TypeScript - Repository pattern with Sequelize - node.js

I'm converting my Express API Template to TypeScript and I'm having some issues with the repositories.
With JavaScript, I would do something like this:
export default class BaseRepository {
async all() {
return this.model.findAll();
}
// other common methods
}
import BaseRepository from './BaseRepository';
import { User } from '../Models';
export default class UserRepository extends BaseRepository {
constructor() {
super();
this.model = User;
}
async findByEmail(email) {
return this.model.findOne({
where: {
email,
},
});
}
// other methods
Now, with TypeScript, the problem is that it doesn't know the type of this.model, and I can't pass a concrete model to BaseRepository, because, well, it is an abstraction. I've found that sequelize-typescript exports a ModelCtor which declares all the static model methods like findAll, create, etc., and I also could use another sequelize-typescript export which is Model to properly annotate the return type.
So, I ended up doing this:
import { Model, ModelCtor } from 'sequelize-typescript';
export default abstract class BaseRepository {
protected model: ModelCtor;
constructor(model: ModelCtor) {
this.model = model;
}
public async all(): Promise<Model[]> {
return this.model.findAll();
}
// other common methods
}
import { Model } from 'sequelize-typescript';
import BaseRepository from './BaseRepository';
import { User } from '../Models';
export default class UserRepository extends BaseRepository {
constructor() {
super(User);
}
public async findByEmail(email: string): Promise<Model | null> {
return this.model.findOne({
where: {
email,
},
});
}
// other methods
}
Ok, this works, TypeScript doesn't complain about methods like findOne or create not existing, but that generates another problem.
Now, for example, whenever I get a User from the repository, if I try to access one of its properties, like user.email, TypeScript will complain that this property does not exist. Of course, because the type Model does not know about the specifics of each model.
Ok, it's treason generics then.
Now BaseRepository uses a generic Model type which the methods also use:
export default abstract class BaseRepository<Model> {
public async all(): Promise<Model[]> {
return Model.findAll();
}
// other common methods
}
And the concrete classes pass the appropriate model to the generic type:
import BaseRepository from './BaseRepository';
import { User } from '../Models';
export default class UserRepository extends BaseRepository<User> {
public async findByEmail(email: string): Promise<User | null> {
return User.findOne({
where: {
email,
},
});
}
// other methods
}
Now IntelliSense lights up correctly, it shows both abstract and concrete classes methods and the model properties (e.g. user.email).
But, as you have imagined, that leads to more problems.
Inside BaseRepository, where the methods use the Model generic type, TypeScript complains that 'Model' only refers to a type, but is being used as a value here. Not only that, but TypeScript also doesn't know (again) that the static methods from the model exist, like findAll, create, etc.
Another problem is that in both abstract and concrete classes, as the methods don't use this anymore, ESLint expects the methods to be static: Expected 'this' to be used by class async method 'all'. Ok, I can just ignore this rule in the whole file and the error is gone. It would be even nicer to have all the methods set to static, so I don't have to instantiate the repository, but maybe I'm dreaming too much.
Worth mentioning that although I can just silence those errors with // #ts-ignore, when I execute this, it doesn't work: TypeError: Cannot read property 'create' of undefined\n at UserRepository.<anonymous>
I researched a lot, tried to make all methods static, but static methods can't reference the generic type (because it is considered an instance property), tried some workarounds, tried to pass the concrete model in the constructor of BaseRepository along with the class using the generic type, but nothing seems to work so far.
In case you want to check the code: https://github.com/andresilva-cc/express-api-template/tree/main/src/App/Repositories
EDIT:
Found this: Sequelize-Typescript typeof model
Ok, I removed some unnecessary code from that post and that kinda works:
import { Model } from 'sequelize-typescript';
export default abstract class BaseRepository<M extends Model> {
constructor(protected model: typeof Model) {}
public async all(attributes?: string[]): Promise<M[]> {
// Type 'Model<{}, {}>[]' is not assignable to type 'M[]'.
// Type 'Model<{}, {}>' is not assignable to type 'M'.
// 'Model<{}, {}>' is assignable to the constraint of type 'M', but 'M' could be instantiated with a different subtype of constraint 'Model<any, any>'.
return this.model.findAll({
attributes,
});
}
import BaseRepository from './BaseRepository';
import { User } from '../Models';
export default class UserRepository extends BaseRepository<User> {
constructor() {
super(User);
}
}
I mean, if I put some // #ts-ignore it at least executes, and IntelliSense lights up perfectly, but TypeScript complains.

We faced the same problem. The solution was to declare returning types with an interface that an abstract repository class implements.
Code for the interface:
export type RepoResult<M> = Promise<Result<M | undefined, RepoError | undefined>>;
export interface IRepo<M> {
save(model: M): RepoResult<M>;
findById(id: string): RepoResult<M>;
search(parameterName: string, parameterValue: string, sortBy: string, order: number, pageSize: number, pageNumber: number): RepoResult<M[]>;
getAll(): RepoResult<M[]>;
deleteById(id: string): RepoResult<M>;
findByIds(ids: string[]): RepoResult<M[]>;
deleteByIds(ids: string[]): RepoResult<any>;
};
Code for the abstract class:
export abstract class Repo<M extends sequelize.Model> implements IRepo<M> {
protected Model!: sequelize.ModelCtor<M>;
constructor(Model: sequelize.ModelCtor<M>) {
this.Model = Model;
}
public async save(doc: M) {
try {
const savedDoc = await doc.save();
return Result.ok(savedDoc);
} catch (ex: any) {
logger.error(ex);
return Result.fail(new RepoError(ex.message, 500));
}
}
public async findById(id: string) {
try {
const doc = await this.Model.findOne({where: {
id: id
}});
if (!doc) {
return Result.fail(new RepoError('Not found', 404));
}
return Result.ok(doc);
} catch (ex: any) {
return Result.fail(new RepoError(ex.message, 500));
}
}
}
Hope it helps. Have a nice day:)
EDIT:
Result is a class that looks like this:
export class Result<V, E> {
public isSuccess: boolean;
public isFailure: boolean;
private error: E;
private value: V;
private constructor(isSuccess: boolean, value: V, error: E) {
if (isSuccess && error) {
throw new Error('Successful result must not contain an error');
} else if (!isSuccess && value) {
throw new Error('Unsuccessful error must not contain a value');
}
this.isSuccess = isSuccess;
this.isFailure = !isSuccess;
this.value = value;
this.error = error;
}
public static ok<V>(value: V): Result<V, undefined> {
return new Result(true, value, undefined);
}
public static fail<E>(error: E): Result<undefined, E> {
return new Result(false, undefined, error);
}
public getError(): E {
if (this.isSuccess) {
throw new Error('Successful result does not contain an error');
}
return this.error;
}
public getValue(): V {
if (this.isFailure) {
throw new Error('Unsuccessful result does not contain a value');
}
return this.value;
}
}
RepoError class:
type RepoErrorCode = 404 | 500;
export class RepoError extends Error {
public code: RepoErrorCode;
constructor(message: string, code: RepoErrorCode) {
super(message);
this.code = code;
}
}
RepoResult type:
export type RepoResult<M> = Promise<Result<M | undefined, RepoError | undefined>>;
You can find more info on the pattern at the link below:
https://khalilstemmler.com/articles/enterprise-typescript-nodejs/functional-error-handling/

Related

NestJS lifecycle methods invoked without implementing their interface

I am having a small question about NestJS. In my code, there is a service which looks something like:
`
import { Inject, Injectable } from '#nestjs/common';
import neo4j, { Driver, int, Result, Transaction } from 'neo4j-driver';
import { Neo4jConfig } from './neo4j-config.interface';
import { NEO4J_CONFIG, NEO4J_DRIVER } from './neo4j.constants';
#Injectable()
export class Neo4jService {
constructor(
#Inject(NEO4J_CONFIG) private readonly config: Neo4jConfig,
#Inject(NEO4J_DRIVER) private readonly driver: Driver,
) {}
onApplicationBootstrap() {
console.log('Hello');
}
getDriver(): Driver {
return this.driver;
}
getConfig(): Neo4jConfig {
return this.config;
}
int(value: number) {
return int(value);
}
beginTransaction(database?: string): Transaction {
const session = this.getWriteSession(database);
return session.beginTransaction();
}
getReadSession(database?: string) {
return this.driver.session({
database: database || this.config.database,
defaultAccessMode: neo4j.session.READ,
});
}
getWriteSession(database?: string) {
return this.driver.session({
database: database || this.config.database,
defaultAccessMode: neo4j.session.WRITE,
});
}
read(
cypher: string,
params?: Record<string, unknown>,
databaseOrTransaction?: string | Transaction,
): Result {
if (databaseOrTransaction instanceof Transaction) {
return (<Transaction>databaseOrTransaction).run(cypher, params);
}
const session = this.getReadSession(<string>databaseOrTransaction);
return session.run(cypher, params);
}
write(
cypher: string,
params?: Record<string, unknown>,
databaseOrTransaction?: string | Transaction,
): Result {
if (databaseOrTransaction instanceof Transaction) {
return (<Transaction>databaseOrTransaction).run(cypher, params);
}
const session = this.getWriteSession(<string>databaseOrTransaction);
return session.run(cypher, params);
}
private onApplicationShutdown() {
console.log('Goodbye')
return this.driver.close();
}
}
`
Then in my main.ts file I have this method called:
`
await app.listen(port);
`
As you can see my service does not implement neither onApplicationBootstrap nor onApplicationShutdown.
How does it come that those methods still get invoked? Should I implement onApplicationBootstrap and onApplicationShutdown or not?
As you can also see I' d like that my onApplicationBootstrap is a private method which would not be possible if I implement the interface.
So, I would like to ask you:
Why the two lifecycle methods get called event without implementing the interface?
Should I implement those interfaces at all or just go on and use the methods which would allow me to define them as private?
I expected those methods to not work without implementing the interfaces
The Typescript interface is there to help us as devs. It doesn't exist at runtime, there's no information about it, so the only thing Nest can do is just check "Hey, does this class have the onModuleInit method?" If yes, add it to a list of classes to call onModuleInit. Do the same with the other lifecycle methods.
The interfaces aren't explicitly necessary, but they do give us devs a better idea of the class by just looking at the export class... line because we can see what is implemented/extended.

Using an Abstract class for mutations in NestJS with dynamic input #Args for the update mutation throws type error

Have been working on adding an abstract class to support versioning for Resolvers for a floorPlan feature and was wondering how I might fix this error. At the moment I do not get any Type errors but when the code compiles NestJs throws this error:
Undefined type error. Make sure you are providing an explicit type for the "floorPlanUpdate" (parameter at index [2]) of the "BaseResolverHost" class.
In a normal resolver this is not a problem as you pass in the type e.g.:
#Args('input') input: UpdateFloorPlanV1Input,
However as I am trying to determine the input type based off the version of the Floorplan using this line:
#Args('input') input: UpdateFloorPlanVersion<FloorPlanVersion>,
So it is not possible to explicitly declare it here. Let me know if this is possible or if anyone has any other approaches
This is the abstract Resolver code:
import { Args, Mutation, Query, Resolver } from '#nestjs/graphql';
import { UserId } from 'src/users/decorators/user-id.decorator';
import { CreateFloorPlanInput } from './v1/dto/create-floor-plan.input';
import { UpdateFloorPlanV1Input } from './v1/dto/update-floor-plan.input';
import { FloorPlanV1 } from './v1/models/floor-plan.model';
import { UpdateFloorPlanV2Input } from './v2/dto/update-floor-plan.input';
import { FloorPlanV2 } from './v2/models/floor-plan.model';
export type FloorPlanVersion = FloorPlanV1 | FloorPlanV2;
export type UpdateFloorPlanVersion<T extends FloorPlanVersion> =
T extends FloorPlanV1 ? UpdateFloorPlanV1Input : UpdateFloorPlanV2Input;
export interface FloorPlanServiceInterface<FloorPlanVersion> {
create(
userId: string,
createFloorPlanInput: CreateFloorPlanInput,
): Promise<FloorPlanVersion>;
findOne(userId: string, floorPlanId: string): Promise<FloorPlanVersion>;
update(
userId: string,
floorPlanId: string,
input: UpdateFloorPlanV1Input,
): Promise<FloorPlanVersion>;
}
export function BaseResolver<T extends Type<FloorPlanVersion>>(
classRef: T,
version: string,
) {
#Resolver({ isAbstract: true })
abstract class BaseResolverHost {
floorPlanService: FloorPlanServiceInterface<FloorPlanVersion>;
constructor(readonly service: FloorPlanServiceInterface<FloorPlanVersion>) {
this.floorPlanService = service;
}
#Mutation(() => classRef, { name: `floorPlanCreate${version}` })
async floorPlanCreate(
#UserId() userId,
#Args('input')
input: CreateFloorPlanInput,
) {
return this.floorPlanService.create(userId, input);
}
#Query(() => classRef, { name: `floorPlan${version}` })
async floorPlan(#UserId() userId, #Args('id') id: string) {
return this.floorPlanService.findOne(userId, id);
}
#Mutation(() => classRef, { name: `floorPlanUpdate${version}` })
async floorPlanUpdate(
#UserId() userId,
#Args('id') id: string,
#Args('input') input: UpdateFloorPlanVersion<FloorPlanVersion>,
) {
return this.floorPlanService.update(userId, id, input);
}
}
return BaseResolverHost;
}
And then it is called like this:
#Resolver(() => FloorPlanV2)
export class FloorPlanV2Resolver extends BaseResolver(FloorPlanV2, 'V2') {
constructor(
readonly floorPlanService: FloorPlanService,
) {
super(floorPlanService);
}
}
So if the ObjectType was FloorPlanV2 we would expect the UpdateFloorPlanInputType to be UpdateFloorPlanV2Input

NestJs - Validate request body using class-validator having 2 options for body class

I have a rest call, which might receive body of type classA or classB.
I need to keep it as 2 different classes.
Example -
// classes -
class ClassA {
#IsString()
#Length(1, 128)
public readonly name: string;
#IsString()
#Length(1, 128)
public readonly address: string;
}
class ClassB {
#IsString()
#Length(1, 10)
public readonly id: string;
}
// my request controller -
#Post('/somecall')
public async doSomething(
#Body(new ValidationPipe({transform: true})) bodyDto: (ClassA | ClassB) // < not validating any of them..
): Promise<any> {
// do something
}
The issue is, that when having more than one class, body is not validated.
How can I use 2 or more classes and validate them using class-validator?
I don't want to use same class..
Thank you all :)
I don't want to use same class..
Then it won't be possible, at least not with Nest's built-in ValidationPipe. Typescript doesn't reflect unions, intersections, or other kinds of generic types, so there's no returned metadata for this parameter, and if there's no metadata that's actionable Nest will end up skipping the pipe.
You could probably create a custom pipe to do the validation for you, and if you have two types you're probably going to have to. You can still call the appropriate class-transformer and class-validator methods inside of the class too.
import { CallHandler, ExecutionContext, Injectable, NestInterceptor } from '#nestjs/common';
import { of } from 'rxjs';
#Injectable()
export class CheckTypeInterceptor implements NestInterceptor {
constructor() {}
async intercept(context: ExecutionContext, next: CallHandler) /*: Observable<any>*/ {
const httpContext = context.switchToHttp();
const req = httpContext.getRequest();
const bodyDto = req.body.bodyDto;
// Need Update below logic
if (bodyDto instanceof ClassA || bodyDto instanceof ClassB) {
return next.handle();
}
// Return empty set
return of([]);
}
}
#UseInterceptors(CheckTypeInterceptor)
export class ApiController {
...
}
Encountered a similar situation where I had to validate some union type request. The solution I ended up with was a custom pipe as Jay McDoniel suggested here. The logic would vary depending on the request body you are dealing with, but per the question in case the following may work
Custom pipe:
import { ArgumentMetadata, BadRequestException, Inject, Scope } from "#nestjs/common";
import { PipeTransform } from "#nestjs/common";
import { plainToInstance } from "class-transformer";
import { validate } from "class-validator";
import { ClassADto } from '../repository/data-objects/class-a.dto';
import { ClassBDto } from '../repository/data-objects/class-b.dto';
export class CustomPipeName implements PipeTransform<any> {
async transform(value: any, { metatype, type }: ArgumentMetadata): Promise<any> {
if (type === 'body') {
const classA = plainToInstance(ClassADto, value);
const classB = plainToInstance(ClassBDto, value);
const classAValidationErrors = await validate(classA);
const classBValidationErrors = await validate(classB);
if (classAValidationErrors.length > 0 && classBValidationErrors.length > 0) {
throw new BadRequestException('some fancy info text');
}
}
return value;
}
}
Controller usage:
#Post('/somecall')
public async doSomething(
#Body(new CustomePipeName()) bodyDto: (ClassA | ClassB)
): Promise<any> {
// do something
}

This expression is not constructable. Type 'xxxx' has no construct signatures

I have a mistake that I can't understand what I'm doing wrong. I am following the instructions set out at https://docs.nestjs.com/techniques/mongodb. The difference is that I created interfaces to build different strategies for implementing a service (and for that I am making a wrapper in Model ).
import { Inject, Injectable } from "#nestjs/common";
import { Model } from "mongoose";
import { Documents } from "src/domain/schemas/documents.schema";
interface IDocumentRepository {
save();
}
interface IDocumentRepositoryFactory {
new(doc?: any): IDocumentRepository;
}
export function createIDocumentRepository(ctor: IDocumentRepositoryFactory, doc?: any): IDocumentRepository {
return new ctor(doc);
}
#Injectable()
export class DocumentRepository implements IDocumentRepository {
constructor(
private doc?: Documents,
#Inject()
private repository?: Model<Documents>
) {}
/* others fields and methods */
save() {
this.doc.save()
}
}
In others code points I call:
someMethod(repository: DocumentRepository) {
/* others codes */
const mdoc = new this.repository(mongoModelDoc); // <<<<---- error
mdoc.save();
/* others codes */
}
It is causing the following error:
This expression is not constructable. Type 'DocumentRepository' has no construct signatures.
return new this.repository(document)
What is wrong and how to resolve to meet this implementation?

How to extend a class in typescript

My service is designed in nodejs.
Below is my scenario
i have two controllers, one will be extending the other. there is a static function in both the controllers where in a static variable will be assigned some value.
depending on the condition of the data, im trying the make a call to the respective controller so that the static variable gets a appropriate assigned value.
Note:
The below code is just a snippet to explain the scenario and not the actual code of the application. But the order / calling / controller structure of this code snippet is exactly same. Also the listOfDept variable will be having separate business logic in the checkStart function of firstController and secondController.
// firstController.ts
firstController implements IFirstController {
private static listOfDept: string[];
static checkStart(){
firstController.listOfDept = // my logic to fill this object
}
constructor (){}
}
getRelevantData(next: (error: string, response: any) => void): void {
var myObject = firstController.listOfDept;
this.myRepository.uniqueData(myObject, next);
}
}
firstController.checkStart();
export = firstController;
//ifirstController.ts
interface IFirstController {
getRelevantData(next: (error: string, response: any) => void): void;
}
// secondController.ts
secondController extends firstController implements iSecondController {
private static listOfDept: string[];
static checkStart(){
firstController.listOfDept = ["Computer Science"];
}
constructor (){
super();
}
}
secondController.checkStart();
export = secondController;
//isecondController.ts
interface ISecondController implements ifirstController{}
//Controller calling the getRelevantData function
//middlewareController
middlewareController implements IMiddlewareController {
constructor(private firstController: IFirstController, private secondController: ISecondController) {
}
getDepData(data: any, next: (error: string, response: any) => void): void {
if(data.url = "fromParent") {
// im expecting this to make a call to checkStart() of firstController
this.firstController.getRelevantData();
} else {
// im expecting this to make a call to checkStart() of secondController
this.secondController.getRelevantData();
}
}
}
Problem faced with the above code
No matter which way the getRelevantData function is getting called, im always getting the value of listOfDept as computer science. It is never going in the checkStart function of first controller.
In general I would discourage using static methods for this kind of initialization and instead inject the required data into constructors or create factory methods for creating object with necessary data.
But, if you do want to use static properties, the problem is that you need to refer to the right parent class in the getRelevantData implementation. The class that constructed the instance can be accessed through constructor property. TypeScript does not process this scenario well, so you have to make a type cast:
// firstController.ts
class firstController implements IFirstController {
// Need to be `protected` to be accessible from subclass
protected static listOfDept: string[];
static checkStart(){
firstController.listOfDept; // my logic to fill this object
}
constructor (){}
getRelevantData(next: (error: string, response: any) => void): void {
// You need to refer to the constructor
let Class = this.constructor as typeof firstController;
var myObject = Class.listOfDept;
// the rest
}
}
firstController.checkStart();
//ifirstController.ts
interface IFirstController {
getRelevantData(next: (error: string, response: any) => void): void;
}
// secondController.ts
class secondController extends firstController implements ISecondController {
// No `listOfDept` definition here
static checkStart(){
secondController.listOfDept = ["Computer Science"];
}
constructor (){
super();
}
}
secondController.checkStart();

Resources