purepdf

Zero-dependency PDF library for Node.js and browsers.

Generate PDF documents from scratch, merge, split, extract text, or convert HTML to PDF — all without any external runtime dependencies.

Features

• Generate PDFs — text, shapes, JPEG images, tables
• 14 built-in fonts — all standard PDF Type1 fonts (Helvetica, Times, Courier, …)
• Merge PDFs — combine any number of PDF files into one
• Split PDFs — extract individual pages or specific page ranges
• Extract text — get the plain text content from any simple PDF
• HTML → PDF — convert basic HTML (headings, paragraphs, lists, bold/italic, tables) to PDF
• ESM + CJS — works in Node.js, Bun, Deno, and modern browsers
• Zero dependencies — no external packages, no native bindings
• TypeScript — full type definitions included

Install

npm install purepdf

Quick start

Generate a PDF

import { PDFDocument } from 'purepdf'
import { writeFileSync } from 'fs'

const doc = new PDFDocument({ size: 'A4' })

// Title
doc.text('Annual Report 2026', 50, 780, {
  font: 'Helvetica-Bold',
  size: 22,
  color: '#1a1a1a',
})

// Paragraph with auto word-wrap
doc.text(
  'This document was generated entirely without external dependencies using purepdf.',
  50, 740,
  { font: 'Helvetica', size: 11, maxWidth: 495 },
)

// Divider line
doc.line(50, 720, 545, 720, { stroke: '#cccccc', lineWidth: 0.5 })

// Coloured rectangle
doc.rect(50, 650, 200, 60, { fill: '#e8f4fd', stroke: '#4a90d9', lineWidth: 1 })

// Table
doc.table(
  {
    headers: ['Product',   'Qty', 'Price'],
    rows:    [
      ['Widget A',  '10', '$5.00'],
      ['Widget B',  '3',  '$12.50'],
      ['Widget C',  '25', '$2.00'],
    ],
  },
  50, 630,
  {
    colWidths:         [250, 80, 80],
    headerBackground:  '#1a73e8',
    headerColor:       '#ffffff',
    alternateRowColor: '#f8f8f8',
    borderColor:       '#dddddd',
  },
)

// Second page
doc.addPage()
doc.text('Page 2', 50, 780, { font: 'Helvetica', size: 14 })

writeFileSync('report.pdf', doc.save())

Merge PDFs

import { mergePDFs } from 'purepdf'
import { readFileSync, writeFileSync } from 'fs'

const a = readFileSync('invoice.pdf')
const b = readFileSync('appendix.pdf')
const merged = mergePDFs([a, b])
writeFileSync('complete.pdf', merged)

Split a PDF

import { splitPDF, extractPages } from 'purepdf'
import { readFileSync, writeFileSync } from 'fs'

const pdf = readFileSync('book.pdf')

// One file per page
const pages = splitPDF(pdf)
pages.forEach((page, i) => writeFileSync(`page-${i + 1}.pdf`, page))

// Extract specific pages (0-based indices)
const firstAndLast = extractPages(pdf, [0, pages.length - 1])
writeFileSync('excerpt.pdf', firstAndLast)

Extract text

import { extractText } from 'purepdf'
import { readFileSync } from 'fs'

const pdf = readFileSync('document.pdf')
const text = extractText(pdf)
console.log(text)

Markdown → PDF

import { markdownToPDF } from 'purepdf'
import { writeFileSync } from 'fs'

const md = `
# Annual Report

This document was generated from **Markdown** using *purepdf*.

## Highlights

- Zero dependencies
- Works in Node.js and browsers
- Supports tables, blockquotes, code blocks

| Quarter | Revenue |
|---------|---------|
| Q1      | $1.2M   |
| Q2      | $1.8M   |

> Revenue grew **50%** quarter over quarter.

\`\`\`ts
import { markdownToPDF } from 'purepdf'
const pdf = markdownToPDF('# Hello')
\`\`\`
`

writeFileSync('report.pdf', markdownToPDF(md))

HTML → PDF

import { htmlToPDF } from 'purepdf'
import { writeFileSync } from 'fs'

const html = `
  <h1>My Document</h1>
  <p>Hello <b>World</b>! This is <i>purepdf</i>.</p>
  <hr/>
  <ul>
    <li>Zero dependencies</li>
    <li>Works in Node.js and browsers</li>
    <li>Full TypeScript support</li>
  </ul>
`

writeFileSync('output.pdf', htmlToPDF(html, { size: 'A4', margin: 60 }))

API

new PDFDocument(options?)

| Option | Type | Default | Description | |--------|------|---------|-------------| | size | PageSize \| [w, h] | 'A4' | Page size |

Methods

| Method | Description | |--------|-------------| | addPage(options?) | Add a new page and make it current. Chainable. | | getPageSize() | Returns { width, height } in points. | | text(str, x, y, options?) | Draw text. Returns the y below the last line. | | rect(x, y, w, h, options?) | Draw a rectangle. | | line(x1, y1, x2, y2, options?) | Draw a line. | | circle(cx, cy, r, options?) | Draw a circle. | | image(jpegData, x, y, options?) | Draw a JPEG image. | | table(data, x, y, options?) | Draw a table. Returns the y below the last row. | | save() | Serialise and return Uint8Array. |

TextOptions

| Option | Type | Default | |--------|------|---------| | font | FontName | 'Helvetica' | | size | number | 12 | | color | string ('#RRGGBB') | '#000000' | | align | 'left' \| 'center' \| 'right' | 'left' | | maxWidth | number | — | | lineHeight | number | 1.2 |

GraphicsOptions

| Option | Type | Default | |--------|------|---------| | stroke | string \| false | '#000000' | | fill | string \| false | false | | lineWidth | number | 1 |

ImageOptions

| Option | Type | Description | |--------|------|-------------| | width | number | Render width in points. Height is auto-calculated. | | height | number | Render height in points. |

TableOptions

| Option | Type | Default | |--------|------|---------| | colWidths | number[] | equal split across 400pt | | rowHeight | number | fontSize + padding*2 | | headerBackground | string | '#1a73e8' | | headerColor | string | '#ffffff' | | alternateRowColor | string | — | | borderColor | string | '#cccccc' | | cellPadding | number | 6 | | font | FontName | 'Helvetica' | | fontSize | number | 10 |

mergePDFs(pdfs: Uint8Array[]): Uint8Array

Merge two or more PDFs. Pages from each input appear in order.

splitPDF(pdf: Uint8Array): Uint8Array[]

Split into individual single-page PDFs.

extractPages(pdf: Uint8Array, indices: number[]): Uint8Array

Extract specific pages (0-based indices) and return as one PDF.

extractText(pdf: Uint8Array): string

Extract plain text from a PDF. Works on simple, unencrypted PDFs.

markdownToPDF(markdown: string, options?): Uint8Array

Convert a Markdown string to PDF.

import { markdownToPDF } from 'purepdf'

const md = `
# My Report

This paragraph has **bold**, *italic*, ~~strikethrough~~, and \`inline code\`.

## Features

- Zero dependencies
- Works in Node.js and browsers

> Important: always validate your input.

| Name  | Score |
|-------|-------|
| Alice | 95    |
| Bob   | 82    |

\`\`\`ts
const pdf = markdownToPDF('# Hello')
\`\`\`
`

const pdf = markdownToPDF(md, { size: 'A4', margin: 60 })

Supported Markdown syntax:

| Feature | Syntax | |---------|--------| | Headings | # H1 … ###### H6, Setext (=== / ---) | | Bold | **text** or __text__ | | Italic | *text* or _text_ | | Bold+italic | ***text*** | | Strikethrough | ~~text~~ | | Inline code | `code` | | Code blocks | ```lang ``` | | Blockquote | > text | | Unordered list | - / * / + | | Ordered list | 1. | | Table (GFM) | \| col \| col \| | | Horizontal rule | --- / *** / ___ | | Link | [text](url) → shows text only | | Image | ![alt](url) → shows [alt] | | Hard line break | Line ending with two spaces |

markdownToHTML(markdown: string): string

Convert Markdown to an HTML string (useful for inspection or pre-processing).

htmlToPDF(html: string, options?): Uint8Array

Convert HTML to PDF.

HtmlToPdfOptions

| Option | Type | Default | |--------|------|---------| | size | PageSize \| [w, h] | 'A4' | | margin | number | 50 | | baseFontSize | number | 11 | | baseFont | FontName | 'Times-Roman' |

Supported HTML tags: h1–h6, p, br, b/strong, i/em, u, ul, ol, li, table, tr, th, td, hr, pre, code, div, span, section, article.

measureText(text, font, size): number

Returns the rendered width of a string in points.

wrapText(text, font, size, maxWidth): string[]

Splits text into lines that fit within maxWidth. Respects \n.

PAGE_SIZES

PAGE_SIZES.A4     // [595.28, 841.89]
PAGE_SIZES.A3     // [841.89, 1190.55]
PAGE_SIZES.A5     // [419.53, 595.28]
PAGE_SIZES.Letter // [612, 792]
PAGE_SIZES.Legal  // [612, 1008]

Built-in fonts (FontName)

| Serif | Sans-serif | Monospace | |-------|-----------|-----------| | Times-Roman | Helvetica | Courier | | Times-Bold | Helvetica-Bold | Courier-Bold | | Times-Italic | Helvetica-Oblique | Courier-Oblique | | Times-BoldItalic | Helvetica-BoldOblique | Courier-BoldOblique |

Also: Symbol, ZapfDingbats.

Limitations

• Fonts: only the 14 standard PDF fonts (no custom/embedded fonts in v1)
• Images: JPEG only (PNG/GIF are not supported in v1)
• HTML: basic structural tags only; complex CSS, <style>, and <script> are ignored
• PDF parsing: traditional XRef tables only (PDF ≤ 1.4 style); PDFs with compressed XRef streams (PDF 1.5+) are not supported
• Encryption: encrypted PDFs are not supported
• Text encoding: Latin-1 / WinAnsiEncoding (chars 32–255). Characters outside this range are replaced with ?

Development

npm install
npm run build       # compile
npm test            # run tests
npm run typecheck   # TypeScript check
npm run test:coverage

License

MIT