- 1. API with NestJS #1. Controllers, routing and the module structure
- 2. API with NestJS #2. Setting up a PostgreSQL database with TypeORM
- 3. API with NestJS #3. Authenticating users with bcrypt, Passport, JWT, and cookies
- 4. API with NestJS #4. Error handling and data validation
- 5. API with NestJS #5. Serializing the response with interceptors
- 6. API with NestJS #6. Looking into dependency injection and modules
- 7. API with NestJS #7. Creating relationships with Postgres and TypeORM
- 8. API with NestJS #8. Writing unit tests
- 9. API with NestJS #9. Testing services and controllers with integration tests
- 10. API with NestJS #10. Uploading public files to Amazon S3
- 11. API with NestJS #11. Managing private files with Amazon S3
- 12. API with NestJS #12. Introduction to Elasticsearch
- 13. API with NestJS #13. Implementing refresh tokens using JWT
- 14. API with NestJS #14. Improving performance of our Postgres database with indexes
- 15. API with NestJS #15. Defining transactions with PostgreSQL and TypeORM
- 16. API with NestJS #16. Using the array data type with PostgreSQL and TypeORM
- 17. API with NestJS #17. Offset and keyset pagination with PostgreSQL and TypeORM
- 18. API with NestJS #18. Exploring the idea of microservices
- 19. API with NestJS #19. Using RabbitMQ to communicate with microservices
- 20. API with NestJS #20. Communicating with microservices using the gRPC framework
- 21. API with NestJS #21. An introduction to CQRS
- 22. API with NestJS #22. Storing JSON with PostgreSQL and TypeORM
- 23. API with NestJS #23. Implementing in-memory cache to increase the performance
- 24. API with NestJS #24. Cache with Redis. Running the app in a Node.js cluster
- 25. API with NestJS #25. Sending scheduled emails with cron and Nodemailer
- 26. API with NestJS #26. Real-time chat with WebSockets
- 27. API with NestJS #27. Introduction to GraphQL. Queries, mutations, and authentication
- 28. API with NestJS #28. Dealing in the N + 1 problem in GraphQL
- 29. API with NestJS #29. Real-time updates with GraphQL subscriptions
- 30. API with NestJS #30. Scalar types in GraphQL
- 31. API with NestJS #31. Two-factor authentication
- 32. API with NestJS #32. Introduction to Prisma with PostgreSQL
- 33. API with NestJS #33. Managing PostgreSQL relationships with Prisma
- 34. API with NestJS #34. Handling CPU-intensive tasks with queues
- 35. API with NestJS #35. Using server-side sessions instead of JSON Web Tokens
- 36. API with NestJS #36. Introduction to Stripe with React
- 37. API with NestJS #37. Using Stripe to save credit cards for future use
- 38. API with NestJS #38. Setting up recurring payments via subscriptions with Stripe
- 39. API with NestJS #39. Reacting to Stripe events with webhooks
- 40. API with NestJS #40. Confirming the email address
- 41. API with NestJS #41. Verifying phone numbers and sending SMS messages with Twilio
- 42. API with NestJS #42. Authenticating users with Google
- 43. API with NestJS #43. Introduction to MongoDB
- 44. API with NestJS #44. Implementing relationships with MongoDB
- 45. API with NestJS #45. Virtual properties with MongoDB and Mongoose
- 46. API with NestJS #46. Managing transactions with MongoDB and Mongoose
- 47. API with NestJS #47. Implementing pagination with MongoDB and Mongoose
- 48. API with NestJS #48. Definining indexes with MongoDB and Mongoose
- 49. API with NestJS #49. Updating with PUT and PATCH with MongoDB and Mongoose
- 50. API with NestJS #50. Introduction to logging with the built-in logger and TypeORM
- 51. API with NestJS #51. Health checks with Terminus and Datadog
- 52. API with NestJS #52. Generating documentation with Compodoc and JSDoc
- 53. API with NestJS #53. Implementing soft deletes with PostgreSQL and TypeORM
- 54. API with NestJS #54. Storing files inside a PostgreSQL database
- 55. API with NestJS #55. Uploading files to the server
- 56. API with NestJS #56. Authorization with roles and claims
- 57. API with NestJS #57. Composing classes with the mixin pattern
- 58. API with NestJS #58. Using ETag to implement cache and save bandwidth
- 59. API with NestJS #59. Introduction to a monorepo with Lerna and Yarn workspaces
- 60. API with NestJS #60. The OpenAPI specification and Swagger
- 61. API with NestJS #61. Dealing with circular dependencies
- 62. API with NestJS #62. Introduction to MikroORM with PostgreSQL
- 63. API with NestJS #63. Relationships with PostgreSQL and MikroORM
- 64. API with NestJS #64. Transactions with PostgreSQL and MikroORM
- 65. API with NestJS #65. Implementing soft deletes using MikroORM and filters
- 66. API with NestJS #66. Improving PostgreSQL performance with indexes using MikroORM
- 67. API with NestJS #67. Migrating to TypeORM 0.3
- 68. API with NestJS #68. Interacting with the application through REPL
- 69. API with NestJS #69. Database migrations with TypeORM
- 70. API with NestJS #70. Defining dynamic modules
- 71. API with NestJS #71. Introduction to feature flags
- 72. API with NestJS #72. Working with PostgreSQL using raw SQL queries
- 73. API with NestJS #73. One-to-one relationships with raw SQL queries
- 74. API with NestJS #74. Designing many-to-one relationships using raw SQL queries
- 75. API with NestJS #75. Many-to-many relationships using raw SQL queries
- 76. API with NestJS #76. Working with transactions using raw SQL queries
- 77. API with NestJS #77. Offset and keyset pagination with raw SQL queries
- 78. API with NestJS #78. Generating statistics using aggregate functions in raw SQL
- 79. API with NestJS #79. Implementing searching with pattern matching and raw SQL
- 80. API with NestJS #80. Updating entities with PUT and PATCH using raw SQL queries
- 81. API with NestJS #81. Soft deletes with raw SQL queries
- 82. API with NestJS #82. Introduction to indexes with raw SQL queries
- 83. API with NestJS #83. Text search with tsvector and raw SQL
- 84. API with NestJS #84. Implementing filtering using subqueries with raw SQL
- 85. API with NestJS #85. Defining constraints with raw SQL
- 86. API with NestJS #86. Logging with the built-in logger when using raw SQL
- 87. API with NestJS #87. Writing unit tests in a project with raw SQL
- 88. API with NestJS #88. Testing a project with raw SQL using integration tests
- 89. API with NestJS #89. Replacing Express with Fastify
- 90. API with NestJS #90. Using various types of SQL joins
- 91. API with NestJS #91. Dockerizing a NestJS API with Docker Compose
- 92. API with NestJS #92. Increasing the developer experience with Docker Compose
- 93. API with NestJS #93. Deploying a NestJS app with Amazon ECS and RDS
- 94. API with NestJS #94. Deploying multiple instances on AWS with a load balancer
- 95. API with NestJS #95. CI/CD with Amazon ECS and GitHub Actions
- 96. API with NestJS #96. Running unit tests with CI/CD and GitHub Actions
- 97. API with NestJS #97. Introduction to managing logs with Amazon CloudWatch
- 98. API with NestJS #98. Health checks with Terminus and Amazon ECS
- 99. API with NestJS #99. Scaling the number of application instances with Amazon ECS
- 100. API with NestJS #100. The HTTPS protocol with Route 53 and AWS Certificate Manager
- 101. API with NestJS #101. Managing sensitive data using the AWS Secrets Manager
- 102. API with NestJS #102. Writing unit tests with Prisma
- 103. API with NestJS #103. Integration tests with Prisma
- 104. API with NestJS #104. Writing transactions with Prisma
- 105. API with NestJS #105. Implementing soft deletes with Prisma and middleware
- 106. API with NestJS #106. Improving performance through indexes with Prisma
- 107. API with NestJS #107. Offset and keyset pagination with Prisma
- 108. API with NestJS #108. Date and time with Prisma and PostgreSQL
- 109. API with NestJS #109. Arrays with PostgreSQL and Prisma
- 110. API with NestJS #110. Managing JSON data with PostgreSQL and Prisma
- 111. API with NestJS #111. Constraints with PostgreSQL and Prisma
- 112. API with NestJS #112. Serializing the response with Prisma
- 113. API with NestJS #113. Logging with Prisma
- 114. API with NestJS #114. Modifying data using PUT and PATCH methods with Prisma
- 115. API with NestJS #115. Database migrations with Prisma
- 116. API with NestJS #116. REST API versioning
- 117. API with NestJS #117. CORS – Cross-Origin Resource Sharing
- 118. API with NestJS #118. Uploading and streaming videos
- 119. API with NestJS #119. Type-safe SQL queries with Kysely and PostgreSQL
- 120. API with NestJS #120. One-to-one relationships with the Kysely query builder
- 121. API with NestJS #121. Many-to-one relationships with PostgreSQL and Kysely
- 122. API with NestJS #122. Many-to-many relationships with Kysely and PostgreSQL
- 123. API with NestJS #123. SQL transactions with Kysely
- 124. API with NestJS #124. Handling SQL constraints with Kysely
- 125. API with NestJS #125. Offset and keyset pagination with Kysely
- 126. API with NestJS #126. Improving the database performance with indexes and Kysely
- 127. API with NestJS #127. Arrays with PostgreSQL and Kysely
- 128. API with NestJS #128. Managing JSON data with PostgreSQL and Kysely
- 129. API with NestJS #129. Implementing soft deletes with SQL and Kysely
- 130. API with NestJS #130. Avoiding storing sensitive information in API logs
- 131. API with NestJS #131. Unit tests with PostgreSQL and Kysely
- 132. API with NestJS #132. Handling date and time in PostgreSQL with Kysely
- 133. API with NestJS #133. Introducing database normalization with PostgreSQL and Prisma
- 134. API with NestJS #134. Aggregating statistics with PostgreSQL and Prisma
- 135. API with NestJS #135. Referential actions and foreign keys in PostgreSQL with Prisma
- 136. API with NestJS #136. Raw SQL queries with Prisma and PostgreSQL range types
- 137. API with NestJS #137. Recursive relationships with Prisma and PostgreSQL
- 138. API with NestJS #138. Filtering records with Prisma
- 139. API with NestJS #139. Using UUID as primary keys with Prisma and PostgreSQL
- 140. API with NestJS #140. Using multiple PostgreSQL schemas with Prisma
- 141. API with NestJS #141. Getting distinct records with Prisma and PostgreSQL
- 142. API with NestJS #142. A video chat with WebRTC and React
- 143. API with NestJS #143. Optimizing queries with views using PostgreSQL and Kysely
- 144. API with NestJS #144. Creating CLI applications with the Nest Commander
- 145. API with NestJS #145. Securing applications with Helmet
- 146. API with NestJS #146. Polymorphic associations with PostgreSQL and Prisma
- 147. API with NestJS #147. The data types to store money with PostgreSQL and Prisma
- 148. API with NestJS #148. Understanding the injection scopes
- 149. API with NestJS #149. Introduction to the Drizzle ORM with PostgreSQL
- 150. API with NestJS #150. One-to-one relationships with the Drizzle ORM
- 151. API with NestJS #151. Implementing many-to-one relationships with Drizzle ORM
- 152. API with NestJS #152. SQL constraints with the Drizzle ORM
- 153. API with NestJS #153. SQL transactions with the Drizzle ORM
- 154. API with NestJS #154. Many-to-many relationships with Drizzle ORM and PostgreSQL
- 155. API with NestJS #155. Offset and keyset pagination with the Drizzle ORM
- 156. API with NestJS #156. Arrays with PostgreSQL and the Drizzle ORM
- 157. API with NestJS #157. Handling JSON data with PostgreSQL and the Drizzle ORM
- 158. API with NestJS #158. Soft deletes with the Drizzle ORM
- 159. API with NestJS #159. Date and time with PostgreSQL and the Drizzle ORM
- 160. API with NestJS #160. Using views with the Drizzle ORM and PostgreSQL
- 161. API with NestJS #161. Generated columns with the Drizzle ORM and PostgreSQL
- 162. API with NestJS #162. Identity columns with the Drizzle ORM and PostgreSQL
- 163. API with NestJS #163. Full-text search with the Drizzle ORM and PostgreSQL
- 164. API with NestJS #164. Improving the performance with indexes using Drizzle ORM
- 165. API with NestJS #165. Time intervals with the Drizzle ORM and PostgreSQL
- 166. API with NestJS #166. Logging with the Drizzle ORM
- 167. API with NestJS #167. Unit tests with the Drizzle ORM
- 168. API with NestJS #168. Integration tests with the Drizzle ORM
When developing our REST APIs, we often focus on implementing the four fundamental operations: creating, reading, updating, and deleting (CRUD). The most basic approach to removing a record from our database is to delete it permanently. In this article, we explore the idea of soft deletes that allow us to keep the removed entities in the database.
Explaining soft deletes
The most straightforward way of implementing the soft deletes feature is through a boolean flag.
1 2 3 4 5 |
CREATE TABLE categories ( id int GENERATED BY DEFAULT AS IDENTITY PRIMARY KEY, name text NOT NULL, is_deleted boolean DEFAULT false ) |
By adding the is_deleted boolean DEFAULT false line above, we add the is_deleted column that has the false value by default.
Whenever we want to remove a record in the categories table, we don’t have to delete it permanently. Instead, we change the value in the is_deleted column to true.
1 2 3 |
UPDATE categories SET is_deleted = true WHERE name = 'NestJS' |
The most important thing to understand about soft deletes is that they affect various queries we perform. Let’s try fetching a list of all the categories, for example.
1 |
SELECT * FROM categories |
The above query returns all categories, including the deleted ones. So, to fix this issue, we need to filter out the removed records.
1 2 |
SELECT * FROM categories WHERE is_deleted = false |
Soft delete benefits
A significant benefit of soft deletes is that we can effortlessly restore the deleted record. This allows for a user experience far superior to restoring backups. For example, in our application, we can implement an undo button that changes the value in the is_deleted column back to false.
We can also take advantage of querying the deleted records from the database. They can prove to be helpful when generating various reports, for example.
Soft deletes can also be useful when dealing with relationships. For example, permanently deleting a record referenced in another table can cause a foreign constraint violation. This does not happen with soft deletes because we don’t remove the entities from the database.
Soft delete drawbacks
An important disadvantage of soft deletes is that we need to consider them in all related queries. Whenever we fetch our data and forget to filter by the is_deleted column, we might show the user the data they shouldn’t have access to. Implementing filtering can also affect our performance.
Another important consideration is related to the unique constraint. Let’s look at the users table we’ve defined in one of the previous parts of this series.
userSchema.prisma
1 2 3 4 5 6 7 8 9 |
model User { id Int @id @default(autoincrement()) email String @unique name String password String address Address? @relation(fields: [addressId], references: [id]) addressId Int? @unique posts Post[] } |
In our model, we require every email to be unique. When we delete a record permanently, we make the email available to other users. However, removing users through soft deletes does not make their email addresses available for reuse.
Implementing soft deletes with Prisma
The first step when implementing soft deletes is to add the appropriate column to our table. A common approach to soft deletes is storing the deletion date instead of a simple boolean flag.
categorySchema.prisma
1 2 3 4 5 6 |
model Category { id Int @id @default(autoincrement()) name String posts Post[] deletedAt DateTime? @db.Timestamptz } |
Above, we’ve added the deletedAt property, which is a timestamp with a timezone. We will assign it with value whenever we want to delete a record from our database.
If you want to know more about managing date and time with PostgreSQL, check out Managing date and time with PostgreSQL and TypeORM
Now, we need to generate our migration using the Prisma CLI.
1 |
npx prisma migrate dev --name category-add-deleted-at-column |
Doing the above creates a new file in the migrations directory.
migrations/20230423200453_category_add_deleted_at_column/migration.sql
1 |
ALTER TABLE "Category" ADD COLUMN "deletedAt" TIMESTAMPTZ; |
The official Prisma documentation suggests implementing soft deletes with middleware. Let’s explain this concept first.
Middleware in Prisma
With middleware, we can perform an action before or after a query runs. To attach a middleware, we need to use the $use method. An appropriate place to do that is in our PrismaService class.
prisma.service.ts
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 |
import { Injectable, OnModuleInit, OnModuleDestroy } from '@nestjs/common'; import { PrismaClient, Prisma } from '@prisma/client'; @Injectable() export class PrismaService extends PrismaClient implements OnModuleInit, OnModuleDestroy { async onModuleInit() { await this.$connect(); this.$use(this.loggingMiddleware); } loggingMiddleware: Prisma.Middleware = async (params, next) => { console.log( `${params.action} ${params.model} ${JSON.stringify(params.args)}`, ); const result = await next(params); console.log(result); return result; }; async onModuleDestroy() { await this.$disconnect(); } } |
1 delete Category {"where":{"id":8}}
1 Result: {id: 8, name: "Category"}
Above, we pass a callback to the $use method. Its first argument we call params, contains available parameters, such as the performed action, the model, and the arguments. The second argument, called next, is a function that calls the next action in the chain.
Middleware for deleting
To implement soft deletes with middleware, we need to modify the Prisma query every time a record of a particular type is deleted. We need to change a delete action into the update action and provide the appropriate date.
prisma.service.ts
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 |
import { Injectable, OnModuleInit, OnModuleDestroy } from '@nestjs/common'; import { PrismaClient, Prisma } from '@prisma/client'; @Injectable() export class PrismaService extends PrismaClient implements OnModuleInit, OnModuleDestroy { async onModuleInit() { await this.$connect(); this.$use(this.categorySoftDeleteMiddleware); } categorySoftDeleteMiddleware: Prisma.Middleware = async (params, next) => { if (params.model !== 'Category') { return next(params); } if (params.action === 'delete') { return next({ ...params, action: 'update', args: { ...params.args, data: { deletedAt: new Date(), }, }, }); } return next(params); }; async onModuleDestroy() { await this.$disconnect(); } } |
If we also use the deleteMany action, we need to handle it separately.
Thanks to the above middleware, every time we call the prismaService.category.delete method, we perform the update action under the hood instead.
Middleware for querying
Now we need to write a middleware that filters out removed categories whenever we fetch them. We need to handle fetching a single category and fetching multiple categories separately.
prisma.service.ts
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 |
import { Injectable, OnModuleInit, OnModuleDestroy } from '@nestjs/common'; import { PrismaClient, Prisma } from '@prisma/client'; @Injectable() export class PrismaService extends PrismaClient implements OnModuleInit, OnModuleDestroy { async onModuleInit() { await this.$connect(); this.$use(this.categorySoftDeleteMiddleware); this.$use(this.categoryFindMiddleware); } categoryFindMiddleware: Prisma.Middleware = async (params, next) => { if (params.model !== 'Category') { return next(params); } if (params.action === 'findUnique' || params.action === 'findFirst') { return next({ ...params, action: 'findFirst', args: { ...params.args, where: { ...params.args?.where, deletedAt: null, }, }, }); } if (params.action === 'findMany') { return next({ ...params, args: { ...params.args, where: { ...params.args?.where, deletedAt: null, }, }, }); } return next(params); }; // ... } |
Thanks to the above approach, whenever we use the findUnique, findFirst, and findMany actions, it filters our deleted categories under the hood.
Summary
An advantage of the above approach is that with correctly written middleware, we won’t forget to handle the deletedAt column properly whenever we fetch or delete our entities.
However, this can make our code quite a bit messy. Instead of interacting with the categories table through the CategoriesService, we spread our business logic into unrelated parts of our application. It might also not be instantly apparent to our teammates that whenever they call the this.prismaService.category.delete method, it runs the update action under the hood.
The middleware approach might be suitable in less structured applications that don’t use the NestJS framework. However, in our case, we put the business logic related to categories in the CategoriesService. Therefore, it might make more sense to ditch middleware and implement soft deletes directly through our service for simplicity and readability. Nevertheless, it’s a very good example to learn how the middleware works in Prisma.
Good afternoon Wanago, do you know how to use prisma extensions with nestjs since the $use method is deprecated
Issue https://github.com/prisma/prisma/issues/18628
Hi, this is a great point, in a new project that I’m using Prisma, the $use method is deprecated, a solution for this would be great
Another drawback of middleware is that it doesn’t apply to nested relations, therefore, implementing soft delete in each service is better now.