|
| From: | John W. Eaton |
| Subject: | Re: multibyte representation of character codes |
| Date: | Sat, 14 Dec 2013 08:59:15 -0600 |
| User-agent: | Mozilla/5.0 (X11; Linux i686; rv:10.0.11) Gecko/20121123 Icedove/10.0.11 |
On 12/14/2013 04:39 AM, Серёжа Плотников wrote:
Oh, I see. I thought it's just about changing the way, how non-ASCII characters are read, because their codes are correctly represented in UTF-8.
Not, it's not a trivial change. If we make a change for wide characters, we'll probably want to be compatible with Matlab, and that means using 16-bit characters and that change will be pervasive, all throughout Octave. That sounds like something for 5.0 or 6.0 or never.
Just kidding about the "never", but I think it will be a lot of work and quite disruptive.
jwe
| [Prev in Thread] | Current Thread | [Next in Thread] |