- Improve overall performance, code factorization.
- Add a lot of robustness around, to avoid Python exceptions but present clear error messages instead, e.g. when .dat XML file is a corrupt XML.
- Add more filtering/cleaning capabilities via input .dat file browsing?
- etc.