Re: [Qemu-devel] [PATCH v2] tests: Use "command -v" instead of which(1) in shell scripts

[Eric Blake <eblake@redhat.com>]

[-- Attachment #1: Type: text/plain, Size: 1578 bytes --]

On 11/19/2014 12:07 AM, Fam Zheng wrote:
> When which(1) is not installed, we would complain "perl not found"
> because it's the first set_prog_path check. The error message is
> wrong.
> 
> Fix it by using "command -v", a native way to query the existence of a
> command.
> 
> Suggested-by: Eric Blake <eblake@redhat.com>
> Signed-off-by: Fam Zheng <famz@redhat.com>

> +++ b/tests/qemu-iotests/common.config
> @@ -47,7 +47,7 @@ export PWD=`pwd`
>  # $1 = prog to look for, $2* = default pathnames if not found in $PATH
>  set_prog_path()
>  {
> -    p=`which $1 2> /dev/null`
> +    p=`command -v $1 2> /dev/null`

Reviewed-by: Eric Blake <eblake@redhat.com>

>      if [ -n "$p" -a -x "$p" ]; then

Unrelated: this line "works" because this is a /bin/bash script, but it
is non-portable.  Use of -a and -o inside [] is a mistake waiting to
happen.  For example, is [ ! "$a" -a "$b" ] supposed to be true or false
for all values of $a and $b?  Naively, this says return true if '! "$a"'
(a is empty) and '"$b"' (b is non-empty); but if $a is '(' and $b is ')'
it could also be parsed as returning the negation of whether the
parenthesized string "-a" is non-empty.

Use of -a and -o in [[]] is a bit better, but I still HIGHLY recommend
that constructs like this be rewritten as [ -n "$p" ] && [ -x "$p" ] for
avoidance of confusion and prevention of copy-pasting the test to
non-bash shells.  But that would be a separate patch.

-- 
Eric Blake   eblake redhat com    +1-919-301-3266
Libvirt virtualization library http://libvirt.org


[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 539 bytes --]