advanced-rbac

Purpose

To implement a scalable permission system where users have roles, and roles have granular permissions.

When to Use

• B2B SaaS applications (Admin, Editor, Viewer).
• Systems with complex access requirements.

Procedure

1. Data Model

• Users: id, name, role_id (or m:n user_roles).
• Roles: id, name (e.g., 'admin', 'editor').
• Permissions: id, action, resource (e.g., create:post, read:user).
• RolePermissions: Join table.

2. Middleware (Enforcement)

Don't check roles (if user.role === 'admin'). Check permissions.

// middleware/authorize.js
const authorize = (requiredPermission) => {
  return async (req, res, next) => {
    const userPermissions = await getUserPermissions(req.user.id); // Cached in JWT or Redis
    if (!userPermissions.includes(requiredPermission)) {
      return res.status(403).json({ error: 'Forbidden' });
    }
    next();
  };
};

// usage
app.post('/posts', authorize('create:post'), createPost);

3. Hierarchy (Optional)

Roles can inherit from others.

• Admin has all permissions of Editor + delete:user.
• This logic is handled in the permission resolution function.

4. Frontend Integration

Send permissions list to frontend on login.

{
  token: "...",
  user: {
    id: 1,
    permissions: ["read:posts", "create:posts"]
  }
}

Use a <Can permission="create:posts"> component to hide/show UI elements.

Constraints

• Granularity: Balance between too broad (manage:all) and too specific (update:post:title).
• Caching: Fetching permissions on every request is slow. Cache them in the JWT or Redis (invalidate on role change).

Expected Output

A flexible security model allowing admins to configure what users can do without changing code.