On 23/03/2018 14:14, Herbert Xu wrote:
On Fri, Mar 23, 2018 at 01:52:10PM +0100, Harald van Dijk wrote:


It doesn't, and I didn't say it did. POSIX doesn't care how it's
implemented, POSIX cares about the results produced. When another approach
produces the same results, both approaches are equally correct.

Right, but restoring the old behaviour for white spaces only is
silly.  It's silly
to have this work:

        set -- A1 B2 C3
        echo ${@%2 C3}

but not this:

        set -- A1 B2 C3
        IFS=:
        echo ${@%2:C3}

Agreed. Both are doable in a POSIX-compatible way, but the second would require a different approach: either scanning the strings to suppress the addition of the separator character in those cases where its presence would cause splitting to produce an additional field, or big changes in how regions are recorded and split. Either way, that's a large cost for a feature that's likely seen very little real-world use.

Cheers,
Harald van Dijk
--
To unsubscribe from this list: send the line "unsubscribe dash" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Reply via email to