cycle/entity-behavior

entity-behavior's Issues

Test OptimisticLock when an entity is deleting

Test OptimisticLock when an entity is deleting (with and without SoftDelete behavior)

Originally posted by @roxblnfk in #24 (comment)

🐛 OptimisticLock schema generation

No duplicates 🥲.

• I have searched for a similar issue in our bug tracker and didn't find any solutions.

What happened?

Using example from this link: https://cycle-orm.dev/docs/entity-behaviors-optimistic-lock

#[Entity]
#[OptimisticLock(
  field: 'version',                         // Required. By default 'version' 
  column: 'version',                        // Optional. By default 'null'. If not set, will be used information from property declaration.
  rule: OptimisticLock::RULE_INCREMENT      // Optional. By default OptimisticLock::RULE_INCREMENT
)]
class Page
{
    #[Column(type: 'primary')]
    private int $id;
    
    #[Column(type: 'integer')]
    private int $version;
}

During schema generation an error was received:
In RegistryModifier.php line 61: Field version must be of type integer

If remove attribute #[Column(type: 'integer')] - then works.

Version

entity-behavior v1.1.1
orm v2.2.1,
PHP 8.1

Supporting for custom EventListener

Add possibility to replace EventListener with custom implementation.
Does anyone need this?

🐛 Undefined class or binding 'Cycle\ORM\ORMInterface` on a listener resolving

No duplicates 🥲.

• I have searched for a similar issue in our bug tracker and didn't find any solutions.

What happened?

The ORM factroy throws an exception on resolving this listener

    public function __construct(
        private ORMInterface $orm,
        private EntityManagerInterface $em,
    ) {
    }

The message of the exception is Spiral\Core\Exception\Container\NotFoundException: Undefined class or binding 'Cycle\ORM\ORMInterface

Version

entity-behavior v1.1.1
orm v2.1.0,
PHP 8.1

UpdatedAt field should be not null

https://cycle-orm.dev/docs/entity-behaviors-timestamps

When I used old cycle/orm: ^1 version updated_at column always was filled with first insert. Now there is no. Bug?