TypeScript icon, indicating that this package has built-in type declarations

0.5.3 • Public • Published

PDF Maker

PDF Maker is a library for generating PDF documents in JavaScript.

  • Easy to use: document contents are defined in plain objects.
  • Works anywhere: Browser, Node.js, Deno.
  • TypeScript support: types included in the npm package.

This project is heavily inspired by pdfmake and builds on pdf-lib and fontkit. It would not exist without the great work and the profound knowledge contributed by the authors of these projects.


The function makePdf() creates PDF data from a given document definition.

Basic Example

const fontData = await readFile('Roboto-Regular.ttf');
const fontDataBold = await readFile('Roboto-Medium.ttf');
const pdfData = await makePdf({
  // Fonts must be registered (see below)
  fonts: {
    Roboto: [{ data: fontData }, { data: fontDataBold, bold: true }],
  // Page margins (`x` is a shorthand for left and right)
  margin: { x: '2.5cm', top: '2cm', bottom: '1.5cm' },
  // Content as an array of blocks
  content: [
    // Blocks can contain text and text attributes
    { text: 'Lorem ipsum', bold: true, textAlign: 'center', fontSize: 24 },
    // Text can also be an array of text ranges with different attributes
      text: [
        'dolor sit amet, consectetur adipiscing elit ',
        { text: 'sed do eiusmod', italic: true },
        ' tempor, incididunt ut labore et dolore magna aliqua.',
await writeFile(`hello.pdf`, pdfData);


All fonts are embedded in the PDF and need to be registered with the fonts attribute. Font data is accepted in .ttf or .otf format, as ArrayBuffer, Uint8Array, or base64-encoded string. Each font family can include different variants that are selected based on the attributes bold and italic.

const documentDefinition = {
  fonts: {
    // The `fontFamily` name of the font
    'DejaVu-Sans': [
      // TTF / OTF font data as Uin8Array or base64 encoded string
      { data: fontDataDejaVuSansNormal },
      { data: fontDataDejaVuSansBold, bold: true },
      { data: fontDataDejaVuSansItalic, italic: true },
      { data: fontDataDejaVuSansBoldItalic, bold: true, italic: true },
    Roboto: [{ data: fontDataRobotoNormal }, { data: fontDataRobotoMedium, bold: true }],
  content: [
    { text: 'lorem ipsum', fontFamily: 'Roboto', bold: true }, // will use Roboto Medium
    { text: 'dolor sit amet' }, // will use DejaVu-Sans (the font registered first), normal


JPG images are supported. All images need to be registered with the images attribute. Images can be used more than once in the document without multiplying the image's footprint in the created PDF. The size of an image can be confined using the width and height attributes.

const documentDefinition = {
  images: {
    'logo': { data: imageData }
  content: [
    { image: 'logo', width: 200, height: 100 },


The content attribute of the document definition accepts an array of top-level blocks that can be text blocks, image blocks, column or row layout blocks. Page breaks will only occur between top-level blocks.


To arrange blocks horizontally, they can be included in a block with a columns attribute. When columns have a width attribute, it will be respected. The remaining space will be distributed evenly across all columns.

  columns: [
    {text: 'Column 1', width: 100}, // 100 pt wide
    {text: 'Column 2'}, // gets half of the remaining width
    {text: 'Column 3'}, // gets half of the remaining width


A rows layout can be used to group multiple rows in a single block, e.g. to apply common attributes to them or to include them in a surrounding columns layout.

  rows: [
    {text: 'Row 1'},
    {text: 'Row 2'},
    {text: 'Row 3'},
  textAlign: 'right',


There is no generated documentation yet, please refer to content.ts for an overview and specification of all supported attributes in a document definition.


A more extensive example is included in the examples/ folder.

Run npm start to generate a PDF in the out/ folder on every change of this file.





Package Sidebar


Weekly Downloads






Unpacked Size

4.26 MB

Total Files


Last publish


  • ralfstx