[fa.info-mac] INFO-MAC Digest V2 #11

info-mac@uw-beaver (03/08/85)

From: Moderator John Mark Agosta <INFO-MAC-REQUEST@SUMEX-AIM.ARPA>


INFO-MAC Digest           Friday, 8 Mar 1985       Volume 2 : Issue 11

Today's Topics:
                        new downloading software
             Pascal version of macput and macget for VAX/VMS
                    Macget and Macput (in C) for VMS
                      Programmers' Deep Shit Alerts
                                  MIDI
                     Compatibility of MS-WORD files
                              New finder?!?


----------------------------------------------------------------------

Date: Thu 7 Mar 85 16:38:27-PST
From: John Mark Agosta <INFO-MAC-REQUEST@SUMEX-AIM.ARPA>
Subject: new downloading software

INFO-MAC now has "macput" and "macget" commands for several DEC 
operating systems: Hurray!

Andy Sweer at SUMEX has TOPS 20 test versions of macput and macget in 
the info-mac directory. They are assembly language translations of the
unix versions, so they function similarly, except you must use a "/" 
instead of a "-" to indicate flags in the argument. Please give 
Sweer@SUMEX your comments about their operation.

I received two versions for VAX/VMS symultaneously!

The one from Kris Kreutzman is in PASCAL, with assorted MEM, DOC and
CLD files.

The other from Stewart Rubenstein @Harvard, whose C code for both 
functions is bundled in a DCL file.

Take your pick! Notice that file names have been prefaced by VMS- to 
help organize the info-mac directory. Expect to see more of this 
pseudo subdirectory naming convention in the future.

-jma

------------------------------

Date: Thu, 28 Feb 85 18:57:48 PST
From: Michael Peirce <peirce@lll-crg.ARPA>
Subject: Pascal version of macput and macget for VAX/VMS


The following is a Vax/VMS Pascal program that an associate of mine 
wrote to allow XMODEM transfers between the Vax running VMS and 
MacTerminal.  Included is a .MEM file that explains it's operation, a 
.CLD file used with SET COMMAND to define XMAC as a command, and the 
Pascal source.  Since Kris is not actually on the net, any questions 
regarding the program should go to me (peirce@lll-crg) and I'll relay 
them along.  Hope this is helpful to someone out there.  One last 
thing, please fill free to distribute XMAC freely.  All Kris asks is 
that his Copyright notice stays put.

Michael Peirce (peirce@lll-crg) 
..................................................................

[ The USER MANUAL is filed in VMS-XMAC.DOC -jma ] [ The COMMAND LINE
DEFINITION FILE has been placed in VMS-XMAC.CLD -jma ] [ XMAC.PAS has
been placed in <info-mac>VMS-XMAC.PAS -jma ]

[ This installation guide is stored in VMS-XMAC.MEM -jma]

TITLE :  Macintosh to DEC VAX using VMS, XMODEM protocol file transfer
utility.  ( Transmitted by Michael Peirce (peirce@lll-crg) )

DESCRIPTION :  This program allows users to transfer files between a
Apple Macintosh and a DEC VAX minicomputer running VMS.  Two types of
transfers are possible, TEXT files and Macintosh binary files.  TEXT
files are editable on both computers, but some restrictions apply (see
documentation).  Macintosh binary files can be transfered to the VAX,
and back.  Binary files are stored on the VAX in one file.  Both types
of transfers use the XMODEM protocol provided in the MacTerminal
program for error detection and correction.

WARNINGS :  This is release number one! so beware!  1) This program
works best at 1200 baud, and worst at 9600 baud.  I will
   fix this in the next revision. (something with the the QIOW system
call) 2) No crash error handling and/or recovery.  3) Works only with
the latest MacTerminal revision.  4) Not tried under VMS version 4.0,
yet.

Following are the three files associated with XMAC.

  XMAC.DOC - the user documentation on how to run XMAC.
  XMAC.CLD - the command line definition file.
  XMAC.PAS - the source file.



Compile the XMAC.PAS file.  Use DEC Pascal version 2.1 or greater.

  $ PAS2 xmac ! or $ PASCAL xmac.  Whatever is set up for version 2.x
  $ LINK xmac


Edit the command line definition file (XMAC.CLD).  Change the
directory location of the image.

  from:  IMAGE user0:[kris.public]xmac.exe
  to:  IMAGE <drive>:[<directory>]xmac.exe

Where <drive> and <directory> specify where XMAC.EXE resides.


Add to your LOGIN.COM file :

  $ SET COMMAND <drive>:[<directory>]xmac

Where <drive> and <directory> specify where XMAC.CLD resides.  When
your login.com file in invoked, you are ready to run XMAC.

------------------------------

From: stew%lhasa.UUCP@harvard.ARPA
Date: 	1 Mar 85 01:10 EST To: harvard!info-mac@sumex
Subject: Macget and Macput (in C) for VMS

Here is my version of macget and macput for VMS.  To unpack this, just
cut at the dotted line and execute the remainder of the file as a DCL 
command procedure.  This works under Version 3.* of VMS.  It should
work under Version 4 as well, but some improvements will be possible;
I will post again if there is enough interest.

Stew Rubenstein Harvard University Chemical Labs 12 Oxford St.  
Cambridge, MA 02138 rubenstein@harvard.arpa 
{ihnp4,seismo,ut-sally}!harvard!rubenstein

- ------------------------------ Cut Here -----------------------------
[ The Whole dcl file may be found in VMS-MACPUT-GET-C.DCL -jma ]

------------------------------

Date: Sun 3 Mar 85 12:41:15-PST
From: Gustavo Fernandez <FERNANDEZ@SU-SCORE.ARPA>
Subject: Programmers' Deep Shit Alerts

Anyone who has read the system error handler section of Inside Mac has
seen the 'system alerts' window which was supposedly for programmers
while the bomb we all know and love was for naive users.
Unfortunately, Apple never released the programmers' alerts.
   Having a very old version of the system (october '83) from which I
was able to grab such goodies as the Monkey, etc, I decided to try to
bring over the old deep shit alerts as well. (That's what they're
called!)
   I had to do a complete disassembly of the file as several updates
had been made since then. I then merged my disassembly with the
published source of the user alerts (Lisa workshop local/init2.text)
The results are now stored in <info-mac>sysalerts.asm (assembly
source) and <info-mac>sysalerts.hqx (BINHEX 4.0 file of resource file
sysalerts containing a single INIT 2 resource to replace the existing
INIT 2 in the system folder - Use REDIT.)

The programmers' alerts have the following advantages...

1) They display the error message as well as the error code. (ID=28 is
   displayed as "Stack overflow," etc.)

2) They give a register dump of A0-A7/D0-D7 and PC.

3) There is a FINDER button that does an ExitToShell call as well as
the
   usual restart and resume buttons.

They have the following disadvantages...

1) They are defenitely NOT for naive users. The register dump is ugly.
   The error messages migh either scare people or make thim jump to
   conclusions

2) The FINDER button may not always work. Currently it does the same
as
   the RESTART button if the ExitToShell trap is patched. (Both the
switcher
   and Hyper do this.)

3) They take up almost 4 times as much memory in the system heap than
do
   the user alerts. I have yet to try these out on a 128K mac where
the
   system heap is a lot smaller.

I took a few liberties and added a few things to the code.

1) The register names are displayed as well as their values. (ie
D0=xxxxxx...)

2) Some of the error messages have been changed to reflect their true
meaning
   in a Mac environment (eg Line 1111 exception changed to Fxxx
Breakpoint.)
   Some other errors which are nearly impossible to create (such as
bus
   error or privelige exception) were deleted to save memory.

3) The grayed out RESUME button, which was faked in by the user
alerts, and
   nonexistent in the system alerts, was added for the sake of
compatibility.
   Some memory can be reclaimed by deleting subroutine PlotGrayBut and
its
   single call at the beginning of routine p4. This will make this
button
   appear ONLY when resume is enabled.

4) Certain parts of the display were shifted slightly

5) The code now comforms to the current conventions which have changed
since
   October, '83.

                                                Have Fun and Happy
Hacking.
                                                Gus Fernandez
                                                FERNANDEZ@SU-SCORE
- -------

Return-Path: <FERNANDEZ@SU-SCORE.ARPA> Date: Mon 4 Mar 85 23:01:34-PST
From: Gustavo Fernandez <FERNANDEZ@SU-SCORE.ARPA> Subject: Re: your
D--- Sh-- alerts codes.  To: INFO-MAC-REQUEST@SUMEX-AIM.ARPA

To install the alerts, simply use RMOVER or the Resource editor to
replace the INIT 2 resource in the system wwith the new one.

                                                        Gus

------------------------------

Date: Thu, 7 Mar 85 10:05:39 est
From: kevin%gvax@Cornell.ARPA (Kevin Karplus)
Subject: MIDI

The baud rate constant can be computed as
  (115,200/baud) - 2 For a baud rate of 31 250, the constant is 1.686.
The nearest integer (2) is 18.6% off, well outside the +-1% allowed in
the MIDI spec.  Apple people really gnashed their teeth when they
heard about the unfortunate choice of baud rates made by the synth
manufacturers.  Your best bet is to provide an outside clock.

I have a student currently building an ADC converter for the Mac that
uses an outside clock.  The hardware is finished, but I don't want to
say it works until the software is working.

Gaining access to the SCC registers is easy.  Sumacc has a bunch of 
constants for the addresses in the memory space.  I believe they have 
names like sccbase.  You will need a data sheet for the 8530 SCC chip.

To use an external clock, you have to change register 11 to set 
receive clock and transmit clock to the TRxC pin instead of the baud
rate generator.  I believe the correct value is 0x28, but I may be
mis-interpreting the TRxC O/I bit.  If you use a faster clock than
31.25kHz, you can play with register 4, changing the clock mode (x1,
x16, x32, x64).  Note, in x32 mode, a 1MHz clock can be used.

The best way I've found to change the SCC parameters is to first open
the desired port with standard serial driver calls, then write
directly to the memory location representing the control port of the
SCC.

[Note: for those who remember my announcing an almost finished ADC
last summer.  This is a different student, the first ADC board had to
be discarded as it was poorly designed and constructed, and could not
be reasonably debugged.  The perennial problem with free labor,...]

                                        Kevin Karplus

------------------------------

Date: 6 Mar 1985  9:13:35 EST (Wednesday)
From: Dick Kalagher <kalagher@mitre>
Subject: Compatibility of MS-WORD files

I read someplace that Microsoft Word files were compatible between the
Mac and the IBM PC.  It's a half-truth.  We sent a WORD file from a 
Mac to a PC and from a PC to a Mac.  WORD opened and loaded the file 
on both machines but each file contained control characters and other
stuff.  Formatting information was lost.  The text was all there and
the other stuff could be edited out.  However, it would have been
better to send a "text only" ASCII file and reformat on the other
machine.  Thus there does not appear to be anything special about WORD
files that make them any eaiser to use on two different PC's.

------------------------------

Date: Mon, 4 Mar 85 23:59:04 est
From: Steve D. Miller <steve@tove>
Subject: New finder?!?


   I recently obtained a disk that has Switcher and MacWrite 3.985 on 
it, and I noticed when I booted off of it that things were a bit 
different than usual.  The trash can was replaced by a skeletal figure
with a scythe, the "Special" menu had a different name, and things
seemed to be running faster. I did an "About the Finder" on it, and it
said it was version 2.1X, of all things.  Has anyone else seen this
thing?

   I don't have a hard disk, so I have no idea if any of the long- 
rumored hard disk improvements are implemented, but the window
operations (redrawing, moving, closing, and the like) were quite
definitely faster.  Now that I think about it, there was also a "New
Folder" option under File; I would assume that it does the obvious,
but haven't had the chance to try it.

   The upshot of all this is that I'd like to install it as the normal
finder on my disks, but I don't want to do so if such an action is 
likely to mung my defenseless data beyond all recognition.  Any pieces
of information would be appreciated.

  Thanks in advance,


        -Steve Miller (steve@maryland.arpa)

------------------------------

End of INFO-MAC Digest
**********************