June 24, 2013

Produce & Publish Status Update 2

This is another update on the current implementation progress of the complete Produce & Publish infrastructure.

Significant progress has been made on the reimplementation of the Produce & Publish Server. The Produce & Publish server provides some web service API methods that all arbitrary application to convert their content to PDF through a central conversion server. The PDF generation is (as in the past) based on one of the following PDF converters:

  • PrinceXML 8
  • PDFreactor 6
  • Anntennahouse Formatter V 6

The general conversion approach remains the same:

  • all conversions are based on HTML/XML as input
  • styling is accomplished through CSS3The new Produce & Publish Server implementation is also past as the older version on Python Pyramid and it provides its services through an XML-RPC API. Legacy support for Apache FOP, XInc and XMLMind converters have been removed (not much used, poor output quality, unmaintained). In addition the server provide conversion support using LibreOffice or OpenOffice for office format A to office format B using the Python-based wrapper unoconv. This API will be used later to provide a simple DOCX conversion and import functionality for the Produce & Publish backend. All API methods support asynchronous processing (e.g. for larger documents). In this case the caller receives a job id that can be used to poll and download the converted document.

The Produce & Publish server remains open-source (published under the GNU Public License V2).