CamelCase Lucid models output in AdonisJs

August 19, 2018

On this page

When working with databases, a popular choice is to make use of snake_case to define table field names.

Also working with snake_case comes with a benefit of accidentally getting bitten by PostgreSQL. Check the following screenshot.

However, Javascript programmers are used to of camelCase and often reach me asking:

Is there any way to auto-convert field names from snake_case to camelCase in Lucid?

Unfortunately, Lucid doesn’t do any automatic conversion of field names, and I want to improve that story in upcoming versions.

In the post, I will share a neat way to make use of a custom serializer, which converts all snake_case fields to camelCase when toJSON is called.

Defining Serializer

You can create custom serializers inside app/Models/Serializers directory and tell your models to use this new serializer vs. the one shipped with Lucid.

'use strict'

const _ = require('lodash')
const VanillaSerializer = require('@adonisjs/lucid/src/Lucid/Serializers/Vanilla')

class CamelCaseSerializer extends VanillaSerializer {
  _camelizeProperties (collection) {
    return _.reduce(collection, (result, value, key) => {
      result[_.camelCase(key)] = value
      return result
    }, {})

  _attachMeta (modelInstance, output) {
    if (_.size(modelInstance.$sideLoaded)) {
      output.__meta__ = this._camelizeProperties(modelInstance.$sideLoaded)

  _getRowJSON (modelInstance) {
    const json = this._camelizeProperties(modelInstance.toObject())
    this._attachRelations(modelInstance, json)
    this._attachMeta(modelInstance, json)
    return json

module.exports = CamelCaseSerializer
  1. We extend the VanillaSerializer shipped with Lucid.
  2. Override a couple of its methods. (I know these are private methods, but that’s the only option).
  3. Use lodash to convert field names using _.camelCase method.

Update models to use new Serializer

const Model = use('Model')

class User extends Model {
  static get Serializer () {
    return 'App/Models/Serializers/CamelCase'

Use a Base Model

Now instead of updating all the models, you can create a BaseModel and make all other models extend it.

const Model = use('Model')

class BaseModel extends Model {
  static get Serializer () {
    return 'App/Models/Serializers/CamelCase'
  static _bootIfNotBooted () {
    if ( !== 'BaseModel') {

The _bootIfNotBooted definition is important since we don’t want our BaseModel to boot.

Finally, update other models to extend the BaseModel.

const BaseModel = use('App/Models/BaseModel')

class User extends BaseModel {


The serializers are invoked only when you call toJSON to a query output. Accessing attributes of the model instance will still have snake_case field names.

This will be improved in future versions of Lucid :)

Discuss on TwitterEdit on GitHub

Harminder Virk

Blog by Harminder Virk.
Sharing my personal learnings :)