args['id'] == Auth::id();
                }
            ]
        ];
    }

    // ...

}

or you can create a class that extends the abstract GraphQL Privacy class:

use Auth;
use Rebing\GraphQL\Support\Privacy;

class MePrivacy extends Privacy
{
    public function validate(array $queryArgs, $queryContext = null): bool
    {
        return $queryArgs['id'] == Auth::id();
    }
}

use MePrivacy;

class UserType extends GraphQLType
{

    // ...

    public function fields(): array
    {
        return [
            'id' => [
                'type'          => Type::nonNull(Type::string()),
                'description'   => 'The id of the user'
            ],
            'email' => [
                'type'          => Type::string(),
                'description'   => 'The email of user',
                'privacy'       => MePrivacy::class,
            ]
        ];
    }

    // ...

}

Query variables

GraphQL offers you the possibility to use variables in your query so you don't need to "hardcode" value. This is done like that:

query FetchUserByID($id: String)
{
    user(id: $id) {
        id
        email
    }
}

When you query the GraphQL endpoint, you can pass a JSON encoded variables parameter.

http://homestead.app/graphql?query=query+FetchUserByID($id:Int){user(id:$id){id,email}}&params={"id":123}

Custom field

You can also define a field as a class if you want to reuse it in multiple types.

namespace App\GraphQL\Fields;

use GraphQL\Type\Definition\Type;
use Rebing\GraphQL\Support\Field;

class PictureField extends Field
{
    protected $attributes = [
        'description'   => 'A picture',
    ];

    public function type(): Type
    {
        return Type::string();
    }

    public function args(): array
    {
        return [
            'width' => [
                'type' => Type::int(),
                'description' => 'The width of the picture'
            ],
            'height' => [
                'type' => Type::int(),
                'description' => 'The height of the picture'
            ]
        ];
    }

    protected function resolve($root, $args)
    {
        $width = isset($args['width']) ? $args['width']:100;
        $height = isset($args['height']) ? $args['height']:100;

        return 'http://placehold.it/'.$width.'x'.$height;
    }
}

You can then use it in your type declaration

namespace App\GraphQL\Types;

use App\GraphQL\Fields\PictureField;
use App\User;
use GraphQL\Type\Definition\Type;
use Rebing\GraphQL\Support\Type as GraphQLType;

class UserType extends GraphQLType
{
    protected $attributes = [
        'name'          => 'User',
        'description'   => 'A user',
        'model'         => User::class,
    ];

    public function fields(): array
    {
        return [
            'id' => [
                'type' => Type::nonNull(Type::string()),
                'description' => 'The id of the user'
            ],
            'email' => [
                'type' => Type::string(),
                'description' => 'The email of user'
            ],
            //Instead of passing an array, you pass a class path to your custom field
            'picture' => PictureField::class
        ];
    }
}

Even better reusable fields

Instead of using the class name, you can also supply an actual instance of the Field. This allows you to not only re-use the field, but will also open up the possibility to re-use the resolver.

Let's imagine we want a field type that can output dates formatted in all sorts of ways.

namespace App\GraphQL\Fields;

use GraphQL\Type\Definition\Type;
use Rebing\GraphQL\Support\Field;

class FormattableDate extends Field
{
    protected $attributes = [
        'description' => 'A field that can output a date in all sorts of ways.',
    ];

    public function __construct(array $settings = [])
    {
        $this->attributes = \array_merge($this->attributes, $settings);
    }

    public function type(): Type
    {
        return Type::string();
    }

    public function args(): array
    {
        return [
            'format' => [
                'type' => Type::string(),
                'defaultValue' => 'Y-m-d H:i',
                'description' => 'Defaults to Y-m-d H:i',
            ],
            'relative' => [
                'type' => Type::boolean(),
                'defaultValue' => false,
            ],
        ];
    }

    protected function resolve($root, $args): ?string
    {
        $date = $root->{$this->getProperty()};

        if (!$date instanceof Carbon) {
            return null;
        }

        if ($args['relative']) {
            return $date->diffForHumans();
        }

        return $date->format($args['format']);
    }

    protected function getProperty(): string
    {
        return $this->attributes['alias'] ?? $this->attributes['name'];
    }
}

You can use this field in your type as follows:

namespace App\GraphQL\Types;

use App\GraphQL\Fields\FormattableDate;
use App\User;
use GraphQL\Type\Definition\Type;
use Rebing\GraphQL\Support\Type as GraphQLType;

class UserType extends GraphQLType
{
    protected $attributes = [
        'name'          => 'User',
        'description'   => 'A user',
        'model'         => User::class,
    ];

    public function fields(): array
    {
        return [
            'id' => [
                'type' => Type::nonNull(Type::string()),
                'description' => 'The id of the user'
            ],
            'email' => [
                'type' => Type::string(),
                'description' => 'The email of user'
            ],

            // You can simply supply an instance of the class
            'dateOfBirth' => new FormattableDate,

            // Because the constructor of `FormattableDate` accepts our the array of parameters,
            // we can override them very easily.
            // Imagine we want our field to be called `createdAt`, but our database column
            // is called `created_at`:
            'createdAt' => new FormattableDate([
                'alias' => 'created_at',
            ])
        ];
    }
}

Eager loading relationships

The Rebing\GraphQL\Support\SelectFields class allows to eager load related Eloquent models. Only the required fields will be queried from the database.

The class can be instantiated by typehinting SelectFields $selectField in your resolve method.

You can also construct the class by typehinting a Closure. The Closure accepts an optional parameter for the depth of the query to analyse.

Your Query would look like:

namespace App\GraphQL\Queries;

use Closure;
use App\User;
use GraphQL;
use GraphQL\Type\Definition\Type;
use GraphQL\Type\Definition\ResolveInfo;
use Rebing\GraphQL\Support\SelectFields;
use Rebing\GraphQL\Support\Query;

class UsersQuery extends Query
{
    protected $attributes = [
        'name' => 'users',
    ];

    public function type(): Type
    {
        return Type::listOf(GraphQL::type('User'));
    }

    public function args(): array
    {
        return [
            'id' => [
                'name' => 'id', 
                'type' => Type::string(),
            ],
            'email' => [
                'name' => 'email', 
                'type' => Type::string(),
            ]
        ];
    }

    public function resolve($root, $args, $context, ResolveInfo $info, Closure $getSelectFields)
    {
        // $info->getFieldSelection($depth = 3);

        // If your GraphQL query exceeds the default nesting query, you can increase it here:
        // $fields = $getSelectFields(11);

        /** @var SelectFields $fields */
        $fields = $getSelectFields();
        $select = $fields->getSelect();
        $with = $fields->getRelations();

        $users = User::select($select)->with($with);

        return $users->get();
    }
}

Your Type for User might look like shown below. The profile and posts relations must also exist in the UserModel's relations. If some fields are required for the relation to load or validation etc, then you can define an always attribute that will add the given attributes to select.

The attribute can be a comma separated string or an array of attributes to always include.

namespace App\GraphQL\Types;

use App\User;
use GraphQL\Type\Definition\Type;
use Rebing\GraphQL\Support\Facades\GraphQL;
use Rebing\GraphQL\Support\Type as GraphQLType;

class UserType extends GraphQLType
{
    /**
     * @var array
     */
    protected $attributes = [
        'name'          => 'User',
        'description'   => 'A user',
        'model'         => User::class,
    ];

    /**
    * @return array
    */
    public function fields(): array
    {
        return [
            'uuid' => [
                'type' => Type::nonNull(Type::string()),
                'description' => 'The uuid of the user'
            ],
            'email' => [
                'type' => Type::nonNull(Type::string()),
                'description' => 'The email of user'
            ],
            'profile' => [
                'type' => GraphQL::type('Profile'),
                'description' => 'The user profile',
            ],
            'posts' => [
                'type' => Type::listOf(GraphQL::type('Post')),
                'description' => 'The user posts',
                // Can also be defined as a string
                'always' => ['title', 'body'],
            ]
        ];
    }
}

At this point we have a profile and a post type as expected for any model

class ProfileType extends GraphQLType
{
    protected $attributes = [
        'name'          => 'Profile',
        'description'   => 'A user profile',
        'model'         => UserProfileModel::class,
    ];

    public function fields(): array
    {
        return [
            'name' => [
                'type' => Type::string(),
                'description' => 'The name of user'
            ]
        ];
    }
}

class PostType extends GraphQLType
{
    protected $attributes = [
        'name'          => 'Post',
        'description'   => 'A post',
        'model'         => PostModel::class,
    ];

    public function fields(): array
    {
        return [
            'title' => [
                'type' => Type::nonNull(Type::string()),
                'description' => 'The title of the post'
            ],
            'body' => [
                'type' => Type::string(),
                'description' => 'The body the post'
            ]
        ];
    }
}

Type relationship query

Note: this only applies when making use of the SelectFields class to query Eloquent models!

You can also specify the query that will be included with a relationship via Eloquent's query builder:

class UserType extends GraphQLType
{

    // ...

    public function fields(): array
    {
        return [
            // ...

            // Relation
            'posts' => [
                'type'          => Type::listOf(GraphQL::type('Post')),
                'description'   => 'A list of posts written by the user',
                'args'          => [
                    'date_from' => [
                        'type' => Type::string(),
                    ],
                 ],
                // $args are the local arguments passed to the relation
                // $query is the relation builder object
                // $ctx is the GraphQL context (can be customized by overriding `\Rebing\GraphQL\GraphQLController::queryContext`
                'query'         => function(array $args, $query, $ctx) {
                    return $query->where('posts.created_at', '>', $args['date_from']);
                }
            ]
        ];
    }
}

Pagination

Pagination will be used, if a query or mutation returns a PaginationType.

Note that unless you use resolver middleware, you will have to manually supply both the limit and page values:

namespace App\GraphQL\Queries;

use Closure;
use GraphQL\Type\Definition\ResolveInfo;
use GraphQL\Type\Definition\Type;
use Rebing\GraphQL\Support\Facades\GraphQL;
use Rebing\GraphQL\Support\Query;

class PostsQuery extends Query
{
    public function type(): Type
    {
        return GraphQL::paginate('posts');
    }

    // ...

    public function resolve($root, $args, $context, ResolveInfo $info, Closure $getSelectFields)
    {
        $fields = $getSelectFields();

        return Post::with($fields->getRelations())
            ->select($fields->getSelect())
            ->paginate($args['limit'], ['*'], 'page', $args['page']);
    }
}

Query posts(limit:10,page:1){data{id},total,per_page} might return

{
    "data": {
        "posts: [
            "data": [
                {"id": 3},
                {"id": 5},
                ...
            ],
            "total": 21,
            "per_page": 10
        ]
    }
}

Note that you need to add in the extra 'data' object when you request paginated resources as the returned data gives you the paginated resources in a data object at the same level as the returned pagination metadata.

Simple Pagination will be used, if a query or mutation returns a SimplePaginationType.

namespace App\GraphQL\Queries;

use Closure;
use GraphQL\Type\Definition\ResolveInfo;
use GraphQL\Type\Definition\Type;
use Rebing\GraphQL\Support\Facades\GraphQL;
use Rebing\GraphQL\Support\Query;

class PostsQuery extends Query
{
    public function type(): Type
    {
        return Type::nonNull(GraphQL::simplePaginate('posts'));
    }

    // ...

    public function resolve($root, $args, $context, ResolveInfo $info, Closure $getSelectFields)
    {
        $fields = $getSelectFields();

        return Post::with($fields->getRelations())
            ->select($fields->getSelect())
            ->simplePaginate($args['limit'], ['*'], 'page', $args['page']);
    }
}

Batching

Batched requests are required to be sent via a POST request.

You can send multiple queries (or mutations) at once by grouping them together. Therefore, instead of creating two HTTP requests:

POST
{
    query: "query postsQuery { posts { id, comment, author_id } }"
}

POST
{
    query: "mutation storePostMutation($comment: String!) { store_post(comment: $comment) { id } }",
    variables: { "comment": "Hi there!" }
}

you could batch it as one

POST
[
    {
        query: "query postsQuery { posts { id, comment, author_id } }"
    },
    {
        query: "mutation storePostMutation($comment: String!) { store_post(comment: $comment) { id } }",
        variables: { "comment": "Hi there!" }
    }
]

For systems sending multiple requests at once, this can help performance by batching together queries that will be made within a certain interval of time.

There are tools that help with this and can handle the batching for you, e.g. Apollo

A note on query batching: whilst it may look like an "only win" situations, there are possible downsides using batching:

All queries/mutations are executed in the same "process execution context".
If your code has side-effects which might not show up in the usual FastCGI environment (single request/response), it may cause issues here.

The "HTTP middleware" is only executed for the whole batch once
In case you would expect it being triggered for each query/mutation included. This may be especially relevant for logging or rate limiting.
OTOH with "resolver middleware" this will work as expected (though the solve different problems).

No limitations on the number of queries/mutations
Currently there's no way to limit this.

Support for batching can be disabled by setting the config batching.enable to false.

Scalar types

GraphQL comes with built-in scalar types for string, int, boolean, etc. It's possible to create custom scalar types to special purpose fields.

An example could be a link: instead of using Type::string() you could create a scalar type Link and reference it with GraphQL::type('Link').

The benefits would be:

a dedicated description so you can give more meaning/purpose to a field than just call it a string type
explicit conversion logic for the following steps:
- converting from the internal logic to the serialized GraphQL output (serialize)
- query/field input argument conversion (parseLiteral)
- when passed as variables to your query (parseValue)

This also means validation logic can be added within these methods to ensure that the value delivered/received is e.g. a true link.

A scalar type has to implement all the methods; you can quick start this with artisan make:graphql:scalar <typename>. Then just add the scalar to your existing types in the schema.

For more advanced use, please refer to the official documentation regarding scalar types.

A note on performance: be mindful of the code you include in your scalar types methods. If you return a large number of fields making use of custom scalars which includes complex logic to validate field, it might impact your response times.

Enums

Enumeration types are a special kind of scalar that is restricted to a particular set of allowed values. Read more about Enums here

First create an Enum as an extension of the GraphQLType class:

namespace App\GraphQL\Enums;

use Rebing\GraphQL\Support\EnumType;

class EpisodeEnum extends EnumType
{
    protected $attributes = [
        'name' => 'episode',
        'description' => 'The types of demographic elements',
        'values' => [
            'NEWHOPE' => 'NEWHOPE',
            'EMPIRE' => 'EMPIRE',
            'JEDI' => 'JEDI',
        ],
    ];
}

Note: within the $attributes['values'] array the key is enum value the GraphQL client will be able to choose from, while the value is what will your server receive (what will enum be resolved to).

The Enum will be registered like any other type in your schema in config/graphq.php:

'schemas' => [
    'default' => [
        'types' => [
            EpisodeEnum::class,
        ],

Then use it like:

namespace App\GraphQL\Types;

use Rebing\GraphQL\Support\Type as GraphQLType;

class TestType extends GraphQLType
{
    public function fields(): array
    {
        return [
            'episode_type' => [
                'type' => GraphQL::type('EpisodeEnum')
            ]
        ];
    }
}

Unions

A Union is an abstract type that simply enumerates other Object Types. The value of Union Type is actually a value of one of included Object Types.

It's useful if you need to return unrelated types in the same Query. For example when implementing a search for multiple different entities.

Example for defining a UnionType:

namespace App\GraphQL\Unions;

use App\Post;
use GraphQL;
use Rebing\GraphQL\Support\UnionType;

class SearchResultUnion extends UnionType
{
    protected $attributes = [
        'name' => 'searchResult',
    ];

    public function types(): array
    {
        return [
            GraphQL::type('Post'),
            GraphQL::type('Episode'),
        ];
    }

    public function resolveType($value)
    {
        if ($value instanceof Post) {
            return GraphQL::type('Post');
        } elseif ($value instanceof Episode) {
            return GraphQL::type('Episode');
        }
    }
}

Interfaces

You can use interfaces to abstract a set of fields. Read more about Interfaces here

An implementation of an interface:

namespace App\GraphQL\Interfaces;

use GraphQL;
use GraphQL\Type\Definition\Type;
use Rebing\GraphQL\Support\InterfaceType;

class CharacterInterface extends InterfaceType
{
    protected $attributes = [
        'name' => 'character',
        'description' => 'Character interface.',
    ];

    public function fields(): array
    {
        return [
            'id' => [
                'type' => Type::nonNull(Type::int()),
                'description' => 'The id of the character.'
            ],
            'name' => Type::string(),
            'appearsIn' => [
                'type' => Type::nonNull(Type::listOf(GraphQL::type('Episode'))),
                'description' => 'A list of episodes in which the character has an appearance.'
            ],
        ];
    }

    public function resolveType($root)
    {
        // Use the resolveType to resolve the Type which is implemented trough this interface
        $type = $root['type'];
        if ($type === 'human') {
            return GraphQL::type('Human');
        } elseif  ($type === 'droid') {
            return GraphQL::type('Droid');
        }
    }
}

A Type that implements an interface:

namespace App\GraphQL\Types;

use GraphQL;
use Rebing\GraphQL\Support\Type as GraphQLType;
use GraphQL\Type\Definition\Type;

class HumanType extends GraphQLType
{
    protected $attributes = [
        'name' => 'human',
        'description' => 'A human.'
    ];

    public function fields(): array
    {
        return [
            'id' => [
                'type' => Type::nonNull(Type::int()),
                'description' => 'The id of the human.',
            ],
            'name' => Type::string(),
            'appearsIn' => [
                'type' => Type::nonNull(Type::listOf(GraphQL::type('Episode'))),
                'description' => 'A list of episodes in which the human has an appearance.'
            ],
            'totalCredits' => [
                'type' => Type::nonNull(Type::int()),
                'description' => 'The total amount of credits this human owns.'
            ]
        ];
    }

    public function interfaces(): array
    {
        return [
            GraphQL::type('Character')
        ];
    }
}

Supporting custom queries on interface relations

If an interface contains a relation with a custom query, it's required to implement public function types() returning an array of GraphQL::type(), i.e. all the possible types it may resolve to (quite similar as it works for unions) so that it works correctly with SelectFields.

Based on the previous code example, the method would look like:

    public function types(): array
    {
        return[
            GraphQL::type('Human'),
            GraphQL::type('Droid'),
        ];
    }

Sharing interface fields

Since you often have to repeat many of the field definitons of the Interface in the concrete types, it makes sense to share the definitions of the Interface. You can access and reuse specific interface fields with the method getField(string fieldName): FieldDefinition. To get all fields as an array use getFields(): array

With this you could write the fields method of your HumanType class like this:

public function fields(): array
{
    $interface = GraphQL::type('Character');

    return [
        $interface->getField('id'),
        $interface->getField('name'),
        $interface->getField('appearsIn'),

        'totalCredits' => [
            'type' => Type::nonNull(Type::int()),
            'description' => 'The total amount of credits this human owns.'
        ]
    ];
}

Or by using the getFields method:

public function fields(): array
{
    $interface = GraphQL::type('Character');

    return array_merge($interface->getFields(), [
        'totalCredits' => [
            'type' => Type::nonNull(Type::int()),
            'description' => 'The total amount of credits this human owns.'
        ]
    ]);
}

Input Object

Input Object types allow you to create complex inputs. Fields have no args or resolve options and their type must be InputType. You can add rules option if you want to validate input data. Read more about Input Object here

First create an InputObjectType as an extension of the GraphQLType class:

namespace App\GraphQL\InputObject;

use GraphQL\Type\Definition\Type;
use Rebing\GraphQL\Support\InputType;

class ReviewInput extends InputType
{
    protected $attributes = [
        'name' => 'reviewInput',
        'description' => 'A review with a comment and a score (0 to 5)'
    ];

    public function fields(): array
    {
        return [
            'comment' => [
                'name' => 'comment',
                'description' => 'A comment (250 max chars)',
                'type' => Type::string(),
                // You can define Laravel Validation here
                'rules' => ['max:250']
            ],
            'score' => [
                'name' => 'score',
                'description' => 'A score (0 to 5)',
                'type' => Type::int(),
                'rules' => ['min:0', 'max:5']
            ]
        ];
    }
}

The Input Object will be registered like any other type in your schema in config/graphq.php:

'schemas' => [
    'default' => [
        'types' => [
            'ReviewInput' => ReviewInput::class
        ],

Then use it in a mutation, like:

// app/GraphQL/Type/TestMutation.php
class TestMutation extends GraphQLType {

    public function args(): array
    {
        return [
            'review' => [
                'type' => GraphQL::type('ReviewInput')
            ]
        ]
    }

}

Type modifiers

Type modifiers can be applied by wrapping your chosen type in Type::nonNull or Type::listOf calls or alternatively you can use the shorthand syntax available via GraphQL::type to build up more complex types.

GraphQL::type('MyInput!');
GraphQL::type('[MyInput]');
GraphQL::type('[MyInput]!');
GraphQL::type('[MyInput!]!');

GraphQL::type('String!');
GraphQL::type('[String]');
GraphQL::type('[String]!');
GraphQL::type('[String!]!');

Field and input alias

It is possible to alias query and mutation arguments as well as input object fields.

It can be especially useful for mutations saving data to the database.

Here you might want the input names to be different from the column names in the database.

Example, where the database columns are first_name and last_name:

namespace App\GraphQL\InputObject;

use GraphQL\Type\Definition\Type;
use Rebing\GraphQL\Support\InputType;

class UserInput extends InputType
{
    protected $attributes = [
        'name' => 'userInput',
        'description' => 'A review with a comment and a score (0 to 5)'
    ];

    public function fields(): array
    {
        return [
            'firstName' => [
                'alias' => 'first_name',
                'description' => 'A comment (250 max chars)',
                'type' => Type::string(),
                'rules' => ['max:250']
            ],
            'lastName' => [
                'alias' => 'last_name',
                'description' => 'A score (0 to 5)',
                'type' => Type::int(),
                'rules' => ['min:0', 'max:5']
            ]
        ];
    }
}

namespace App\GraphQL\Mutations;

use Closure;
use App\User;
use GraphQL;
use GraphQL\Type\Definition\Type;
use GraphQL\Type\Definition\ResolveInfo;
use Rebing\GraphQL\Support\Mutation;

class UpdateUserMutation extends Mutation
{
    protected $attributes = [
        'name' => 'updateUser'
    ];

    public function type(): Type
    {
        return GraphQL::type('User');
    }

    public function args(): array
    {
        return [
            'id' => [
                'type' => Type::nonNull(Type::string())
            ],
            'input' => [
                'type' => GraphQL::type('UserInput')
            ]
        ];
    }

    public function resolve($root, $args, $context, ResolveInfo $resolveInfo, Closure $getSelectFields)
    {
        $user = User::find($args['id']);
        $user->fill($args['input']));
        $user->save();

        return $user;
    }
}

JSON columns

When using JSON columns in your database, the field won't be defined as a "relationship", but rather a simple column with nested data. To get a nested object that's not a database relationship, use the is_relation attribute in your Type:

class UserType extends GraphQLType
{
    // ...

    public function fields(): array
    {
        return [
            // ...

            // JSON column containing all posts made by this user
            'posts' => [
                'type'          => Type::listOf(GraphQL::type('Post')),
                'description'   => 'A list of posts written by the user',
                // Now this will simply request the "posts" column, and it won't
                // query for all the underlying columns in the "post" object
                // The value defaults to true
                'is_relation' => false
            ]
        ];
    }

    // ...
}

Field deprecation

Sometimes you would want to deprecate a field but still have to maintain backward compatibility until clients completely stop using that field. You can deprecate a field using directive. If you add deprecationReason to field attributes it will become marked as deprecated in GraphQL documentation. You can validate schema on client using Apollo Engine.

namespace App\GraphQL\Types;

use App\User;
use GraphQL\Type\Definition\Type;
use Rebing\GraphQL\Support\Type as GraphQLType;

class UserType extends GraphQLType
{
    protected $attributes = [
        'name'          => 'User',
        'description'   => 'A user',
        'model'         => User::class,
    ];

    public function fields(): array
    {
        return [
            'id' => [
                'type' => Type::nonNull(Type::string()),
                'description' => 'The id of the user',
            ],
            'email' => [
                'type' => Type::string(),
                'description' => 'The email of user',
            ],
            'address' => [
                'type' => Type::string(),
                'description' => 'The address of user',
                'deprecationReason' => 'Deprecated due to address field split'
            ],
            'address_line_1' => [
                'type' => Type::string(),
                'description' => 'The address line 1 of user',
            ],
            'address_line_2' => [
                'type' => Type::string(),
                'description' => 'The address line 2 of user',
            ],
        ];
    }
}

Default field resolver

It's possible to override the default field resolver provided by the underlying webonyx/graphql-php library using the config option defaultFieldResolver.

You can define any valid callable (static class method, closure, etc.) for it:

'defaultFieldResolver' => [Your\Klass::class, 'staticMethod'],

The parameters received are your regular "resolve" function signature.

Macros

If you would like to define some helpers that you can re-use in a variety of your queries, mutations and types, you may use the macro method on the GraphQL facade.

For example, from a service provider's boot method:

namespace App\Providers;

use GraphQL\Type\Definition\Type;
use Illuminate\Support\ServiceProvider;
use Rebing\GraphQL\Support\Facades\GraphQL;

class AppServiceProvider extends ServiceProvider
{
    /**
     * Bootstrap any application services.
     *
     * @return void
     */
    public function boot()
    {
        GraphQL::macro('listOf', function (string $name): Type {
            return Type::listOf(GraphQL::type($name));
        });
    }
}

The macro function accepts a name as its first argument, and a Closure as its second.

Automatic Persisted Queries support

Automatic Persisted Queries (APQ) improve network performance by sending smaller requests, with zero build-time configuration.

APQ is disabled by default and can be enabled in the config via apc.enabled=true or by setting the environment variable GRAPHQL_APQ_ENABLE=true.

A persisted query is an ID or hash that can be generated on the client sent to the server instead of the entire GraphQL query string. This smaller signature reduces bandwidth utilization and speeds up client loading times. Persisted queries pair especially with GET requests, enabling the browser cache and integration with a CDN.

Behind the scenes, APQ uses Laravels cache for storing / retrieving the queries. They are parsed by GraphQL before storing, so re-parsing them again is not necessary. Please see the various options there for which cache, prefix, TTL, etc. to use.

Note: it is advised to clear the cache after a deployment to accomodate for changes in your schema!

For more information see:

Note: the APQ protocol requires the hash sent by the client being compared with the computed hash on the server. In case a mutating middleware like TrimStrings is active and the query sent contains leading/trailing whitespaces, these hashes can never match resulting in an error.

In such case either disable the middleware or trim the query on the client before hashing.

Notes

The error descriptions are aligned with apollo-server.

Client example

Below a simple integration example with Vue/Apollo, the createPersistedQueryLink automatically manages the APQ flow.

// [example app.js]

require('./bootstrap');

window.Vue = require('vue');

Vue.component('example-component', require('./components/ExampleComponent.vue').default);

import { ApolloClient } from 'apollo-client';
import { ApolloLink } from 'apollo-link';
import { createHttpLink } from 'apollo-link-http';
import { createPersistedQueryLink } from 'apollo-link-persisted-queries';
import { InMemoryCache } from 'apollo-cache-inmemory';
import VueApollo from 'vue-apollo';

const httpLinkWithPersistedQuery = createPersistedQueryLink().concat(createHttpLink({
    uri: '/graphql',
}));

// Create the apollo client
const apolloClient = new ApolloClient({
    link: ApolloLink.from([httpLinkWithPersistedQuery]),
    cache: new InMemoryCache(),
    connectToDevTools: true,
})

const apolloProvider = new VueApollo({
    defaultClient: apolloClient,
});

Vue.use(VueApollo);

const app = new Vue({
    el: '#app',
    apolloProvider,
});

<!-- [example TestComponent.vue] -->

<template>
    <div>
        <p>Test APQ</p>
        <p>-> <span v-if="$apollo.queries.hello.loading">Loading...</span>{{ hello }}</p>
    </div>
</template>

<script>
    import gql from 'graphql-tag';
    export default {
        apollo: {
            hello: gql`query{hello}`,
        },
        mounted() {
            console.log('Component mounted.')
        }
    }
</script>

Misc features

Detecting unused variables

By default, 'variables' provided alongside the GraphQL query which are not consumed, are silently ignored.

If you consider the hypothetical case you have an optional (nullable) argument in your query, and you provide a variable argument for it but you make a typo, this can go unnoticed.

Example:

mutation test($value:ID) {
  someMutation(type:"falbala", optional_id: $value)
}

Variables provided:

{
  // Ops! typo in `values`
  "values": "138"
}

In this case, nothing happens and optional_id will be treated as not being provided.

To prevent such scenarios, you can enable the config option detect_unused_variables and set it to true.

Guides

Upgrading from v1 to v2

Although version 2 builds on the same code base and does not radically change how the library itself works, many things were improved, sometimes leading to incompatible changes.

Step 0: make a backup!
Re-publish the configuration file to learn about all the new settings
The order and arguments/types for resolvers has changed:
- before: resolve($root, $array, SelectFields $selectFields, ResolveInfo $info)
- after: resolve($root, $array, $context, ResolveInfo $info, Closure $getSelectFields)
- If you now want to use SelectFields, you've to first request it: $selectFields = $getSelectFields();. The primary reason for this is performance. SelectFields is an optional feature but consumes resources to traverse the GraphQL request AST and introspect all the types for their configuration to apply its magic. In the past it was always constructed and thus consumed resources, even when not requested. This has been changed to an explicit form.
Many method signature declarations changed to improve type safety, which have to be adapted:
- The signature of the method fields changed:
  - from public function fields()
  - to public function fields(): array
- The signature of the method toType changed:
  - from public function toType()
  - to public function toType(): \GraphQL\Type\Definition\Type
- The signature of the method getFields changed:
  - from public function getFields()
  - to public function getFields(): array
- The signature of the method interfaces changed:
  - from public function interfaces()
  - to public function interfaces(): array
- The signature of the method types changed:
  - from public function types()
  - to public function types(): array
- The signature of the method type changed:
  - from public function type()
  - to public function type(): \GraphQL\Type\Definition\Type
- The signature of the method args changed:
  - from public function args()
  - to public function args(): array
- The signature of the method queryContext changed:
  - from protected function queryContext($query, $variables, $schema)
  - to protected function queryContext()
- The signature of the controller method query changed:
  - from function query($query, $variables = [], $opts = [])
  - to function query(string $query, ?array $variables = [], array $opts = []): array
- If you're using custom Scalar types:
  - the signature of the method parseLiteral changed (due to upgrade of the webonxy library):
    - from public function parseLiteral($ast)
    - to public function parseLiteral($valueNode, ?array $variables = null)
The UploadType now has to be added manually to the types in your schema if you want to use it. The ::getInstance() method is gone, you simple reference it like any other type via GraphQL::type('Upload').
Follow Laravel convention and use plural for namspaces (e.g. new queries are placed in App\GraphQL\Queries, not App\GraphQL\Query anymore); the respective make commands have been adjusted. This will not break any existing code, but code generates will use the new schema.
Be sure to read the Changelog for more details

Migrating from Folklore

https://github.com/folkloreinc/laravel-graphql, formerly also known as https://github.com/Folkloreatelier/laravel-graphql

Both code bases are very similar and, depending on your level of customization, the migration may be very quick.

Note: this migration is written with version 2.* of this library in mind.

The following is not a bullet-proof list but should serve as a guide. It's not an error if you don't need to perform certain steps.

Make a backup before proceeding!

composer remove folklore/graphql
if you've a custom ServiceProvider or did include it manually, remove it. The point is that the existing GraphQL code should not be triggered to run.
composer require rebing/graphql-laravel
Publish config/graphql.php and adapt it (prefix, middleware, schemas, types, mutations, queries, security settings, graphiql)
- Removed settings
  - domain
  - resolvers
- schema (defaul schema) renamed to default_schema
- middleware_schema does not exist, it's defined within a schema.<name>.middleware now
Change namespace references:
- from Folklore\
- to Rebing\
See Upgrade guide from v1 to v2 for all the function signature changes
The trait ShouldValidate does not exist anymore; the provided features are baked into Field
The first argument to the resolve method for queries/mutations is now null (previously its default was an empty array)

Performance considerations

Lazy loading of types

Lazy loading of types is a way of improving the start up performance.

If you are declaring types using aliases it is not supported. If that is not the case, you can enable it with lazyload_types set to true.

Example of aliasing not supported by lazy loading

I.e. you cannot have a query class ExampleQuery with the $name property example but register it with a different one; this will not work:

'query' => [
    'aliasedEXample' => ExampleQuery::class,
],

Wrap Types

You can wrap types to add more information to the queries and mutations. Similar as the pagination is working you can do the same with your extra data that you want to inject (see test examples). For instance, in your query:

public function type(): Type
{
    return GraphQL::wrapType(
        'PostType',
        'PostMessageType',
        \App\GraphQL\Types\WrapMessagesType::class,
    );
}

public function resolve($root, $args)
{
    return [
        'data' => Post::find($args['post_id']),
        'messages' => new Collection([
                new SimpleMessage("Congratulations, the post was found"),
                new SimpleMessage("This post cannot be edited", "warning"),
        ]),
    ];
}

Name		Name	Last commit message	Last commit date
Latest commit History 1,327 Commits
.github		.github
config		config
docker		docker
example		example
resources/views		resources/views
src		src
tests		tests
tmp/phpstan		tmp/phpstan
.editorconfig		.editorconfig
.gitattributes		.gitattributes
.gitignore		.gitignore
.php_cs.dist		.php_cs.dist
CHANGELOG.md		CHANGELOG.md
LICENSE		LICENSE
README.md		README.md
composer.json		composer.json
phpstan-baseline.neon		phpstan-baseline.neon
phpstan.neon.dist		phpstan.neon.dist
phpunit.xml.dist		phpunit.xml.dist

License

Rasool-deldar/graphql-laravel

Folders and files

Latest commit

History

Repository files navigation