SY.CHRISTINE@CU20B.COLUMBIA.EDU (Christine M Gianone) (06/03/87)
Info-Kermit Digest Wed, 3 Jun 1987 Volume 6 : Number 12
Today's Topics:
New IBM PC Kermit with Tektronix Emulation
QK-Kermit Answers
Macros for VAX/VMS EDT in MS-Kermit 2.28
Apple II Kermit Answers
Sending BREAK from Toshiba Lap Tops
UNIX V.3 & Kermit
IBM PC Kermit 2.29b Screen Color Bug
Handshake Bug in MS Kermit 2.29
Inconsistent Length of BREAK in MS-Kermit
Patches for HP-150 Kermit Problems
Kermit Has Made It To the Board Level
Does Perkin-Elmer Kermit Work?
----------------------------------------------------------------------
Date: Monday, 18 May 1987 15:37:59
From: BJH6@UK.AC.CAMBRIDGE.PHOENIX (BJH6%CAM.PHX@UK.AC.CAM.ENG-ICF)
Via: SYSKERMIT%vax1.central.lancaster.ac.uk@Cs.Ucl.AC.UK
Subject: New IBM PC Kermit with Tektronix Emulation
Keywords: Tektronix Emulation, MS-Kermit
I have developed an IBM TEK emulation for KERMIT, and am sending you the
.BOO file. The notes below are intended for a .BWR file - I have called the
version MSTEK.BOO.
This is a version of KERMIT 2.29 with a built-in Tektronix emulator. Please
note:
1. This version will work with IBM CGA graphics, or any software that emulates
it. It will not work with Hercules, but should work on an EGA card, but only in
CGA resolution.
2. The TEK emulation is not complete - in particular, the characters are IBM's
own character set and there is no MARGIN 1. All vectors should, however, be
correctly positioned in relation to one another, though the aspect ratio may
not be perfect. In addition, a cross-hair facility is provided. Use arrow keys
to move the cross-hairs, and SHIFT-arrow for faster movement.
3. TEK mode is entered from VT or other mode upon receipt of the sequence ESC
FormFEED (hex 1B 0C); you can exit from TEK mode either when ESC-US (hex 1B 1F)
is sent by the host, or by pressing the Kermit escape key (CTRL-], by default).
4. Please send any comments/problems to:
Brian Holley
Faculty of Economics and Politics
University of Cambridge
BJH6%CAM.PHX@UK.AC.CAM.ENG-ICF
[Ed. - Thanks, Alan, for passing this along, and thanks to Brian for
contributing his work. The .BOO file only, along with this message (no
source) is available in Kermit distribution as MSTIBT.*. Some alternate
versions of Tektronix emulation may also appear for IBM PC Kermit, and there
is some reason to hope that one of these may be incorporated into the next
real release, 2.30, although if new code keeps showing up at the present rate,
the next release may never see the light of day...]
------------------------------
Date: Thu, 21 May 87 16:53 EDT
From: VIC@QUCDN.BITNET
Subject: QK-Kermit Answers
Keywords: QK-Kermit
In response to the item about QK-Kermit in Info-Kermit V6 #11, all the problems
you mentioned are related to the use of the Graphic ToolBox and the use of
overlays. I have a new version of Qk-kermit which eliminates the use of the
Graphic Toolbox and overlays.
I will send this version off to Columbia as soon as I make sure that the CP/M
code will be compatable with the MS-DOS code.
Victor Lee (613)-545-2033
[Ed. - Thanks Victor! We will include it as soon as Columbia receives it.]
------------------------------
Date: Wed 1 Apr 87 12:23:47-EST
From: D. M. Rosenblum <DR01@TE.CC.CMU.EDU>
Subject: Macros for VAX/VMS EDT in MS-Kermit 2.28
Keywords: VAX/VMS EDT, MS-DOS Kermit
I finally wrote up some macros that let me get at VAX/VMS EDT in MS-DOS
Kermit 2.28 for the IBM-PC. (They may be totally inappropriate for other
keyboards.) Like actual Zenith-19 terminals, these emulate the EDT keys by
position, rather than by name. (I realize that 2.29 is doing VT100-family
support, but CMU isn't yet using 2.29, and I presume a lot of folks out
there still use 2.28 or earlier versions.) To use them, go into EDT change
mode, escape back to KERMIT, issue a DO SETEDT command, and connect back to
the VAX; then when you're done, after you exit EDT, escape back to KERMIT,
issue a DO CLEAREDT command, and connect back to the VAX. The documentation
of which keys do what is in the command file (at the end of this message).
By the way, the point that I made some time back about the kind of emulation
that I'd like to see is that I'd like to have MS-Kermit, if it's doing
Zenith-19 emulation, automatically do the user-defined macro SETEDT when it
receives an ESC = sequence from the host, and do the user-defined macro
CLEAREDT when it receives an ESC > from the host. (Of course, the names of
the macros don't have to be SETEDT and CLEAREDT, but the idea is that the
user should be able to define key settings to be invoked and revoked
respectively on receipt of ESC = and ESC >.) I believe that the VT100 uses
those same old VT52 escape sequences to get the alternate and normal keypad
(these are apparently different from the ANSI standards, which, according to
another source, are allegedly ESC [ > 7 h to go into alternate keypad mode
and ESC [ > 7 l to go back to normal keypad mode).
[Ed. - Yes, may people have requested such macro definitions. They have
been put into a file called KER:MSIEDT.INI.]
------------------------------
Date: Thu, 21 May 87 22:54:49 EDT
From: friedman@topaz.rutgers.edu (Gadi )
Subject: Apple II Kermit Answers
Keywords: Apple II Kermit
[Ed. - This is in response to the "Apple ][ Kermit Comments" in V6#11
from Martin J Carter.]
You'll probably get lots of replies, but.. Your problem (1), in which the
name gets mangled, is caused by the lack of a lower case display chip in your
apple. Without one, all lower case characters are displayed as symbols. If
you want to still use this kermit, (the newer one is MUCH better). You should
make sure your file names are upper case.
Gadi
ARPA: friedman@topaz.rutgers.edu
UUCP: {harvard, seismo, ut-sally, sri-iu}!rutgers!friedman
CMS: RUTGERS!SYSOP (CMS is DOWN. Long live CMS)
------------------------------
Date: Mon, 25 May 87 11:53:50 EDT
From: "Roger Fajman" <RAF@NIHCU>
Subject: Sending BREAK from Toshiba Lap Tops
Keywords: Toshiba, MS-DOS Kermit
> I ported MSKERMIT to my Toshiba 3100, but BREAK does not seem to work.
> Does anybody have any experience with this problem? I am using the Toshiba
> internal modem.
The Toshiba internal modem for the Toshiba 1100+ and 3100 does not transmit
break signals. Toshiba says that, yes, it works that way. A company called
Megahertz makes an internal modem for the 1100+ and 3100 called the T1200
that does send break signals. I have tried it for a short time and found no
problems. We just purchased an 1100+ with a T1200 modem, but the modem was
backordered. After it arrives, I will be able to say more. Inability to
send a break signal seems to be a fairly common problem among the cheap 1200
bps modems. I have encountered it several times before.
The address for Megahertz is
Megahertz Corporation
2681 Parley's Way, Bldg. 2-102
Salt Lake City, Utah 84109
Telephone: 800-33-TURBO
801-485-8857
I believe that the list price for the T1200 is $400.
[Ed. - In general, Kermit does not work with internal modems unless the
modem mimics the serial port exactly, or there is explicit code in the
Kermit source which knows about that particular internal modem. MS-DOS
Kermit 2.29B reportedly does work with Hayes and Hayes compatible full and
half-card internal modems, however.]
------------------------------
Date: 26 May 87 19:38:11 GMT
From: gatech!mcdchg!heiby@RUTGERS.EDU (Ron Heiby)
Subject: UNIX V.3 & Kermit
Keywords: C-Kermit
Russell Nelson <bh01@clutx.bitnet> says:
> Have you heard anything about using kermit with Unix V.3? AT&T has a new
> terminal driver (uugetty) for modem lines that does not work well with
> kermit.
>
> I've finally got 'cu' working. kermit dies because uucp owns the port.
It isn't the driver that's new. The program "uugetty" is a replacement for
"/etc/getty" that has knowledge of the same lock files that cu and uucp use.
Use of "uugetty" allows a port to be used as either incoming or outgoing on
demand, without administrator intervention.
One solution to Russell's problem is not to use "uugetty" on the line(s)
he wants to use kermit with. On a system with very few modems this is
pretty limiting. When I brought kermit up on my SVR3 system, I made the
kermit program suid to uucp. This allowed it to manipulate the lock files
and the port just fine. It seems, though, that kermit doesn't really expect
to be running suid, so this "solution" just creates a fairly large security
hole. (I turned off the suid bit when I found out.) As I mentioned in a
note a couple of weeks ago, the make target for "att3bx" is only used for
the HDB lock file protocol. Since uugetty is only available with HDB (as
far as I know), perhaps the same target and #define could be used to insert
code to deal with kermit running suid. My guess is that all that is needed
is a "setuid(getuid());" after the lock is created and port opened
successfully. Probably belongs shortly after the call to ttlock() in ttopen().
I haven't completely tested it, but this seems to work ok on my system. I
also had to comment out the call to access() that checks to see if the lock
directory is writable, as access() uses the real uid of the caller instead
of the effective uid, which is what we want. We then depend on the actual
creat of the lock file's return code to indicate a problem.
These changes might not be a bad idea for all UNIX versions of kermit.
Ron Heiby, heiby@mcdchg.UUCP Moderator: comp.newprod & comp.unix
"Small though it is, the human brain can be quite effective when used properly"
[Ed. - Thanks for the info. We'll try to do something about this in the
next release of Unix Kermit.]
------------------------------
Date: Mon, 18 May 87 15:52:37 edt
From: snorthc@nswc-g.ARPA
Subject: IBM PC Kermit 2.29b Screen Color Bug
Keywords: MS-DOS Kermit
If you set the color with the mskermit.ini to bright white on blue
(set term 1, 37, 44), the color will change during certain full screen
applications. We first discovered this with the 4BSD program more. The
color is fine until you press the space bar for the next segment of text,
then the text changes to dull white or grey on blue. We have an office
automation program called Office Power that causes the same color change.
Further information: We have reproduced these results on IBM ATs, Compaq
286s and Z-248s with various brands of EGA cards and monitors. If you press
the ESC-chr and then c and then another c, your color will be reset.
[Ed. - From JRD: snorth@nswc-g.arpa (no real name) today commented that
telling Kermit to use a Connect mode screen of bright white on blue (Set Term
Color 1 37 44) later resulted in ordinary white on blue as an application ran
on the host machine. This is correct since the host machine is sending a
command to set the screen that way. An escape sequence of ESC [ 0 m meaning
reset all video attributes does this. The difficulty regarding Kermit is that,
first it is running on systems which designate ordinary white to be a poor
light grey (yes, on my AT+ega+Multisync system too), and second the bold
colors still need to be under control of the host for other emphasis.]
------------------------------
Date: Thu, 21 May 87 14:46:42 MEZ
From: C0034008%DBSTU1.BITNET@wiscvm.wisc.edu
Subject: Handshake Bug in MS Kermit 2.29
Keywords: MS-DOS Kermit
I think I have found a bug in MSDOS-Kermit version 2.29 for IBM-PC. Our
configuration is an IBM PC/AT-02 connected via interface to an IBM /370 with
VM/SP Rel.4.
Using CMS-Kermit version 2.1 in the server mode everything was allright
except the GET command failed. But talking to the CMS-Kermit 3.1 server
mostly each command ended in the message 'Invalid server command'. Only the
REMOTE commands worked correctly. An observation with a linemonitor pointed
out that MSDOS-Kermit sends a NUL byte at the beginning of a packet.
Actually at the GET command it was the second packet which had a leading NUL
byte. Version 2.28 has not this bug ( but it is not as comfortable as 2.29).
Because I'm not a great programmer in MASM I added a condition to the outchr
routine in module msxibm, which skips if the character is a NUL.
Unfortunately now it is impossible to set padding char NUL.
Of course this isn't a satisfactory way and I hope that you can give me a
better solution to this problem.
Thank's in advance for you efforts
Matthias Brocks
[Ed. - This is a known bug of MS-DOS Kermit 2.29. The latest release
(2.29B) fixes the problem. It's in MSTIBM.BOO in the Kermit distribution.]
------------------------------
Date: Fri 22 May 1987 09:08:05 EDT
From: <SS@LL.ARPA>
Subject: Inconsistent Length of BREAK in MS-Kermit
Keywords: MS-DOS Kermit
I had problems with the break character on my COMPAQ 386 using MSKERMIT.
MSKERMIT for the IBM-PC and compatables uses an instruction loop to time the
length of a break character (based on the timing of a 4.77Mhz 8088) and the
faster the machine the shorter the break. I patched the assemly program to
calebrate the break on startup correctly.
Hope this helps with your problem.
[Ed. - The next release will generate BREAKs of constant duration,
independent of the CPU clock speed.]
------------------------------
Date: Thu, 21 May 87 11:59:41 EDT
From: uwvax!seismo!wucs1!wuphys!hpuslma!coalson@ucbvax.Berkeley.EDU
Subject: Patches for HP-150 Kermit Problems
Keywords: HP Kermit
I contacted Bill MacAllister who had mentioned a fix for 8-bit transfers
using kermit on an hp150 in msvhp1.bwr. The following is a dif of my
msxhp1.asm file after making the changes from a listing he sent me. I have
tried it as a local kermit and it seems to work. I haven't tried it with
the hp150 working as a remote kermit or as server.
[Ed. - The patch has been added to the file KER:MSVHP1.BWR and forwarded
to Joe Doupnik for the next release.]
------------------------------
Date: Wed, 27 May 87 11:49:03 EDT
From: rmcqueen (Robert C McQueen) @ sitvxb
Subject: Kermit Has Made It To the Board Level
Keywords: Terminal Emulation
According to the May 18th Network World:
AST Research, Inc. recently unveiled a 2 port terminal emulation board. The
board will do VT220 terminal emulation, and file transfer using the Xmodem
or Kermit protocols. The article states the board has an 80186 on it and
128kb memory. The board will handle 5 windows (2 VT220 sessions, a DOS
session and two notepads.).
Different.
Bob
------------------------------
Date: 20 May 87 15:24:38 GMT
From: msmith@gauss.rutgers.edu (Mark Smith)
Subject: Does Perkin-Elmer Kermit Work?
Keywords: Perkin-Elmer Kermit
Hi! Is there anyone out there that is successfully running Kermit in any
version on a Perkin-Elmer? Has anyone ever heard of anyone running it
successfully on a Perkin-Elmer. If so, please respond by E-mail to me or
call me at (201) 894-7732 between 9am and 4:445 pm EDT.
Thanks.
mark
[Ed. - Reportedly, the Perkin-Elmer Kermit versions work. Is there reason
to believe this is not the case?]
------------------------------
End of Info-Kermit Digest
*************************
-------