Fix pdf-lib "Input document to PDFDocument.load is encrypted"

The error

Error: Input document to `PDFDocument.load` is encrypted. You can use
`PDFDocument.load(..., { ignoreEncryption: true })` if you wish to load the document anyways.

pdf-lib (tested with 1.17.1) checks whether the PDF has an encryption dictionary and refuses to load it by default. This happens with both kinds of PDF protection:

• an open (user) password, where you need a password to view the file, and
• an owner (permissions) password only, where the file opens normally but printing or editing is restricted. Many bank statements and published reports are like this.

Why ignoreEncryption: true is a trap

The error message suggests ignoreEncryption: true, and many answers online stop there. It does make the error go away:

const doc = await PDFDocument.load(bytes, { ignoreEncryption: true });
doc.getPage(0).drawText('Hello');
const out = await doc.save();   // no error...

But pdf-lib still can’t decrypt anything. It copies the encrypted objects as if they were plain data. We tested both kinds of encrypted input: pdf-lib loaded and saved them without complaint, and every output file was corrupt (MuPDF: corrupt object stream, no readable pages).

Only use ignoreEncryption if you need metadata such as the page count and never save the result.

The fix: decrypt first, then use pdf-lib

With qpdf (command line)

# Owner-password-only files: no password needed
qpdf --decrypt locked.pdf unlocked.pdf

# Files with an open password
qpdf --decrypt --password=THE_PASSWORD locked.pdf unlocked.pdf

With pikepdf (Python, bundles qpdf)

import pikepdf

with pikepdf.open("locked.pdf", password="THE_PASSWORD") as pdf:  # omit password for owner-only files
    pdf.save("unlocked.pdf")

Then load it in pdf-lib

import { PDFDocument } from 'pdf-lib';
import fs from 'fs';

const doc = await PDFDocument.load(fs.readFileSync('unlocked.pdf'));
doc.getPage(0).drawText('Hello', { x: 50, y: 50, size: 20 });
fs.writeFileSync('edited.pdf', await doc.save());

In our test, both decrypted files loaded normally, and the edited output kept the original text plus the new text.

Want the result protected again? Re-encrypt the final file with qpdf --encrypt, pikepdf, or PyMuPDF. pdf-lib can’t write encrypted PDFs.

In the browser

If decryption must happen client-side, pdf-lib alone can’t do it. The options are:

• decrypt on a server or in a build step with qpdf or pikepdf, or
• use a library with encryption support, such as a WebAssembly build of qpdf or MuPDF, before handing the bytes to pdf-lib.

Only work with files you’re allowed to modify

Removing a permissions password from your own documents, or documents you’re authorised to edit, is fine. Bypassing restrictions on other people’s documents may breach copyright or terms of use. Non-developers can see How to remove a PDF password.