dash.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Tim Ruehsen <tim.ruehsen@gmx.de>
To: Harald van Dijk <harald@gigawatt.nl>
Cc: dash@vger.kernel.org
Subject: Re: ':' noop results in ':: not found'
Date: Fri, 28 Oct 2016 15:08:15 +0200	[thread overview]
Message-ID: <1749484.J6nWIO8AN1@blitz-lx> (raw)
In-Reply-To: <c5b4a25d-bdf6-eb44-2401-c9af897f9b23@gigawatt.nl>

[-- Attachment #1: Type: text/plain, Size: 2127 bytes --]

On Friday, October 28, 2016 1:22:20 PM CEST Harald van Dijk wrote:
> On 28/10/16 12:54, Tim Ruehsen wrote:
> > Hi,
> > 
> > maybe you can enlighten me :-)
> > 
> > I try to use dash (0.5.8-2.3) on Debian unstable for executing ./configure
> > scripts. Since Debian builds with --disable-lineno, all ./configure
> > scripts
> > silently fall back to bash (I created bug report #842242 to get this
> > fixed).
> As you found there, this is intentional, specifically to force configure
> scripts to continue to be run with bash, which is what happened with
> older dash versions. I asked at the time why they don't just call bash
> explicitly (bash configure instead of ./configure) if that's the goal,
> but didn't get a response.
> 
> > Next I rebuilt the package from the (Debian) sources without
> > --disable-lineno and installed it. Now I see error messages from dash
> > when it comes to the do nothing operator :.
> > 
> > $ dash -c 'test -n "a" && :'
> > dash: 1: :: not found
> > $ echo $?
> > 127
> > 
> > $ bash -c 'test -n "a" && :'
> > $ echo $?
> > 0
> > 
> > Since this : construct is wildly used and I know that dash on other
> > systems
> > work with it, I wonder what is wrong here.
> > 
> > Is there a known bug, maybe fixed on recent dash versions ?
> > Or has this some simple reason my stupidity doesn't see ?
> 
> This is likely to happen if you have the LC_ALL environment variable set
> when building dash, and I can reproduce your results with 0.5.8 by
> setting LC_ALL to en_GB.UTF-8. It was fixed in 0.5.9. (Specifically,
> where a fixed sort order was needed, 0.5.8 only forced LC_COLLATE=C, but
> that still allowed LC_ALL to override it.)

Harald !!!
That's it !
I wish you could see me grinning from ear to ear !

> dash won't be the only program that has problems with this; if you're
> building software yourself, setting LANG should generally be okay, but
> LC_ALL is best avoided. Unless you're doing it specifically to find and
> report/fix bugs, anyway.

In fact, I use LC_ALL to find bugs. Especially turkish and greek have some 
nifty casing pitfalls ;-)

Thank you so much, Harald !

Tim

[-- Attachment #2: This is a digitally signed message part. --]
[-- Type: application/pgp-signature, Size: 801 bytes --]

      reply	other threads:[~2016-10-28 13:08 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-10-28 10:54 ':' noop results in ':: not found' Tim Ruehsen
2016-10-28 11:22 ` Harald van Dijk
2016-10-28 13:08   ` Tim Ruehsen [this message]

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=1749484.J6nWIO8AN1@blitz-lx \
    --to=tim.ruehsen@gmx.de \
    --cc=dash@vger.kernel.org \
    --cc=harald@gigawatt.nl \
    /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;
as well as URLs for NNTP newsgroup(s).