ID Generator

A service that generates unique, random numeric IDs for multiple consuming applications with pre-generated pools, Verhoeff checksum, and multi-ID-type support.

Overview

The ID Generator service produces unique, random numeric IDs for applications such as social registries, farmer registries, health systems, and more. Each consuming application operates within its own ID type, with an independent pool of pre-generated IDs.

The service is built with FastAPI, uses PostgreSQL for storage, and is designed for horizontal scaling on Kubernetes.

Key capabilities

• Multi-ID-type support — Each application (farmer ID, household ID, national ID, etc.) gets an independent ID pool with configurable ID length.

• Pre-generated pool — IDs are generated in advance and stored in PostgreSQL, ensuring fast issuance with no generation delay at request time.

• 10 filter rules — Comprehensive ID quality filters (inspired by MOSIP UIN generation filters) plus Verhoeff checksum ensure ID quality.

• Kubernetes-native — Horizontally scalable with zero-contention ID issuance and coordinated pool replenishment across pods.

• Structural validation — Downstream systems can validate any ID without a database lookup.

How it works

1. On startup → create tables, fill ID pools to minimum threshold
2. On request → atomically issue one pre-generated ID (AVAILABLE → TAKEN)
3. In background → monitor pool levels, replenish when low

ID lifecycle

IDs follow a simple two-state model:

AVAILABLE  →  TAKEN

• AVAILABLE — In pool, ready to be issued.

• TAKEN — Issued to a caller; permanently consumed. Never re-issued.

Detailed documentation

Page	Description
Functional Specifications	ID types, generation rules, 10 filter rules, pool management, exhaustion handling, configuration
API Reference	REST API endpoints, request/response formats, error codes, OpenAPI spec
Technical Architecture	Async stack, database design, concurrency patterns, Kubernetes deployment
Deployment	Helm chart installation, Docker setup, local development, environment variables
Testing	Test categories, execution commands, HTML reporting

Source code

• GitHub: https://github.com/OpenG2P/id-generator

Technology stack

Component	Choice	License
Language	Python 3.11+	PSF License (permissive)
Web Framework	FastAPI	MIT
ASGI Server	Uvicorn	BSD-3-Clause
DB Driver	asyncpg	Apache 2.0
ORM	SQLAlchemy 2.x (async)	MIT
Config	Pydantic Settings	MIT
Database	PostgreSQL	PostgreSQL License (permissive)
Deployment	Kubernetes + Helm	Apache 2.0

All components use permissive open-source licenses. No copyleft (GPL) dependencies.