https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=42948 David Cook <dcook@prosentient.com.au> changed: What |Removed |Added ---------------------------------------------------------------------------- Summary|Diacritics corrupted when |Authority records default |saving authority records |to MARC-8 instead of UTF-8 |(UTF-8 mangled into |if missing MARC21 leader |unrelated Unicode | |characters, e.g. é→ř, | |ú/ñ→ž) | Status|NEW |CONFIRMED --- Comment #8 from David Cook <dcook@prosentient.com.au> --- So if I change the Leader to use MARC-8 encoding instead of UTF-8 I get "Přez Zį͠ga, Juan" which is different to what you said on Bugzilla but is the same to what I see in your screenshots with the Z39.50. And looking at your screenshots again... you've got a misconfigured framework for your "Authority type". You're missing the "000 - LEADER". And if there is no "000 - LEADER", it looks like it doesn't default to UTF-8 and instead defaults to MARC-8, which then leads to your encoding issue. -- I had wondered a bit at the start if this was more of a support issue than a Koha bug, but I thought it worth investigating. I should've double-checked the screenshots earlier... I think... we can probably still classify this one as a bug, because all MARC21 records should have a leader. Maybe we do need to make certain fields unhideable (e.g. 000/Leader, 005, 008) as hiding these fields leads to bad records. I'm going to rename this one from 'Diacritics corrupted when saving authority records (UTF-8 mangled into unrelated Unicode characters, e.g. é→ř, ú/ñ→ž)' to 'Authority records default to MARC-8 instead of UTF-8 if missing MARC21 leader' -- You are receiving this mail because: You are the assignee for the bug. You are watching all bug changes.