Models

As Strapi is a headless Content Management System (CMS), creating a data structure for the content is one of the most important aspects of using the software. Models define a representation of the data structure.

There are 2 different types of models in Strapi:

• content-types, which can be collection types or single types, depending on how many entries they manage,
• and components that are data structures re-usable in multiple content-types.

If you are just starting out, it is convenient to generate some models with the Content-type Builder directly in the admin panel. The user interface takes over a lot of validation tasks and showcases all the options available to create the content's data structure. The generated model mappings can then be reviewed at the code level using this documentation.

Model creation

Content-types and components models are created and stored differently.

Content-types

Content-types in Strapi can be created:

• with the Content-type Builder in the admin panel,
• or with Strapi's interactive CLI strapi generate command.

The content-types has the following models files:

• schema.json for the model's schema definition. (generated automatically, when creating content-type with either method)
• lifecycles.js for lifecycle hooks. This file must be created manually.

These models files are stored in ./src/api/[api-name]/content-types/[content-type-name]/, and any JavaScript or JSON file found in these folders will be loaded as a content-type's model (see project structure).

Components

Component models can't be created with CLI tools. Use the Content-type Builder or create them manually.

Components models are stored in the ./src/components folder. Every component has to be inside a subfolder, named after the category the component belongs to (see project structure).

Model schema

The schema.json file of a model consists of:

• settings, such as the kind of content-type the model represents or the table name in which the data should be stored,
• information, mostly used to display the model in the admin panel and access it through the REST and GraphQL APIs,
• attributes, which describe the data structure of the model,
• and options used to defined specific behaviors on the model.

Model settings

General settings for the model can be configured with the following parameters:

Parameter	Type	Description
tableName	String	Database table name in which the data should be stored
kind Optional, only for content-types	String	Defines if the content-type is: a collection type (collectionType) or a single type (singleType)

// ./api/[api-name]/content-types/restaurant/schema.json

{
  "kind": "collectionType",
  "tableName": "Restaurants_v1",
}

Model information

The info key in the model's schema describes information used to display the model in the admin panel and access it through the Content API. It includes the following parameters:

Parameter	Type	Description
displayName	String	Default name to use in the admin panel
singularName	String	Singular form of the collection type name. Used to generate the API routes and databases/tables collection. Should be kebab-case.
pluralName	String	Plural form of the collection type name. Used to generate the API routes and databases/tables collection. Should be kebab-case.
description	String	Description of the model
icon Optional, only for Components	String	FontAwesome (opens new window) (v5) icon name to use for the component's icon in the admin panel

// ./src/api/[api-name]/content-types/restaurant/schema.json

  "info": {
    "displayName": "Restaurant",
    "singularName": "restaurant",
    "pluralName": "restaurants",
    "description": ""
  },

Model attributes

The data structure of a model consists of a list of attributes. Each attribute has a type parameter, which describes its nature and defines the attribute as a simple piece of data or a more complex structure used by Strapi.

Many types of attributes are available:

• scalar types (e.g. strings, dates, numbers, booleans, etc.),
• Strapi-specific types, such as:
  • media, for files uploaded through the Media library
  • relation to describe a relation between content-types
  • component to define a component (i.e. a data structure usable in multiple content-types)
  • dynamiczone to define a dynamic zone (i.e. a flexible space based on a list of components)
  • and the locale and localizations types, only used by the Internationalization (i18n) plugin

The type parameter of an attribute should be one of the following values:

Type categories	Available types
String types	string text richtext enumeration email password uid
Date types	date time datetime timestamp
Number types	integer biginteger float decimal
Other generic types	boolean array json
Special types unique to Strapi	media relation component dynamiczone
Internationalization (i18n)-related types Can only be used if the i18n plugin is installed	locale localizations

Validations

Basic validations can be applied to attributes using the following parameters:

Parameter	Type	Description	Default
required	Boolean	If true, adds a required validator for this property	false
max	Integer	Checks if the value is greater than or equal to the given maximum	-
min	Integer	Checks if the value is less than or equal to the given minimum	-
minLength	Integer	Minimum number of characters for a field input value	-
maxLength	Integer	Maximum number of characters for a field input value	-
private	Boolean	If true, the attribute will be removed from the server response. 💡 This is useful to hide sensitive data.	false
configurable	Boolean	If false, the attribute isn't configurable from the Content-type Builder plugin.	true

// ./src/api/[api-name]/content-types/restaurant/schema.json

{
  // ...
  "attributes": {
    "title": {
      "type": "string",
      "minLength": 3,
      "maxLength": 99,
      "unique": true
    },
    "description": {
      "default": "My description",
      "type": "text",
      "required": true
    },
    "slug": {
      "type": "uid",
      "targetField": "title"
    }
    // ...
  }
}

Database validations and settings

✋ 🚧 This API is considered experimental.

These settings should be reserved to an advanced usage, as they might break some features. There are no plans to make these settings stable.

Database validations and settings are custom options passed directly onto the tableBuilder Knex.js function during schema migrations. Database validations allow for an advanced degree of control for setting custom column settings. The following options are set in a column: {} object per attribute:

Parameter	Type	Description	Default
name	string	Changes the name of the column in the database	-
defaultTo	string	Sets the database defaultTo, typically used with notNullable	-
notNullable	boolean	Sets the database notNullable, ensures that columns cannot be null	false
unsigned	boolean	Only applies to number columns, removes the ability to go negative but doubles maximum length	false
unique	boolean	Enforces database level unique, caution when using with draft & publish feature	false
type	string	Changes the database type, if type has arguments, you should pass them in args	-
args	array	Arguments passed into the Knex.js function that changes things like type	[]

// ./src/api/[api-name]/content-types/restaurant/schema.json

{
  // ...
  "attributes": {
    "title": {
      "type": "string",
      "minLength": 3,
      "maxLength": 99,
      "unique": true,
      "column": {
        "unique": true // enforce database unique also
      }
    },
    "description": {
      "default": "My description",
      "type": "text",
      "required": true,
      "column": {
        "defaultTo": "My description", // set database level default
        "notNullable": true // enforce required at database level, even for drafts
      }
    },
    "rating": {
      "type": "decimal",
      "default": 0,
      "column": {
        "defaultTo": 0,
        "type": "decimal", // using the native decimal type but allowing for custom precision
        "args": [
          6,1 // using custom precision and scale
        ]
      }
    }
    // ...
  }
}

uid type

The uid type is used to automatically prefill the field value in the admin panel with a unique identifier (UID) (e.g. slugs for articles) based on 2 optional parameters:

• targetField (string): If used, the value of the field defined as a target is used to auto-generate the UID.
• options (string): If used, the UID is generated based on a set of options passed to the underlying uid generator (opens new window). The resulting uid must match the following regular expression pattern: /^[A-Za-z0-9-_.~]*$.

Relations

Relations link content-types together. Relations are explicitly defined in the attributes of a model with type: 'relation' and accept the following additional parameters:

Parameter	Description
relation	The type of relation among these values: oneToOne oneToMany manyToOne manyToMany
target	Accepts a string value as the name of the target content-type
mappedBy and inversedBy Optional	In bidirectional relations, the owning side declares the inversedBy key while the inversed side declares the mappedBy key

One-to-One relationships are useful when one entry can be linked to only one other entry.

They can be unidirectional or bidirectional. In unidirectional relationships, only one of the models can be queried with its linked item.

Unidirectional use case example:

• A blog article belongs to a category.
• Querying an article can retrieve its category,
• but querying a category won't retrieve the owned article.

// ./src/api/[api-name]/content-types/article/schema.json

const model = {
  attributes: {
    category: {
      type: 'relation',
      relation: 'oneToOne',
      target: 'category',
    },
  },
};

Bidirectional use case example:

• A blog article belongs to a category.
• Querying an article can retrieve its category,
• and querying a category also retrieves its owned article.

// ./src/api/[api-name]/content-types/article/schema.json

const model = {
  attributes: {
    category: {
      type: 'relation',
      relation: 'oneToOne',
      target: 'category',
      inversedBy: 'article',
    },
  },
};


// ./src/api/[api-name]/content-types/category/schema.json

const model = {
  attributes: {
    article: {
      type: 'relation',
      relation: 'oneToOne',
      target: 'article',
      mappedBy: 'category',
    },
  },
};

One-to-Many relationships are useful when:

• an entry from a content-type A is linked to many entries of another content-type B,
• while an entry from content-type B is linked to only one entry of content-type A.

One-to-many relationships are always bidirectional, and are usually defined with the corresponding Many-to-One relationship:

Example:

A person can own many plants, but a plant is owned by only one person.

// ./src/api/[api-name]/content-types/plant/schema.json

const model = {
  attributes: {
    owner: {
      type: 'relation',
      relation: 'manyToOne',
      target: 'api::person.person',
      inversedBy: 'plants',
    },
  },
};

// ./src/api/person/models/schema.json

const model = {
  attributes: {
    plants: {
      type: 'relation',
      relation: 'oneToMany',
      target: 'api::plant.plant',
      mappedBy: 'owner',
    },
  },
};

Many-to-One relationships are useful to link many entries to one entry.

They can be unidirectional or bidirectional. In unidirectional relationships, only one of the models can be queried with its linked item.

Unidirectional use case example:

A book can be written by many authors.

// ./src/api/[api-name]/content-types/book/schema.json

const model = {
  attributes: {
    author: {
      type: 'relation',
      relation: 'manyToOne',
      target: 'author',
    },
  },
};

Bidirectional use case example:

An article belongs to only one category but a category has many articles.

// ./src/api/[api-name]/content-types/article/schema.json

const model = {
  attributes: {
    author: {
      type: 'relation',
      relation: 'manyToOne',
      target: 'category',
      inversedBy: 'article',
    },
  },
};


// ./src/api/[api-name]/content-types/category/schema.json

const model = {
  attributes: {
    books: {
      type: 'relation',
      relation: 'oneToMany',
      target: 'article',
      mappedBy: 'category',
    },
  },
};

Many-to-Many relationships are useful when:

• an entry from content-type A is linked to many entries of content-type B,
• and an entry from content-type B is also linked to many entries from content-type A.

Many-to-many relationships can be unidirectional or bidirectional. In unidirectional relationships, only one of the models can be queried with its linked item.

Unidirectional use case example:

const model = {
  attributes: {
    categories: {
      type: 'relation',
      relation: 'manyToMany',
      target: 'category',
    },
  },
};

Bidirectional use case example:

An article can have many tags and a tag can be assigned to many articles.

// ./src/api/[api-name]/content-types/article/schema.json

const model = {
  attributes: {
    tags: {
      type: 'relation',
      relation: 'manyToMany',
      target: 'tag',
      inversedBy: 'articles',
    },
  },
};


// ./src/api/[api-name]/content-types/tag/schema.json

const model = {
  attributes: {
    articles: {
      type: 'relation',
      relation: 'manyToMany',
      target: 'article',
      mappedBy: 'tag',
    },
  },
};

Components

Component fields create a relation between a content-type and a component structure. Components are explicitly defined in the attributes of a model with type: 'component' and accept the following additional parameters:

Parameter	Type	Description
repeatable	Boolean	Could be true or false depending on whether the component is repeatable or not
component	String	Define the corresponding component, following this format: <category>.<componentName>

// ./src/api/[apiName]/restaurant/content-types/schema.json

{
  "attributes": {
    "openinghours": {
      "type": "component",
      "repeatable": true,
      "component": "restaurant.openinghours"
    }
  }
}

Dynamic zones

Dynamic zones create a flexible space in which to compose content, based on a mixed list of components.

Dynamic zones are explicitly defined in the attributes of a model with type: 'dynamiczone'. They also accepts a components array, where each component should be named following this format: <category>.<componentName>.

// ./src/api/[api-name]/content-types/article/schema.json

{
  "attributes": {
    "body": {
      "type": "dynamiczone",
      "components": ["article.slider", "article.content"]
    }
  }
}

Model options

The options key is used to define specific behaviors and accepts the following parameter:

Parameter	Type	Description
privateAttributes	Array of strings	Allows treating a set of attributes as private, even if they're not actually defined as attributes in the model. It could be used to remove them from API responses timestamps. The set of privateAttributes defined in the model are merged with the privateAttributes defined in the global Strapi configuration.
populateCreatorFields	Boolean	Toggles including the created_by and updated_by fields in the API response. Default value: false
draftAndPublish	Boolean	Enables the draft and publish feature. Default value: false

// ./src/api/[api-name]/content-types/restaurant/schema.json

{
  "options": {
    "privateAttributes": ["id", "created_at"],
    "populateCreatorFields": true,
    "draftAndPublish": false
  }
}

Lifecycle hooks

Lifecycle hooks are functions that get triggered when Strapi queries are called. They are triggered automatically when managing content through the administration panel or when developing custom code using queries·

Lifecycle hooks can be customized declaratively or programmatically.

✋ CAUTION

Lifecycles hooks are not triggered when using directly the knex (opens new window) library instead of Strapi functions.

Available lifecycle events

The following lifecycle events are available:

• beforeCreate
• beforeCreateMany
• afterCreate
• afterCreateMany
• beforeUpdate
• beforeUpdateMany
• afterUpdate
• afterUpdateMany
• beforeDelete
• beforeDeleteMany
• afterDelete
• afterDeleteMany
• beforeCount
• afterCount
• beforeFindOne
• afterFindOne
• beforeFindMany
• afterFindMany

Hook event object

Lifecycle hooks are functions that take an event parameter, an object with the following keys:

Key	Type	Description
action	String	Lifecycle event that has been triggered (see list)
model	Object	Model object
em	EntityManagerObject	EntityManager
params	Object	Accepts the following parameters: data select where orderBy limit offset populate
result	Object	Optional, only available with afterXXX events Contains the result of the action.
state	Object	Query state, can be used to share state between beforeXXX and afterXXX events of a query.

Declarative and programmatic usage

To configure a content-type lifecycle hook, create a lifecycles.js file in the ./api/[api-name]/content-types/[content-type-name]/ folder.

Each event listener is called sequentially. They can be synchronous or asynchronous.

// ./src/api/[api-name]/content-types/restaurant/lifecycles.js

module.exports = {
  beforeCreate(event) {
    const { data, where, select, populate } = event.params;

    // let's do a 20% discount everytime
    event.params.data.price = event.params.data.price * 0.8;
  },

  afterCreate(event) {
    const { result, params } = event;

    // do something to the result;
  },
};

Using the database layer API, it's also possible to register a subscriber and listen to events programmatically:

// ./src/api/[api-name]/content-types/restaurant/lifecycles.js

// registering a subscriber
strapi.db.lifecycles.subscribe({
  models: [], // optional;

  beforeCreate(event) {
    const { data, where, select, populate } = event.params;

    event.state = 'doStuffAfterWards';
  },

  afterCreate(event) {
    if (event.state === 'doStuffAfterWards') {
    }

    const { result, params } = event;

    // do something to the result
  },
});

// generic subscribe for generic handling
strapi.db.lifecycles.subscribe((event) => {
  if (event.action === 'beforeCreate') {
    // do something
  }
});