JSZip — Create and Read ZIP Files in JavaScript

Introduction

JSZip lets developers create, read, and modify ZIP archives entirely in JavaScript. It runs in the browser without server-side processing, enabling features like client-side file bundling, in-memory archive inspection, and offline export of multi-file downloads—all with a clean promise-based API.

What JSZip Does

• Creates ZIP archives from strings, blobs, array buffers, and typed arrays
• Reads and extracts contents from existing ZIP files
• Supports nested folder structures within archives
• Generates output as Blob, ArrayBuffer, Uint8Array, base64, or Node.js Buffer
• Handles DEFLATE compression with configurable compression level

Architecture Overview

JSZip maintains an in-memory tree of entries representing files and folders. Each entry stores its content and metadata (name, date, compression method). When reading a ZIP, the library parses the central directory and local file headers per the PKZIP specification, inflating compressed entries on demand. When generating, it serializes the entry tree into a valid ZIP byte stream, compressing entries with the pako DEFLATE library. All operations return promises, allowing non-blocking processing of large archives.

Setup & Configuration

• Install via npm or load from a CDN for browser use
• Create a new archive with new JSZip() and add files with .file(name, content)
• Read an existing ZIP with JSZip.loadAsync(data) where data is a Blob, ArrayBuffer, or base64 string
• Set compression options in generateAsync(): { compression: 'DEFLATE', compressionOptions: { level: 6 } }
• Access individual files with zip.file('path/to/file').async('string') for lazy extraction

Key Features

• Pure JavaScript with no native dependencies or WebAssembly
• Reads and writes ZIP64 archives for files larger than 4 GB
• Streaming generation via generateInternalStream() for memory-efficient output
• Compatible with password-protected ZIP reading via community extensions
• Works identically in browser, Node.js, and Deno environments

Comparison with Similar Tools

• Archiver (Node.js) — streaming ZIP creation for servers; JSZip also works in the browser
• fflate — faster compression via WebAssembly; JSZip has a simpler API and wider adoption
• pako — raw DEFLATE compression only; JSZip provides the full ZIP container format on top
• StreamSaver.js — handles downloading large files; pair with JSZip to stream ZIP downloads in the browser

FAQ

Q: Can JSZip handle large archives in the browser? A: Yes. Use generateInternalStream() for streaming output and loadAsync() for lazy parsing. Very large archives may still be constrained by browser memory limits.

Q: Does JSZip support password-protected ZIP files? A: The core library does not encrypt or decrypt. Community forks and plugins add password support using the traditional ZIP encryption scheme.

Q: Can I add binary files like images to a ZIP? A: Yes. Pass the content as a Blob, ArrayBuffer, or Uint8Array to zip.file() and set { binary: true } if needed.

Q: How does JSZip compare to the native Compression Streams API? A: The Compression Streams API provides raw DEFLATE/gzip compression. JSZip adds the ZIP container format, file metadata, and folder structure on top of compression.

Sources

• https://github.com/Stuk/jszip
• https://stuk.github.io/jszip/