There are some hard-to-handle edge cases when doing display length truncation in Unicode, e.g. the character U+FDFD or "﷽" is four bytes but can be very long de

There are some hard-to-handle edge cases when doing display length truncation in Unicode, e.g. the character U+FDFD or "﷽" is four bytes but can be very long depending on the typeface*, so "completely" solving it is quite hard and has to depend on feedback from your rasterization engine. (*Rendered version on Wikipedia: https://commons.wikimedia.org/wiki/File:Lateef_unicode_U%2BFDFD_2020-03-09_122519.png https://commons.wikimedia.org/wiki/File:Lateef_unicode_U%2BF... )