Re: Libtool-patches Digest, Vol 25, Issue 12

[Ralf Wildenhues]

* Akim Demaille wrote on Thu, Dec 09, 2004 at 09:42:47AM CET:
> >>> "Ralf" == Ralf Wildenhues <address@hidden> writes:
> 
>  >> Well, that reminds me when portability myths existed in the Autoconf
>  >> world.  Most were untrue, including those on functions.
> 
>  > SVR 3 Bourne shell:
>  > $ echo $( ls )
>  > Syntax error: `(' not expected.
> 
> Thanks!  How old is this system?

It's new: AIX 5.

I cheated a little: this is not the system shell, merely a Bourne shell
provided as /usr/bin/bsh.  /usr/bin/sh is a POSIX.2 compliant 1988 Korn
Shell (/bin is a symlink to /usr/bin).

But from Sven's list I mentioned, one can guess there might be systems
where this poses a real problem, so I'd recommend to stay on the safe
side for now.  Maybe one could put a spy on this into libtool?

>  > Other than that, a problem ("feature") that comes up in case of
>  > converting `..` style to $(..) is the difference w.r.t backslashes:
>  > $ echo $(echo '\$') `echo '\$'`
>  > \$ $
> 
> That's definitely a feature!

Yes.

But it is the same thing with my patch implementing basename and dirname
in POSIX shell:  I could make these implementations conform to XSI specs
of the respective tools, but then I'd have to audit every use within
ltconfig for possible problems (that might eventually be necessary
anyway, but one only has finite spare time).

Back to my original patch: OK to apply to HEAD?

Regards,
Ralf