All of lore.kernel.org
 help / color / mirror / Atom feed
From: Eric Blake <eblake@redhat.com>
To: Herbert Xu <herbert@gondor.hengli.com.au>
Cc: "Taylan Ulrich B." <taylanbayirli@gmail.com>,
	dash <dash@vger.kernel.org>,
	zsh-workers@zsh.org
Subject: Re: 'continue' does not work in files sourced with dotcmd
Date: Thu, 07 Jul 2011 09:18:12 -0600	[thread overview]
Message-ID: <4E15CE34.40403@redhat.com> (raw)
In-Reply-To: <20110707144635.GA21272@gondor.apana.org.au>

[-- Attachment #1: Type: text/plain, Size: 2328 bytes --]

[adding zsh list]

On 07/07/2011 08:46 AM, Herbert Xu wrote:
> On Thu, Jul 07, 2011 at 07:26:50AM -0600, Eric Blake wrote:
>>
>> zsh rejected the continue statement (no containing loop within the
>> current scoping of the dot script), and as a result also aborted the
>> overall dot script - but in doing so, it properly emitted an error message.
> 
> dash does not warn in this case because it never prints a warning
> on continue anywhere.
> 
> This behaviour (not warning) is consistent with that of ksh.

Interesting!  And bash indeed behaves differently: in POSIX mode it is
silent, while in bash mode it issues a warning but does not affect exit
status - in line with the fact that in POSIX mode, bash obeys the rule
that if an application is diagnosing something to stderr, then the exit
status should be non-zero (and warnings don't affect exit status, hence
warnings should not be issued in POSIX mode).

Zsh is treating it as a syntax error, which is different from all the
other shells:

$ bash -c 'continue; echo $?'
bash: line 0: continue: only meaningful in a `for', `while', or `until' loop
0
$ bash --posix -c 'continue; echo $?'
0
$ dash -c 'continue; echo $?'
0
$ ksh -c 'continue; echo $?'
0
$ zsh -c 'emulate sh; continue; echo $?'
zsh:continue:1: not in while, until, select, or repeat loop
$ echo $?
1

POSIX states "If n is greater than the number of enclosing loops, the
outermost enclosing loop shall be exited."  But this is admittedly
silent on the behavior when there is no outermost enclosing loop, so I'm
not sure whether zsh treating this as a syntax error is appropriate.

Meanwhile, I still have to wonder about dash behavior - if dash is _not_
treating continue as a syntax error, then why is it aborting the dot
script?  Compare:

$ dash -c 'continue
echo $?'
0
$ dash -c 'printf "continue\\necho 0.\$?\\n" > f; for i in 1; do . ./f;
echo 1.$?; done; echo 2.$?'
1.0
2.0

In isolation, execution after an un-nested continue resumed with the
next statement, but in the dot script, execution after the "un-nested"
continue (at least, from the dot script perspective) aborted the dot
script, but without affecting exit status.

-- 
Eric Blake   eblake@redhat.com    +1-801-349-2682
Libvirt virtualization library http://libvirt.org


[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 619 bytes --]

  reply	other threads:[~2011-07-07 15:18 UTC|newest]

Thread overview: 9+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-02-26 18:23 'continue' does not work in files sourced with dotcmd Taylan Ulrich B.
2011-07-07  3:37 ` Herbert Xu
2011-07-07 13:26   ` Eric Blake
2011-07-07 14:46     ` Herbert Xu
2011-07-07 15:18       ` Eric Blake [this message]
2011-07-08  8:09         ` Herbert Xu
2011-07-09 13:07           ` Jilles Tjoelker
2011-07-09 14:07             ` Herbert Xu
2011-07-10 19:03               ` Jilles Tjoelker

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=4E15CE34.40403@redhat.com \
    --to=eblake@redhat.com \
    --cc=dash@vger.kernel.org \
    --cc=herbert@gondor.hengli.com.au \
    --cc=taylanbayirli@gmail.com \
    --cc=zsh-workers@zsh.org \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.