[NUT-devel] Names in Nut Packets

[Michael Niedermayer]

Hi

On Sun, Feb 19, 2006 at 02:13:45PM +0200, Oded Shimon wrote:
> On Sun, Feb 19, 2006 at 10:47:03AM +0100, Michael Niedermayer wrote:
> > Hi
> > 
> > currently Names are UTF-8, IMHO we should limit them to [a-zA-Z0-9] at least
> > or maybe just [a-z]
> > furthermore its a little tempting to either use the unused bytes for common
> > syllables or store things in 5-6bit per char
> > or does that syllable table need to be extendible but not storeable too? in
> > which case iam obviously not agreeing
> 
> You mean, names of fields in info packets?..

yes


> i think not limit them, allow the user to have whatever extra weird fields 
> he wants, there's no big disadvantage in allowing full utf-8...

id say require names and types to be ASCII and in english or at least suggest
it if possible, otherwise we might end up with X-?????? stuff written in your
favorite asian or whatever language

values should be UTF-8 where appropriate and ASCII where not
(Author,Title, ... should be UTF-8 while Language and Disposition just need 
ASCII)

but i dont have a strong oppionion on these, making all UTF-8 is fine for
me too

[...]
-- 
Michael