I noticed that when you construct a pymarc.Record object it gets the default file encoding of iso8859-1:
https://gitlab.com/pymarc/pymarc/-/blob/main/pymarc/record.py?ref_type=heads#L94
I think this was done for historical reasons, but I wonder if it’s time to change it to uff-8? I’m assuming most ILS systems handle utf-8 ok now?
I’ll be honest, I’m a bit wary of playing with encoding parameters much mostly because it’s not always clear to me how to_unicode, force_utf8, utf8_handling work together. Maybe simplifying this is something to keep in mind for a future major release of pymarc instead of playing with it now…
//Ed