typescript-api-types

When to Use

Apply when defining API request/response types, DTOs, and shared types between frontend and backend.

Patterns

Pattern 1: Request/Response Types

// Source: https://www.typescriptlang.org/docs/handbook/utility-types.html
// Base entity
interface User {
  id: string;
  email: string;
  name: string;
  createdAt: Date;
  updatedAt: Date;
}

// Create DTO (omit auto-generated fields)
type CreateUserDto = Omit<User, 'id' | 'createdAt' | 'updatedAt'>;

// Update DTO (all fields optional except id)
type UpdateUserDto = Partial<Omit<User, 'id'>> & Pick<User, 'id'>;

// Response (dates as strings from JSON)
type UserResponse = Omit<User, 'createdAt' | 'updatedAt'> & {
  createdAt: string;
  updatedAt: string;
};

Pattern 2: API Response Wrapper

// Source: Best practice pattern
interface ApiResponse<T> {
  data: T;
  meta?: {
    page?: number;
    limit?: number;
    total?: number;
  };
}

interface ApiError {
  error: {
    code: string;
    message: string;
    details?: Record<string, string[]>;
  };
}

type ApiResult<T> = ApiResponse<T> | ApiError;

// Type guard
function isApiError(result: ApiResult<unknown>): result is ApiError {
  return 'error' in result;
}

Pattern 3: Zod Schema as Single Source

// Source: https://zod.dev/
import { z } from 'zod';

// Schema is source of truth
const UserSchema = z.object({
  id: z.string().uuid(),
  email: z.string().email(),
  name: z.string().min(1),
  role: z.enum(['user', 'admin']),
});

// Infer types from schema
type User = z.infer<typeof UserSchema>;

const CreateUserSchema = UserSchema.omit({ id: true });
type CreateUserDto = z.infer<typeof CreateUserSchema>;

const UpdateUserSchema = UserSchema.partial().required({ id: true });
type UpdateUserDto = z.infer<typeof UpdateUserSchema>;

Pattern 4: Shared Types Package

// packages/shared-types/src/user.ts
export interface User { /* ... */ }
export type CreateUserDto = Omit<User, 'id'>;

// Frontend: import { User } from '@myapp/shared-types';
// Backend:  import { User } from '@myapp/shared-types';

Pattern 5: API Endpoint Type Map

// Source: Best practice pattern
interface ApiEndpoints {
  'GET /users': { response: User[] };
  'GET /users/:id': { params: { id: string }; response: User };
  'POST /users': { body: CreateUserDto; response: User };
  'PUT /users/:id': { params: { id: string }; body: UpdateUserDto; response: User };
  'DELETE /users/:id': { params: { id: string }; response: void };
}

// Type-safe API client
async function api<K extends keyof ApiEndpoints>(
  endpoint: K,
  options?: Omit<ApiEndpoints[K], 'response'>
): Promise<ApiEndpoints[K]['response']> {
  // Implementation
}

Pattern 6: NoInfer for Strict Type Matching (TypeScript 5.4+)

// Source: https://www.typescriptlang.org/docs/handbook/utility-types.html
// Ensure parameter matches exact union, don't expand it
function validateStatus<S extends string>(
  validStatuses: S[],
  currentStatus: NoInfer<S>
): boolean {
  return validStatuses.includes(currentStatus);
}

// S inferred as 'pending' | 'active' | 'done'
validateStatus(['pending', 'active', 'done'], 'active');  // OK
validateStatus(['pending', 'active', 'done'], 'invalid'); // Error

Anti-Patterns

• Duplicate types - Single source of truth (Zod or interface)
• Manual JSON date parsing - Use consistent date handling
• any for API responses - Type everything
• Frontend/backend type drift - Use shared types package

Verification Checklist

• DTOs derived from base type (Omit, Pick, Partial)
• Zod schemas validate at runtime
• Types inferred from Zod (no duplication)
• API error type defined and handled
• Date serialization consistent