Skip to content

Develop semantic data field xml as basis for visu refactoring

The idea is to introduce a new processing modell and semantoc data field tree structure.

This should serve as a basis for refactoring the visu code and then allow w.r.t. to i18n, codelist translation, custom lang and general configurability while keeping overall code complexity simple.

Status quo

Only keypoints:

  • We have a semantic data model -- without actual data

  • We have data formats (UBL and CII) -- without actual semantic model data

  • Transformations from UBL to XR, from CII to XR,

  • Completly different transformations from XR to HTML and one from XR to PDF

  • The value of XR Transformation is to combine the actual data with the actual model

  • The disadvantage is the XR format is properiatry and gives too many people the idea, that this would be the XRechnung format of choice

The new general process

Divide and Conquer:

  1. Step: Add semantic data to the actual data format
  2. Step: Take original data format from step one (or old legacy XR)

The data structure

Something like the following combines the semantic data model information with the actual data and context

<sem-data>
   <type>term</type>
   <name>Invoice_number</name>
   <id>BT-1</id>
   <xpath>/ubl:Invoice/cbc:ID</xpath>
   <value>12345</value>
   <pos>1</pos>
</sem-data>

By producing something like this above as only a non serialized doc tree, one can also easily detect if the input document is either XR, sem-annotated UBL or CII.

Edited by Renzo Kottmann