Charsets (was: RE: netmd-dev digest, Vol 1 #260 - 3 msgs)

On Fri, Jan 31, 2003 at 03:00:57PM +0100, Alf Steinbach wrote:
> Eric Woudenberg, Minidisc.org Editor, wrote:
> > I think I got this information from John Watkinson's "The Art of
> > Digital Audio" or maybe Pohlman's book on Digital Audio.
> >
> > But just checking on the web, ISO-8895 seems like an 8-bit ASCII-like
> > encoding, so maybe this isn't right, or there's a typo somewhere.
> 
> 
> Just a thought, but ISO-8859-1 (note the digit reversal) is ISO Latin-1,
> which constitutes the first 256 characters of Unicode.  It's also a
> subset of Windows ANSI (Western), the default 8-bit encoding in Windows.

ISO 8859 defines a whole series of single byte charsets, of which
ISO 8859-1 (Latin 1) is just one... 
http://czyborra.com/charsets/iso8859.html has quite a nice overview.

All the refrences I saw on Google for ISO-8895 were typos from
ISO-8859 so I'm going to ignore that for now :)

However I am pretty sure this is *not* what is stored in the UTOC,
or at least not what is returned via NetMD (for a start, I'm
pretty sure there is no ISO-8859 variant which can represent
hiragana and katakana).

The single byte charset appears to be a slight variation
of JIS x0201, as decoded at
http://www.floofy-skirts.org/~chimerae/projects/netmd-dev/misc/minidisc_single.txt

I'v not looked at the wchar stuff very much, but it looks an awful
lot like UTF-8, especially all the recurrent 0xCn bytes (which
are used in UTF-8 to prefix multibyte sequences).

Pete.