lmi
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [lmi] [lmi-commits] odd/multiarch c912adf 5/5: Avoid 'whence' in /bi


From: Greg Chicares
Subject: Re: [lmi] [lmi-commits] odd/multiarch c912adf 5/5: Avoid 'whence' in /bin/sh scripts
Date: Mon, 15 Apr 2019 17:46:45 +0000
User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:60.0) Gecko/20100101 Thunderbird/60.5.1

On 2019-04-15 17:12, Vadim Zeitlin wrote:
> On Fri, 12 Apr 2019 08:32:59 -0400 (EDT) Greg Chicares <address@hidden> wrote:
> 
> GC> branch: odd/multiarch
> GC> commit c912adf0ac998a99961826364234df860926cc57
[...]
> GC>     Avoid 'whence' in /bin/sh scripts
[...]
> GC> -whence wx-config
> GC> +command -v wx-config
> 
>  Just FYI, I know that there is a bug in dash with using "command -v": it
> can return a wrong, non-executable, command if there is one in any
> directories in the PATH before the actual executable one, see the still
> open bug https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=874264

Thanks for pointing that out--I would never have known.

>  It looks unlikely that it would affect lmi here, and I'm not sure if the
> alternative ("type") works better, but I wanted to mention it just in case
> you run into it to prevent you from potentially wasting a lot of time
> trying to figure out was wrong -- as I did, when this happened to me (which
> is, of course, how I know about this problem in the first place).

Trunk and this branch are close to converging. I had inserted this
'command -v' line only for debugging, and everything's pretty well
debugged by now. To be sure, let's set the architecture by the method
proposed for general use, and check which 'wx-config' we get:

/opt/lmi/src/lmi[0]$LMI_HOST=i686-w64-mingw32 ; . ./set_arch.sh
/opt/lmi/src/lmi[0]$whence wx-config                           
/opt/lmi/i686-w64-mingw32/local/bin/wx-config

/opt/lmi/src/lmi[0]$LMI_HOST=x86_64-w64-mingw32 ; . ./set_arch.sh
/opt/lmi/src/lmi[0]$whence wx-config                             
/opt/lmi/x86_64-w64-mingw32/local/bin/wx-config

Okay, it's well and truly debugged. I still find it hard to read
  i686-w64-mingw32   as meaning 32-bit msw, and
  x86_64-w64-mingw32 as meaning 64-bit msw.
But that's the way it's (mis)named, and it's not going to change:
  https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=622276



reply via email to

[Prev in Thread] Current Thread [Next in Thread]