Detailed explanation of Nest.js parameter validation and custom return data format
0x0 Parameter verificationMost of the parameter verification business is implemented using the pipeline method in Nest.js. For details, please refer to the documentation. However, I encountered some problems during the writing process, although the documentation is rather obscure. Make a query interface, which contains some parameters and makes it into dto structure data:
import { ApiProperty } from '@nestjs/swagger'
export class QueryUserDto {
@ApiProperty({
required: false,
description: 'Page number'
})
readonly currentPage: number
@ApiProperty({
required: false,
description: 'Number of items'
})
readonly pageSize: number
@ApiProperty({
required: false,
description: 'User account'
})
readonly username?: string
@ApiProperty({
required: false,
description: 'User status'
})
readonly activeStatus: number
@ApiProperty({
required: false,
description: 'Sort by: ASC, DESC'
})
readonly order: 'DESC' | 'ASC'
}
TYPESCRIPT
When the corresponding parameters are passed in the @Query request, it is found that the data types obtained are all String. Then, after consulting the relevant documents, it is realized that the Type of class-transformer is also needed for conversion:
import { ApiProperty } from '@nestjs/swagger'
import { Type } from 'class-transformer'
export class QueryUserDto {
@ApiProperty({
required: false,
description: 'Page number'
})
@Type(() => Number)
readonly currentPage: number = 1
@ApiProperty({
required: false,
description: 'Number of items'
})
@Type(() => Number)
readonly pageSize: number = 10
@ApiProperty({
required: false,
description: 'User account'
})
readonly username?: string
@ApiProperty({
required: false,
description: 'User status'
})
@Type(() => Number)
readonly activeStatus: number = 3
@ApiProperty({
required: false,
description: 'Sort by: ASC, DESC'
})
readonly order: 'DESC' | 'ASC' = 'DESC'
}
Then enable the transform option in the ValidationPipe pipeline method:
app.useGlobalPipes(
new ValidationPipe({
transform: true
})
)
Or inject in app.modules.ts:
import { ValidationPipe } from '@nestjs/common'
import { APP_PIPE } from '@nestjs/core'
@Module({
imports: [
// ...
],
controllers: [AppController],
providers:
{
provide: APP_PIPE,
useValue: new ValidationPipe({
transform: true
})
}
]
})
The difference between the two methods of use is whether the program is a mixed application type. To save trouble, I write it directly in the global method. The data finally obtained by the service is the data processed by the pipeline business, and there is no need to perform a lot of data type judgment at the service layer. 0x1 Customize the return data formatThe data returned by the controller comes from the database table structure:
{
"id": "d8d5a56c-ee9f-4e41-be48-5414a7a5712c",
"username": "Akeem.Cremin",
"password": "$2b$10$kRcsmN6ewFC2GOs0TEg6TuvDbNzf1VGCbQf2fI1UeyPAiZCq9rMKm",
"email": "[email protected]",
"nickname": "Wallace Nicolas",
"role": "user",
"isActive": true,
"createdTime": "2021-03-24T15:24:26.806Z",
"updatedTime": "2021-03-24T15:24:26.806Z"
}
If you need to define the data format of the final return interface, for example:
{
"statusCode": 200,
"message": "Get Success",
"data": {
"id": "d8d5a56c-ee9f-4e41-be48-5414a7a5712c",
"username": "Akeem.Cremin",
"password": "$2b$10$kRcsmN6ewFC2GOs0TEg6TuvDbNzf1VGCbQf2fI1UeyPAiZCq9rMKm",
"email": "[email protected]",
"nickname": "Wallace Nicolas",
"role": "user",
"isActive": true,
"createdTime": "2021-03-24T15:24:26.806Z",
"updatedTime": "2021-03-24T15:24:26.806Z"
}
}
Here you need to make a custom success request interceptor: nest g in shared/interceptor/transform
import { CallHandler, ExecutionContext, Injectable, Logger, NestInterceptor } from '@nestjs/common'
import { Observable } from 'rxjs'
import { map } from 'rxjs/operators'
import { Request } from 'express'
interface Response<T> {
data: T
}
@Injectable()
export class TransformInterceptor<T> implements NestInterceptor<T, Response<T>> {
intercept(context: ExecutionContext, next: CallHandler<T>): Observable<any> {
const request = context.switchToHttp().getRequest<Request>()
Logger.log(request.url, 'Normal interface request')
return next.handle().pipe(
map(data => {
return {
data: data,
statusCode: 200,
message: 'Request successful'
}
})
)
}
}
Then import it in app.module.ts and use it:
import { ValidationPipe } from '@nestjs/common'
import { APP_INTERCEPTOR } from '@nestjs/core'
import { TransformInterceptor } from '@/shared/interceptor/transform.interceptor'
@Module({
imports: [
// ...
],
controllers: [AppController],
providers:
{
provide: APP_INTERCEPTOR,
useClass: TransformInterceptor
}
]
})
However, please pay attention to the order of APP_INTERCEPTOR. It is best to put TransformInterceptor first, otherwise it will fail. Error Filters: nest gf shared/filters/httpException
import { ArgumentsHost, Catch, ExceptionFilter, HttpException, Logger } from '@nestjs/common'
import { Response, Request } from 'express'
@Catch(HttpException)
export class HttpExceptionFilter implements ExceptionFilter {
catch(exception: HttpException, host: ArgumentsHost) {
const context = host.switchToHttp()
const response = context.getResponse<Response>()
const request = context.getRequest<Request>()
const status = exception.getStatus()
const message = exception.message
Logger.log(`${request.url} - ${message}`, 'Abnormal interface request')
response.status(status).json({
statusCode: status,
message: message,
path: request.url,
timestamp: new Date().toISOString()
})
}
}
Then import it in app.module.ts and use it:
import { ValidationPipe } from '@nestjs/common'
import { APP_FILTER } from '@nestjs/core'
import { HttpExceptionFilter } from '@/shared/filters/http-exception.filter'
@Module({
imports: [
// ...
],
controllers: [AppController],
providers:
{
provide: APP_FILTER,
useClass: HttpExceptionFilter
}
]
})
0x2 Hide a field in the entity classI originally wanted to use the @Exclude attribute to hide some sensitive fields in the database, but found that it could not meet special needs. If a single instance is returned, hiding can be achieved, but I have a findAll that cannot be implemented. The above is explained in great detail in the Serialization | NestJS - A progressive Node.js framework document, but there is another way here. First, add attributes to the strength-sensitive data fields:
import { BaseEntity, Entity, Column, PrimaryGeneratedColumn } from 'typeorm'
@Entity('user')
export class UserEntity extends BaseEntity {
@PrimaryGeneratedColumn('uuid', {
comment: 'User ID'
})
id: string
@Column({
type: 'varchar',
length: 50,
unique: true,
comment: 'Logged in user'
})
username: string
@Column({
type: 'varchar',
length: 200,
select: false,
comment: 'Password'
})
password: string
select: false can hide this field when returning query results, but all queries involving this field must add this field, such as in my login query in user.service.ts:
const user = await getRepository(UserEntity)
.createQueryBuilder('user')
.where('user.username = :username', { username })
.addSelect('user.password')
.getOne()
.addSelect('user.password') Adding this attribute query will include the password field, otherwise the normal query method will not include this field. SummarizeThis is the end of this article about Nest.js parameter validation and custom return data format. For more information about Nest.js parameter validation and data format, please search for previous articles on 123WORDPRESS.COM or continue to browse the following related articles. I hope you will support 123WORDPRESS.COM in the future! |
<<: Sample code for configuring nginx to support https
>>: A complete guide on how to query and delete duplicate records in MySQL
Recommend
MySQL 8.0 upgrade experience
Table of contents Preface 1. First completely uni...
Sample code for seamless scrolling with flex layout
This article mainly introduces the sample code of...
Mysql 5.6 adds a method to modify username and password
Log in to MySQL first shell> mysql --user=root...
How to use regular expressions to automatically match wildcard domain names in nginx
Nginx uses regular expressions to automatically m...
A brief discussion on whether too many MySQL data queries will cause OOM
Table of contents Impact of full table scan on th...
How to install MySQL and Redis in Docker
This article is based on the CentOS 7.3 system en...
How to check disk usage in Linux
1. Use the df command to view the overall disk us...
HTML symbol to entity algorithm challenge
challenge: Converts the characters &, <, &...
Detailed explanation of how to select all child elements using CSS
How to recursively select all child elements usin...
Tutorial on how to modify element.style inline styles
Preface When we were writing the web page style a...
How to place large images in a small space on a web page
Original source: www.bamagazine.com There are nar...
How to use filters to implement monitoring in Zabbix
Recently, when I was working on monitoring equipm...
MySQL 8 new features: Invisible Indexes
background Indexes are a double-edged sword. Whil...
A brief discussion on MySQL large table optimization solution
background The amount of new data in the business...
Solution to the problem of Failed when configuring IDEA to connect to MYSQL database
This article mainly introduces the solution to th...