⚡️ session-cache-utils

Super lightweight, zero-dependency utility to cache data in sessionStorage — with automatic TTL expiration, type safety, and full control.

🚀 Why use this?

• ✅ Simple: One function to read, one to write, one to invalidate. That’s it.
• 🧠 Smart: Auto-expires stale data based on TTL (in minutes).
• 🧱 Type-safe: Written in TypeScript, strongly typed all the way.
• 🌐 Browser-native: Uses sessionStorage, nothing extra.
• 📦 Framework-agnostic: Works in React, Next.js, Svelte, or plain JavaScript apps.

📦 Installation

npm install session-cache-utils

Usage/Examples

import { writeCache, readCache, invalidateCache } from "session-cache-utils";

//📝 Write to cache
writeCache("user-profile", { id: 42, name: "Akshit" });

//🔍 Read from cache (if not older than 30 minutes)
const result = readCache<{ id: number; name: string }>("user-profile", 30);
console.log(result?.data); // { id: 42, name: "Akshit" }

//❌ Invalidate cache
invalidateCache("user-profile");

📖 API Reference

| Method | Description | Example | | ---------------------- | ----------------------------- | ---------------------------- | | readCache<T>(key, ttlMinutes) | Reads cached data and returns it only if it’s still fresh (not older than ttlMinutes) | readCache<User>("user-data", 30) → { data, created_at, updated_at } or null | | writeCache<T>(key, value) | Writes or updates the cache entry with current timestamps. | writeCache("user-data", { id: 1, name: "Akshit" }) → true | | invalidateCache(key) | Removes the specified cache entry from sessionStorage | invalidateCache("user-data") → true |

📦 CachedDataShape

Internal structure used to store cached entries:

readCache api will always return this type for cached entry else null

type CachedDataShape<T> = {
  data: T;
  created_at: number;
  updated_at: number;
};

🧠 Use Case Examples

⚙️ Store UI filters temporarily

📄 Cache API response for current tab

🧭 Save navigation state across routes

💾 Cache form data between page reloads

⛔ Things to know

Only works client-side (window required).

TTL is capped at 24 hours internally for sanity.

Falls back to null and auto-cleans expired data.

Authors

✨ Made with focus by Akshit

• @akshitworkspace

License

MIT – free for personal and commercial use.