Skip to content
Marshal.ts the TypeScript serializer for frontend, backend, http-transport, query parameter, DTOs, message queues, and database, with built-in validations, NestJS support, and MongoDB abstraction, for Browser/NodeJS. The better and 4-6x faster alternative to class-transformer
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Type Name Latest commit message Commit time
Failed to load latest commit information.
packages v0.8.8 Apr 22, 2019
.dockerignore added typedoc Mar 16, 2019
.npmignore docs Mar 17, 2019
.travis.yml Add node 12 to Node.js matrix May 15, 2019
Dockerfile Consolidate all type decators to one @field() decorator to make thing… Mar 17, 2019
heroku.yml added heroku.yml Mar 17, 2019
package.json fixed build Mar 23, 2019
tsconfig.json moved mongo/typeorm to peer deps, improved readme Mar 23, 2019
tslint.json add non-convertable query also to field index Feb 8, 2019
typedoc.js remove benchmark package from docs Mar 18, 2019


Build Status npm version Coverage Status

Marshal is a library written in and for TypeScript to marshal JSON-representable data from JSON object to class instance to database records and vice versa.

Marshal introduces the concept of decorating your entity class once with all necessary annotations (like type declaration, indices, and relations) using only Marshal's TypeScript decorators agnostic to any serialization target by saving only the meta data, and then use it everywhere: frontend, backend, http-transport, query parameter, DTOs, and database, including validations.

The goal is to support all types of structures/use-cases where you need to serialize and validate data in a very user-friendly way while providing the fastest possible serializer for all platforms, NodeJS and browsers.

Marshal shines particularly when you have an application written in Typescript entirely, frontend, CLI, and backend. This allows you to save tons of time by moving your entities, DTO, query parameter signature, etc all as Marshal decorated classes in a common package (super simple with Lerna). You then use and import these classes in your frontend, cli, backend, or whatever you develop.

By using a new client-server framework entirely written in and for TypeScript like glut.ts which is based on Marshal.ts you enter a new world of developing web applications by not caring anymore at all about hand-made serialization and validation. However Marshal.ts helps also for traditional REST APIs


  • Supported types: String, Number, Boolean, Date, Binary, custom classes, Array, object maps, any.
  • Fast marshalling of class instance from and to JSON object
  • Fast marshalling of class instance from and to MongoDB record
  • Constructor support (required property can be placed in constructor)
  • Validation: Built-in, custom class and inline validators
  • Decorated property value (e.g. JSON uses plain Array, class instance uses a custom Collection class)
  • Patch marshalling (ideal for serialising JSON Patch and the like)
  • Complex models with parent references
  • Supports getters
  • Entity definition export to TypeORM (currently columns + indices), so you don't have to decorate twice.
  • NestJS validation pipe
  • MongoDB database abstraction


  • Add type support for: Map<T, K>, Set (WeakMap<T, K>, and Set)
  • Add type support for: BigInt, Big?Array, Float?Array, Int?Array, Uint?Array,
  • Add more built-in validators
  • Support discriminators (union class types)
  • Add support for TypeORM types completely (so we support MySQL, Postgres, SQLiTe, etc.) including relations.
  • Further performance boost by creating JIT serialization functions
  • Add automatic tests IE11+ (help is welcome)



npm install @marcj/marshal reflect-metadata

Install buffer as well if you want to have Binary support.

Example Entity

import {
} from '@marcj/marshal';
import {Buffer} from 'buffer';

class SubModel {
    label: string;

export enum Plan {

class SimpleModel {
    id: string = uuid();

    tags: string[] = [];

    @Field(Buffer) //binary
    picture?: Buffer;

    type: number = 0;

    plan: Plan = Plan.DEFAULT;

    created: Date = new Date;

    children: SubModel[] = [];

    childrenMap: {[key: string]: SubModel} = {};

        public name: string
    ) {}

//data comes usually from files or http request
const instance = plainToClass(SimpleModel, {
    name: 'myName',
    tags: ['foo', 'bar'],
    plan: 'PRO',
    created: 'Sat Oct 13 2018 14:17:35 GMT+0200',
    children: [{label: 'foo'}],
    childrenMap: {'foo': {label: 'foo'}},
    SimpleModel {
      id: 'f2ee05ad-ca77-49ea-a571-8f0119e03038',
      name: 'myName',
      tags: ['foo', 'bar']
      type: 0,
      plan: 1,
      created: 2018-10-13T17:02:34.456Z,
      children: [ SubModel { label: 'foo' } ],
      childrenMap: { foo: SubModel { label: 'bar' } }



Make sure you have experimentalDecorators and emitDecoratorMetadata enabled in tsconfig.json:

  "compilerOptions": {
    "experimentalDecorators": true,
    "emitDecoratorMetadata": true

If you use Webpack's UglifyJsPlugin, make sure names are not mangled (mangle: false), which is the default. This is important to support constructor assignment.


Once your have defined your entity (see above) using the @Field decorators you can use one of Marshal's core methods to transform data.

Note: Class fields that aren't annotated either by @Field() or any other decorator won't be serialized. Their value will be dropped.


  • JSON object to class instance (plainToClass).

  • JSON object to mongo record (plainToMongo).

  • class instance to JSON object (classToPlain).

  • class instance to mongo record (classToMongo).

  • mongo record to class instance (mongoToClass).

  • mongo record to JSON object (mongoToPlain).

Note: 'JSON object' is not a string, but an object with valid JSON values, which can then be used to serialise to JSON string using JSON.stringify(classToPlain(SimpleModel, ...)).


You can validate incoming object literals or an class instance.

First make sure you have some validators attached to your fields you want to validate.

import {Field, validate, ValidationError, validatedPlainToClass, plainToClass} from '@marcj/marshal';

class Page {
    name: string;
    age: number;

const errors = validate(Page, {name: 'peter'});
expect(errors[0].message).toBe('Required value is undefined');
if (errors.length === 0) {
    const page = plainToClass(Page, {name: 'peter'});

//or do both at the same time and throw error if validations fails
const page = validatedPlainToClass(Page, {name: 'peter'});

You can also custom validators

import {Field, AddValidator, PropertyValidator, PropertyValidatorError, ClassType} from '@marcj/marshal';

class MyCustomValidator implements PropertyValidator {
     validate<T>(value: any, target: ClassType<T>, propertyName: string): PropertyValidatorError {
         if (value.length > 10) {
             return new PropertyValidatorError('Too long :()');

class Entity {
    name: string;

or inline validators

import {Field, AddValidator, InlineValidator, ClassType, PropertyValidatorError} from '@marcj/marshal';

class Entity {
    @InlineValidator((value: any) => {
        if (value.length > 10) {
            return new PropertyValidatorError('too_long', 'Too long :()');
    name: string;

Partial serialization

Most of the time, you want to have full class instances so the internal state is always valid. However, if you have a patch mechanism, use JSON Patch, or just want to change one field value in the database, you might have the need to serialize only one field.

  • partialPlainToClass
  • partialClassToPlain
  • partialClassToMongo
  • partialPlainToMongo
  • partialMongoToPlain

Types / Decorators

Class fields are annotated using mainly @Field decorators. You can define primitives, class mappings, relations between parents, and indices for the database (currently MongoDB).

Most of the time @Field is able to detect the primitive type by reading the emitted meta data from TypeScript when you declared the type correctly in Typescript.

Example valid decoration:

class Page {
    @Field() //will be detected as String
    name?: string;
    @Field([String]) //will be detected as String array
    name: string[] = [];
    @Field({String}) //will be detected as String map
    name: {[name: string]: string} = {};

Example not valid decorators:

class Page {
    @Field() //can't be detected, you get an error with further instructions

See documentation of @marcj/marshal for all available decorators. (Search for "Category: Decorator")

Available type decorators:

@FieldMap() //alternative to @Field({Type})
@FieldArray() //alternative to @Field([Type])
@EnumField(MyEnumClass) //implicitly calls @Field()
@UUIDField() //implicitly calls @Field()
@IDField() //necessary entities that should be stored in database, implicitly calls @Field()
@MongoIdField() //for database only, implicitly calls @Field()
@Index() //for database only. Can be used at class and class property.


The meta information about your entity can be exported to TypeORM EntitySchema.

// typeorm.js
import {getTypeOrmEntity} from "@marcj/marshal-mongo";

const TypeOrmSchema = getTypeOrmEntity(MyEntity);
module.exports = {
    type: "mongodb",
    host: "localhost",
    port: 27017,
    database: "test",
    useNewUrlParser: true,
    synchronize: true,
    entities: [TypeOrmSchema]

Marshal.ts uses only TypeORM for connection abstraction and to generate a EntitySchema for your typeOrm use-cases. You need in most cases only to use the @Field decorator with some other Marshal decorators (like @EnumField, @IDField, @UUIDField, @Index) on your entity.

You can generate a schema for Typeorm using getTypeOrmEntity and then pass this to your createConnection call, which makes it possible to sync the schema defined only with Marshal decorators with your database managed by Typeorm.


@Exclude() lets you exclude properties from a class in a certain direction. Per default it excludes to export to *toPlain and *toMongo. You can also use @ExcludeToMongo or @ExcludeToPlain to have more control. Note: Fields that are not decorated with @Field, @MongoIdField, @EnumField or @UUIDField are not mapped and will be omitted.

class MyEntity {
    id: string;
    internalState: string;


@ParentReference together with @Field is used for all *ToClass functions and allows you to have the parent from instance of class given in @Field assigned as reference. Properties that used @ParentReference are automatically excluded in *ToPlain and *ToMongo functions.

class Page {
    name: string;
    children: Page[] = [];
    parent?: PageClass;

const root = plainToClass(Page, {
    name: 'Root',
    children: [
        {name: 'Child 1'},
        {name: 'Child 2'},

root.children[0].parent === root; //true

OnLoad(options?: {fullLoad?: boolean})

With @OnLoad you can register one or multiple callbacks for the onLoad lifetime event. The registered method is called when the class has been instantiated with the *ToClass functions.

If fullLoad is true, the callback is called when the whole chain of objects has been created, which means when all parents and siblings are fully initialised.

class Page {
    name: string;
    onLoad() {

Value decorator

@Decorated lets you transform the actual class into something different. This is useful if you have in the actual class instance (plainToClass or mongoToClass) a wrapper for a certain property, for example string[] => ChildrenCollection.

class ChildrenCollection {
    items: string[];
    constructor(items: string[]) {
        this.items = items;
    public add(item: string) {

class MyEntity {
    id: string;
    //in *toMongo and *toPlain is children the value of ChildrenCollection::items
    children: ChildrenCollection = new ChildrenCollection([]);

ChildrenCollection is now always used in *toClass calls. The constructor of ChildrenCollection receives the actual value as first argument.

const entity = new MyEntity();
const result = classToPlain(MyEntity, entity);
result = {
    id: 'abcde',
    children: ['Foo', 'Bar']

If you read values from mongo or plain to class (mongoToClass, plainToClass) your decorator will be used again and receives as first argument the actual property value:

const entity = plainToClass(MyEntity, {
    id: 'abcde',
    children: ['Foo', 'Bar']
entity.children instanceof ChildrenCollection; //true

//so you can work with it again

Patch transformations

If you work with rather big entities, your probably want to utilise some kind of patch mechanism. Marshal supports to transform partial objects as well with deep path properties. All of following partial* methods maintain the structure of your object and only transform the value. We resolve the dot symbol to retrieve type information, so you can use this also in combination with JSON-Patch.


const converted = partialPlainToClass(SimpleModel, {
    id: 'abcde',
    ['childrenMap.item.label']: 3 

converted['childrenMap.item.label'] === '3' //true

const i2 = partialPlainToClass(SimpleModel, {
    'children': [{'label': 3}]

partialClassToPlain / partialClassToMongo

toPlain and toMongo differ in the way, that latter will transform @ObjectID and @UUID in different way, suitable for Mongo's binary storage.

const plain = partialClassToPlain(SimpleModel, {
    'children.0': i.children[0],
    'stringChildrenCollection': new StringCollectionWrapper(['Foo', 'Bar']),
    'childrenCollection': new CollectionWrapper([new SubModel('Bar3')]),
    'childrenCollection.1': new SubModel('Bar4'),
    'stringChildrenCollection.0': 'Bar2',
    'childrenCollection.2.label': 'Bar5',

expect(plain['stringChildrenCollection']).toEqual(['Foo', 'Bar']);
expect(plain['childrenCollection']).toEqual([{label: 'Bar3'}]);
expect(plain['childrenCollection.1']).toEqual({label: 'Bar4'});

Mongo Database

Marshal's MongoDB database abstraction makes it super easy to retrieve and store data from and into your MongoDB. We make sure the data from your JSON or class instance is correctly converted to MongoDB specific types and inserted IDs are applied to your class instance.

See documentation for more information;

npm install @marcj/marshal-mongo
import {plainToClass} from "@marcj/marshal";
import {Database} from "@marcj/marshal-mongo";
import {createConnection} from "typeorm";

(async () => {
const connection = await createConnection({
    type: "mongodb",
    host: "localhost",
    port: 27017,
    database: "testing",
    useNewUrlParser: true,
const database = new Database(connection, 'testing');

const instance = new SimpleModel('My model');

await, instance);

const list: SimpleModel[] = await database.find(SimpleModel);
const oneItem: SimpleModel = await database.get(
    {id: 'f2ee05ad-ca77-49ea-a571-8f0119e03038'}

NestJS / Express

It's super common to accept data from a frontend via HTTP, transform the body into your class instance, work with it, and then store that data in your MongoDB or somewhere else. With Marshal this scenario is super simple and you do not need any manual transformations.

npm install @marcj/marshal-nest
import {
    Controller, Get, Param, Post, Body
} from '@nestjs/common';

import {SimpleModel} from "@marcj/marshal/tests/entities";
import {Database, classToPlain} from "@marcj/marshal";
import {ValidationPipe} from "@marcj/marshal-nest";

class MyController {
    private database: Database;
    private async getDatabase() {
        if (!this.database) {
            const connection = await createConnection({
                type: "mongodb",
                host: "localhost",
                port: 27017,
                database: "testing",
                useNewUrlParser: true,
            this.database = new Database(connection, 'testing');
        return this.database;
    async save(
        @Body(ValidationPipe({transform: true})) body: SimpleModel,
    ) {
        body instanceof SimpleModel; // true;
        const versionNumber = await (await this.getDatabase()).save(SimpleModel, body);
    async get(@Param('id') id: string) {
        const instance = await (await this.getDatabase()).get(SimpleModel, {_id: id});

        return classToPlain(SimpleModel, instance);

Development of this package

How to release:

# remove packages/*/lib folders
# develop changes, run tests
./node_modules/.bin/jest --coverage

# build TS
lerna run tsc

# publish
lerna publish patch
You can’t perform that action at this time.