mirror of
https://github.com/Floriansylvain/GoAntiSqlPOC.git
synced 2026-08-19 11:43:24 +02:00
86 lines
2.8 KiB
Markdown
86 lines
2.8 KiB
Markdown
# GoAntiSqlPOC - SQL-Free Database Management in Go
|
|
|
|
This project demonstrates a proof of concept for a database solution in Go that uses zero SQL or text-based query languages. Instead, it leverages [BadgerDB](https://github.com/dgraph-io/badger), a fast key-value store written in Go, to provide structured data storage with API-style access patterns.
|
|
|
|
## Disclaimer ⚠️
|
|
|
|
This project is not clean architecture friendly, yet.
|
|
|
|
## Core Features
|
|
|
|
- **No SQL or Query Languages** - Pure programmatic data access through Go functions
|
|
- **Strong Schema Enforcement** - Type safety via Go structs
|
|
- **Relational-Like Data Modeling** - Links between entities using IDs and indexes
|
|
- **Clean API-Style Access** - Simple function calls instead of writing queries
|
|
|
|
## Project Structure
|
|
|
|
- **main.go**: Example usage demonstrating CRUD operations and relationships
|
|
- **store/db.go**: Database connection management with in-memory storage
|
|
- **store/user.go**: User entity operations (create, find, update, list)
|
|
- **store/address.go**: Address entity operations with user relationships
|
|
|
|
## How It Works
|
|
|
|
### Database Management
|
|
|
|
The project uses BadgerDB as the underlying storage engine. The `store.InitDB()` function sets up an in-memory database connection with a singleton pattern using `sync.Once` to ensure it's only initialized once.
|
|
|
|
### Data Models
|
|
|
|
The project defines strongly typed Go structs to enforce schema:
|
|
|
|
- `User`: Stores user information with ID, name, email, and password
|
|
- `Address`: Stores address information with reference to a user via UserID
|
|
|
|
### Data Access Pattern
|
|
|
|
Instead of SQL queries, all data access is done through simple function calls:
|
|
|
|
```go
|
|
// Create a user
|
|
user, _ := store.CreateUser("Bob", "bob@example.com", "hunter2")
|
|
|
|
// Add an address
|
|
addr, _ := store.AddAddressToUser(user.ID, "123 Main St", "Paris", "75001")
|
|
|
|
// List addresses for a user
|
|
addresses, _ := store.ListAddressesByUserID(user.ID)
|
|
|
|
// Update user data
|
|
user.Name = "Bobby"
|
|
store.UpdateUser(user)
|
|
```
|
|
|
|
### "Relational" Modeling
|
|
|
|
The project demonstrates relationship modeling between users and addresses:
|
|
|
|
1. Each address has a `UserID` field linking it to its owner
|
|
2. The `ListAddressesByUserID` function retrieves all addresses for a specific user
|
|
3. Updates to either entity maintain relationship integrity
|
|
|
|
### Data Consistency
|
|
|
|
The project implements data consistency features:
|
|
|
|
- Email uniqueness enforcement using an email-to-ID index
|
|
- Transaction support for atomic operations
|
|
- Error handling for constraint violations
|
|
|
|
### Terminal Demo
|
|
|
|
The main.go file provides a demonstration that showcases:
|
|
|
|
1. User creation with validation
|
|
2. Address management (create, read, update)
|
|
3. Relationship handling (users have multiple addresses)
|
|
4. Constraint enforcement (unique emails)
|
|
5. Entity retrieval by ID and other fields
|
|
|
|
## Running the Demo
|
|
|
|
```bash
|
|
go run main.go
|
|
```
|