SY.CHRISTINE@CU20B.COLUMBIA.EDU (Christine M Gianone) (04/26/88)
Info-Kermit Digest Mon, 25 Apr 1988 Volume 7 : Number 11
ANNOUNCEMENTS -
Announcing Version 2.31-test4 of IBM PC Kermit
New Sliding Windows Kermit for the IBM PC
Announcing Commodore 64/128 Kermit Version 2.1(68)
Announcing a Version of Kermit for MAI Basic Four (MBF) Minis
C-Kermit 4E(070) Diffs for Apple Mac II's A/UX
Masscomp C-Kermit Fixes for RTU 4.0b
Fixes for C-Kermit 4E(70) for VMS
Send digest submissions to Info-Kermit@CU20B, requests for addition to or
deletion from the Info-Kermit subscriber list to Info-Kermit-Request@CU20B.
Kermit files may be obtained over networks and by mail order. On the
Internetwork, use FTP to log in to host CU20B, CU20B.COLUMBIA.EDU, or
CU20B.CC.COLUMBIA.EDU (a DECSYSTEM-20), as user ANONYMOUS, using any password,
and GET the desired files from logical device KER:. 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 KER: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: Thu, 21 Apr 88 21:31 MDT
From: <JRD%USU.BITNET@CUVMA.COLUMBIA.EDU> (Joe Doupnik)
Subject: Announcing Version 2.31-test4 of IBM PC Kermit
Keywords: MS-DOS Kermit 2.31, IBM PC Kermit
A new version of MS Kermit is available for testing on IBM PC's and
compatibles. It is labelled 2.31-test4. It is an early "alpha test" of the
next release, 2.31, for which a release date can't even be estimated. Until
then matters are subject to change, but what is present now is most unlikely
to be withdrawn.
Items of interest beyond 2.30 (a subset):
1. Selection of a serial port has been extended to permit specification of
Bios level interaction with Kermit rather than letting Kermit go directly
to the port hardware. The commands are
SET PORT BIOSn (n = 1 to 4) for Bios interaction
and
SET PORT COMn (n = 1 - 4) for Hardware interaction
and
SET PORT NET host-name for NetBios interaction
The Bios spec is present specifically to support networks which intercept the
normal Bios serial port interrupt 14H. Kermit expects registers to be used
exactly like the IBM Bios, and i/o is not interrupt driven. If there are
problems I am most interested in fixing them post haste.
Further, when the serial port hardware UART is employed the interrupt
request line (IRQ) parameter is now obtained automatically by Kermit so
that the IRQ may be 3 or 4 for different ports, as the user sees fit.
Previously, IRQ had to be 4,3,4,3 for COM1-4; now PS/2's with 4,3,3,3
should work automatically. An arcane detail, to be sure, but in the past
one causing much user frustration.
Until now MS Kermit checked the Bios work area for port info even if the BIOS
were used for comms. Since various LAN mfrs do an intercept it could happen
that they leave COM1 and COM2 intact and grab COM3, so to speak. If there were
no hardware for COM3 the Bios work word would be empty, more often than not,
and port 3 would be unusable even for networking. This version skips checking
up on matters if the Bios is chosen (voluntarily by SET PORT BIOSn, or
automatically be lack of a suitable UART chip). Thus, networking intercepts
can be shuffled down to BIOS3 or BIOS4 to free the real serial ports. The
consequence of selecting BIOSn when there is nothing there is to read lots of
echos of our own characters, harmlessly.
2. File attributes packets are available to transfer the size, creation time
and date of a file to cooperating hosts. Few Kermits implement attributes
packets but the new IBM CMS Kermit and the PDP-11 Kermits do so. C-Kermit
will support them in the near future. Two new commands result from attributes
packets:
SET ATTRIBUTES ON | OFF (default is ON)
to suppress attribute negotiations if the host might be confused by them, and
MAIL filename username@host
MAIL is exactly the same as a SEND command except the file attributes packet
requests the host submit the file to the local Mail utility with the indicated
addressing (can include spaces) rather than store the file on disk.
Presently, the command is unique to MS Kermit; expect selected C-Kermit
support in the near future. MAIL will fail right away (without sending the
file) if the remote host cannot understand the request.
3. A Transaction log is available to record the filename, direction of
transfer, size, starting time, date, and final status of each regular file
exchange. The log entry is a terse two line description ready for printing or
automated processing.
LOG TRANSACTION <optional filename, TRANSACT.LOG is the default>
turns on this logging,
CLOSE TRANSACTION
turns it off, and
SHOW LOG
indicates the status of logging. Material is alway appended to log files to
prevent losing older information. Similarly, there are separate SESSION
(Connect mode) and PACKET log files.
4. Kermit can be operated directly from DOS files, such as
KERMIT < todo.lst
or Preprocessor | Kermit
These files or pipes contain the same information as would be typed by hand,
yet Kermit knows a file is being read and will exit to DOS when the file
becomes empty (it will not hang waiting for keystrokes which never arrive).
Kermit TAKE files can be run this way if desired.
5. The command parser permits definition and use of Kermit "variables" such
as \%1, \%2 and so on. Variables have names of backslash, percent sign,
character '0' and above (with upper and lower cases merged). Variables are
defined as Macros of that name and the definition can be any string. Such
strings can contain other variable names. These variables substitute the
definition text for their name. For example
DEFINE dial AT\13,input OK,pause,output ATDT \%1,inp CONNECT, Connect
DO dial 1-800-555-1212
places the word "1-800-555-1212" in variable \%1 and that text appears
in the Hayes ATDT command during dialing.
Thus, variables \%1,..\%9 are automatically defined by adding words after
the macro name in the DO command, much like command line arguments to
programs or Batch files.
Other variables are defined by the DEFINE <macro name> <mac definition>
command, as usual. Of note is that when a variable name appears in a Kermit
command it is immediately replaced by its definition; if the definition
contains other variables they are also expanded immediately. No definition
is the same as a null length string.
DEF test echo first \%1, second \%2, third \%3,
DEF \%e echo first \%1, second \%2, third \%3,
yield the same display, via
DO test word word word
or DO \%e this that another
or \%e foo bar bletch
Macros can be displayed by SHOW MACRO <optional mac name abbreviation>.
6. The Tektronix 4010 emulator has been improved and made more like a real
Tek 4010 terminal. The user may allow Kermit to select the active graphics
display adapter (default) or may override that choice via the command
SET TERMINAL GRAPHICS Auto-sensing | CGA | EGA | VGA | HERCULES | ATT
ATT includes Olivetti M24/M28/DEC VAXmate/Toshiba T3100/AT&T 6300/6300+.
VGA invokes a 640x480 dot color display but the bottom 79 lines are not
saved due to display adapter memory limits.
7. Screen scrolling in the VT102 emulator has been improved twenty fold for
the case where top and bottom display lines do not scroll. Emacs and many
applications scroll interior lines; this version should do so much faster.
Whole screen scrolling is limited by the display adapter board rather than
Kermit.
8. Error messages have been improved to be more helpful. For example,
typing the command
Kermit-MS> SAT Timer on
shows the message '?Word "SAT" is not usable here', and
Kermit-MS> SET Timer on and off and maybe on the table
does indeed Set Timer ON and also shows the message
?Ignoring extra characters "and off and maybe on the table"
And so forth. After all, if the machine knows when an error has been made
then maybe it can fix the safe kind without bothering people.
9. The general screen status display is still present but has been supplemented
by a number of individual "SHOW keyword" displays to assist finding things.
SHOW ? indicates all the current keywords for this command. Command STATUS
gives a screenful of the most interesting things. SHOW is under active
development to find decent groupings and display formats.
10. The Kermit command line interface now permits full 8-bit character inputs,
with only NUL, ESC, DEL/BS, ^W (delete word), ^U (delete line), and ^C being
special. This is to enhance support for various languages and keyboards.
11. A new Kermit verb for Connect mode is \Kholdscrn which acts like the
DEC VT10x Hold Screen key: display is suspended immediately by ignoring
new characters (depends on flow control at the serial port level to XOFF
a chatty host). This verb is not assigned to a key presently. It is a toggle
and is cleared by resetting the terminal (\Kreset, ALT =).
12. The VT102 emulator supports the escape sequence pair ESC [ 1 2 h/l to
control local echoing of outgoing material. ANSI spec.
13. The VT102 "25th" line, or status line, is under better control in that
a fossil of the line (written by the host) does not appear unexpectedly.
The status display indicates when the line is on and owned by the host.
14. Another networking item is that files opened read-only now have the DOS
DENY-NONE bit set so that competing tasks may access them simultaneously,
such as when they are run by Pushing or RUN within Kermit.
15. The command SPACE no longer requires presence of CHKDSK.COM. The disk
space calculation is done internally. Empty disk drives yield a nice message.
16. Script commands have been enhanced. A new command is
WAIT [timeout] \CD | \CTS | \DSR
WAIT 04:15 \CD\DSR ; wait for carrier detect and modem ready
to wait for the presence of one or more of these modem signals, or fail with
a timeout.
Commands WAIT, PAUSE, INPUT accept timeouts of seconds and now HH:MM:SS
(truncate from the right but leave at least one colon) which means essentially
"until" that time of day (24 hour clock). The hh:mm:ss must be less that
12 hours from the present so the software can distinguish early from late. A
carriage return will force a timeout, to avoid rebooting a somnolent machine.
17. If a host echoes our XOFF Kermit now detects the situation and ignores
the arriving XOFF. Also, while waiting for the host to send XON Kermit tests
new characters at a rapid rate to improve response time.
18. CD is now a synomym for CWD (and for SET DEFAULT).
19. PAUSE now stops pausing when characters are typed at the keyboard.
This allows you to put things like this into your script files:
ECHO Type any character when ready
PAUSE 1000
20. A HELP command was added -- nothing fancy, just enough to get users
started (memory is a consideration here).
21. Finally, MS Kermit works well under MicroSoft Windows. Previously, it
ran fine under Windows but was unable to show Connect mode material which
was voluntarily rolled back on-screen. Screen scrolling now works perfectly.
Tested on MS Windows 2.03 on AT's and MS Windows 386.
[Ed. - Thanks once again, Joe, for your tireless efforts. The new test
version is now in the Kermit Distribution as MSTIBM.BOO, along with this
message, MSTIBM.HLP, which can serve, for now, as an addendum to the MS-DOS
Kermit User Guide.]
------------------------------
Date: Fri 15 Apr 88 16:06:45-EST
From: Frank da Cruz <SY.FDC@CU20B.COLUMBIA.EDU>
Subject: New Sliding Windows Kermit for the IBM PC
Keywords: MS-DOS Kermit, Sliding Windows, BIX
This is to announce a version of Kermit for the IBM PC by Terje Mathisen of
Norsk Hydro Data, announced by him on BIX in November 87 (his BIX mail ID is
"terjem"). It's written in Turbo Pascal, and claims to include long packets
AND sliding windows, and a nonstandard "fast mode", in which data fields are
not encoded at all (it can only use this to talk to itself). There is dumb
terminal emulation, which seems to take place only in a small horizontal
window under the command menu.
The source files arrived with nonstandard naming conventions, and have
therefore been packed into a single file, TP4KER.PAS, with name markers like
<<< KERMIT.PAS >>> at the head of each file. The executable program is encoded
as TP4KER.BOO (which can be decoded back into TP4KER.EXE using any of the
MSBPCT programs). There is no manual, but a very short summary is in
TP4KER.HLP. Comments & reviews welcome, as would be results of testing the
long packet and sliding window features against other Kermit implementations
that have them.
------------------------------
Date: Mon, 18 Apr 88 00:54:48 -0500
From: ray@j.cc.purdue.edu (Ray Moody)
Subject: Announcing Commodore 64/128 Kermit Version 2.1(68)
Keywords: Commodore 64, C64
Commodore 64/128 Kermit Version 2.1 (68) is now available. This new version
has many new features, improvements, and even some bug fixes. Here are the
main new features/improvements/bug fixes:
+ Enhanced DEC VT-100 terminal emulation with support for
the VT-100 keypad. Kermit should now work fine on
operating systems such as VMS. V2.1 has some VT-102
features added as well: Insert/Delete Line/Character.
+ Limited Tektronix 4010 graphics terminal emulation. Ker-
mit can plot both graphics and text with a resolution of
320 x 200 (C-64) or 640 x 200 (C-128).
+ Commodore 128 grey key and numeric keypad support. The
grey keys and the numeric keypad, not normally accessible
on a C-128 in C-64 mode, are active within Kermit.
+ A special file-type for transferring C Power (now Power
C) source code files. You can now download C source code
ASCII files as well as upload C Power source files. All
necessary character translations are handled automati-
cally.
+ A screen driver for the Batteries Included BI-80 80
column card. This device gives a display as nice as the
Commodore 128's 80-column hardware, but lacks many attri-
butes such as blinking and bold.
In addition to these major improvements, many smaller changes were made.
Several bugs were squashed that affected VT-100 emulation, the STATUS
command, and other things.
Obtaining Kermit on a Floppy Disk:
A copy of Commodore Kermit may be obtained by sending $5.00 postage and
handling to:
Dr. Evil Laboratories
P.O. Box 190
St. Paul, IN 47272
We stress that Commodore Kermit is absolutely free, the $5.00 is only
used to cover the cost of the disk, the mailer, postage, and handling. The
disk will contain Commo- dore Kermit version 2.1, a copy of the preliminary
documentation broken into several files small enough to be viewed with a
good wordprocessor, an initialization file, and, for C128 users, an autoboot
sector. (All of the files on the disk are available for download from the
Kermit archives. There is a program in the Kermit archives that will create
the proper autoboot sector for people that want to have one.)
Sometime this summer a final printed (and maybe even bound) version of
the manual will be available at a reasonable cost. Stay tuned.
Also available from Dr. Evil Laboratories is a custom ROM for the
Batteries Included 80-column card that provides access to the entire ASCII
character set. Currently we ask $5.00 for the custom ROM. This price is
subject to change and installation is required.
Dr. Evil Laboratories is a small software company in which Kent and I
are partners. The company has the facili- ties to distribute Kermit much
more efficiently than either of us could personally. Also, Dr. Evil Labs
has a permanent address, something that we, being in college, don't have.
All orders to Dr. Evil Laboratories must be in U.S. funds. Indiana
residents must add 5% sales tax.
Kermit can also be downloaded from the Kermit archives on ARPANET,
BITNET, and other places. For complete downloading instructions, see the
file KER:AANETW.HLP.
Ray Moody
ray@j.cc.purdue.edu
ihnp4!pur-ee!j.cc.purdue.edu!ray
moody@purccvm.BITNET
Kent Sullivan
Qlink: corvairkid
The files are as follows:
c64ker.ann - announcement of new kermit
c64ker.hlp - how to download from cu20b. How to get 1541 disk
c64ker.doc - documentation (a total rewrite of the old docs)
c64ker.hex - the downloadable binary itself
c64ker.m65 - source code. Too big to mail, so split into parts
c64ker.rom - replacement character rom for BI-80 80-column card
c64ker.ini - program to create kermit.ini file
c64auto.bas - program to create an autoboot sector
[Ed. - Thanks Ray. The files have been updated as you suggested. The new
files have replaced the old ones in KER:C64*.*]
------------------------------
Date: Thur 07 Apr 88 14:31:00-PDT
From: Edward V. Wastrodowski, Sphere Holdings Limited (SHL)
Subject: Announcing a Version of Kermit for MAI Basic Four (MBF) Minis
Keywords: MAI, Basic Four
This is to announce the availablity of Kermit for the MAI Basic Four
minicomputers, series model numbers 7000,8000, and 9000; it is
written using the latest release of BOSS/VS Business Basic (tm)
called BB86.
[Ed. - Thanks for putting this together and sending it to us on tape.
It has been installed in the Kermit Distribution as MBF*.*, and available
by mail order on Tape D. Also included is a short Basic program to read
ANSI-D format tapes (like the ones we distribute Kermit on).]
------------------------------
Date: Thu, 07 Apr 88 15:57:22 PST
Subject: C-Kermit 4E(070) Diffs for Apple Mac II's A/UX
From: Marion Hakanson <hakanson@mist.cs.orst.edu>
Keywords: C-Kermit, Macintosh A/UX, A/UX
Below are the changes necessary to make C-Kermit 4E(070) compile and run
properly on an Apple Mac II running A/UX (Apple's Unix). A/UX is mostly
System V, but with BSD compatibility grafted on in some rather unique ways.
Here is a short description of the changes I made (quite minor, really).
ckuker.mak: Add an "aux" target, which is identical to the "sys3"
target with an added "-DAUX" flag. See below.
ckufio.c: Near the top, in a #ifdef UXIII/#endif pair, is a
#define MAXNAMLEN DIRSIZ. Apparently A/UX has its own
MAXNAMLEN definition, so surrounding the above #define
in #ifndef MAXNAMLEN/#endif takes care of that problem
in a "portable" manner.
ckutio.c: I really wanted to avoid adding an AUX flag, but this
is quite unique (i.e. "strange"). A/UX has SIGTSTP,
so the job-control code in C-Kermit gets compiled in
properly, but C-Kermit expects 4.2bsd signal semantics
when this happens. The easiest workaround seems to be
to add to sysinit() a call to A/UX's set42sig() library
routine, which enables BSD-style reliable signals, along
with the corresponding TTY job control. In other words,
it allows C-Kermit to be suspended with ^Z and restarted
with "fg" from the C-shell, instead of being killed when
you try to restart it.
The only other trick I had to discover to use C-Kermit was that you may need
to "set line /dev/modem" (and not tty0) to use the modem port. Note that
I'm communicating with a hard-wired port selector, and not a "real" modem,
but modem control lines are used here (perhaps not the correct ones,
though).
Apple seems to distribute A/UX with C-Kermit 4E(066) installed in the
/usr/bin directory (this was the C-Kermit with the serious long packet bug),
and sources in /usr/src/kermit. Other than a similar, but less portable
than mine, change to ckufio.c, I could detect no modifications they made to
the distribution 4E(066), which of course did NOT do the right thing with
Ctrl-Z, etc.
Please let me know if problems occur with these changes, or if other
modifications are indicated. We've only had this A/UX system up for a
couple of days, and Kermit was a necessity for connecting it to something
other than itself. Here are the patches:
[Ed. - Many thanks! For now, your changes have been added to the CKUKER.BWR
file, and they will certainly be incorporated into the next release, which,
given the number of changes that have arrived in recent weeks, will have to
be pretty soon...]
------------------------------
Date: 15 Apr 88 13:59:27 GMT
From: dalesys%lamont.Columbia.edu@lamont (dale chayes)
Subject: Masscomp C-Kermit Fixes for RTU 4.0b
Keywords: Masscomp, RTU, C-Kermit
I have butchered the recent release of C-Kermit to accomodate the notion
of "dir.h" in the beta version of Masscomp's RTU (RTU-4.0.b1) operating system.
(For RTU-3.1, "make rtu" with the distributed sources works fine.)
I did it in a rather crude way. It would be more appropriate to make a new
entry in the makefile "rtu4" and use ifdefs, but that's not how it
happened.... (at least I put in some comments (:-))
Dale Chayes Lamont-Doherty Geological Observatory of Columbia University
usmail: Route 9W, Palisades, N.Y. 10964
voice: (914) 359-2900 extension 434 fax: (914) 359-6817
usnet: ...philabs!lamont!dale
[Ed. - Thanks, Dale. Your changes are in CKUKER.BWR for now, and should
find their way into the next release. It sure would be nice if makers of
Unix systems could agree on where their header files go, and whether they
must be included or not...]
------------------------------
Date: Wed 20 Apr 88 15:42:14-PDT
From: Ted Shapin <BEC.SHAPIN@ECLA.USC.EDU>
Subject: Fixes for C-Kermit 4E(70) for VMS
To: info-kermit@CU20B.COLUMBIA.EDU
Phone: (714)961-3393; Mail:Beckman Instruments, Inc.
Mail-addr: 2500 Harbor Blvd., X-11, Fullerton CA 92634
Here are some fixes for C-Kermit 4E(70) we made for our VAX VMS system.
The fixes are from Will Wood, Beckman Instruments, Fullerton, CA.
[Ed. - Listings omitted, but added to CKVKER.BWR, and forwarded to Jamie
Hanrahan, who's working on a new release of C-Kermit for VMS. The fixes
include a more effective way of having the C-Kermit server log itself out when
it gets a BYE command, and a fix for the function that returns the current
default device:[directory], and some terminal i/o improvements.]
------------------------------
End of Info-Kermit Digest
*************************
-------