Niels Lohmann 92c729f657 Accept ill-formed UTF-8 in all binary readers again ...

RFC 8949 and the MessagePack/BSON/UBJSON/BJData specs leave UTF-8
well-formedness checking up to the decoder, so following #5529 the
binary readers are lenient by default again, as in release 3.12.0
(the reader-side check was added by #5185/#5531, not in any release);
reader-side validation becomes opt-in in a follow-up PR. The writers
stay strict and throw type_error.316 for ill-formed UTF-8. BON8 is
unchanged, since UTF-8 lead bytes are structural there.

Signed-off-by: Niels Lohmann <mail@nlohmann.me>

2026-10-01 07:59:14 +02:00

..

nlohmann

Accept ill-formed UTF-8 in all binary readers again

2026-10-01 07:59:14 +02:00