Character Encodings and Livecode fields

[Graham Samuel]

The first 255 characters on Unicode will occupy double the number of bytes as the first 255 characters of ascii: this means that a transformation has to take place, however trivial (the extra byte in each case is perhaps all zeros - I have not looked at this yet). This is what I call 'promoting' - it transforms a mean and restrictive encoding to a generous and universal one. 'Promotion' seems to me a good word for this transformation. Or one could of course stick to the word 'transformation'.

If you're saying that ascii can become Unicode without any manipulation at all, then I see that I don't understand the basic format of Unicode, which I thought was basically 16 bits (2 bytes) per character. Maybe I am being fooled by my history as a binary bit-twiddler whose experience pre-dates even the word 'byte'. Octal, anyone?

G

On 26 Jan 2014, at 17:23, Richmond <richmondmathewson at gmail.com> wrote:

> I don't know what you mean by "'promoting' non-Unicode character strings to Unicode; that sounds a bit odd: as far as I know the ASCII
> set is subsumed as the first 255 chars on Unicode so that is neither here nor there.