No more discussion on this topic, for now.

Let us make progress one step at a time, and discuss compliance one command
at a time.

Ram
On May 18, 2012 11:19 PM, "Lionel Cons" <[email protected]>
wrote:

> On 17 May 2012 15:10, Joerg Schilling
> <[email protected]> wrote:
> > Irek Szczesniak <[email protected]> wrote:
> >
> >> Joerg, please stop promoting your Joerg Bourne shell here. Thank you.
> >> Also please stop changing wikipedia articles (e.g. "comparison of
> >> command shells") for promotional purposes, too. Just saying that here
> >> before you quote that article as "proof" for Bourne shell vs Korn
> >> shell capabilities.
> >
> > Some people live in a world that has been true in 1979 and these people
> don't
> > like information to be updated to even to the state of 1995.
> >
> > Some people seem to believe that other people will believe them when
> they use
> > the term "promotional" for attempts to remove false claims.
> >
> > I am not interested in these people, as I live in the present and as I
> don't
> > like false claims.
> >
> > Given the fact that I added the command line history editor I wrote
> 1982..1984
> > for my "bsh" to the Bourne Shell in December 2006, it would not even be
> > promotional if I did add a related hint to the wiki article as we now
> have 2012.
>
> *sigh*
> You are diverting to other topics. See below.
> >
> >
> >> > It seems that you did not understand the ideas behind the POSIX
> standard.
> >> >
> >> > POSIX with good reason does not standardize on pathnames (except for
> >> > /dev/null). This allows a POSIX compliant system to put a POSIX
> compliant shell
> >> > wherever it likes in order to keep a traditional Bourne Shell in
> /bin/sh.
> >> > I however even plan to have ksh in /bin and the Bourne Shell in
> /sbin/sh
> >> > even though the current version of ksh93 is not fully POSIX compliant.
> >>
> >> Joerg, I ran the SUS test suite last week against ast-open.2012-05-04.
> >> ksh93, part of this beta release, passed the tests. Could you please
> >> send the bug to *this* list which makes ksh93 not fully POSIX
> >> compliant? Thank you.
> >
> > Passing the SUS tests does not verify POSIX compliance. Look e.g. in
> special at
> > Mac OS X that passed the tests even though bash is the only half-way
> POSIX
> > shell that was definitely not POSIX compliant when the tests have been
> run.
> > Bash did not implement the -e option correctly (important for correct
> behavior
> > of make(1)) and bash most likely still does job-control on scripts that
> causes
> > nested make calls to continue in the background if you hit ^C.
>
> Joerg, would you please come with with detailed proof that ksh (ksh93)
> violates the POSIX standards. Please don't divert to other topics,
> just present us the technical facts. bash, your Bourne shell, Mac OS X
> or anything else does not matter in this discussion.
>
> Lionel
> _______________________________________________
> belenix-dev mailing list
> [email protected]
> http://mail.opensolaris.org/mailman/listinfo/belenix-dev
>
_______________________________________________
belenix-discuss mailing list
http://mail.opensolaris.org/mailman/listinfo/belenix-discuss
http://groups.google.com/group/belenix-discuss

Reply via email to