Orionjs schemas seamlessly integrate with MongoDB through the @orion-js/mongodb package, providing type-safe database operations.

​
Defining Collections

To create a MongoDB collection with schema validation: 
import {createCollection} from '@orion-js/mongodb'
import {schemaWithName, InferSchemaType} from '@orion-js/schema'

// Define your schema
export const UserSchema = schemaWithName('UserSchema', {
  _id: {type: String},
  name: {type: String},
  email: {type: String},
  createdAt: {type: Date}
})

// Infer the TypeScript type from the schema
export type UserType = InferSchemaType<typeof UserSchema>

// Create a typed collection
const Users = createCollection({
  name: 'users',
  schema: UserSchema
})

// Now you have type-safe database operations
const user: UserType = await Users.findOne({email: 'user@example.com'})

​
TypedId Support

Orionjs MongoDB has built-in support for typed IDs with automatic prefixing: 
import {typedId} from '@orion-js/mongodb'
import {schemaWithName, InferSchemaType} from '@orion-js/schema'

export const ProductSchema = schemaWithName('ProductSchema', {
  _id: {type: typedId('product')}, // Will create IDs like 'product-abc123'
  name: {type: String},
  price: {type: Number}
})

export type ProductType = InferSchemaType<typeof ProductSchema>

// The type for _id will be `product-${string}`

​
Collection Operations

All MongoDB collection operations are typed based on your schema: 
import {createCollection} from '@orion-js/mongodb'
import {OptionalId} from '@orion-js/mongodb'

// OptionalId makes the _id field optional for insert operations
const newUser: OptionalId<UserType> = {
  name: 'John Doe',
  email: 'john@example.com',
  createdAt: new Date()
}

// Insert with type safety
const userId = await Users.insertOne(newUser)

// Update with type checking
await Users.updateOne(
  {_id: userId},
  {$set: {name: 'John Smith'}}
)

// Find with type safety
const user = await Users.findOne({_id: userId})
console.log(user.name) // TypeScript knows the structure

​
Schema Validation in MongoDB

When you create a collection with a schema, Orionjs automatically:
  1. Creates a MongoDB validation schema based on your Orionjs schema
  2. Applies validation on all insert and update operations
  3. Enforces types and constraints defined in your schema
// This will throw a validation error
try {
  await Users.insertOne({
    name: 123, // Type error: should be string
    email: 'invalid-email' // Validation error if email type is used
  })
} catch (error) {
  console.error('Validation failed:', error)
}

​
Repository Pattern

For better organization, you can create a repository class for each collection: 
import {createCollection} from '@orion-js/mongodb'
import {schemaWithName, InferSchemaType} from '@orion-js/schema'
import {Service} from '@orion-js/services'

export const UserSchema = schemaWithName('UserSchema', {
  _id: {type: typedId('user')},
  name: {type: String},
  email: {type: String},
  createdAt: {type: Date}
})

export type UserType = InferSchemaType<typeof UserSchema>

@Service()
export class UserRepository {
  private users = createCollection({
    name: 'users',
    schema: UserSchema
  })

  async findById(id: string) {
    return this.users.findOne({_id: id})
  }

  async findByEmail(email: string) {
    return this.users.findOne({email})
  }

  async create(user: Omit<UserType, '_id'>) {
    return this.users.insertOne(user)
  }

  async updateName(id: string, name: string) {
    await this.users.updateOne({_id: id}, {$set: {name}})
  }
}

​
Index Management

You can define indices for your collections: 
import {createCollection} from '@orion-js/mongodb'
import {UserSchema} from './schemas/UserSchema'

const Users = createCollection({
  name: 'users',
  schema: UserSchema,
  indexes: [
    {
      keys: {email: 1},
      options: {unique: true}
    },
    {
      keys: {name: 1, createdAt: -1}
    }
  ]
})