cmg@WATSUN.CC.COLUMBIA.EDU (Christine M Gianone) (02/05/91)
Info-Kermit Digest Mon, 4 Feb 1991 Volume 13 : Number 2
Today's Topics:
Info-Kermit Digest Delivery
Last Test Prerelease of MS-DOS Kermit 3.02
DEC Rainbow Kermit 3.02 Prerelease Available for Testing
MS-DOS Kermit 3.02 in ROM on the HP Portable Plus
Using the MS-DOS Kermit Terminal Emulator with PICK
CP/M Kermit for the Ampro Little Board
Digest submissions may be sent to Info-Kermit@WATSUN.CC.COLUMBIA.EDU,
requests for addition to or deletion from the Info-Kermit subscriber list to
Info-Kermit-Request@WATSUN.CC.COLUMBIA.EDU or to KERMIT@CUVMA.BITNET.
Kermit files may be obtained over networks and by mail order. On the
Internetwork, use FTP to log in to host WATSUN.CC.COLUMBIA.EDU, a SUN-4/280
running UNIX (SUNOS 4.1), IP host number 128.59.39.2. Login as user anonymous
(note, lower case), any password, and GET or MGET (MULTIPLE GET) the desired
files. The Kermit files are in directories kermit/a, kermit/b, kermit/c,
kermit/d, and kermit/e. Test versions are in kermit/test. Binaries are in
kermit/bin (use ftp in binary mode). You can also get Kermit files over the
BITNET/EARN network; to get started send a message with text HELP to KERMSRV,
the Kermit file server, at host CUVMA. For detailed instructions, read the
file kermit/a/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: Mon, 4 Feb 1991 11:33:24 EST
From: Christine M Gianone <cmg@watsun.cc.columbia.edu>
Subject: Info-Kermit Digest Delivery
Keywords: Info-Kermit Digest
Many Info-Kermit subscribers report that they have not been receiving
Info-Kermit issues regularly. Others report receiving two or more copies of
each one. These problems seem to be caused by our local UNIX mail delivery
agent, sendmail, and our postmaster is investigating. It is very likely that
the enormous number and geographic dispersion of Info-Kermit subscribers is
greater than sendmail ever expected to cope with.
I would encourage those of you who subscribe to Info-Kermit directly to move
your subscription to the I-KERMIT listserver on BITNET/EARN. You can do this
even your mail address is not on a BITNET host. To do this,
send e-mail to LISTSERV@CUVMA.BITNET containing the following text:
SUBSCRIBE I-KERMIT <name>
where <name> is your personal name, which can be in upper and lower case,
contain spaces, etc. Once you receive a digest issue from I-KERMIT, you can
send e-mail to Info-Kermit@watsun.cc.columbia.edu asking to have your direct
subscription cancelled because now you are receiving the digest through
LISTSERV. If you ever want to cancel your I-KERMIT subscription, send e-mail
to LISTSERV@CUVMA.BITNET containing the text:
SIGNOFF I-KERMIT
Users of Unix systems that receive the regular newsgroup postings can also
read comp.protocols.kermit, which is simply a reposting of the Info-Kermit
Digest to the Unix netnews network (how or by whom I have never known).
Meanwhile, for those of you who are wondering which issues you might have
missed, Volume 12 (July through December, 1990) had 7 issues (1 through 7).
We're now in Volume 13 (Jan-Jun 1991), and this is the second issue. Missing
issues can be retrieved from kermit/e/v12.* and v13.* on watsun via anonymous
ftp on the Internet, or V12 * and V13 * from KERMSRV at CUVMA on BITNET/EARN.
Sorry for the inconvenience, hope all is better soon.
------------------------------
Date: Mon, 4 Feb 1991 12:46:58 EST
From: Christine M Gianone <cmg@watsun.cc.columbia.edu>
Subject: Last Test Prerelease of MS-DOS Kermit 3.02
Keywords: MS-DOS Kermit 3.02
We're nearing the final release of the next version of MS-DOS Kermit, which up
till now we have been calling 3.02. When it is finally released, the version
number will be 3.10 to reduce confusion about test versions versus real
versions. As of this announcement, development of this release is frozen.
Please continue to send in bug reports and other comments, but suggestions for
new features will be kept for a post-3.10 release. This message, extracted
from Joe Doupnik's MSR302.UPD file, lists the changes since the last posting,
in Info-Kermit V13 #1. Subsequent messages describe work on some of the
non-IBM-PC versions.
The files are available in the Kermit Distribution "test" area, kermit/test
on watsun.cc.columbia.edu for anonymous FTP access on the Internet, and the
T: area of KERMSRV at CUVMA on BITNET. They include:
BOO File Documentation Systems Supported
mstibm.boo msr302.upd,.doc IBM PC, PS/2, and compatibles
mstgen.boo Generic MS-DOS
mstgri.boo mstgri.doc,.hlp Grid Compass
msthp1.boo msthp1.doc,.hlp Hewlett Packard 150
msthpx.boo msthpx.doc,.hlp Hewlett Packard 110 and Portable Plus
mstrb1.boo mstrb1.doc DEC Rainbow-100
mstv90.boo Victor 9000 (Sirius 1)
mstz10.boo mstz10.doc Heath/Zenith-100
The "boo" files are executable (.EXE) files encoded in "BOO" format, which
can be decoded back into .EXE format using any of the MSBPCT programs that
can be found in kermit/a/msbpct.* on watsun or MSBPCT * on KERMSRV.
The non-IBM versions listed above are approximately up to date with the IBM
version, but they lack many of the system-dependent features such as network
support, international character-set translation, and others, but they do
incorporate sliding windows, script language programming, and other
system-independent aspects of MS-DOS Kermit 3.02. The non-IBM versions have
not been tested in the current edit, although they were tested at earlier
points in the 3.02 development cycle.
Thanks to Joe Doupnik for his skill, patience, endurance, generosity, etc,
in pulling this release together, and to other MS-DOS Kermit programming
volunteers including Robert Weiner, Gary Stebbins, John Nyenhuis, Terry
Kennedy, Bert Tyler, and many others, and to the hundreds or thousands of
testers. Please report bugs or problems in this latest program edit to
Info-Kermit or directly to Joe at JRD@CC.USU.EDU (Internet) or JRD@USU
(BITNET/EARN). Joe's notes follow:
112. Add a feature suggested by John Klensin and Frank da Cruz, from C Kermit.
New substitution variable notation \$(item) means replace \$(item) by
the string on the right hand side of the line starting as ITEM= in the
DOS Environment. Any word may be used for ITEM. The replacement will be
empty if either the string or the item is absent in the Environment.
Leading and trailing spaces and tabs are removed from the string.
Examples:
Echo \$(path)
shows the string which occurs after "PATH=" in the Environment, such
as C:\;C:\WP;C:\DOS.
IF eq \$(video) CGA echo Upgrade to VGA
will show the string "Upgrade to VGA" if someone has provided the line
VIDEO=CGA in the Environment, perhaps by typing SET VIDEO=CGA or
written there by a program.
ASSIGN \%p \$(path)
gives variable \%p the definition of the right hand side of PATH=.
Semicolons occurring in \$(..) replacement strings will be treated as
ordinary characters and will not start a comment. Comments starting
with a semicolon are now restricted to ordinary Macros and Take files.
Changes are in files mssker.asm and msscmd.asm.
113. Add variable \v(speed) to the list of substitution variables. \v(speed)
is the ascii value of the communications line speed, in bits/second. It
may be the word "unknown", without the quotes, if the value is not
available, and it may contain non-numerical characters such as "." and
"/" as in 45.5, 134.5, and 75/1200. Network connections will usually
return "unknown". Changes are in file msscmd.asm.
114. Add variable "program" to the WRITE command and let it be variable
\v(program). It is the string MS-DOS_KERMIT. Notice the underscore.
Changes in file msster.asm.
115. Correct two keyboard related problems: reporting \v(keyboard) as 88 keys
for such keyboards, and retaining the SET KEY {ON, OFF, LK250} state if
an LK250 keyboard is requested (by this command) but is not present.
Changes in file msuibm.asm.
116. Make small corrections to network-presence tests which previously
said no-network if the net's interrupt pointed to segment 0f000h, system
ROM. With the advent of memory mappers such as QEMM, 386MAX, etc holes
in ROM can be filled by network drivers. The new test procedures allow
networks to be detected in the ROM holes. File msxibm.asm.
117. Correct a problem with screen rollback storage when system memory is
really short. An error in the code allowed storage (and a system crash)
if the area were absent due to insufficient memory. File msyibm.asm.
118. Correct a presence signature of the STB/EM Plus board when changing
between 80 and 132 columns, from Terry Kennedy. File msyibm.asm.
119. Correct a strange protocol problem. If receiving characters from an IBM
mainframe host (or equiv) which requires use of a "handshake" character,
typically Control-Q, and an end of packet character arrives prematurely
then the bad partial packet is marked and delivered as a good one.
Change in file msscom.asm.
120. Add code from Bert Tyler to manually set IBM XGA display boards in 80
or 132 column mode because (apparently) the first boards lack support
in their Bios. File msyibm.asm.
121. Add command SHOW VARIABLES [\v([name])] which displays the strings
associated with variables of the form \v(variable-name). Leading "\v("
and trailing ")" may be omitted. SHOW VARIABLE with no argument shows
all variables; abbreviations are allowed here. Files changed are
mssker.asm, msscmd.asm, msssho.asm.
122. Add 80/132 column support for the ATI VGA Wonder board, as a extended
case of the EGA board. Tnx to Steve Wood. File msyibm.asm.
123. Permit abbreviations in \v(variable) phrases, such as \v(ver) rather
than the full form of \v(version). File msscmd.asm
124. Correct a bug in the file receiver code which could reject a file if
SET ATTRIBUTES LENGTH OFF were stated. File mssrcv.asm, done 16 Jan 91.
125. When the device holding the Session log (Connect mode) becomes full or
reports an error during writing then put the error message on the status
line, and suspend logging. Previously the error message was written on
the main part of the screen where it could be overwritten or missed, and
logging was reattempted every 128 bytes. Now logging is attempted again
when reentering Connect mode. File msster.asm
126. Changes to \v(variable) area.
a) Truncate the fractions of seconds field from \v(time).
b) Add \v(ndate) which reports the date with all digits as YYYYMMDD
for easier testing in scripts. \v(date) reports MM-DD-YYYY.
c) Add \v(status) which reports the value of the Kermit status word;
0 is success, any other numeric is a failure. Most Kermit commands
clear this word result before execution. The current values are
listed below; "user intervention" may be added to other values.
Condition Status Comment
success 0 initial value for most commands
send failure 1 includes OUTPUT and TRANSMIT commands
receive failure 2 includes INPUT command timeout
remote cmd failure 4 REMOTE command failed for any reason
Take file failure 8 Take file not found
general failure 16 many causes
user intervention 256 usually by typing Control-C
These changes also apply to WRITE statements.
Changes affect most mss*.asm files.
127. Add two tests to the script IF statement. These are lexical comparisons
between two words. The curly brace {..} operators are not usable here.
IF LLT word1 word2 <command>
executes <command> if word1 has fewer characters (after
removing leading spaces) than word2, or if at the first
character where the words differ word1 has a character
occurring earlier (less than) in the ascii collating sequence
than that in word 2.
IF LGT word1 word2 <command>
opposite of above, for word1 longer than word2 or the first
differing character in word1 occurs later (greater than) that
in word2 in the ascii collating sequence.
Combining these with IF EQUAL and the NOT modifier yields all logical
comparison conditions, as shown in the table below. Case sensitivity
of the tests is controlled by SET INPUT CASE {IGNORE, OBSERVE}, with
IGNORE being the default; when case is observed uppercase letters are
"less than" lower case letters.
Lexical Test Kermit IF statement
word1 = word2 IF EQUAL word1 word2 ...
word1 != word2 IF NOT EQUAL word1 word2 ...
word1 < word2 IF LLT word1 word2 ...
word1 > word2 IF LGT word1 word2 ...
word1 => word2 IF NOT LLT word1 word2 ...
word1 <= word2 IF NOT LGT word1 word2 ...
Changes are in file mssscp.asm
128. Speedup response of the VT300 terminal emulator to VMS SET TERM/INQUIRE,
and correct SET DEBUG SESSION to not invoke packet logging. Files
mszibm.asm and msscom.asm, resp.
129. Send an XON when opening the "serial port", to free a host blocked by
an XOFF from whatever cause. Correct REINPUT not timing out automatically
if the history buffer is full and a string match is not found.
Files msxibm.asm and mssscp.asm, resp. 25 Jan 1991
130. For macros invoked while in terminal emulation mode, such as TERMINALS,
PRODUCT, user defined macros, etc., make variable \%0 be the macro name.
Changes in file msyibm.asm.
131. Add special feature of calling Macro ON_EXIT when Kermit is exiting to
DOS (but not when PUSHing/shelling to DOS). The user needs to define
the Macro and it may contain any commands but normally ones wanted to
put the system back into a standard form. For example
DEFINE ON_EXIT run cols80.bat,echo Goodbye from MS-DOS Kermit.
This Macro will be executed to switch the screen to 80 column mode
when Kermit command EXIT or QUIT is given.
Changes are in file mssker.asm.
132. Permit specification of the IRQ value of a UART serial port. The command
SET COM3 <port address> <IRQ value>
has the new optional second field to hold the IRQ (2 through 15 decimal).
Example:
SET COM3 \x03e8 \2
The equivalent command for the DOS Environment would be
C> SET KERMIT=com3 \x03e8 \2
If the IRQ value is omitted then an IRQ of 4 or 3 will be used for
testing. In all cases, a test is run to see if the serial port responds
with this port address and IRQ value. If it does not then the port is
tested again at the given address but using IRQ 4 or 3. If that fails
the port is assigned an IRQ of 4 for COM1 and COM3 and 3 for COM2 and
COM4; PS/2 machines are assigned 3 for COM2..4. XT's and other machines
based on 8x88/8x86 cpu chips can not use an IRQ above 7.
BEWARE!! It is very easy to interfere with other equipment in the
computer, such as your harddisk, if the IRQ is used by other boards.
The hard disk on XT's uses IRQ 5, and on AT's and above it typically
uses IRQ 14. Parallel printers rarely use IRQs so by swithing off IRQ
generation on the parallel port(s) one may recover IRQ 7 and 5 (AT's).
Check carefully before experimenting. You use this facility only at your
own risk.
Changes are in file mssset.asm and msxibm.asm.
[Ed. - Notice -- this item was added only reluctantly by Joe, in response to
hundreds of requests for from those who have strange setups for their serial
ports -- nonstandard boards, unusual mixtures of serial devices, etc etc.
Joe's (and our) advice is to use Kermit's IRQ-setting option only as a last
resort. It's much better and safer to try to rearrange your equipment
configuration to use a standard IRQ of 3 or 4 for all serial ports.
Nevertheless, it is important that those of you who have been requesting the
ability to specify your own IRQ test this feature and report back as quickly
as possible, good or bad. Thanks!]
133. Change name of \v(path) and WRITE <log> PATH to be \v(directory) and
WRITE <log> DIRECTORY, respectively, to avoid possible confusion with
the PATH= clause in the DOS Environment.
Files msscmd.asm and msster.asm.
134. Small revisions to keep CTTY COMn happy. Shut down the serial port
when at the Kermit prompt to let DOS style i/o have a chance at the
serial communications line.
File mssker.asm.
135. Reduce the WRITE command to syntax of WRITE <dest> text. Date and time
etc are available via variables \v(date) and so on. Now the script
input-buffer is displayed as part of the command SHOW SCRIPT.
Files msssho.asm and msster.asm.
------------------------------
Date: Thu, 10 Jan 91 02:43:04 EST
From: Robert Weiner <rweiner@watsun.cc.columbia.edu>
Subject: DEC Rainbow Kermit 3.02 Prerelease Available for Testing
Keywords: Rainbow, DEC Rainbow, MS-DOS Kermit 3.02
Here's the latest release of DEC Rainbow 100 MS-Kermit version 3.02 dev
27-Jan-1991 built from the current source on watsun in ~kermit/test/mss*.
This brings the Rainbow version in sync & up to date with the IBM PC version
as far as the MS-DOS Kermit system independent parts are concerned.
I haven't seen much (if any) feedback about the Rainbow versions. I'm using
it on a Rainbow 100B with alot of memory and a V20 processor. I'd like to
hear from someone who has used it on a machine with less memory, etc, or on a
Rainbow 100A. I don't expect anything should have broke on a 'smaller'
Rainbow, but who knows.
This keeps the Rainbow version up-to-date with the current IBM PC version.
There is now also some documentation for the new Rainbow version in
mstrb1.doc.
[Ed. - Thanks, Robert! The new files are in kermit/test/mstrb1.boo and .doc,
with a binary version in kermit/bin/mstrb1.exe.]
------------------------------
Date: Sun, 13 Jan 91 20:40:21 -0500
From: nyenhuis@ecn.purdue.edu (John Nyenhuis)
Subject: MS-DOS Kermit 3.02 in ROM on the HP Portable Plus
Keywords: MS-DOS Kermit 3.02, HP Portable Plus
Joe Doupnik suggested that I pass along my experiences in installing MS-DOS
Kermit 3.02 in ROM on the HP Portable Plus.
Over the holidays, I had some fun with the HP Portable Plus. I got a ROM
burner and put Kermit in ROM. In the Portable Plus architecture, it is
possible to run a program form ROM. (The operating system puts the ROM
starting at 9000:0000.) The idea is to have the code segments in ROM and to
have the data and stack segments in RAM. To do this, I had to make a few
patches to mssker.asm. Code like "mov ax,data" needs to be changed since it
is not possible for MS DOS to make segment fixups at load time if the program
is in ROM. It was also necessary to do the approximately 120 relocation fixes
that arise from the two code segments. (I wrote a small Turbo Pascal program
to do this.) I wrote a small loader program in assembler which calls Kermit
3.02. The loader program has the responsibility of setting up the data and
stack segments, copying Kermit's data segment to RAM, and informing Kermit of
the values of DS and SS. Kermit and the loader program fit in a pair of
27C512 EPROMS with about 35K to spare.
Kermit in ROM makes the Portable Plus, which has no internal drives, stand
alone. i.e., if you don't have an external drive, just use the Kermit from ROM
to load your files. Since the program is run from ROM, Kermit will run on a
Portable with as little as 128 K RAM.
I am willing to be of help in supplying Kermit in ROM to HP Portable Plus
users. I can e-mail the source code and image files to the interested. For a
fee (say $50) to cover the cost of EPROMS, shipping, and time, I can do custom
burns and also include instructions on how to install the ROMs. (Please
advise if this proposed service would cause a problem with Columbia's
copyright on the program. I am deeply appreciative of the efforts that Joe
Doupnik and others have put into MS-DOS Kermit and under no circumstances do I
want to take unfair advantage of their efforts.) Any thoughts you have on
distributing MS DOS Kermit ROMS for the Portable Plus are most welcome.
[Ed. - John, you are most welcome to offer this service.]
Also, I have passed along versions of MS DOS Kermit 3.02 for the GRiD Compass,
HP 150, and HP Portable and Portable Plus to Joe Doupnik. (All are RAM
based.) He said they would be put in /kermit/test in watsun.
[Ed. - They are, as listed earlier in this issue.]
Sincerely,
John Nyenhuis Purdue University School of Electrical Engineering
West Lafayette IN 47907 (317)494-3524 nyenhuis@ecn.purdue.edu
------------------------------
Date: Sat, 19 Jan 1991 13:12:22 EST
From: Christine M Gianone <cmg@watsun.cc.columbia.edu>
Subject: Using the MS-DOS Kermit Terminal Emulator with PICK
Keywords: PICK, MS-DOS Kermit and PICK
David L. Clark of History Database in Malibu, CA, USA, has sent in a document
describing how to use MS-DOS 3.0 Kermit's VT320 emulator to access the PICK
operating system in general, and the History Database in particular. This
document includes detailed instructions for the PICK end, plus MS-DOS Kermit
initialization files for the PC. This document has been put with the other
PICK Kermit files in kermit/d/pickrm.vt (PICKRM.VT on KERMSRV), and in
kermit/a/msipic.doc (MSIPIC.DOC on KERMSRV). Thanks to David for this
excellent document.
------------------------------
Date: Tue, 8 Jan 91 18:08:00 EST
From: Mike Freeman <freeman@watsun.cc.columbia.edu>
Subject: CP/M Kermit for the Ampro Little Board
Keywords: CP/M Kermit
Here are the system dependent files to adapt CP/M-80 Kermit 4.10 for the Ampro
Little Board. The code was contributed courtesy of Jay S. Rouman of Mt.
Pleasant, MI (dexter!jsr@sharkey.cc.umich.edu). He tells me that it works up
thru 4800 bps and file transfers work fine at 9600 bps though the screen
doesn't keep up (presume QUIET mode doesn't present a problem). I also
submitted cpxsy2.asm with a comment fixed and cpaaaa.hlp and cpk410.ann
updated to show CP/M Kermit support for the Ampro Little Board.
There are also a number of bugs fixed:
(1) leading white space in commands and keyword searches is now ignored
(whether from keyboard, CP/M command-line or TAKEfile -- this was the
original intent of the code but it didn't work).
(2) A further fix to the "disk full" reporting code in CPSPK1.ASM per
report from Russel Lang is hopefully the last.
(3) A RET has been put at the end of OUTCONN in CPXNOR.ASM -- D.P.Arnot
of Scottish Agricultural College in Auchincurive, Ayr, UK reported
this.
(4) Mr. Arnot also says the Amstrad Kermits don't work properly -- they
put garbage in received files (4.09 didn't work, apparently, either).
(5) The CPSCMD.ASM now zaps the entire fcb (in CMIL). This fixes a bug
where the COPY command did not copy files correctly after one
successful COPY on my HP-125.
Re the Amstrad: Am trying to get more info -- don't have one here. Sorry for
the continuing fixes but suggestions keep coming in. I presume this will
settle down pretty soon! It's interesting to see CP/Mers come out of the
closet. Ciao!
Some bugs in Remote commands and in the Transmit comands are fixed
(syntax for Remote commands is enforced more) and the 8K buffer is used
for the COPY, TYPE and PRINT commands (the big buffer). I hadn't
intended to submit the TYPE and COPY changes but had these bug fixes to
put in so thought I'd put these enhancements in also.
[Ed. - The new files are in kermit/test/cp*.* on watsun and in T:CP* * on
CUVMA.]
------------------------------
End of Info-Kermit Digest
*************************