Re: system/2208: Incorrect 'join' behaviour?

[Tobias Weingartner <weingart@natasha.tepid.org>]

On Tuesday, November 27, Andrew Pinski wrote:
>  
>  Sun's SunOS 4.1.4 (pre-Solaris) returns the something different:
>  ./joint
>  b 1
>  x 2 apple
>  y 2 orange
>  %/>uname -a
>  SunOS physung.p 4.1.4 9 sun4m

This one is wrong, the posix spec clearly indicates how the
lines that do join are supposed to be output.  Not surprising
really, considering the age of this platform/os.


>  Also Mac OS X returns the same as OpenBSD:
>  [zhivago:~/testjoin] pinskia% ./joint
>  1 a b
>  x 2 apple
>  y 2 orange

Chances are this is a "bug" or difference in the 4.4BSD
version of join.  Unfortunately, the posix spec I consulted
did not say much about the lines that did not join.  By
extension, and for consistancy, I'd say that this is a bug
and should be fixed.  It would be nice to have a standard
to point to though...  anybody have something better than
what I saw?


>  HPUX returns the same as most other systems.

Another one that does it (IMHO) correctly...

--Toby.