README.md

Speck - Let you create your domain entities with reactive validation based on React propTypes

This package let you create entities with schema validation based on React PropTypes.

• Installing
• Using
• Examples

Installing

$ npm install speck-entity

Using

Sample Entities

const Joi = require('joi')
const joiAdapter = require('validatorAdapters')('joi', Joi)
const Speck = require('speck-entity')

class MyEntity extends Speck {
  static SCHEMA = {
    field: joiAdapter(Joi.string()),
    otherField: {
      validator: joiAdapter(Joi.number()),
      defaultValue: 10
    }
  }
}

class FatherEntity extends Speck {
  static SCHEMA = {
    children: {
      validator: joiAdapter(Joi.array().items(Joi.object().type(MyEntity)))
      type: MyEntity
    }
  }
}

Get default values

const niceInstance = new MyEntity();
console.log(niceInstance.toJSON()); // { field: undefined, otherField: 10 }
console.log(niceInstance.errors); // {}

Validations

const buggedInstance = new MyEntity({ field: 10, otherField: 'value' });
console.log(buggedInstance.toJSON()); // { field: 10, otherField: 'value' }
console.log(buggedInstance.errors); /* or buggedInstance.getErrors() -- but... getErrors also includes children errors
  {
    field: {
      errors: [ 'Invalid undefined `field` of type `number` supplied to `MyEntityEntity`, expected `string`.' ]
    },
    otherField: {
      errors: [ 'Invalid undefined `otherField` of type `string` supplied to `MyEntityEntity`, expected `number`.' ]
    }
  }
*/

Validate on change value

const otherInstance = new MyEntity({ field: 'myString' });
console.log(otherInstance.errors); // {}
console.log(otherInstance.valid); // true

otherInstance.field = 1;
console.log(otherInstance.errors); // {field: { errors: [ 'Invalid undefined `field` of type `number` supplied to `MyEntityEntity`, expected `string`.' ] }}
console.log(otherInstance.valid); // false

Parse children to Entity

const fatherInstance = new FatherEntity({
  children: [{
    field: 'A',
    otherField: 2
  }, {
    field: 'B',
    otherField: 3
  }]  
})
console.log(fatherInstance.children[0]); //An instance of MyEntity
console.log(fatherInstance.children[1].toJSON());
//{ field: 'B', otherField: 3 }

Builder

When you need to create objects with custom verification like

    const elementList = {
        elements: [{
          type: 'product',
          name: true,
          price:  true
        }, {
          type: 'default',
          isDefault: true
        }]
      };

In such cases you can define a builder as follows:

class ElementList extends Speck {}
ElementList.SCHEMA = {
  elements: {
    validator: noop,
    builder: (dataList) => dataList.map(data => {
      if (data.type === 'product') return new ProductEntity(data);
      if (data.type === 'default') return new FakeEntityWithBoolean(data);
    })
  }
};

By defining builder you tell Speck Entity that you take the responsibility of instansitating and returning a new object of the type which suits you the best. This is a powerful concept as it lets users dynamically create new types on the fly.

Clean unexpected values

const anotherInstance = new MyEntity({ field: 'myString', fake: 'fake' });
console.log(anotherInstance.toJSON()); // { field: 'myString', otherField: 10 }

To understand the validators React PropTypes

Well known issues

• Create helpers for relationships validations(Like, mininum, maximum)
• Create identifier and equal comparison
• Type builders and/or custom builders are not being applied on instance setters

Contextual validation

  class FakeEntityWithExcludeContext extends Speck {
      static SCHEMA = {
          id: joiAdapter(Joi.number().required()),
          requiredProp1: joiAdapter(Joi.number().required()),
          requiredProp2: joiAdapter(Joi.number().required()),
          requiredProp3: joiAdapter(Joi.number().required())
      }

      static CONTEXTS = {
        create: { exclude: [ 'requiredProp2', 'requiredProp3' ] },
        edit: { include: [ 'id', 'requiredProp1', 'requiredProp2' ] },
        onlyId: { include: [ 'id' ] }        
      }
  }

   const myEntity = new FakeEntityWithIncludeContext({ id: 1 });

   const contextCreate = myEntity.validateContext('create');
   console.log(contextCreate.errors); // { requiredProp1: { errors: [ ... ] } }
   console.log(contextCreate.valid); // false

   const contextEdit = myEntity.validateContext('edit');
   console.log(contextEdit.errors); // { requiredProp1: { errors: [ ... ] }, requiredProp2: { errors: [ ... ] } }
   console.log(contextEdit.valid); // false

   const contextOnlyId = myEntity.validateContext('onlyId')
   console.log(contextOnlyId.errors); // {}
   console.log(contextOnlyId.valid); // true

Each context (create and edit in example above), could have include property OR exclude, the include property receives the properties that will be validated in this context, and the exclude property represents the properties that will be ignored on validation.

In the example, the create context, will only check the 'requiredProp1' and 'requiredProp2' fields, and the edit context will check 'requiredProp1', 'requiredProp2' and 'id' properties.

You can't combine include and exclude in the same context definition

##Custom validation You can validate your entity adding the property in fields and setting the new validator

  class Entity extends Speck {
    static SCHEMA = {
      id: joiAdapter(Joi.number().required()),
      requiredProp1: joiAdapter(Joi.number().required())
    }

    static CONTEXTS = {
      create: {
        fields: {
          requiredProp1: (obj, field) => {
            if(obj[field] === -1) return new Error('Error -1');
          }
        }
      }
    }
  }

  const entity = new Entity({
    id: 1,
    requiredProp1: -1
  });

  const contextValidated = entity.validateContext('create');
  console.log(entity.errors.requiredProp1); // undefined
  console.log(contextValidated.requiredProp1); // { errors: [Error: Error -1] }

Hooks

class EntityWithHook extends Speck {
    static SCHEMA = {
        fieldWithHook: {
          validator: joiAdapter(Joi.number()),
          hooks: {
            afterSet(data, fieldName) {
              // data is the whole data of the instance
              // fieldName the current fieldName
              // DO WHATEVER YOU WANT
            }
          }
        },
        anotherFieldWithHook: {
          validator: joiAdapter(Joi.number()),
          hooks: {
            afterSet(data, fieldName) {
              return { anotherField: data[fieldName] * 2 } // if the afterSet hook returns an object is merged to data
            }
          }
        },
        anotherField: joiAdapter(Joi.number()),
    }
}

const myEntity = new EntityWithHook({ fieldWithHook: 'foo', anotherFieldWithHook: 'bar', anotherField: null });

myEntity.anotherFieldWithHook = 10 //according to the after set hook anotherField newValue will be 20