Thank you very much for your response. I just looked at the source code of xlsx and the approach is indeed very similar, but there are a few differences. Docxtemplater has tags inside the word template and is in this sense an engine and not a recipe. We envision the process as follows:
- User prepares a word template with custom placeholders/tags as described here: https://docxtemplater.readthedocs.io/en/latest/tag_types.html
- He uploads the word template to jsreport
- A custom engine renders a new docx file
- This file will be passed optionally to a custom recipe to convert it to PDF
We probably need to implement a custom engine (docx) as well as a custom recipe (docx-to-pdf). Any view on this?