On 9/17/16 7:38 AM, Joerg Schilling wrote:
> Chet Ramey <chet.ra...@case.edu> wrote:
>>> But please explain (character by character) why you create an output that 
>>> is in 
>>> conflict with the reference implementation (a patched ksh88).
>> `Character by character'? Really?
>> (And obviously the so-called `reference implementation' doesn't reflect
>> the intent of the standard either, as revealed by the discussion.)
> This claim does not match the general POSIX rules.

So it doesn't really matter what the standard says or what its intent is
determined to be? It only matters what the so-called `reference
implementation' does?

>> Anyway, bash, in its default mode, treats embedded single- and double-
>> quoted strings as special, and skips over them according to the "skipping
>> over enclosed quoted strings" portion of the text in 2.6.2 that's been
>> there since at least 1992.
> /usr/xpg4/bin/sh
> $ echo "${foo-'a"b'"}"
> This is a certified POSIX shell....

OK.  If this specific behavior is part of the certification test, then
either the test needs to be changed or the standard does.  If it's not,
then it really doesn't matter what a `certified' shell does.

I'm done with this discussion unless we want to reopen interpretation 221.
There's no path forward from here that's not a huge waste of time

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

Reply via email to