DASH Shell discussions
 help / color / mirror / Atom feed
From: Cristian Ionescu-Idbohrn <cristian.ionescu-idbohrn@axis.com>
To: Herbert Xu <herbert@gondor.apana.org.au>
Cc: Jesper Bengtsson <jesper.bengtsson@axis.com>, dash@vger.kernel.org
Subject: Re: Parameter expansion fails when assigning a local variable
Date: Sun, 28 Jun 2009 23:39:16 +0200 (CEST)	[thread overview]
Message-ID: <0906282220250.1491@somehost> (raw)
In-Reply-To: <20090622085729.GA22431@gondor.apana.org.au>

On Mon, 22 Jun 2009, Herbert Xu wrote:

> Jesper Bengtsson <jesper.bengtsson@axis.com> wrote:
> >
> > I've found a problem with expanding $@.
> > If I declare a local variable and assigns the expanded positional
> > parameters:
> >        local v="$@"
> > It fails with something like:
> >        local: 20: \/: bad variable name
>
> This is expected behaviour.

Hmm...  Why is it so?  Because it's the only "natural" expectation?

> local is just like any other utility

Utility?  Not built-in?  Documentation refers to it as command.  Same
thing?

> in how it handled parameters.  So "$@" will have been split before
> it reaches local.  To get what you want, you should use "$*".

But wait a second.  Shouldn't quotes '"' around $@ protect the whole $@?
It does so in different other context.  Why not here?

> > I've tried this on dash 0.5.5.1 and 0.5.3 and both fail in the same way.
> > I've also tried on bash and ash (Busybox) and both handles declaration
> > and assignment on one line.
>
> Bash and older versions of dash (like the one in Busybox) has a hack

Hack.  Is that a hack?  It is not expected behaviour.  It's a hack?
Even newer/recent/latest versions of (busybox) ash have that "hack".
And dash claims to be the only POSIX "true" implementation?

> that disables field splitting for local and certain other
> commands.  This is not guaranteed by POSIX.

Sure.  That's what it looks like.  POSIX doesn't even mention it, does it?
Still.  Quotes around $@ should protect it, shouldn't it?

	local v="$@"

should expand to:

	v="'arg1' 'arg2' ..."

shouldn't it?

There's som parallel unnatural behaviour I noticed:

	local var=$(some forked thing here)

$(...)
^^   ^ doesn't seem to protect the resulting string. There seems to be a
difference between:

	local h=$(grep --help)

and:

	local h="$(grep --help)"

Can anything motivate that?  Is that POSIX too?


Cheers,

-- 
Cristian

  reply	other threads:[~2009-06-28 22:02 UTC|newest]

Thread overview: 10+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2009-06-22  8:13 Parameter expansion fails when assigning a local variable Jesper Bengtsson
2009-06-22  8:57 ` Herbert Xu
2009-06-28 21:39   ` Cristian Ionescu-Idbohrn [this message]
2009-06-29  0:47     ` Herbert Xu
2009-06-29 17:10       ` Cristian Ionescu-Idbohrn
2009-06-30  1:38         ` Herbert Xu
2009-06-30 16:17           ` Cristian Ionescu-Idbohrn
2009-07-01  7:42             ` Oleg Verych
2009-07-02 16:53               ` Cristian Ionescu-Idbohrn
2009-07-03 21:14                 ` Oleg Verych

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=0906282220250.1491@somehost \
    --to=cristian.ionescu-idbohrn@axis.com \
    --cc=dash@vger.kernel.org \
    --cc=herbert@gondor.apana.org.au \
    --cc=jesper.bengtsson@axis.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox