LCBP3-DMS App Summary

What It Is

LCBP3-DMS is a document management system for the Laem Chabang Port Phase 3 construction project. Repo docs describe it as a system for managing construction documents, approvals, workflows, and cross-organization communication. The codebase is split into a Next.js frontend and a NestJS backend, with MariaDB, Redis, and Elasticsearch in the application stack.

Who It's For

Primary users are document-heavy construction project teams working across multiple organizations. Repo evidence points especially to Document Control, Org Admin, Engineer/Reviewer, Superadmin, consultants, supervisors, and contractors.

What It Does

Manages correspondence records between organizations.
Supports RFA workflows for technical approval requests.
Tracks contract and shop drawings.
Handles transmittals and circulation sheets.
Applies 4-level RBAC / CASL-based access control.
Generates document numbers with Redis-backed locking rules.
Supports file upload and attachment handling with cleanup jobs.
Exposes admin screens for users, orgs, projects, workflows, numbering, and audit logs.
Provides search through an Elasticsearch-backed search module.
Includes notification processing and a migration area for legacy document import.

\newpage

How It Works

Frontend

frontend/ is a Next.js 16 App Router app.
Route groups in frontend/app/ show three main surfaces: (auth), (dashboard), and (admin).
Installed frontend libraries indicate form validation and data fetching with React Hook Form, Zod, TanStack Query, and Zustand.
frontend/app/api/auth/[...nextauth]/route.ts: Not found in repo via direct read during this task, but the route file exists in the app tree.

Backend

backend/src/main.ts boots a NestJS 11 app with Helmet, CORS, 50 MB body limits, global validation, response transformation, exception filtering, and Swagger.
backend/src/app.module.ts wires Config, TypeORM for MariaDB, BullMQ, Redis, scheduling, throttling, logging, monitoring, resilience, and feature modules.
Feature modules present in code include auth, user, project, organization, contract, correspondence, RFA, drawing, transmittal, circulation, workflow-engine, document-numbering, search, notification, audit-log, dashboard, master, and migration.

Data Flow

Browser -> Next.js UI -> backend API under /api.
Backend validates requests, applies guards/interceptors, and persists entities to MariaDB through TypeORM.
Redis is used in the running architecture for BullMQ queues and Redis module integration; repo specs also tie Redis to locking, caching, and idempotency patterns.
Search requests flow through the NestJS search module into Elasticsearch.
Notification work is queued through BullMQ and also exposes a WebSocket gateway.
Migration services write through backend modules and the shared file-storage module rather than direct client-side data access.

How To Run

Install workspace dependencies with pnpm install at repo root.
Start local infra from backend/docker-compose.yml to bring up MariaDB, Redis, and Elasticsearch.
Prepare env files: frontend/.env.example -> frontend/.env.local; backend example env file: Not found in repo.
Load the SQL schema and seed data using the commands documented in README.md.
Run backend with pnpm run start:dev in backend/.
Run frontend with pnpm run dev in frontend/.

Notes

Minimal dev URLs from repo docs: frontend http://localhost:3000, backend http://localhost:3001, Swagger /docs.
Exact production deployment topology is documented, but this summary keeps only the minimum local-start path.

3.7 KiB Raw Blame History