factur-x

Factur-X and Order-X: electronic invoicing and ordering standards


Keywords
e-invoice, ZUGFeRD, Factur-X, Order-X, e-procurement, odoo
License
BSD-3-Clause
Install
pip install factur-x==2.1

Documentation

Factur-X and Order-X Python library

Factur-X is a Franco-German e-invoicing standard which complies with the European e-invoicing standard EN 16931. The Factur-X specifications are available on the FNFE-MPE website in English and French. The Factur-X standard is also called ZUGFeRD 2.2 in Germany.

Order-X is the equivalent of Factur-X for purchase orders. The Order-X specifications are available in English on the FNFE-MPE website and on the FeRD website.

The main feature of this Python library is to generate Factur-X invoices and Order-X orders from a regular PDF document and a Factur-X or Order-X compliant XML file.

This lib provides additionnal features such as:

  • extract the XML file from a Factur-X or Order-X PDF file,
  • check a Factur-X or Order-X XML file against the official XML Schema Definition.

Some of the features provided by this lib also work for ZUGFeRD 1.0 (the ancestor of the Factur-X standard).

Installation

To install it on Linux, run:

sudo pip3 install --upgrade factur-x

Usage

from facturx import generate_from_file

generate_from_file(regular_pdf_file, xml_file)

The PDF file regular_pdf_file will be updated to Factur-X/Order-X. If you want to write the resulting Factur-X/Order-X PDF to another file, use the argument output_pdf_file.

To have more examples, look at the docstrings in the source code or look at the source code of the command line tools located in the bin subdirectory.

Command line tools

Several command line tools are provided with this lib:

  • facturx-pdfgen: generate a Factur-X or Order-X PDF file from a regular PDF file and an XML file
  • facturx-pdfextractxml: extract the XML file from a Factur-X or Order-X PDF file
  • facturx-xmlcheck: check a Factur-X or Order-X XML file against the official XML Schema Definition

All these commande line tools have a --help option that explains how to use them and shows all the available options.

Tutorial: generate a Factur-X invoice under Windows

Download the last version of Python for Windows from python.org/downloads.

Launch the installer. On the first screen of the installer, enable the option Add python.exe to PATH. At the end of the installation process, the installer displays a screen with the message Setup was successful ; at that step, it may propose you Disable path length limit with a help message that says Changes your machine configuration to allow programs, including Python, to bypass the 260 character "MAX_PATH" limitation. You must accept this proposal (otherwise the installation of the factur-x library will fail): click on the label Disable path length limit and follow the instructions.

Open a Windows command prompt as Administrator and enter the following command to download and install the factur-x library:

pip3 install --upgrade factur-x

Look at the installation logs and make sure there are no error messages. Close the Windows command prompt.

Open a new Windows command prompt (not as Administrator) and enter the following command (adapt the path to your filesystem):

python C:\Users\Alexis\AppData\Local\Programs\Python\Python311\Scripts\facturx-pdfgen --help

It should display the help of the command facturx-pdfgen.

Enter the following command to generate a Factur-X invoice:

python C:\Users\Alexis\AppData\Local\Programs\Python\Python311\Scripts\facturx-pdfgen C:\Users\Alexis\Documents\invoice.pdf C:\Users\Alexis\Documents\fx.xml C:\Users\Alexis\Documents\invoice-facturx.pdf

where:

  • C:\Users\Alexis\Documents\invoice.pdf is the original PDF invoice,
  • C:\Users\Alexis\Documents\fx.xml is the Factur-X XML file,
  • C:\Users\Alexis\Documents\invoice-facturx.pdf is the Factur-X PDF invoice that will be generated.

Webservice

This project also provides a webservice to generate a Factur-X or Order-X PDF file from a regular PDF file, the XML file and additional attachments (if any). This webservice uses Flask. To run the webservice, run facturx-webservice available in the bin subdirectory of the project. To query the webservice, you must send an HTTP POST request in multipart/form-data using the following keys:

  • pdf -> PDF file (required)
  • xml -> Factur-X or Order-X file (any profile, required)
  • attachment1 -> First attachment (optional)
  • attachment2 -> Second attachment (optional)
  • ...

To deploy this webservice in production, follow the guidelines of the official Flask documentation: you should use a WSGI server (such as Gunicorn) and a reverse proxy (such as Nginx or Apache). You will certainly have to increase the default maximum upload size (default value is only 1MB under Nginx!): use the parameter client_max_body_size for Nginx and LimitRequestBody for Apache.

I recommend this tutorial (in French) which explains how to deploy a Flask application with Gunicorn and Nginx on Ubuntu.

You can use curl, a command line tool to send HTTP requests (on Linux Ubuntu/Debian, just install the curl package) to generate the request:

curl -X POST -F 'pdf=@/home/me/regular_invoice.pdf' -F 'xml=@/home/me/factur-x.xml' -F 'attachment1=@/home/me/delivery_note.pdf' -o /home/me/facturx_invoice.pdf https://ws.fnfe-mpe.org/generate_facturx

A public instance of this webservice is available on a server of FNFE-MPE at the URL https://ws.fnfe-mpe.org/generate_facturx.

Licence

This library is published under the BSD licence (same licence as pypdf on which this lib depends).

Contributors

Changelog

  • Version 3.1 dated 2023-08-13
    • Keep bookmarks, annotations, etc. from input PDF file. For that, we use the method clone_document_from_reader() of pypdf instead of append_pages_from_reader()
    • Fix bug on xml type parsing (bug introduced in version 3.0)
    • raise explicit error when trying to generate a ZUGFeRD 1.x PDF invoice
  • Version 3.0 dated 2023-08-13
    • Replace dependency on PyPDF4 by pypdf. The development focus is back on pypdf and the forks PyPDF2, PyPDF3 and PyPDF4 are not maintained any more, cf this article.
    • Remove support for Python 2.7
    • In the scripts, replace /usr/bin/python3 by /usr/bin/env python
  • Version 2.5 dated 2023-03-24
    • Add support for ZUGFeRD 1.0 in get_level()
    • xml_check_xsd(): avoid warning Use specific 'len(elem)' or 'elem is not None' test instead.
  • Version 2.4 dated 2023-03-13
    • Update Factur-X XSD of all profiles to version 1.0.6
    • Update Order-X XSD of all profiles to version 1.0.0
  • Version 2.3 dated 2021-04-12
    • Fix wrong flavor argument passed by generate_facturx_from_file() to generate_from_file()
  • Version 2.2 dated 2021-04-08
    • Make method generate_from_binary() accessible via the lib
  • Version 2.1 dated 2021-04-07
    • Update Order-X XSD to the latest version provided to me by FNFE-MPE
  • Version 2.0 dated 2021-04-04
    • Add support for Order-X. This implies several changes:
      • method check_facturx_xsd() deprecated in favor of the new method xml_check_xsd() but still operates with a warning
      • method get_facturx_flavor() deprecated in favor of the new method get_flavor() but still operates with a warning
      • method generate_facturx_from_binary() deprecated in favor of the new method generate_from_binary() but still operates with a warning
      • method generate_facturx_from_file() deprecated in favor of the new method generate_from_file() but still operates with a warning
      • new optional argument orderx_type for methods generate_from_file() and generate_from_binary() with default value autodetect
      • new method get_orderx_type() to get the Order-X type (order, order change or order response)
      • new method get_xml_from_pdf() that work both on Factur-X and Order-X (the method get_facturx_xml_from_pdf() still exists and only operates on Factur-X)
      • scripts updated
    • Add lang argument to methods generate_from_file() and generate_from_binary() to set the lang of the PDF. This is one of the requirements for PDF accessibility, which is important for people with disabilities: it allows PDF speech synthesizers for blind people to choose the right language.
    • Add ability to choose the AFRelationship PDF property for the Factur-X/Order-X XML file and also for the additionnal attachments:
      • new argument afrelationship for methods generate_from_file() and generate_from_binary()
      • new key afrelationship for the attachments dict as argument of generate_from_file() and generate_from_binary()
    • Argument additional_attachments was deprecated in method generate_facturx_from_file() in version 1.8: it doesn't operate any more and only displays a warning.
    • Replace the optparse lib by the argparse lib in scripts.
  • Version 1.12 dated 2020-07-16
    • Compress attachments and XMP metadata using Flate compression
  • Version 1.11 dated 2020-05-11
    • Fix crash UnicodeEncodeError on Python 2.7
  • Version 1.10 dated 2020-04-14
    • Update XSD of all profiles to Factur-X version 1.0.5
  • Version 1.9 dated 2020-02-11
    • Improve Python3 support in get_facturx_xml_from_pdf()
  • Version 1.8 dated 2020-01-16
    • New tool facturx-webservice which implements a REST webservice using Flask to generate a Factur-X PDF invoice via a simple POST request.
    • New argument 'attachments' for generate_facturx_from_file() which replaces argument additional_attachments:
      • Possibility to set a filename for the attachment different from filename of the filepath
      • Possibility to set creation dates for attachments
      • Update script facturx-pdfgen to use the new attachments argument
  • Version 1.7 dated 2020-01-13
    • Fix bug in release 1.6 in XMP: variables were not replaced by their real value
  • Version 1.6 dated 2020-01-09
    • Generate XMP (XML-based PDF metadata) via string replacement instead of using XML lib
  • Version 1.5 dated 2019-11-13
    • Fix bug in generate_facturx_from_file() when using argument additional_attachments
  • Version 1.4 dated 2019-07-24
    • Update Factur-X XSD to the final version of Factur-X v1.0.04
    • Support XML extraction with ZUGFeRD invoices using 'zugferd-invoice.xml' filename (instead of the filename 'ZUGFeRD-invoice.xml' specified by the standard)
  • Version 1.3 dated 2019-06-12
    • Add XSD files for Extended profile in the Python package
  • Version 1.2 dated 2019-06-12
    • add support for the Extended profile
    • validate XML for Minimum and Basic WL profiles with the XSD of profile EN 16931, as asked by Cyrille Sautereau
    • minor improvements in the code for /Kids
  • Version 1.1 dated 2019-04-22
    • Improve support for embedded files extraction by adding support for /Kids
  • Version 1.0 dated 2019-01-26
  • Version 0.9 dated 2019-01-25
  • Version 0.8 dated 2018-06-10
    • Make pretty_print work for XMP file, for better readability of that file
  • Version 0.7 dated 2018-05-24
    • Fix XMP structure under /x:xmpmeta/rdf:RDF/rdf:Description (use XML tags instead of XML attributes)
    • declare PDF-1.6 instead of PDF-1.3 (still declared by default by pyPDF2)
  • Version 0.6 dated 2018-05-01
    • Now fully PDF/A-3 compliant with additionnal attachments (tested with veraPDF)
    • facturx-pdfgen: don't overwrite by default and add --overwrite option
    • Add factur-x library version number in metadata creator entry
  • Version 0.5 dated 2018-03-29
    • Fix XMP metadata structure
    • Now fully PDF/A-3 compliant when the input PDF file is PDF/A compliant (tested with veraPDF). This implied copying /OutputIntents and /ID datas from source PDF to Factur-X PDF.
    • Fix support for additionnal attachments: they can now all be saved with Acrobat Reader
    • Improve XML extraction from PDF Factur-x file
  • Version 0.4 dated 2018-03-27
    • Factur-x specs say /AFRelationship must be /Data (and not /Alternative)
    • Update Factur-X XSD to v1.0 final
    • Add support for additionnal attachments
    • Add factur-x lib version in Creator metadata table
    • Add /PageMode = /UseAttachments, so that the attachments are displayed by default when opening Factur-X PDF invoice with Acrobat Reader
    • Improve and enrich PDF objects (ModDate, CheckSum, Size)