Browser Usage

The package works with browser-friendly binary values. You can generate a PDF, wrap it in a Blob, and preview or download it.

import { createDocument } from "@criston/zeropdf";

const doc = createDocument();
const page = doc.addPage();

page.text("Preview me", {
  x: 56,
  y: 760,
  fontSize: 16
});

const blob = doc.toBlob();
const url = URL.createObjectURL(blob);

iframe.src = url;

Streaming to a Response

Use WebStreamSink with a TransformStream to stream a PDF straight to the browser without buffering the whole document:

import { WebStreamSink } from "@criston/zeropdf";

const { readable, writable } = new TransformStream<Uint8Array>();
const sink = new WebStreamSink(writable);
void doc.writeTo(sink).then(() => sink.close());

const response = new Response(readable, {
  headers: { "Content-Type": "application/pdf" }
});

BufferSink also works in the browser when you just need the bytes.

Browser limitations

The core engine — generation, parsing, editing, fonts, images, tables, templates, annotations, and tagged PDF — runs in the browser with no Node dependencies. Two areas are Node-only:

• Disk-backed output. TempFileSink, NodeStreamSink, and writeToFile rely on node:fs and are not available in the browser. Use toBlob(), toArrayBuffer(), toUint8Array(), BufferSink, or WebStreamSink instead.
• AES encryption. AES-encrypting a document requires Node's node:crypto; in the browser it throws "AES PDF encryption requires Node.js crypto support in this build." Generate unencrypted PDFs client-side and encrypt server-side if you need protection. (Secure random bytes fall back to the Web Crypto API where available.)

The Node-only sinks load node:fs through a lazy dynamic import(), so bundlers building for the browser will not pull them into the core path.

Playground

The repo includes a browser playground in examples/website: write template code on the left and see the rendered PDF on the right, all in-browser.

npm run example:web

Open http://localhost:4173/, edit the code, and press Run (or ⌘/Ctrl+Enter). Every zeropdf export is in scope; your code just needs to build a document and return it.