@schorts/firestore-dao

🔥 Firestore DAO

This module provides a type-safe, domain-driven abstraction over Firestore persistence. It integrates tightly with the Model, Entity, Criteria and UnitOfWork constructs from the Shared Kernel, enabling expressive, consistent, and testable data access across bounded contexts.

🚀 Installation

npm install --save @schorts/firestore-dao

🧩 Purpose

Firestore DAOs encapsulate all persistence logic for domain entities, including:

• 🔍 Querying by ID, criteria, or full collection.
• 🧠 Translating domain filters into Firestore constraints.
• 🧼 Formatting primitives for Firestore compatibility.
• 🔁 Coordinating writes via FirestoreUnitOfWork for atomicity.

They are not responsible for business logic, validation, or orchestration — only persistence.

🏗️ Architecture

Each DAO implements the following shared kernel interfaces:

• Model<Entity> - defines the contract for persistence operations.
• Entity - domain object with identity and behavior.
• UnitOfWork - optional batching mechanism for transactional consistency.

🧪 Example Usage

import { initializeApp, getFirestore, EntityRegistry } from "@schorts/firestore-dao";

// You need to use the internal firebase/firestore packages and register you entity via EntityRegistry

EntityRegistry.register("users", User);

const user = new User({ id: "abc123", name: "Alice" });

await userDAO.create(user); // direct write

const uow = new FirestoreUnitOfWork(firestore);
await userDAO.create(user, uow); // batched write
await uow.commit();

const found = await userDAO.findByID("abc123");
const results = await userDAO.search(Criteria.where("status", "EQUAL", "active"));

Supports:

• Standard filters (EQUAL, IN, GREATER_THAN, etc.)
• Geo-radius queries with post-filtering via distanceBetween.
• Ordering, limits, and pagination (startAfter).

🚧 Future Extensions

• Soft deletes via status: "deleted".

🧠 Philosophy

This layer reflects a commitment to:

• Domain-driven design.
• Type safety and semantic clarity.
• Separation of concerns.
• Scalable, testable architecture.