A simple JS/TS client for interacting with a Gotenberg API.
Gotenberg is a Docker-powered stateless API for converting HTML, Markdown and Office documents to PDF.
- HTML and Markdown conversions using Google Chrome headless
- Office conversions (.txt, .rtf, .docx, .doc, .odt, .pptx, .ppt, .odp and so on) using unoconv
- Assets: send your header, footer, images, fonts, stylesheets and so on for converting your HTML and Markdown to beautiful PDFs!
- Easily interact with the API using Go and PHP libraries (and now - JavaScript too ;)
$ yarn add gotenberg-js-clientOr using npm
$ npm install --save gotenberg-js-clientimport { pipe, gotenberg, convert, html, please } from 'gotenberg-js-client'
const toPDF = pipe(
gotenberg('http://localhost:3000'),
convert,
html,
please
)
// --- 8< ---
// convert file from disk
const pdf = await toPDF('file://index.html')
// or convert stream
const pdf = await toPDF(fs.createReadStream('index.html'))
// or convert string!
const pdf = await toPDF('<html>...</html>')
// library returns NodeJS.ReadableStream,
// so you can save it to file, if you want, for example
pdf.pipe(fs.createWriteStream('index.pdf'))
// or you can send it as response in Express application
app.get('/pdf', function (req, res) {
//...
pdf.pipe(res)
})You can define any source like string, Buffer, file link, stream.Readable, or URL (for url conversions).
Detailed sources format you can find here.
You can define sources as array or object, for example:
// `toPDF` function is defined above ↑↑↑
// as object
const pdf = await toPDF({
'index.html': 'file://index.html',
'header.html': 'file://header.html',
'footer.html': 'file://footer.html',
'style.css': 'file://style.css',
'img.png': 'file://img.png',
'font.wof': 'file://font.wof',
})
// as array of tuples
const pdf = await toPDF([
['index.html', 'file://index.html'],
['header.html', 'file://header.html'],
['footer.html', 'file://footer.html'],
])
// as even 1-dimensional array of files
// in that case filenames will be retrieved from file path
const pdf = await toPDF([
'file://index.html',
'file://header.html',
'file://footer.html',
])Instead of array you can use any iterable, like Map, Set, arguments, iterator from generator function, or any object with [Symbol.iterator] defined.
Detailed sources format you can find here.
When converting HTML or Markdown, you can use to helper, to set paper size, margins and orientation:
import {
pipe,
gotenberg,
convert,
html,
please,
to,
a4,
landscape,
} from 'gotenberg-js-client'
const toPDF = pipe(
gotenberg('http://localhost:3000'),
convert,
html,
to(a4, landscape),
please
)You can use simple object(s) for to argument(s) as well:
//...
to({
paperWidth: 8.27,
paperHeight: 11.69,
marginTop: 0,
marginBottom: 0,
marginLeft: 0,
marginRight: 0,
landscape: true,
})
//...
// or
to([8.27, 11.69], [0, 0, 0, 0], { landscape: true })
//...
// or
to({ paper: [8.27, 11.69], margins: [0, 0, 0, 0], landscape: true })
//...
// or
to({ width: 8.27, height: 11.69 }, { landscape: true })
//...
// or
to({ top: 0, bottom: 0 })
//...
// or any other combinationWhen using array for paper size, order should be [width, height]
When using array for margins, order should be [top, right, bottom, left] (just like in CSS)
You can set common options, like resultFilename, or waitTimeout, or, actually, you can override any option, using set helper:
//...
set({
resultFilename: 'foo.pdf',
waitTimeout: 2.5,
})
//...There are some modifiers functions as well, like filename, timeout, delay, webhook and googleChromeRpccBufferSize:
//...
set(filename('foo.pdf'), timeout(2.5))
//...Also you can specify page ranges using set(range) (will not work with merge):
//...
set(range('1-1'))
//...or scale, using set(scale) (works with HTML, Markdown and URL conversions):
//...
set(scale(0.75))
//...Markdown // Gotenberg documentation
import { pipe, gotenberg, convert, markdown, please } from 'gotenberg-js-client'
const toPDF = pipe(
gotenberg('http://localhost:3000'),
convert,
markdown,
please
)
// --- 8< ---
const pdf = await toPDF({
'index.html': `
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>My PDF</title>
</head>
<body>
{{ toHTML .DirPath "content.md" }}
</body>
</html>`,
'content.md': `
# My awesome markdown
...
`,
})Note: I use strings here as an example, remind that you can use other supported source type.
Office // Gotenberg documentation
import {
pipe,
gotenberg,
convert,
office,
to,
landscape,
set,
filename,
please,
} from 'gotenberg-js-client'
const toPDF = pipe(
gotenberg('http://localhost:3000'),
convert,
office,
to(landscape),
set(filename('result.pdf')),
please
)
// --- 8< ---
const pdf = await toPDF('file://document.docx')Note: I use file link here as an example, remind that you can use other supported source type, say, Buffer, or stream.Readable:
https.get(
'https://file-examples.com/wp-content/uploads/2017/02/file-sample_100kB.docx',
async (document) => {
const pdf = await toPDF({ 'document.docx': document })
// ...
}
)Url // Gotenberg documentation
import { pipe, gotenberg, convert, url, please } from 'gotenberg-js-client'
const toPDF = pipe(
gotenberg('http://localhost:3000'),
convert,
url,
please
)
// --- 8< ---
// you can use link as string
const pdf = await toPDF('https://google.com')
// or URL object
const pdf = await toPDF(new URL('https://google.com'))Note: The only supported source for Url conversion is text url or instance of URL class.
You can set remote url header (for example, for authentication or host specifying) with helper add(header) (or add(headers), or both):
const toPDF = pipe(
gotenberg('http://localhost:3000'),
convert,
url,
add(
header('Foo-Header', 'Foo'),
header('Bar-Header', 'Bar'),
headers({ 'Baz1-Header': 'Baz1', 'Baz2-Header': 'Baz2' })
),
please
)(This also applies for Webhook headers, just use webhookHeader instead of header and webhookHeaders instead of headers).
Merge // Gotenberg documentation
Like you would think:
import { pipe, gotenberg, merge, please } from 'gotenberg-js-client'
const toMergedPDF = pipe(
gotenberg('http://localhost:3000'),
merge,
please
)There is special function adjust, which you can use to modify any field in prepared internal Request object. You can check internal Request object structure in types. Any object, passed to adjust, will be merged with prepared Request.
For example, you can modify url, if your Gotenberg instance is working behind reverse proxy with some weird url replacement rules:
import { pipe, gotenberg, convert, html, adjust, please } from 'gotenberg-js-client'
// Original Gotenberg HTML conversion endpoint is
// -> /convert/html
// But your reverse proxy uses location
// -> /hidden/html/conversion
const toPDF = pipe(
gotenberg('http://localhost:3000'),
convert,
html,
adjust({ url: '/hidden/html/conversion' }),
please
)But, using that function, remember about Peter Parker principle:
"With great power comes great responsibility"
If you happen to use this package from JavaScript, you will, obviously, lost type safety, but in return, you can use proposed pipe operator (with Babel plugin), to get beauty like this:
const toPDF = source =>
source
|> gotenberg('http://localhost:3000')
|> convert
|> html
|> to(a4, noMargins)
|> set(filename('out.pdf'))
|> pleaseIf you don't like to have simple imported names in your namespace, you can use import * syntax:
import * as got from 'gotenberg-js-client'
const toPDF = got.pipe(
got.gotenberg('http://localhost:3000'),
got.convert,
got.html,
got.please
)