Re: PIPESTATUS differs from $? for compound command

[microsuxxor]

a COMMAND is a WORD
says somewhere .. somewhere ..

On Thu, Dec 12, 2024, 5:56 PM Mike Jonkmans <bashbug@jonkmans.nl> wrote:

> On Thu, Dec 12, 2024 at 03:56:52AM -0500, Lawrence Velázquez wrote:
> > On Thu, Dec 12, 2024, at 2:01 AM, Mike Jonkmans wrote:
> > > Yes, the posix grammar doesn't allow it.
> > > But it is not clear, from the bash documentation,
> > > that a pipeline is not a command.
> > Sure, the documentation doesn't explicitly say that.
>
> AFAIAC the documentation is what this is about.
> Posix or yacc grammar terms do not correspond exactly to the notions
> in the documentation of bash.
> I.e. 'command' is used but not defined in the doc's SHELL GRAMMAR.
>
> > But it also doesn't say that pipelines are composed of pipelines,
> > the way it says that lists are composed of pipelines.
> > So why would you make that assumption?
>
> Given the doc's definition of a pipeline:
>         [time [-p]] [ ! ] command [ [|⎪|&] command2 ... ]
>
> and note that nowhere else is a '!' found in the doc's grammar.
>
> So the only way - given the doc - to parse '! ! false'
> is by having a pipeline as command.
>
> > > Moreover, bash accepts e.g.:
> > >     ! ! false
> > > How can you parse that, without a pipeline being a command?
> > The yacc grammar uses two separate non-terminals to realize the
> > colloquial notion of a "pipeline":
>
> Yes, I am familiar with yacc. But it is about the documentation's grammar.
>
> > > And, as said, ksh allows the construct (! true | ! true).
> > > It is not impossible, it just is not posix.
> > Of course it's not impossible, it's just goofy.  A pipeline like:
> >       ! true | ! true | ! true
> > looks like it's composed of several segments that are each being
> > individually "negated", but it's logically equivalent [*] to:
> >       ! ! ! true | true | true
> > so why allow the confusion?  (POSIX doesn't permit the latter either,
> > which is consistent at least.)
> > [*]: Yes, I know that in bash ! ! just cancels out, so it probably
> >      wouldn't be *exactly* the same.
>
> PIPESTATUS and pipefail would make it different.
> Useful? Maybe. But one can easily workaround it.
> Note that I am not advocating a change in code in this thread.
>
> --
> Regards, Mike Jonkmans
>
>