ميزات البرنامج · إزاحات حدود UTF-16

def utf16_offsets(text):
    if type(text) is not str or len(text) > 4096:
        raise ValueError("bounded text")
    offsets = [0]
    for char in text:
        value = ord(char)
        if 0xD800 <= value <= 0xDFFF:
            raise ValueError("unpaired surrogate")
        offsets.append(offsets[-1] + (2 if value > 0xFFFF else 1))
    return offsets