Using non-ASCII as identifier could lead to intersting things:
eacute=(e\\U301 \\UE9); for i in {0..7};do
printf -v gen 'g%bn%bralit%b' "${eacute[i>>2]}" "${eacute[(i>>1)&1]}"
"${eacute[i&1]}"
eval "function $gen () { echo $RANDOM ;}"
done
set | grep --no-group-separator -A3 '^g.*lit.*()' | sed 's/^/ /'
généralité ()
{
echo 16740
}
généralité ()
{
echo 10077
}
généralité ()
{
echo 27993
}
généralité ()
{
echo 24670
...
(généralité;généralité;généralité;généralité;généralité;généralité;généralité;généralité)|xargs
16740 10077 27993 24670 12601 7528 9292 6933
So you may store eight different functions with near same name. ( could be
usefull for obfuscation ;-)
(you could do same using many filesystems, storing different files with near
same name)
( NOTE: Don't try to copy/paste accented part of this mail in your console! )
Le Mon, Jun 01, 2026 at 04:26:21PM -0400, Chet Ramey a écrit :
> On 5/20/26 12:10 PM, Ótzï fón GlüϷgor wrote:
> > Hello,
> > I would like to propose an opt-in Bash option to allow non-ASCII characters
> > in shell identifiers.
>
> This has come up before. I'm not inclined to add this feature -- in my
> view, the cost of implementation and maintenance outweighs the value.
>
> Chet
> --
> ``The lyf so short, the craft so long to lerne.'' - Chaucer
> ``Ars longa, vita brevis'' - Hippocrates
> Chet Ramey, UTech, CWRU [email protected] http://tiswww.cwru.edu/~chet/
--
Félix Hauri - <[email protected]> - http://www.f-hauri.ch