[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [PATCH] tests: use printf, not echo in init.sh's warn_ function
From: |
Jim Meyering |
Subject: |
Re: [PATCH] tests: use printf, not echo in init.sh's warn_ function |
Date: |
Wed, 15 Jun 2011 22:15:35 +0200 |
James Youngman wrote:
> I'd doubt that there are gnulib tests for which this will really make
> a difference, but
There is none in coreutils, but there *are* tests there that set IFS.
Just none that use these relatively new functions in an affected scope.
> ( IFS=' '; printf '%s\n' "$*"; )
>
> is perhaps slightly more reproducible.
Good suggestion. Setting IFS does seem prudent.
However, I'm inclined to use a subshell only if necessary:
warn_ ()
{
case $IFS in
' '*) printf '%s\n' "$*" 1>&$stderr_fileno_ ;;
*) ( IFS=' '; printf '%s\n' "$*" 1>&$stderr_fileno_ ) ;;
esac
}
Thanks!
- [PATCH] tests: use printf, not echo in init.sh's warn_ function, Jim Meyering, 2011/06/14
- Re: [PATCH] tests: use printf, not echo in init.sh's warn_ function, Eric Blake, 2011/06/14
- Re: [PATCH] tests: use printf, not echo in init.sh's warn_ function, Jim Meyering, 2011/06/14
- Re: [PATCH] tests: use printf, not echo in init.sh's warn_ function, James Youngman, 2011/06/14
- Re: [PATCH] tests: use printf, not echo in init.sh's warn_ function, Jim Meyering, 2011/06/17
- RE: [PATCH] tests: use printf, not echo in init.sh's warn_ function, Voelker, Bernhard, 2011/06/20
- Re: [PATCH] tests: use printf, not echo in init.sh's warn_ function, Jim Meyering, 2011/06/20
- RE: [PATCH] tests: use printf, not echo in init.sh's warn_ function, Voelker, Bernhard, 2011/06/20