Re: Proposed new POSIX sh policy
* Mike Hommey (firstname.lastname@example.org) [061106 22:00]:
> On Mon, Nov 06, 2006 at 09:01:27AM -0800, Russ Allbery <email@example.com> wrote:
> > Mike Hommey <firstname.lastname@example.org> writes:
> > > Russ Allbery <email@example.com> wrote:
> > >> + <item>the <tt>-a</tt> and <tt>-o</tt> <tt>test</tt> operators
> > >> + must be supported</item>
> > > Why is that needed ?
> > It's in widespread use in both Debian scripts and in upstream scripts.
> > When we tried to warn about this behavior in lintian, it turned up
> > hundreds of packages and we got a lot of objections to the check on the
> > grounds that dash supports this construct and the only shell that doesn't
> > is posh. It seemed like the general consensus was that requiring that all
> > those scripts be modified to require bash was more trouble than it was
> > worth.
> Well we got bug reports for that on firefox, IIRC, and we changed it,
> that was not a real problem to replace [ some != test -a other = test ]
> with [ some != test ] && [ other = test ]...
I agree -a/-o should be replaced, but I don't think we really consider
it an RC bug. So, a shell *must* support the operators, but I don't
think that we should encourage people to use them.