Robert Elz <k...@munnari.oz.au> wrote:
> Date: Tue, 14 Aug 2018 12:30:22 +0200
> From: Joerg Schilling <joerg.schill...@fokus.fraunhofer.de>
> Message-ID:
> <5b72af3e.ws0hf/j4O/wuiekr%joerg.schill...@fokus.fraunhofer.de>
>
> | break is a command that has an exit code that is the previous saved exit
> code
>
> No, it isn't, its exit status is 0, it says so clearly, page 2387 line 76359
Sorry, the exist status is cleared before a builtin command runs. I was
confusing this with ";;".
> | and that has the side effect of leaving the loop.
>
> Yes. The big question is when that happens, does it
> happen immediately, or does it happen after ...
>
> | The ! statement negates the current saved exit code and
> | makes it the new saved exit code.
>
> that happens, and what in the standard tells us, one way
> or the other (or even makes it unspecified, which I suspect
> at this point is all we could do.)
>
> From your earlier message ...
>
> | Interesting is that dash returns 1 while mksh and posh return 0.
>
> I am not sure why, as best I can tell, dash and mksh share no
> heritage whatever, mksh derives (I believe) from pdksh (way
> back) which was a ksh88 clone (of a kind). posh I am know
> nothing about.
posh is derved from ash and it is a fun as it requires glibc specific bugs
in order to work.
> dash comes from ash which was a bourne shell clone (of a kind).
>
> | So it seems that older versions of ash return 0 while dash has
> | been reworked at this point.
>
> No, just the opposite, dash (and the NetBSD sh) are "old ash" and
> return 1, the FreeBSD sh (all of these ash descendants) has been
> reworked to return 0.
So you believe that posh is from a newer ash? This seems to be unlikely from
looking into the code.
Jörg
--
EMail:jo...@schily.net (home) Jörg Schilling D-13353 Berlin
joerg.schill...@fokus.fraunhofer.de (work) Blog: http://schily.blogspot.com/
URL: http://cdrecord.org/private/ http://sf.net/projects/schilytools/files/'
