QueryBundle

latest stable version

2.4 (master)	2.3	2.2

Run tests

• ./runtests.sh run all unit tests
• ./agile.sh generate testdox documentation
• ./coverage.sh generate and open html coverage

Plain symfony project for query-bundle

The purpose of this project is to see how studiomado/query-bundle works and can be installed in a plain symfony project.

• database configuration
• install query-bundle
• create at least one entity

Database configuration

Remember to update parameter.yml file, parameter.yml.dist and config.yml file. In config.yml file also remember that the drive MUST be changed in pdo_sqlite to enable doctrine to work with this database.

This is just an example: for this example we use sqlite but in production you can use mysql or postgres or any other database supported by doctrine.

prompt> ./bin/console doctrine:database:create
Created database /path/to/project/var/data/data.sqlite for connection named default

Install query-bundle

prompt> composer require studiomado/query-bundle

Create at least one entity

Create at least one entity ...

prompt> ./bin/console doctrine:generate:entity

In this example I created an entity Task following command steps.

created ./src/AppBundle/Entity/
created ./src/AppBundle/Entity/Task.php
> Generating entity class src/AppBundle/Entity/Task.php: OK!
> > Generating repository class src/AppBundle/Repository/TaskRepository.php: OK!

... and update the schema ...

    prompt> ./bin/console doctrine:schema:update
    ATTENTION: This operation should not be executed in a production environment.
                         Use the incremental update to detect changes during development and use
                         the SQL DDL provided to manually update your database in production.

    The Schema-Tool would execute "1" queries to update the database.
    Please run the operation by passing one - or both - of the following options:
            doctrine:schema:update --force to execute the command
            doctrine:schema:update --dump-sql to dump the SQL statements to the screen

The schema update works only with force option

prompt> ./bin/console doctrine:schema:update --force
Updating database schema...
Database schema updated successfully! "1" query was executed

Just take a look of the database content (that now is simply empty).

prompt> ./bin/console doctrine:query:dql "select t from AppBundle:Task t"

The query will return an empty array of result

 array (size=0)
   empty

Just add first task ...

prompt> ./bin/console doctrine:query:sql "insert into task values (null, 'complete this guide', 'todo') "

and take a look of the content

prompt> ./bin/console doctrine:query:dql "select t from AppBundle:Task t"

array (size=1)
  0 =>
    object(stdClass)[507]
      public '__CLASS__' => string 'AppBundle\Entity\Task' (length=21)
      public 'id' => int 1
      public 'description' => string 'complete this guide' (length=19)
      public 'status' => string 'todo' (length=4)

Complete installation

First of all install vendors

prompt> composer require jms/serializer-bundle
prompt> composer require willdurand/hateoas-bundle
prompt> composer require white-october/pagerfanta-bundle
prompt> composer require friendsofsymfony/rest-bundle

and then, … add vendors in your app/AppKernel

new FOS\RestBundle\FOSRestBundle(),
new JMS\SerializerBundle\JMSSerializerBundle(),
new Bazinga\Bundle\HateoasBundle\BazingaHateoasBundle(),

Complete configuration and use the bundle

Once everything is done, you can add new endpoints using the query-bundle to query the database.

namespace AppBundle\Controller;

use Sensio\Bundle\FrameworkExtraBundle\Configuration\Route;
use Symfony\Bundle\FrameworkBundle\Controller\Controller;
use Symfony\Component\HttpFoundation\Request;
use Symfony\Component\HttpFoundation\Response;

class DefaultController extends Controller
{
    /** @Route("/", name="homepage") */
    public function indexAction(
        Request $request,
        \Doctrine\ORM\EntityManager $em,
        \JMS\Serializer\Serializer $serializer
    ) {
        $data = $em->getRepository('AppBundle:Task')
            ->setRequest($request)
            ->findAllPaginated();

        $content = $serializer->serialize($data, 'json');

        return new Response($content, 200);
    }
}

Configure your entity repository

Now be sure that your repository extends the right BaseRepository.

namespace AppBundle\Repository;

class TaskRepository extends \Mado\QueryBundle\Repositories\BaseRepository
{
    // to do …
}

namespace AppBundle\Entity;

/** @ORM\Entity(repositoryClass="AppBundle\Repository\TaskRepository") */
class Task
{
    // to do …
}

Customize entity serialization

Now if you want to customize responses add

use JMS\Serializer\Annotation as JMS;

On top of your entities and complete your JMS configurations. See JMS documentation to get all the complete documentation.

Here some examples:

• http://127.0.0.1:8000/?filtering[status]=todo
• http://127.0.0.1:8000/?filtering[status|contains]=od
• http://127.0.0.1:8000/?filtering[status|endswith]=gress

Find All No Paginated

Added a new method in BaseRepository
When you need results applying filter and sort without pagination

public function findAllNoPaginated();

This feature was needed to create an Excel Report, injecting results into the Excel Report

Example without pagination

In Controller:

public function getTasksExcelReportAction(Request $request)
    {
        $tasks = $this->getDoctrine()
            ->getRepository('AppBundle:Task')
            ->findAllNoPaginated();

        $reportExcel = new TasksReport($tasks);
        $reportExcel->createReport();

        $excelContent = $reportExcel->printReport();

        return new Response(
            $excelContent,
            200, [
                'Content-Type' => 'application/vnd.openxmlformats-officedocument.spreadsheetml.sheet',
            ]
        );        
    } 

Example with pagination

In Controller:

    public function getTasksAction(Request $request)
    {
        return $this->getDoctrine()
            ->getRepository('AppBundle:Task')
            ->setRequest($request)
            ->findAllPaginated();
    }

Queries

AND Conditions

If you want to create an AND condition with this library you can create it from the client for example with a simple GET request like this:

/api/foo?filtering[name|eq]=bar&filtering[surname|eq]=bar

This request will produce a query like this:

SELECT f0_.id AS id_0, f0_.name AS name_1, f0_.surname AS surname_2" .
FROM foo f0_" .
WHERE f0_.name = "bar" AND f0_.surname = "bar"

OR Conditions

If you want to create an OR condition with this library you can create it from the client for example with a simple GET request like this:

/api/foo?filtering_or[name|eq]=bar&filtering_or[surname|eq]=bar

This request will produce a query like this:

SELECT f0_.id AS id_0, f0_.name AS name_1, f0_.surname AS surname_2" .
FROM foo f0_" .
WHERE ((f0_.name = "bar" OR f0_.surname = "bar"))

Instead, if you want to have more OR conditions separated you can do something like this:

 /api/foo?filtering_or[name|eq|1]=bar&filtering_or[surname|eq|1]=bar&filtering_or[group|contains|2]=baz&filtering_or[role|contains|2]=baz

This request will produce a query like this:

SELECT f0_.id AS id_0, f0_.name AS name_1, f0_.surname AS surname_2, f0_.group AS group_3, f0_.role AS role_4" .
FROM foo f0_" .
WHERE (f0_.name = "bar" OR f0_.surname = "bar") AND (f0_.group LIKE "%baz%" OR f0_.role LIKE "%baz%")

This can be done by using a counter after the operator separated by |

Search into relations

If you want to search inside an entity where a condition is inside another entity you can do this:

/api/users?filtering[_embedded.group.name|contains =bar

This request will produce a query like this:

SELECT u0_.id AS id_0 u0_.username AS username_1,  u0_.group_id AS group_id_2 " .
FROM User u0_
LEFT JOIN Group g1_ ON u0_.group_id = g1_.id " .
WHERE g1_.name LIKE "%bar%"

To do this you need to add inside the user entity some Hateoas annotations like this:

 * @Hateoas\Relation(
 *     "groups",
 *     embedded = @Hateoas\Embedded(
 *         "expr(object.getGroups())",
 *          exclusion = @Hateoas\Exclusion(
 *              excludeIf = "expr(object.getGroups().isEmpty() === true)",
 *              groups={"groups"},
 *              maxDepth = 1
 *          )
 *     ),
 *     exclusion = @Hateoas\Exclusion(
 *              excludeIf = "expr(object.getGroups().isEmpty() === true)",
 *              groups={"groups"},
 *              maxDepth = 1
 *          )
 * )

If you add Hateoas annotations correctly, you can search deeper than only "one level". Here an example:

/api/users?filtering[_embedded.profile.location.country.name|contains]=italy

In this example you search all users that have a profile with a country location name: Italy.
Profile, location and country are entities and name is the field.

You can use _embedded filter also into filtering_or conditions.