Import of papers fails when it contains unicode characters that require 4 bytes in UTF-8 encoding.
Caused by database using utf8mb3 character set. Actual fix would be to convert to utf8mb4 at a carefully chosen moment after testing.
Temporal fix: use HTML entities in abstract field of papers (these are markdown, and may contain HTML). For titles (we only had 1 occurrence up till now), replace the character with a
Submitted by Elmer van Chastelet on 14 June 2022 at 08:21?
so the import won’t fail at the cost of this character getting lost.