cmg@CUNIXC.CC.COLUMBIA.EDU (Christine M Gianone) (01/07/89)
Info-Kermit Digest Fri, 6 Jan 1989 Volume 9 : Number 1
Today's Topics:
Announcing MS-DOS Kermit 2.32
New German MS-DOS Kermit Help File
MS-Kermit and EIA flow control (RTS/CTS)
MS-Kermit and the Public Domain MNP software emulation package.
MS-Kermit and the Hyperion?
Kermit Parity Negotiation?
Kermit with Telebit Modem Support?
VT100 setting file for MacKermit and the IBM 7171 protocol converter
Some Responses to V8 #11 and Kermit-370
Bug in Kermit-80 for Northstar Advantage
Send digest submissions to Info-Kermit@CUNIXC.CC.COLUMBIA.EDU, requests for
addition to or deletion from the Info-Kermit subscriber list to
Info-Kermit-Request@CUNIXC.CC.COLUMBIA.EDU.
Kermit files may be obtained over networks and by mail order. On the
Internetwork, use FTP to log in to host CUNIXC, CUNIXC.CC.COLUMBIA.EDU, a
VAX 8700 running UNIX (Ultrix). The IP host number is 128.59.40.130. You
should be able to to FTP to CUNIXC, login as user ANONYMOUS (any password)
and GET the desired files. The Kermit files are in directories kermit/a,
kermit/b, kermit/c, kermit/d, and kermit/e. You can also get Kermit files
over BITNET/EARN; to get started send a message with text HELP to KERMSRV,
the Kermit file server, at host CUVMA. For detailed instructions, read the
file k1/aanetw.hlp (AANETW HLP on KERMSRV). To order by mail, request a
complete list of Kermit versions and an order form from Kermit Distribution,
Columbia University Center for Computing Activities, 612 West 115th Street,
New York, NY 10025 USA.
----------------------------------------------------------------------
Date: Wed Dec 21 13:14:20 1988 EST
From: Christine Gianone <cmg@cunixc.cc.columbia.edu>
Subject: Announcing MS-DOS Kermit 2.32
Keywords: MS-DOS Kermit 2.32, IBM PC, DEC Rainbow, HP-150
Keywords: Grid Compass II, Victor 9000, Hebrew, Arabic
This is to announce version 2.32 of MS-DOS Kermit for the IBM PC and PS/2
families and compatibles, the DEC Rainbow, the HP-150 and 110, the Grid
Compass II, the Victor 9000, and for "generic DOS". This release replaces
version 2.31 of July 1988. The majority of the work was done by Professor
Joe R. Doupnik of Utah State University, who also prepared the 2.29, 2.30,
and 2.31 releases.
Like previous releases, version 2.32 includes an advanced and efficient
implementation of the Kermit file transfer protocol, emulation of the VT102,
VT52, and Heath 19 terminals, as well as the Tektronix 4010/4014 graphics
terminals, a script programming language, and a wide variety of options,
settings, and other features.
New to version 2.32 are many bug fixes and internal improvements, and one
major new feature -- support, during CONNECT mode, for languages like Hebrew
and Arabic in which characters are displayed from right to left, plus
mechanisms to facilitate switching between character sets (and their
associated directions). These new language features were adapted from work
done by Baruch Cochavy, IIT, Haifa, Israel, whose test version of "Hebrew
Kermit" was announced in Info-Kermit V8 #9. The enhanced language support is
available through the new command SET TERMINAL DIRECTION, plus a new
ALTERNATE-ROM option for the SET TERMINAL CHARACTER-SET command, plus the
association of special predefined macro names with new private escape
sequences.
Other improvements include:
- A new ASK command, prompts user to input data to a variable.
- A new IF [NOT] EQUAL command compares string variables (use with ASK).
- A new ASSIGN commands copies one variable to another.
- Screens cleared by ESC [ 2 J are now saved in the rollback buffer.
- Removal of Timeout and Hangup screen messages during script/macro execution.
- A way to pass multi-word strings as single arguments to a macro.
- A way to disable server timeout NAKs (while server waits to answer a call).
- REMOTE LOGIN command (for use with Kermit servers that support it).
- Miscellaneous improvements in the file transfer display.
- Better color support during CONNECT.
Among the major bugs that were fixed are:
- Problems with GOTOs and labels in script programs.
- Ungermann-Bass Net/1 support.
- Loss of data when receiving files to a printer.
- Crashes resulting from overflow of script or macro buffers.
- Crashes resulting from stray interrupts.
- Victor 9000 now can do Tektronix graphics again.
- More COM1/COM2 confusion cleared up (e.g. for PCjr).
- Filesize and disk space computation with file attribute packets.
- SET SEND TIMEOUT interference with file encoding and decoding.
- Program should now be assemblable by MASM 5.1 as well as 5.0 and 4.0.
The changes since version 2.31 are described fully in the file MSR232.UPD.
The user manual, MSKERM.DOC (.MSS, .PS) has been revised to reflect the new
features.
This new release is available in both source and binary form. It may be
ordered on diskette or magnetic tape from Columbia, and it may be obtained via
network file transfer over the Internet or BITNET/EARN. On the Internet, use
FTP (user anonymous) to host CUNIXC.CC.COLUMBIA.EDU [128.59.40.130], CWD (or
CD) to the directory kermit/a. On BITNET/EARN, send an interactive or e-mail
message to KERMSRV@CUVMA (the BITNET Kermit file server), containing the text
HELP, to get started.
The executable files are stored in a special printable format, called "BOO
files", suitable for BITNET file transfer, electronic mail, etc. These are
decoded back into .EXE files using any of the MSBPCT "BOO-file decoder"
programs. The documentation is available online in plain ASCII text format,
in Scribe text formatter source format, and in Postscript format. Following
is a synopsis of the files. The KERMSRV name is the same as the CUNIXC name,
except the CUNIXC name is lowercase with a period between the filename and
filetype, wherease the KERMSRV name is uppercase with a space separating the
filename and filetype, e.g. msaaaa.hlp on CUNIXC is MSAAAA HLP on CUVMA.
CUNIXC Name Size Description
msaaaa.hlp 7K Explanation of file naming conventions
msbaaa.hlp 4K Explanation of bootstrapping procedure
msb*.* 174K total BOO-file encoding/decoding programs, many languages
msvibm.boo 115K IBM PC Kermit, BOO-encoded executable
msvrb1.boo 83K DEC Rainbow Kermit BOO file
msvgen.boo 78K Generic MS-DOS Kermit BOO file
msvhp1.boo 77K HP-150 (untested)
msvgri.boo 78K Grid Compass II (untested)
mskerm.doc 333K MS-DOS 2.32 Kermit manual, plain ASCII text
mskerm.ps 489K MS-DOS 2.32 Kermit manual, for Postscript printers
mskerm.mss 323K Scribe text formatter source for manual
mskerm.hlp 16K A summary of MS-Kermit commands
mskerm.bwr 14K List of known restrictions, bugs, etc.
mss*.* 776K total System-independent MASM Source files (12 files)
msg*.* 128K System-dependent source (Tektronix graphics)
msu*.* 70-85K each Sys-depn source (keyboard support, all systems)
msx*.* 25-174K each Sys-depn source (port i/o, etc, all systems)
msy*.* 22-124K each Sys-depn source (terminal emulation, IBM only)
msz*.* 66-189K each Sys-depn source (term emul, cont'd, IBM only)
msv*.mak 2K each Microsoft MAKE files for each version
msv*.bat 2K each Batch files to build each version
msv*.lnk 1K each LINK command files for each version
The utility program MSUCHK.C (and .BOO), contributed by Phil Benchoff, allows
convenient determination of MS-Kermit's keyboard codes on the IBM PC
family. And the files MSIXSE.* are the XSEND program, contributed by Mark
Zinzow, for constructing MS-Kermit command files that send entire directory
trees from one DOS system to another.
For the benefit of those who are still running version 2.29 and haven't
converted to later versions because the SET KEY syntax changed so drastically,
there is also a new file, MSVIBM.KEY, which lists the old (2.29) and new (2.30
and later) IBM keyboard scan codes.
Be sure to read the MSKERM.BWR file before trying to use the new version, or
reporting any problems with it.
Here are the minimum files needed for the new release ("xxx" stands for the
specific version, IBM, RB1, HP1, HPX, V90, GEN, etc):
1. For everybody: The documentation -- MSKERM.DOC, MSKERM.HLP, MSKERM.BWR,
MSR232.UPD.
2. The new Kermit program itself: MSVxxx.BOO (where xxx denotes the particular
MS-DOS system: IBM, RB1, HP1, etc).
3. If you don't have an MSBPCT "BOO-file decoder", also get at least MSBPCT.BAS
(the slow BASIC version) and MSBPCT.BOO (an .EXE version from C, which
you can decode with MSBPCT.BAS). Use MSBPCT.EXE to decode MSVxxx.BOO.
4. For those who want to make modifications to the sources:
MSS*.*, MSGxxx.* (if any), MSXxxx.*, MSYxxx.* (if any), MSZxxx.* (if any),
MSVxxx.MAK (or .BAT if you don't have MAKE), and MSVxxx.LNK.
The systems for which we don't yet have the new version ready are still in
the Kermit distribution as before, under the MSV, MSX, and MSY prefixes. These
will be replaced if and when new ones appear.
The IBM PC, DEC Rainbow, and HP-150 versions may also be ordered on diskette
from Columbia, along with typeset, printed copies of the manual. The IBM
version is available on 5.25-inch 360K DS DD diskettes, and on 3.5-inch 720K DS
diskettes for the PS/2 family. The Rainbow version is on RX50. The HP-150
version is on 3.5" diskette.
FINAL WORDS:
Like any Kermit program, MS-DOS Kermit is for everyone to use and share.
There is no software license, and the program and documentation may be
reproduced and redistributed without restriction, so long as this is not done
for profit. Please take or order the files you need. If you're getting files
over a network, please be judicious -- don't ask for MS*.*, or the networks
will be choked for months! Once you get a working copy at your site, share it
with others, rather than having them make further network requests.
Send complaints, bug reports, suggestions, comments, or even praise (where
it's deserved) to Info-Kermit@CUNIXC (.CC.COLUMBIA.EDU). And thanks once
again on behalf of the world's thousands or millions of MS-DOS Kermit users to
Joe Doupnik for his skill, generosity, and patience. And thanks also to the
beta testers.
------------------------------
Date: Fri, 23 Dec 88 17:01:16 SET
From: RECK@DBNUAMA1.BITNET
Subject: New German MS-DOS Kermit Help File
Keywords: MS-DOS Kermit 2.32, German
Enclosed is the updated German version of the MS-Kermit help file.
Here are a few notes on seeming glitches in the help file; none of them
serious, of course:
Section top-level commands:
CLOSE: option "ALL" is missing
Section remote commands:
REMOTE LOGIN is missing
REMOTE SEND should really be REMOTE MESSAGE, I presume
Commands honoured by MS-Kermit server:
I cannot check this, since I don't have a suitable connection to a PC;
but I guess MS-Kermit will honour a HELP command?
Section SEND/RECEIVE parameters:
There is no SET SEND DELAY command - the appropriate command is just SET
DELAY.
TIMEOUT : time given is in *seconds*
Section SHOW commands:
TRANSLATION does not accept "INPUT" as a subcommand. TRANSLATION by
itself suffices.
Pretty minor stuff...
\Gisbert
[Ed. - Thanks, Gisbert, all corrections valid, and all are now in the new
(English) MSKERM.HLP. The German version is in MSKGER.HLP.]
------------------------------
Date: Wed, 23 Nov 88 14:17:13 EST
From: John C Klensin <KLENSIN@INFOODS.MIT.EDU>
Subject: MS-Kermit and EIA flow control (RTS/CTS)
Keywords: MS-DOS Kermit 2.31
Our otherwise-several-times-qualified new ISDN phone system seems to strongly
prefer RTS/CTS to XON/XOFF, especially at higher speeds. I looked at the
possibility of adding this to MSKERMIT many years ago (before you got actively
involved) and decided I was not motivated enough. I'm now feeling motivated,
since AT&T claims significantly higher data throughput when hardware flow
control is used.
On the other hand, I don't want to do anything that would foul up or delay
the brilliant job you are doing, or whatever release schedule you have. So:
should I pick up the sources for 2.31A from somewhere and go to work on them?
Or should I do this against the "released" 2.31 sources, or should I wait
until after 2.31A becomes "official"?
While I haven't looked at the code in 2-3 years, I assume that this type of
enhancement should impact only a single module, plus or minus the table(s)
that know what all of the options are. If you can point me to the right
place, it would save a bit of time and be appreciated.
Thanks, John Klensin, MIT
[From jrd - The place to look for hardware vs software flow control in all in
one file: msxibm.asm. We have just canned version 2.32 (nee 2.31/A) and
Columbia is sorting through the mountain of files I sent to them.
Hardware flow control is slightly more messy because we need to probe the
hardware for a signal level and to select sending a character vs wiggling a
wire. It would be interlocked with the real UART port, as contrasted with the
zillions of other port options now present (Bios and networks). It is do-able
fairly easily and cleanly. Have a go, with my blessing.]
Thanks. I'll wait until the 2.32 sources appear, then go at it. Your comment
about the messy-ness parallels a note that I sent to the local AT&T folks
yesterday: my experience with EIA flow control is that it works very well when
one has a controller organized s.t. one can get interrupts when the state of
the lines change. When, as with the IBM PC UART, about the only thing one can
do is poll them, one is basically using EIA control to simulate Xon/Xoff
logic, with no real gain in performance and at a higher nusiance level. But,
who am I to argue with the hardware folks....
On the tab stuff, I think that something like 1:8, or some other way of saying
"start with 1, increment by 8 until you get to the end" would be a lovely way
to spell "default", as well as solving many other problems (there are a few
word processing packages that could use similar suggestions).
[jrd: Yes, I understand. The tabs set by the host can be overridden by a
terminal reset, \Kreset, while in Connect mode. This is much like a real VT102
where the "setup" values are restored by a reset. When we set tabs at the
Kermit prompt level that is essentially the "setup" state for the emulator. A
default condition could be added to the SET TERMINAL command, where default
means every 8; actually I'd like an implied repeat count on the tabs 1:8 start
in col 1, repeat by skipping 8 columns, or something like that.]
Thanks again.
john
------------------------------
Date: Sat, 17 Dec 88 23:43:06 +0200
From: Ze'ev "Steed" Shtadler <steed%TECHUNIX.BITNET@cuvmb.cc.columbia.edu>
Subject: MS-Kermit and the Public Domain MNP software emulation package.
Keywords: MS-DOS Kermit, MNP Emulation Software
Looking in the SIMTEL-20 archives I found a software package that can make any
asynchronous modem an MNP Class-2 error-free modem. This package is written
in Microsoft C and Assembler. The telephone lines around here are noisy, and
it is very difficult to use Kermit as a terminal emulator (and I don't put the
blame on Kermit :-). The modem on the other end of the line is an MNP modem,
but I couldn't use its error free option since I can only afford a regular
modem. By putting the MNP emulation into Kermit I can have an error free
terminal too.
Did anyone try to combine Kermit with this package? How complicated is
changing the serial I/O routines?
Ze'ev.
ARPANET: steed%techunix.bitnet@cunyvm.cuny.edu
UUCP: ...!psuvax1!techunix.bitnet!steed
BITNET: steed@techunix
Domain: steed@techunix.technion.ac.il
CSNET: steed%techunix.bitnet@csnet-relay.csnet
[Ed. - This would probably be a BIG job, and would tax Kermit's terminal
emulator, which already does about 100 things with each character (see the
new diagram in the MS-DOS Kermit 2.32 User Guide). However, you could
conceivably use this program "under" Kermit, in conjunction with an ANSI.SYS
or NANSI.SYS or similar console driver, and then pop back up to Kermit when
you need to transfer files.]
------------------------------
Date: Thu, 15 Dec 1988 15:52 EDT
From: Mark Bramwell 519 661-3714 <MBramwel@business.uwo.ca>
Subject: MS-Kermit and the Hyperion?
Keywords: Hyperion, MS-DOS Kermit
I have recently picked up a Hyperion portable computer. This is a MS-DOS
machine whose day has come and gone. It has 256k, built-in monitor, serial,
parallel, and 2 360k floppy drives. The unit boots ok with PC-DOS 3.1 and
PC-DOS 4.0
The serial port is not exactly IBM compatible. If I put a break-out on
the serial port and type in the command DIR > COM1:
I can see data coming from the port. However, kermit says COM1: NOT
AVAILABLE.
Question: Is there ANY version of kermit that will run on the hyperion?
Does anyone know the IRQ and BASE address of this box so I
can try patching a comm package?
There were two versions of the hyperion, and I have one of each. Slight
BIOS change between the two.
[Ed. - Several things to try: (1) Use Generic DOS Kermit, rather than the
IBM PC version; (2) Use the IBM PC version, but try giving the command
"SET PORT 2", or if that doesn't work, "SET PORT BIOS1" or "SET PORT BIOS2";
(3) get the latest source code and add Hyperion support to it.]
------------------------------
Date: Mon, 28 Nov 88 10:56 MDT
From: Pete Klammer 303/556-3915 <PKLAMMER%CUDNVR@vaxf.colorado.edu>
Subject: Kermit Parity Negotiation?
Keywords: Kermit Protocol, Parity
Just about the last frontier of KERMIT user friendliness is parity. Would it
be possible for KERMIT to automatically sense or set or negotiate parity?
Getting parity right is the "final frustration".
We have MS-KERMIT v2.31, VMS KERMIT-32 v3.3.111, PRIME KERMIT v1.04.1,
C-KERMIT on various UNIXen, etc. Our "least-common-denominator" parity
setting recommendation is 7-BIT EVEN. But when we log into VMS, it is through
a DECserver-500 ethernetted to the VAX, and the parity is effectively
concealed from VMS: SET TERM and SHOW TERM will not budge VMS's stubborn
notion that our terminal parity is NONE. So when we crank up VMS KERMIT into
SERVER mode, it is thinking PARITY NONE, while our PC KERMIT is still thinking
PARITY EVEN. Any GET, SEND, or REMOTE commands now retry fruitlessly.
Couldn't the guy in PARITY NONE mode step down into EVEN or ODD for a couple
of retries?
* --poko Pete Klammer, Systems Programmer, (303)556-3915
* CU-Denver Computing Services / Campus Box 169
* 1200 Larimer St NC2506 / Denver CO 80204-5300
* BITNET%"PKLAMMER@CUDENVER"
* INTERNET%"PKLAMMER%CUDNVR@VAXF.COLORADO.EDU"
[Ed. - Agreed. As you noted, the system that is running the Kermit
program cannot necessarily be aware of what parity is being used, even by
itself. Thus, negotiation wouldn't necessarily work. But a heuristic can be
used instead -- anyone who wants to is welcome to try it. The leading control
fields of the packet (MARK, LEN, SEQ, and TYPE) are always 7-bit characters.
Thus the receiver of a packet can deduce that parity is in use if any of these
characters have their 8th bit set. In fact, by finding any two characters in
this set that would have opposite parity sense, you could even determine which
kind of parity (even, odd, or mark) is being used. The only pitfall here is
distinguishing "space" parity from no parity. This trick also assumes,
possibly incorrectly, that a system wants you to send it the same kind of
parity that it is sending to you.]
------------------------------
Date: 16 Dec 88 08:51:06 GMT
From: versatc!tran@sun.com (Tony Tran)
Subject: Kermit with Telebit Modem Support?
Keywords: Telebit, Modems
Does anybody have a version of kermit which supports the Telebit Trailblazer
modem at 19.2kb ? Ours only supports up to 9.6kb.
I would appreciate any hint or sources sent to my following e-mail address.
Tony Tran
Versatec, Inc.
Santa Clara, Calif.
{sun|ames|pyramid|ubvax}!versatc!tran
[Ed. - All versions of Kermit already work OK with the Telebit. But
Unix Kermit only lets you give SET SPEED commands up to 9600, whereas the
Telebit can run at 19,200. The trick is just to add the (nonstandard) symbols
for 19,200 baud to the C-Kermit source and recompile, e.g. as outlined by
asg@space.mit.edu (Sergei A. Gourevitch), whose message on the subject has
been added to the ckuker.bwr file. The Telebit modem, by the way, actually
executes the Kermit protocol between the modem and the computer, and its own
"PEP" protocol between the modems. Thus the Kermit you run (say) on your PC
is actually trading packets with the Kermit in the modem, not the Kermit on
the other end of the connection.]
------------------------------
Date: Fri, 16 Dec 88 00:48:48 IST
From: "Jonathan B. Owen" <GDAU100%BGUVM.BITNET@cuvmb.cc.columbia.edu>
Subject: VT100 setting file for MacKermit and the IBM 7171 protocol converter
Keywords: MacKermit 0.9(40)
This file is a Kermit 0.9(40) document containing key definitions
for emulating a VT100 terminal to be used with a 7171 IBM protocol
converter. Note that this is for use with an extended keyboard only.
The keys defined are as follows:
F1 thru F12 -- Use to access IBM Function keys
Clear key -- Clear Screen
Arrow keys -- Moves cursor around screen
Help/ins -- Toggle Insert/Overstrike Mode
Home -- Moves to first field in Panel
Page Up/Down -- Same as pressing F7/F8 respectivly
End -- Sends <CR> BOTTOM <CR> (useful in XEDIT)
Del -- Delete Char. under cursor
F13 -- Toggles between CP and VM modes
F14 -- Same as Clear Key (See above)
Also, definitions useful for modem users:
<Option><ESC> -- Sends +++ to the modem (Hayes escape sequence)
<Option> h -- Sends "at h" for hanging up line
Hope this helps out...
(--) /--) /-(\ Email: gdau100@bguvm (bitnet)
\ / /--K | \|/\ /\/) /|-\ Snail: 55 Hovevei Zion
_/_/o /L__)_/o \/\__/ \X/ \_/ | |_/ Tel-Aviv, 63346 ISRAEL
(/ Jonathan B. Owen Voice: (03) 281-422
[Ed. - Thanks! Your settings file has been added to kermit/b/ckmker.set,
which is a collection of settings files for MacKermit 0.9.]
------------------------------
Date: Fri, 16 Dec 88 01:22:11 EST
From: "Roger Fajman" <RAF%NIHCU.BITNET@cuvmb.cc.columbia.edu>
Subject: Some Responses to V8 #11 and Kermit-370
Keywords: Kermit-370
> J M Hicks asked A Question about the Kermit Protocol:
>
> > I was under the impression that characters between packets should be
> > completely ignored.
> >
> > What ought Kermit to do? If such characters between packets appear often,
> > one end can give up due to all the bad packets it thinks it has received.
>
> While the editor pointed out (correctly) that Kermits *should* ignore
> characters between packets, half-duplex systems (such as IBM mainframes)
> receive RECORDS, rather than PACKETS. In such circumstances, an extra CR is
> evidence of a lost or garbled packet, and, considering that some IBM
> mainframe
> Kermit applications are unable to time out, it is clearly risky to ignore a
> bare CR.
> John
It's also risky not to ignore bare CRs. Some systems, such as IBM TSO/TCAM,
echo a CR/LF when they receive a CR. If the Kermit on the other end insists
on treating that CR as the end of a packet, you'll never get a file transfer
going without changing the EOL character. We experienced this exact problem
with older releases of C Kermit talking to NIH TSO Kermit. It took quite a
while to figure out what was happenning.
------------------------------
Date: Mon, 14 Nov 1988 21:08 MST
From: Keith Petersen <W8SDZ@wsmr-simtel20.army.mil>
Subject: Bug in Kermit-80 for Northstar Advantage
Keywords: Kermit-80, Northstar Advantage, CP/M-80 Kermit
Relayed from GEnie:
Item 1411308 88/11/13 15:35
From: J.LABUTE John P. Labute
To: W8SDZ Keith Petersen, CP/M SysOp
Sub: Kermit-80
There seems to be a bug in the version of Kermit-80 for the Northstar
Advantage (CP409ADV.COM). This version echos back to the remote anything
that it receives, even with the local echo off (as I verified with my
laptop). I have tried two versions for my APPLE II with success. Can
you help me or put me in contact with somebody who can. Thanks.
[Ed. - Anybody???]
------------------------------
End of Info-Kermit Digest
*************************