But you can assume non-UTF-8 upon seeing an invalid UTF-8 byte sequence. From there, it can be application-specific depending on what encoding you expect to se

But you can assume non-UTF-8 upon seeing an invalid UTF-8 byte sequence. From there, it can be application-specific depending on what encoding you expect to see. (And of course UTF-16 if there's a BOM)