Stacks Codebase Design

Codebase design

Design deep modules: a lot of behaviour behind a small interface, placed at a clean seam, testable through that interface. Use this language and these principles wherever Stacks code is being designed or restructured. The aim is leverage for callers, locality for maintainers, and testability for everyone.

Stacks is already built this way, which is why the vocabulary is worth sharing: defineModel() puts a schema, validation, factories, relationships and a REST surface behind one call. The driver packages (cache, queue, storage, email, ai, chat, search-engine, dns) are ports with two or more adapters each. Use the words below rather than inventing new ones per package.

Credit: the deep-module model here is adapted from Matt Pocock's codebase-design skill (MIT), https://github.com/mattpocock/skills, itself building on Ousterhout and Feathers.

Glossary

Use these terms exactly. Do not substitute "component", "service", "API" or "boundary". Consistent language is the whole point.

Module: anything with an interface and an implementation. Deliberately scale-agnostic: a function, an action, a model, a @stacksjs/* package, a tier-spanning slice. Avoid: unit, component, service.

Interface: everything a caller must know to use the module correctly. The type signature, but also invariants, ordering constraints, error modes, required config, and performance characteristics. For a Stacks action the interface includes its route, its middleware and its request shape, not just its handle signature. Avoid: API, signature (too narrow, they name only the type surface).

Implementation: what is inside a module. Distinct from adapter: a thing can be a small adapter over a large implementation (the S3 storage driver) or a large adapter over a small one (an in-memory fake). Reach for "adapter" when the seam is the topic, "implementation" otherwise.

Depth: leverage at the interface. How much behaviour a caller or a test can exercise per unit of interface they have to learn. A module is deep when a lot of behaviour sits behind a small interface, shallow when the interface is nearly as complex as the implementation.

Seam (Michael Feathers): a place where you can alter behaviour without editing in that place. The location at which a module's interface lives. Where to put the seam is its own design decision, distinct from what goes behind it. In Stacks the app/ override model is a seam the framework hands you for free: app/Actions/Cms/PostIndexAction.ts replaces the default without editing it. Avoid: boundary (overloaded with DDD's bounded context).

Adapter: a concrete thing that satisfies an interface at a seam. Describes role (which slot it fills), not substance (what is inside). config/cache.ts picking memory or redis is adapter selection.

Leverage: what callers get from depth. More capability per unit of interface learned. One implementation pays back across N call sites and M tests.

Locality: what maintainers get from depth. Change, bugs, knowledge and verification concentrate in one place rather than spreading across callers. Fix once, fixed everywhere.

Deep versus shallow

Deep = small interface, lots of implementation:

┌─────────────────────┐
│   Small interface   │  ← few entry points, simple params
├─────────────────────┤
│                     │
│  Deep implementation│  ← complexity hidden
│                     │
└─────────────────────┘

Shallow = large interface, little implementation. Avoid:

┌─────────────────────────────────┐
│       Large interface           │  ← many methods, complex params
├─────────────────────────────────┤
│  Thin implementation            │  ← mostly passes through
└─────────────────────────────────┘

When designing an interface, ask: can I reduce the number of entry points? Can I simplify the params? Can I hide more complexity inside?

The useApi trait is the canonical deep interface in this framework: one config object, and the model gains five actions, five routes, an OpenAPI entry and a dashboard view. useSearch, useAuth and useSoftDeletes are the same trade. When you find yourself writing the fifth near-identical action, the question is whether a trait wants to be born.

Principles

• Depth is a property of the interface, not the implementation. A deep module can be internally composed of small, swappable parts. They just are not part of the interface. A module can have internal seams (private, used by its own tests) as well as the external seam at its interface.
• The deletion test. Imagine deleting the module. If complexity vanishes, it was a pass-through. If complexity reappears across N callers, it was earning its keep.
• The interface is the test surface. Callers and tests cross the same seam. If you want to test past the interface, the module is probably the wrong shape.
• One adapter means a hypothetical seam. Two adapters means a real one. Do not introduce a seam unless something actually varies across it. Production plus test counts as two.

Designing for testability

1. Accept dependencies, do not create them.

   // testable
   export async function processOrder(order: Order, gateway: PaymentGateway) {}
   
   // hard to test
   export async function processOrder(order: Order) {
     const gateway = new StripeGateway()
   }
   

2. Return results, do not produce side effects.

   // testable
   function calculateDiscount(cart: Cart): Discount {}
   
   // hard to test
   function applyDiscount(cart: Cart): void { cart.total -= discount }
   

3. Small surface area. Fewer entry points means fewer tests. Fewer params means simpler setup.

In a Stacks app the third dependency is almost always the database, and it does not need injecting: refreshDatabase() plus model factories give you a real one per suite. See stacks-tdd for where that line sits.

Relationships

• A module has exactly one interface, the surface it presents to callers and tests.
• Depth is a property of a module, measured against its interface.
• A seam is where a module's interface lives.
• An adapter sits at a seam and satisfies the interface.
• Depth produces leverage for callers and locality for maintainers.

Rejected framings

• Depth as a ratio of implementation lines to interface lines (Ousterhout): rewards padding the implementation. Use depth-as-leverage instead.
• "Interface" as the TypeScript interface keyword or a class's public methods: too narrow. Interface here includes every fact a caller must know.
• "Boundary": overloaded with DDD's bounded context. Say seam or interface.

Going deeper

• Deepening a cluster given its dependencies: DEEPENING.md covers the dependency categories, seam discipline, and replace-do-not-layer testing.
• Exploring alternative interfaces: DESIGN-IT-TWICE.md runs parallel sub-agents to design one interface several radically different ways, then compares on depth, locality and seam placement.

Downstream

Reach for stacks-tdd to write the tests at the seam you chose, and stacks-domain-modeling when the module needs a name the project does not have yet.