Warren Moore

You mean internally? Because it does accept in/out utf8 interfaces to set/grab the data

Yes, but also the API is 'sort-of' messed up UTF-16, e.g. length is returned as number of "code units" (number of unichars) and characterAtIndex() returns a single unichar, but unichar is uint16_t, so some characters return 2 code units. Same mess as on Windows.

I bet when they created NSString, they thought that a simple 16-bit-per-character encoding ought-to-be-enough-for-everybody, and later they hacked this into UTF-16. I would've preferred ASCII hacked into UTF-8 instead ;)

...what I'd need is a simple way to loop over UTF-32 characters in the NSString, I'm aware that I can decode into an NSData object with any encoding, but that's a bit awkward :D

You can read more about the history of NSString here: https://www.objc.io/issues/9-strings/unicode/ …, but generally speaking, a list of code points (≠ characters) isn’t what you want, regardless of encoding. Unless you’re writing your own shaping/layout engine, which is a Very Hard Problem.

Well, he's doing UI so very likely he actually wants code units

What do you do with code points once you get them? If the answer is “index into an atlas or glyph array,” you’re Doing It Wrong.

You mean because of combining characters?