Re: different exit codes in $? and ${PIPESTATUS[@]}

[Chet Ramey]

On 10/12/12 5:36 PM, Wladimir Sidorenko wrote:
> Ok, thanks. I'll notice this for the future. It's of course arguable,
> whether PIPESTATUS stores a true value, since the '!' inversion
> keyword gets ignored in this case. But if it was intended, than it's
> ok.

The case where PIPESTATUS stores only a single exit status is the
degenerate case.  Pipelines have multiple elements.

What do you think should happen in the following case?

! exit 1 | exit 2 | exit 3

$? reflects only the status of the pipeline, which in this case is the
(inverted) status of the final command.  Why should $PIPESTATUS not
reflect the actual exit statuses?

Chet
-- 
``The lyf so short, the craft so long to lerne.'' - Chaucer
                 ``Ars longa, vita brevis'' - Hippocrates
Chet Ramey, ITS, CWRU    chet@case.edu    http://cnswww.cns.cwru.edu/~chet/