SDK reference

@stafyniaksacha/facturx exposes a small, focused API. Everything is async and works with strings, Buffers or already-parsed documents.

Imports

The main entry point exports the functions and a Models namespace:

import {
  check,
  extract,
  generate,
  invoiceToXml,
  Models,
  validateSchematron,
  xmlToInvoice,
} from '@stafyniaksacha/facturx'

The model classes are also available individually from the /models subpath (handy when building an invoice from scratch):

import { AmountType, CrossIndustryInvoiceType, IDType /* … */ } from '@stafyniaksacha/facturx/models'

Exports

Export	Kind	Description
generate(options)	() => Promise<Uint8Array>	Embed XML into a PDF → PDF/A-3 bytes.
extract(options)	() => Promise<ExtractResult>	Pull the embedded XML out of a PDF.
check(options)	() => Promise<CheckResult>	XSD (+ optional Schematron) validation.
validateSchematron(options)	() => Promise<{ valid, errors }>	EN 16931 business-rule validation (Factur-X only).
xmlToInvoice(xml)	() => Promise<CrossIndustryInvoiceType>	Parse XML into a typed model.
invoiceToXml(invoice)	() => Promise<XMLDocument>	Serialize a model back to XML.
Models	namespace	All Cross Industry Invoice model classes & types.

Typical flows

Produce a Factur-X PDF from an existing PDF + XML:

import { readFile, writeFile } from 'node:fs/promises'
import { generate } from '@stafyniaksacha/facturx'

const pdf = await readFile('./invoice.pdf')
const xml = await readFile('./factur-x.xml')

const bytes = await generate({ pdf, xml }) // validates, then embeds
await writeFile('./facturx.pdf', bytes)

Read the data out of a received invoice:

import { readFile } from 'node:fs/promises'
import { extract, xmlToInvoice } from '@stafyniaksacha/facturx'

const pdf = await readFile('./received.pdf')
const { xml, flavor, level } = await extract({ pdf })

const invoice = await xmlToInvoice(xml)
const seller = invoice.supplyChainTradeTransaction
  .applicableHeaderTradeAgreement.sellerTradeParty.name?.value

Conventions

• flavor / level are optional on every function that takes XML — they're autodetected from the document when omitted.
• Inputs accept string | Buffer (and generate/check also accept a parsed XMLDocument; generate/extract also accept a PDFDocument).
• Errors are thrown as standard Errors (e.g. No attachment found, Invalid XML).