qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Eric Blake <eblake@redhat.com>
To: Kamil Rytarowski <n54@gmx.com>,
	Peter Maydell <peter.maydell@linaro.org>,
	qemu-devel@nongnu.org
Cc: patches@linaro.org
Subject: Re: [Qemu-devel] [PATCH] tests: Handle $RANDOM not being supported by the shell
Date: Fri, 14 Jul 2017 07:02:10 -0500	[thread overview]
Message-ID: <64d381ca-ce9a-a19b-0e6f-8ad287d27361@redhat.com> (raw)
In-Reply-To: <910a8148-915a-ab66-1c6c-f18145c13e37@gmx.com>

[-- Attachment #1: Type: text/plain, Size: 1209 bytes --]

On 07/14/2017 06:27 AM, Kamil Rytarowski wrote:
> On 14.07.2017 12:45, Peter Maydell wrote:
>> In various places in our test makefiles and scripts we use the
>> shell $RANDOM to create a random number. This is a bash
>> specific extension, and doesn't work on other shells.
> 
> This is supported on other shells like ksh (Korn Shell), but still as an
> extension.
> 
>> With dash the shell doesn't complain, it just effectively
>> always evaluates $RANDOM to 0:
>>   echo $((RANDOM + 32768))     => 32768
>>
>> However, on NetBSD the shell will complain:
>>   "-sh: arith: syntax error: "RANDOM + 32768"
>>
> 
> I will make sure whether this behavior is correct in our sh(1).

The question is what your shell does for:

unset foo
echo $(( foo % 255 ))

if it reliably prints 0, then it would do the same when RANDOM is an
undefined (and not a magic) variable.  Presumably, where sh is going
wrong is that it is treating an undefined variable not as 0, but as a
syntax error, and you'd get that behavior regardless of whether RANDOM
is in the mix.

-- 
Eric Blake, Principal Software Engineer
Red Hat, Inc.           +1-919-301-3266
Virtualization:  qemu.org | libvirt.org


[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 604 bytes --]

  parent reply	other threads:[~2017-07-14 12:02 UTC|newest]

Thread overview: 9+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-07-14 10:45 [Qemu-devel] [PATCH] tests: Handle $RANDOM not being supported by the shell Peter Maydell
2017-07-14 11:27 ` Kamil Rytarowski
2017-07-14 11:50   ` Kamil Rytarowski
2017-07-14 12:02   ` Eric Blake [this message]
2017-07-14 11:41 ` Fam Zheng
2017-07-14 12:03 ` Eric Blake
2017-07-17  9:43 ` Stefan Hajnoczi
2017-07-20 11:26 ` Markus Armbruster
2017-07-20 15:34 ` Peter Maydell

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=64d381ca-ce9a-a19b-0e6f-8ad287d27361@redhat.com \
    --to=eblake@redhat.com \
    --cc=n54@gmx.com \
    --cc=patches@linaro.org \
    --cc=peter.maydell@linaro.org \
    --cc=qemu-devel@nongnu.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 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).