Closed steingod closed 4 years ago
When harvesting from data centres, what according to the harvest stream should be UTF-8 is really ISO 8859-1. This cause downstream hickups. Robust handling of character encoding issues in harvest_metadata.py is required.
First attempt is commited and this solved some issues previously detected.
This is resolved and implemented.
When harvesting from data centres, what according to the harvest stream should be UTF-8 is really ISO 8859-1. This cause downstream hickups. Robust handling of character encoding issues in harvest_metadata.py is required.