Perhaps it could be noted. But it fails
with both musl and glibc. I think we need new
regex engine, perhaps adapted from musl,
that supports NUL bytes and UTF-8 transparently.
maandree
On Sun, 3 Apr 2016 22:30:45 +0200
FRIGN <dev_AT_frign.de> wrote:
> On Wed, 30 Mar 2016 19:01:16 +0200
> Mattias Andrée <maandree_AT_kth.se> wrote:
>
> Hey Mattias,
>
> > $ echo äö | ./grep [å]
> > äö
> >
> > This is not want one expects from
> > a program that supports UTF-8.
>
> I assume the proper way here would be to add a note on
> the manpage that this highly depends on the libc you are
> using (and esp. the Regex engine).
>
> Cheers
>
> FRIGN
>
Received on Sun Apr 03 2016 - 23:35:21 CEST