Code Smell 70 - Anemic Model Generators

Code Smell 70 - Anemic Model Generators

Anemic Models are the problem. Anemic Model Generators are the Meta Problem.

TL;DR: Do not create anemic objects. Much less with automatic tools.

Problems

  • Anemic Objects

  • Coupling to Implementation

  • Harder to Debug

Solutions

  1. Create your objects manually.

  2. Focus on essential behavior instead of accidental data storage.

Sample Code

Wrong

<?

AnemicClassCreator::create(
    'Employee',
    [
        new AutoGeneratedField('id', '$validators->getIntegerValidator()'),
        new AutoGeneratedField('name', '$validators->getStringValidator()'),
        new AutoGeneratedField('currentlyWorking', '$validators->getBooleanValidator()')
    ]);

class Employee extends AutoGeneratedObjectEmployee {
}

//We have magic setters and getters
//getId() , setId(), getName()
//Validation is not implicit
//Class are loaded via an autoloader

$john = new Employee;
$john->setId(1);
$john->setName('John');
$john->setCurrentlyWorking(true);

$john->getName(); //returns 'John'

Right

<?

final class Employee {
    private $name;
    private $workingStatus;

    public function __construct(string $name, WorkingStatus $workingStatus) {
        //..
    }

    public function name(): string {
        return $this->name;
        //This is not a getter. It is Employee's responsibility to tell us her/his name
    }
}

//We have no magic setters or getters
//all methods are real and can be debugged
//Validations are implicit

$john = new Employee('John', new HiredWorkingStatus());

$john->name(); //returns 'John'

Detection

Often, anemic models are generated with meta-programming.

We need to track these magic code generators.

Tags

  • Anemic

Conclusion

Code Wizards, Meta-programming, and anemic models are all code smells.

We need to avoid these dark techniques.

Having to write explicitly the code makes us reflect on every piece of data we encapsulate.

Relations

More info

Credits

Photo by Lenny Kuhne on Unsplash


The best smells are something that's easy to spot and most of time lead you to really interesting problems. Data classes (classes with all data and no behavior) are good examples of this. You look at them and ask yourself what behavior should be in this class.

Martin Fowler


This article is part of the CodeSmell Series.