ZipIt ⚡️

The professional, high-performance ZIP streaming and resumable downloader for modern browsers. No server-side zipping. No RAM spikes. Zero compromise.

→ Try the Demo · API Reference · GitHub

⚡️ Why ZipIt?

Traditional file downloads in the browser are broken for large batches. ZipIt solves this by using Origin Private File System (OPFS) for staging and on-the-fly streaming compression.

| The Problem | Traditional Approach | ZipIt Solution | | :--- | :--- | :--- | | Server Load | Zipping on server blocks CPU/RAM | ✅ 0% Server CPU. Zipped in browser. | | RAM Usage | Loading all files into memory | ✅ 5MB Working Set. Streams to disk. | | Persistence | Refresh loses everything | ✅ Resumable. IDB tracks byte-level progress. | | Structure | Flat file lists | ✅ Folder Preservation. Recreates folders. | | Large Files | Crash on 1GB+ batches | ✅ Unlimited. Limited only by disk space. |

🚀 Getting Started

1. Install

pnpm add @khatiwadaprashant/zipit-core fflate

2. Basic Usage (5 Lines)

import { createZipIt } from '@khatiwadaprashant/zipit-core';

const ds = createZipIt({ concurrency: 4 });

// Add files with folder structure
ds.add('https://api.example.com/photo.jpg', { folder: 'travel/2024' });

// Stream-zip directly to user's disk
await ds.zip('my-archive.zip'); 

3. Native Folder Save (Zero ZIP needed)

// Recreates the folder structure directly on the user's local disk
await ds.start({ saveToFolder: true });

📘 Core Architecture

1. Client-Side ZIP Streaming

ZipIt uses a high-performance worker-based pipeline. As bytes are fetched from the network, they are fed into a fflate stream, which pumps compressed chunks directly to the user's disk via the File System Access API.

2. OPFS Staging Pipeline

To handle network instability, ZipIt stages files in the Origin Private File System. This allows for:

• Resumable Downloads: If the user refreshes, we know exactly how many bytes we have.
• Backpressure: We pause the network fetch if the disk write speed falls behind.

3. Byte-Level Resumability

// On app mount: restore the previous session automatically
const pending = await ds.hydrate();
if (pending.length > 0) {
  console.log(`Resuming ${pending.length} files...`);
  ds.start();
}

🛠 API Reference

createZipIt(options)

Creates a new manager instance.

• concurrency: Number of parallel download workers (Default: 3).
• dbName: IndexedDB name for state persistence (Default: 'zipit_v1').
• onProgress: Global progress callback.

Instance Methods

• ds.add(url, options): Queue a file.
• ds.start(options): Begin downloading/transferring.
• ds.pause() / ds.resume(): Control the queue.
• ds.zip(filename): Generate and download a ZIP stream.
• ds.hydrate(): Restore session from IndexedDB.

🌐 Browser Compatibility

| Feature | Chrome / Edge | Firefox | Safari | | :--- | :--- | :--- | :--- | | Streaming ZIP | ✅ 102+ | ✅ 111+ | ✅ 15.4+ | | OPFS Staging | ✅ | ✅ | ✅ 16+ | | Folder Save | ✅ (Native) | ⚠️ (ZIP Fallback) | ⚠️ (ZIP Fallback) | | Resumability | ✅ | ✅ | ✅ |

🤝 Contributing

We love contributions! Please read our Contributing Guide and follow Conventional Commits.

📝 License

MIT © Prashant Khatiwada