Does postgresql have a function that will give me the numeric unicode code point for a character in a unicode (aka utf8) database text string? That is, something like ascii() but that works for unicode characters?