LibPDF

A modern PDF library for TypeScript. Parse, modify, and generate PDFs with a clean, intuitive API.

Beta Software: LibPDF is under active development. APIs may change between minor versions. Not yet recommended for production use.

Why LibPDF?

LibPDF was born from frustration. At Documenso, we found ourselves wrestling with the JavaScript PDF ecosystem:

• PDF.js is excellent for rendering, but it's read-only
• pdf-lib has a great API, but chokes on slightly malformed documents
• pdfkit only generates, no parsing at all

We kept adding workarounds. A patch here for a malformed xref table. A hack there for an encrypted document. Eventually, we decided to build what we actually needed:

• Lenient like PDFBox and PDF.js: opens documents other libraries reject
• Intuitive like pdf-lib: clean, TypeScript-first API
• Complete: encryption, digital signatures, incremental saves, form filling

Features

| Feature | Status | Notes | | ------------------ | ------ | ------------------------------------------ | | Parse any PDF | Yes | Graceful fallback for malformed documents | | Create PDFs | Yes | From scratch or modify existing | | Encryption | Yes | RC4, AES-128, AES-256 (R2-R6) | | Digital Signatures | Yes | PAdES B-B, B-T, B-LT, B-LTA | | Form Filling | Yes | Text, checkbox, radio, dropdown, signature | | Form Flattening | Yes | Bake fields into page content | | Merge & Split | Yes | Combine or extract pages | | Attachments | Yes | Embed and extract files | | Text Extraction | Yes | With position information | | Font Embedding | Yes | TTF/OpenType with subsetting | | Images | Yes | JPEG, PNG (with alpha) | | Incremental Saves | Yes | Append changes, preserve signatures |

Installation

npm install @libpdf/core
# or
bun add @libpdf/core

Quick Start

Parse an existing PDF

import { PDF } from "@libpdf/core";

const pdf = await PDF.load(bytes);
const pages = await pdf.getPages();

console.log(`${pages.length} pages`);

Open an encrypted PDF

const pdf = await PDF.load(bytes, { credentials: "password" });

Fill a form

const pdf = await PDF.load(bytes);
const form = await pdf.getForm();

form.fill({
  name: "Jane Doe",
  email: "[email protected]",
  agreed: true,
});

const filled = await pdf.save();

Sign a document

import { PDF, P12Signer } from "@libpdf/core";

const pdf = await PDF.load(bytes);
const signer = await P12Signer.create(p12Bytes, "password");

const signed = await pdf.sign({
  signer,
  reason: "I approve this document",
});

Merge PDFs

const merged = await PDF.merge([pdf1Bytes, pdf2Bytes, pdf3Bytes]);

Draw on a page

import { PDF, rgb } from "@libpdf/core";

const pdf = PDF.create();
const page = pdf.addPage({ size: "letter" });

page.drawText("Hello, World!", {
  x: 50,
  y: 700,
  fontSize: 24,
  color: rgb(0, 0, 0),
});

page.drawRectangle({
  x: 50,
  y: 600,
  width: 200,
  height: 100,
  color: rgb(0.9, 0.9, 0.9),
  borderColor: rgb(0, 0, 0),
  borderWidth: 1,
});

const output = await pdf.save();

Runtime Support

LibPDF runs everywhere:

• Node.js 20+
• Bun
• Browsers (modern, with Web Crypto)

Known Limitations

Some features are not yet implemented:

| Feature | Status | Notes | | --------------------------- | ---------------- | -------------------------------------- | | Signature verification | Not implemented | Signing works; verification is planned | | TrueType Collections (.ttc) | Not supported | Extract individual fonts first | | JBIG2 image decoding | Passthrough only | Images preserved but not decoded | | JPEG2000 (JPX) decoding | Passthrough only | Images preserved but not decoded | | Certificate encryption | Not supported | Password encryption works | | JavaScript actions | Ignored | Form calculations not executed |

These limitations are documented to set expectations. Most don't affect typical use cases like form filling, signing, or document manipulation.

Philosophy

Be lenient

Real-world PDFs are messy. Export a document through three different tools and you'll get three slightly different interpretations of the spec. LibPDF prioritizes opening your document over strict compliance. When standard parsing fails, we fall back to brute-force recovery, scanning the entire file to rebuild the structure.

Two API layers

• High-level: PDF, PDFPage, PDFForm for common tasks
• Low-level: PdfDict, PdfArray, PdfStream for full control

Documentation

Full documentation at libpdf.dev

Sponsors

LibPDF is developed by Documenso, the open-source DocuSign alternative.

Contributing

We welcome contributions! See our contributing guide for details.

# Clone the repo
git clone https://github.com/libpdf/core.git
cd libpdf

# Install dependencies
bun install

# Run tests
bun run test

# Type check
bun run typecheck

License

MIT

The src/fontbox/ directory is licensed under Apache-2.0 as it is derived from Apache PDFBox.