Xref: utzoo comp.unix.wizards:23685 comp.unix.questions:25058
Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!sun-barr!lll-winken!uunet!auspex!guy
From: guy@auspex.auspex.com (Guy Harris)
Newsgroups: comp.unix.wizards,comp.unix.questions
Subject: Re: get terminal speed from shell script
Message-ID: <3977@auspex.auspex.com>
Date: 28 Aug 90 21:01:12 GMT
References: <90Aug12.135618edt.18763@me.utoronto.ca> <1990Aug13.005849.23223@jarvis.csri.toronto.edu> <849@hls0.hls.oz>
Followup-To: comp.unix.wizards
Organization: Auspex Systems, Santa Clara
Lines: 30

>By the way, the SUN manuals point out quite clearly that
>
>	"the settings are reported on the standard error."

Yes, but they also point out equally clearly that

  Special Requests
     speed       The terminal speed alone is printed on the stan-
                 dard output.

     size        The terminal (window) sizes are printed  on  the
                 standard output, first rows and then columns.

                 size and speed always report on the settings  of
                 /dev/tty,  and always report the settings to the
                 standard output.

so "stty speed" and "stty size", in *both* environments, report the
speed and size to the standard output.

The problem is that the BSD-environment "stty" normally fetches a bunch
of mode settings from the standard output when it starts up, and if that
fails, it quite reasonably reports this fact. 

However, the failure is of no consequence if you're just doing "stty
speed" or "stty size", since the only data it uses comes from its
"ioctl" on "/dev/tty", so it shouldn't worry about the "ioctl"s on the
standard output (or shouldn't even do them in the first place!).

I think this bug is fixed in 4.1.