On Wed, 12 Nov 2025 21:39:09 +0100
Daniel Kiper <[email protected]> wrote:

> On Sun, Nov 09, 2025 at 04:31:52PM -0600, Glenn Washburn wrote:
> > The script assumes that it is run from the root of the source tree,
> > which is where it is located. So this should be enforced to prevent
> > accidental misuses.
> >
> > Signed-off-by: Glenn Washburn <[email protected]>
> > ---
> >  linguas.sh | 5 +++++
> >  1 file changed, 5 insertions(+)
> >
> > diff --git a/linguas.sh b/linguas.sh
> > index b95ad4f7d9c2..45cdf2ac14ac 100755
> > --- a/linguas.sh
> > +++ b/linguas.sh
> > @@ -1,5 +1,10 @@
> >  #!/bin/sh
> >
> > +SDIR=$(realpath -e "$0")
> 
> SDIR=$(readlink -f "$0")

I'm curious, why do you prefer this? On Debian they both come from the
coreutils package. They both resolve symlinks, so functionally I
believe they are the same. According to the Debian readlink man page,
"Note realpath(1) is the preferred command to use for canonicalization
functionality." I find the name a little more intuitive, as we want the
real path and the argument might not be a link to be read, which is
what readlink suggests to me. Is readlink an older util, and thus more
widely supported?

> 
> > +SDIR=${SDIR%/*}
> 
> SDIR=$(dirname "$SDIR")

Also, why this is preferable? My guess is that it is more
readable. I chose the pure shell approach because its one less exec and
external binary dependency. That syntax is supported on both sh and
dash, so I expect wide compatibility.

Glenn

> 
> Daniel

_______________________________________________
Grub-devel mailing list
[email protected]
https://lists.gnu.org/mailman/listinfo/grub-devel

Reply via email to