2 ms·
> And on Windows it's "potentially ill-formed UTF-16" On Windows, and in many Windows APIs, the return value is a "sequence of UTF-16 code units" (which is ess
by gsnedders 11y ago
> And on Windows it's "potentially ill-formed UTF-16"
On Windows, and in many Windows APIs, the return value is a "sequence of UTF-16 code units" (which is essentially a fancy way of saying "a sequence of 16-bit words"). This goes back to Windows originally using UCS-2, where any sequence of 16-bit words is valid (because there wasn't any concept of a surrogate, yet alone a lone surrogate), given its support predates the Unicode consortium extending Unicode to 21-bits and along with it UTF-16's existence (1996).