Console scoansi emulation problem
Mike Brown
The console emulation works differently, in all "mkdev scoansi"
modes. It looks like the application is missing a CR-LF on a
couple of screens, which causes the last character of the line
to be over written by every character of what should be the
next line. From a PC, termlite and TinyTerm are doing a
"autowrap", as is the terminal window from the X screen,
so everything looks fine. Only the text screens fail
( scoansi, scoansi-old, ansi ... ).
It is obvious that the real fault is in the application,
but is there a way to modify the console emulation?
Thanks,
Mike
--
Michael Brown
The Kingsway Group
Paulo Couceiro
Terminal Emulation Control
Choose 4. Use old (pre 5.0.6) in-kernel scoansi emulator
Relink the kernel
That should do the trick
"Mike Brown" <mi...@tkg.ca> wrote in message news:3C390869...@tkg.ca...
Mike Brown
The PRE 5.0.6 emulation does seem to be different, but
does not fix this one issue. The application fails in
all combinations of old/new emulation and old/new terminfo.
It looks like the same machine running 5.0.5 works fine,
so the PRE 5.0.6 emulation is different from 5.0.5 in this
one case
Paulo Couceiro
http://stage.caldera.com/cgi-bin/ssl_reference?112082
http://stage.caldera.com/cgi-bin/ssl_reference?109521
Hope this helps
"Mike Brown" <mi...@tkg.ca> wrote in message news:3C3B9CDD...@tkg.ca...
Mike Brown
Paul, thanks for the references. I had check the Caldera site,
and have gone through the various "mkdev scoansi" settings. All
versions basically work, and scoadmin and the applications are
fine EXCEPT for the two small display problems. It looks like
the old version ( ie 5.0.5 and older ) did a line wrap, but
no matter what mode I put the 5.0.6 into ( following the articles )
it still displays incorrectly at these two points. The XTERM
emulator works fine.
I will submit a bug report to Caldera.
Thanks
Bela Lubkin
> Paul, thanks for the references. I had check the Caldera site,
> and have gone through the various "mkdev scoansi" settings. All
> versions basically work, and scoadmin and the applications are
> fine EXCEPT for the two small display problems. It looks like
> the old version ( ie 5.0.5 and older ) did a line wrap, but
> no matter what mode I put the 5.0.6 into ( following the articles )
> it still displays incorrectly at these two points. The XTERM
> emulator works fine.
Mike --
The OpenServer console has a pair of escape sequences to turn on and off
autowrap. ``ESC [ ? 7 h'' turns autowrap on; ``ESC [ ? 7 l'' turns it
off.
Try issuing one of those before running your app.
If that makes no difference, run your app under a capturer (e.g.
script(TC), or over a serial or telnet connection with the other end set
to capture). Look at the output. Is the _app itself_ issuing the
``7l'' autowrap-off sequence? If so, look at the termcap or terminfo
entry being used by the app. If it has a reference to that sequence,
clone it, take out the ``7l'' reference, and use that entry -- see if
it's any better.
Mind you, the 7h/7l sequences have been around for a long time and as
far as I know their behavior and initial default have not changed; so I
don't see _why_ this would happen...
>Bela<
Mike Brown
Bela,
I revisited the escape sequence issue based on your reply, I had tried
this earlier. A little strange ...
For this test I created three files, with an "hd" output of:
on:
1b 5b 3f 37 68
off:
1b 5b 3f 37 6c
test:
6a 6a 6a 6a 6a 6a 6a 6a 6a 6a 6a 6a 6a 6a 6a
6a 6a 6a 6a 6a 6a 6a 6a 6a 6a 6a 6a 6a 6a 6a
6a 6a 6a 6a 6a 6a 6a 6a 6a 6a 6a 6a 6a 6a 6a
6a 6b 6b 6b 6b 6b 6b 6b 0a
so, I have the autowrap on string, an autowrap off string, and a file
that has 49 'j' and 7 'k' characters.
When I "cat on", autowrap is off, and when I "cat off" autowrap is on, as
tested by "cat test". I checked "man console", and it does say that
" Please note that this is the inverse of what you would expect. "
It would seem that ESC[?7h = reset = NO autowrap ( overwrite ), and
ESC[?7l = set = autowrap.
BUT
When I run the same test from an XTERM session, it works opposite
to above, and "on" ESC[?7h = autowrap, "off" ESC[?7l = NO autowrap.
AND
When I run the same test from a termlite session, it works opposite
to the console also.
Best that I can surmise, default autowrap ( automargin ) has changed
from 5.0.5 to 5.0.6, and by the documentation the 5.0.6 console is
correct, but XTERM, termlite, Windows telnet, century tiny term ....
are all wrong, as is the handling of ESC[?7h and ESC[?7l. I am not
going to get into which method is right, but consistency would be
nice.
I believe I tried a "cat on" to fix the problem with the application,
which did not help, and I will try a "cat off" to see what happens.
Mike Brown
> test:
>
> 6a 6a 6a 6a 6a 6a 6a 6a 6a 6a 6a 6a 6a 6a 6a
> 6a 6a 6a 6a 6a 6a 6a 6a 6a 6a 6a 6a 6a 6a 6a
> 6a 6a 6a 6a 6a 6a 6a 6a 6a 6a 6a 6a 6a 6a 6a
> 6a 6b 6b 6b 6b 6b 6b 6b 0a
>
> so, I have the autowrap on string, an autowrap off string, and a file
> that has 49 'j' and 7 'k' characters.
>
No, its 82 'j' and 7 'k' characters, so it does wrap the line
Bob Rasmussen
that "rmam and smam were reversed on the console". I believe I verified it at
the time (I can't right now).
The rmam and smam refer to terminfo capabilities.
I believe only the console driver is wrong. I believe the following all agree
with each other:
1) the terminfo entries
2) the X terminal behavior
3) the ANSI standard
4) VTxxx terminal behavior
5) probably every terminal emulator there
How to fix it? One solution would be to create a different terminfo entry for
the console, with rmam and smam reversed, and use it.
(It's difficult to program emulations when the target has bugs. I have written
to THE PERSON at SCO several times on this subject, and been met with
deafening silence.)
Regards,
....Bob Rasmussen, President, Rasmussen Software, Inc.
personal e-mail: r...@anzio.com
company e-mail: r...@anzio.com
voice: (US) 503-624-0360 (9:00-6:00 Pacific Time)
fax: (US) 503-624-0760
web: http://www.anzio.com
Mike Brown
I will give that a try, now that I understand they are reversed.
Did this change between 5.0.[0-5] and 5.0.6?
Thanks
Bob Rasmussen
> I will give that a try, now that I understand they are reversed.
>
> Did this change between 5.0.[0-5] and 5.0.6?
I have verified that rmam and smam are reversed in the console, but not in the
GUI term, on SCO 5.0.6. I'm fairly sure the problem showed up first in 5.0.6;
prior to that, no changes had been made to the console for several years.
Mike Brown
Given that this may be termed a "bug", what is the best channel/method
for getting it fixed?
Jean-Pierre Radley
| On Fri, 11 Jan 2002, Mike Brown wrote:
|
| > I will give that a try, now that I understand they are reversed.
| >
| > Did this change between 5.0.[0-5] and 5.0.6?
|
| I have verified that rmam and smam are reversed in the console, but not in the
| GUI term, on SCO 5.0.6. I'm fairly sure the problem showed up first in 5.0.6;
| prior to that, no changes had been made to the console for several years.
I think the Facetwin people told me this inversion appeared on 5.0.6, and
was OK on 5.0.5.
--
JP
Jean-Pierre Radley
| Bob Rasmussen wrote:
|
| > On Fri, 11 Jan 2002, Mike Brown wrote:
| >
| > > I will give that a try, now that I understand they are reversed.
| > >
| > > Did this change between 5.0.[0-5] and 5.0.6?
| >
| > I have verified that rmam and smam are reversed in the console,
| > but not in the GUI term, on SCO 5.0.6. I'm fairly sure the problem
| > showed up first in 5.0.6; prior to that, no changes had been made to
| > the console for several years.
|
| for getting it fixed?
Bob and I are mounting an assault on the the proper parties at SCO...
--
JP
Mike Brown
Thanks, I won't duplicate the issue with SCO.