Re: `printf -v foo ""` does not set foo=
Linda Walsh writes: > I have a feeling this would be hard to fix, since how can printf > tell the difference between > printf -v foo "" > and > printf -v foo > ?? The latter has $# -eq 2. Andreas. -- Andreas Schwab, SUSE Labs, sch...@suse.de GPG Key fingerprint = 0196 BAD8 1CE9 1970 F4BE 1748 E4D4 88E3 0EEA B9D7 "And now for something completely different."
Aw: Re: `printf -v foo ""` does not set foo=
Thats one of the reasons I suggested the following syntax
printf -v test "%s" ""
It doesn't have this problem it also saves other problems as well.
of if you want to expand back slashes etc.
printf -v test "%b" ""
Gesendet: Montag, 17. Juni 2013 um 08:33 Uhr
Von: "Linda Walsh"
An: bug-bash@gnu.org
Betreff: Re: `printf -v foo ""` does not set foo=
Mike Frysinger wrote:
> simple test code:
> unset foo
> printf -v foo ""
> echo ${foo+set}
>
> that does not display "set". seems to have been this way since the
feature
> was added in bash-3.1.
> -mike
Indeed:
> set -u
> unset foo
> printf -v foo ""
> echo $foo
bash: foo: unbound variable
> foo=""
> echo $foo
I have a feeling this would be hard to fix, since how can printf
tell the difference between
printf -v foo ""
and
printf -v foo
??
(with nothing after it?) it seems the semantic parser would have
already
removed the quotes by the time the args are passed to printf, even
this:
> set -u
> printf -v foo "$(echo "$'\000'")"
> echo $foo
still leaves foo gutless: without content (even if were null)
Re: currently doable? Indirect notation used w/a hash
On Sat, Jun 15, 2013 at 12:36:22PM -0700, Linda Walsh wrote:
> John Kearney wrote:
> >There is also a backdoor approach that I don't really advise.
> >val="${ArrayName}[Index]"
> >echo "${!val}"
> -
> Don't advise? Any particular reason? or stylistic?
I'd shared this advice ("don't use it"), because I cannot for the
life of me tell whether this is a bug or a feature. As near as I
can tell, it is an unforeseen consequence of the parser implementation,
not documented anywhere. As such, I would not rely on it to continue
working in future Bash releases.
P.S. you meant printf -v, not -V.
Aw: Re: currently doable? Indirect notation used w/a hash
Like I said its a back door aproach, it circumvents the parser. which
doesn't allow this syntax
${${Name}[1]}
I didn't actually find this myself it was reproted on this list a long
time ago. I do remember Chet saying he wouldn't break it. But other
than that I can't remember the discussion all that well. As always with
this topic it was a pretty lively debate.
Yhea its a constant fight getting my email clients to stop
capitialising various things in code.
Gesendet: Montag, 17. Juni 2013 um 13:57 Uhr
Von: "Greg Wooledge"
An: "Linda Walsh"
Cc: "John Kearney" , bug-bash
Betreff: Re: currently doable? Indirect notation used w/a hash
On Sat, Jun 15, 2013 at 12:36:22PM -0700, Linda Walsh wrote:
> John Kearney wrote:
> >There is also a backdoor approach that I don't really advise.
> >val="${ArrayName}[Index]"
> >echo "${!val}"
> -
> Don't advise? Any particular reason? or stylistic?
I'd shared this advice ("don't use it"), because I cannot for the
life of me tell whether this is a bug or a feature. As near as I
can tell, it is an unforeseen consequence of the parser implementation,
not documented anywhere. As such, I would not rely on it to continue
working in future Bash releases.
P.S. you meant printf -v, not -V.
