Nautilus 1992 July/Nautilus-3-8/Nautilus-3-8.bin/Tools & Utilities/Techy Stuff/Doco ƒ/CSMP ƒ/CSMP-V1-088.TXT

Text File | 1992-06-30 | 48.1 KB | 1,260 lines

C.S.M.P. Digest             Thu, 21 May 92       Volume 1 : Issue 88

Today's Topics:

    Popup menus - menuitems at runtime
    Hooking hypercard to internet
    no BusErrors when writing to nonexistent NuBusSlot with Quadra
    Segment loader question.
    Action! information?
    (Q): Does anyone know about 'ictb' rsrcs?
    COMBINATORIX...math-type programming question?
    Programming Apple OneScanner?
    Serial Port Programming


The Comp.Sys.Mac.Programmer Digest is moderated by Michael A. Kelly.

These digests are available (by using FTP, account anonymous, your email
address as password) in the pub/mac/csmp-digest directory on ftp.cs.uoregon.
edu.  This is also the home of the comp.sys.mac.programmer Frequently Asked
Questions list.  The last several issues of the digest are available from
sumex-aim.stanford.edu as well.

These digests are also available via email.  Just send a note saying that you
want to be on the digest mailing list to [e-mail], and you will
automatically receive each new digest as it is created.

The digest is a collection of articles from the internet newsgroup comp.sys.
mac.programmer.  It is designed for people who read c.s.m.p. semi-regularly
and want an archive of the discussions.  If you don't know what a newsgroup
is, you probably don't have access to it.  Ask your systems administrator(s)
for details.  (This means you can't post questions to the digest.)

The articles in these digests are taken directly from comp.sys.mac.programmer.
They are not edited; all articles included in this digest are in their original
posted form.  The only articles that are -not- included in these digests are
those which didn't receive any replies (except those that give information
rather than ask a question).  All replies to each article are concatenated
onto the original article in the order in which they were received.  Article
threads are not added to the digests until the last article added to the
thread is at least one month old (this is to ensure that the thread is dead
before adding it to the digests).

Send administrative mail to [e-mail].

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

From: [e-mail]
Subject: Popup menus - menuitems at runtime
Organization: School of Chemistry, University of Sydney
Date: Fri, 17 Apr 1992 02:02:58 GMT


I want to use the new control for popup menus (IM VI 3-16) but I want to
set up the menu-items at run-time. Easy enough, once you can extract the
MenuHandle from the control's contrlData field then you can do,

    for (i=0;i<menuItemsCount;++i)
        AppendMenu(popupMenu,ArrayStrPtr[i]);
    ShowControl(popupCntl);

Unfortunately, the control is NOT updated until you've called TrackControl,
which leaves the control Rect blank until the user clicks on it. Has
anyone figured out a workaround for this problem, or I am just completely
wrong about this.


- -----------------------------------------------------------------------------
Adrian Wong, Dept.of Theoretical Chemistry     [e-mail]
University of Sydney NSW 2006 Australia        061-2-692 4137


+++++++++++++++++++++++++++

From: [e-mail] (Chris Wysocki)
Date: 16 Apr 92 21:06:19 GMT
Organization: Harvard University, Cambridge, MA

In article <[e-mail]>, [e-mail]
writes:

> I want to use the new control for popup menus (IM VI 3-16) but I want to
> set up the menu-items at run-time. Easy enough, once you can extract the
> MenuHandle from the control's contrlData field then you can do,
>
>     for (i=0;i<menuItemsCount;++i)
>         AppendMenu(popupMenu,ArrayStrPtr[i]);
>     ShowControl(popupCntl);
>
> Unfortunately, the control is NOT updated until you've called TrackControl,
> which leaves the control Rect blank until the user clicks on it. Has
> anyone figured out a workaround for this problem, or I am just completely
> wrong about this.

The way I got around this problem was involved puting a single blank item in
the MENU resource and adding the items at runtime with code that looked like:

     for (i=0;i<menuItemsCount;++i)
   if (i == 0)
    SetItem(popupMenu,1,ArrayStrPtr[i]);
   else
          AppendMenu(popupMenu,ArrayStrPtr[i]);
     ShowControl(popupCntl);

Not particularly elegant, I admit, but it works.

Chris Wysocki
[e-mail]


+++++++++++++++++++++++++++

From: [e-mail] (Quinn "The Eskimo!")
Organization: The University of Western Australia
Date: Tue, 21 Apr 1992 02:25:46 GMT

In article <[e-mail]>, [e-mail] writes:
>
> I want to use the new control for popup menus (IM VI 3-16) but I want to
> set up the menu-items at run-time. Easy enough, once you can extract the
> MenuHandle from the control's contrlData field then you can do,
>
>     for (i=0;i<menuItemsCount;++i)
>         AppendMenu(popupMenu,ArrayStrPtr[i]);
>     ShowControl(popupCntl);
>
> Unfortunately, the control is NOT updated until you've called TrackControl,
> which leaves the control Rect blank until the user clicks on it. Has
> anyone figured out a workaround for this problem, or I am just completely
> wrong about this.

I think the appropriate magic is...

  SetCtlMax(...,CountMItems(...))

Quinn "The Eskimo!"   <[e-mail]>  "Real Coke, Diet .sig"
Department of Computer Science, The University of Western Australia
  -- Scarred for life by ResEdit, MenuInfo.menuProc<>nil and popup menus.


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

From: [e-mail] (Daniel M. van der Zee (I89))
Subject: Hooking hypercard to internet
Date: 16 Apr 92 14:46:13 GMT
Organization: FWI, University of Amsterdam

For the sake of a simple user interface i would like to be able to hook
up a hypercard stack to a unix BBS running on the internet. The goal is
to give a user of our set of stacks (course design system) a very simple
interface to the BBS as an intergral part of the stack by adding a menu
to the menubar with items like "post message", "mail message" and "read mail".
A comand like "post message" should bring up a floating window and enable to
edit a message there (+ copy/paste from the stack) and (at the user's request)
open a connection to the bbs, post the message and logout.

To handle commands like this i'm looking for external commands like
  OpenInternetConnection
  CloseInternetConnection
  SendText
  RecieveText
Does someone know where I can get such commands and, if not, where i should
look for documentation on writing such code (maybe i should look at the
NCSA telnet source).
Oh yeah, users of our stack use low-end macs (mac+/classic 2.5 Mb RAM) and
therefore have not installed system 7 (still running 6.X) and the externals
should be able to run on their hard/software.

If you want to reply, please send me a copy of your message throug e-mail
because our system admin set up the news system to delete messages within 2/3
DAYS and a easterweekend is starting tomorrow.

I will ofcourse post a summary to the net and let you know how i solved the
problem.

(Excuse my poor spelling etc)

Thank you already,
Daniel van der Zee
[e-mail]

+++++++++++++++++++++++++++

From: [e-mail]
Date: 19 Apr 92 16:55:46 GMT
Organization: New Jersey Institute of Technology

> For the sake of a simple user interface i would like to be able to hook
> up a hypercard stack to a unix BBS running on the internet. The goal is
> to give a user of our set of stacks (course design system) a very simple
> interface to the BBS as an intergral part of the stack by adding a menu
> to the menubar with items like "post message", "mail message" and "read mail".
> A comand like "post message" should bring up a floating window and enable to
> edit a message there (+ copy/paste from the stack) and (at the user's request)
> open a connection to the bbs, post the message and logout.
>
> To handle commands like this i'm looking for external commands like
>   OpenInternetConnection
>   CloseInternetConnection
>   SendText
>   RecieveText
> Does someone know where I can get such commands and, if not, where i should
> look for documentation on writing such code (maybe i should look at the
> NCSA telnet source).
> Oh yeah, users of our stack use low-end macs (mac+/classic 2.5 Mb RAM) and
> therefore have not installed system 7 (still running 6.X) and the externals
> should be able to run on their hard/software.
>
        I would suggest rethinking Hypercard as the appropriate front
end for your BBS. You could do this much more quickly with the new
LORAN modules for Microphone II. I could probably get a prototype up
and running in a couple of days although I am fairly familiar with the
environment already. Your mileage may vary. The big advantage is that
Microphone is already written to handle _exactly_ this sort of task. No
XCMDs or XFCNs would need to be written. The disadvantage is that
Microphone II 4.0 is $215 or so a pop, but quantity discounts may (or
may not) be available. Also there is a currently unreleased runtime
version of it, hopefully soon available for small licensing fees. I
know of one information service using this already and perhaps Software
Ventures Corporation, the publishers of Microphone II, would be willing
to let you in on this deal too. If you're a small group then it's
probably cheaper to buy everyone copies of MP II than to do the sort of
programming you would need to do to accomplish the same task in
HyperCard.


Elliotte Rusty Harold        Department of Applied Mathematics
[e-mail]        New Jersey Institute of Technology
[e-mail]        Newark, NJ 07103

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

From: [e-mail] (Joachim Ostheimer)
Subject: no BusErrors when writing to nonexistent NuBusSlot with Quadra
Date: 14 Apr 92 10:09:52 GMT
Organization: Techn. University of Berlin, Germany



With the introduction of the quadra, apple changed
the way the mac reacts to a BusError.

The Change: when you try to _write_  to a nonexistent
address -> nothing happens, no BusError as with
any other Mac.
For a hardware projekt, I have to make _even_the_Quadra_
generate a BusError on illegal writes in NuBus
address space.



Joachim Ostheimer
Technical University Berlin, Germany.

+++++++++++++++++++++++++++

From: [e-mail] (Matthew T. Russotto)
Date: Tue, 14 Apr 92 15:14:25 GMT
Organization: College of Engineering, University of Maryland, College Park

In article <[e-mail]> [e-mail] (Joachim Ostheimer) writes:
>
>
>With the introduction of the quadra, apple changed
>the way the mac reacts to a BusError.
>
>The Change: when you try to _write_  to a nonexistent
>address -> nothing happens, no BusError as with
>any other Mac.
>For a hardware projekt, I have to make _even_the_Quadra_
>generate a BusError on illegal writes in NuBus
>address space.

On the 68030 and 68851, if the memory address is not in the
translation table (or is marked invalid), a bus error is generated
whether it is a write or a read.  Apple would not have been able to
change this-- therefore, either the 68040 changed it, or a bus error
DOES happen, but the default handler just skips the offending
instruction.  If the latter, all you have to do is install your own
bus error.


- --
Matthew T. Russotto    [e-mail]    [e-mail]
Some news readers expect "Disclaimer:" here.
Just say NO to police searches and seizures.  Make them use force.
(not responsible for bodily harm resulting from following above advice)

+++++++++++++++++++++++++++

From: [e-mail] (Joachim Ostheimer)
Organization: Techn. University of Berlin, Germany
Date: Tue, 14 Apr 1992 21:17:02 GMT

i have my own buserror routine which filters all read and write busErrors
in empty NubusSlots and this works fine with every 68030 mac.

But with the Quadras i can catch only reads.

You can Check it out with Macsbug:
   dm FS000000    (  S stands for SlotNr.  )
 ->    Unable to acces that address    (68030 or 68040)

   sm FS000000 abcd
 ->    Unable to acces that address   (68030)
 ->        ( nothing )                (68040)

 I even tested all other Vektors ( Interrupt 0-7 ....)
 ... there seems to be a Difference.

Joachim Ostheimer
Technical University Berlin, Germany.


+++++++++++++++++++++++++++

From: [e-mail] (Russell J. Davoli)
Date: 15 Apr 92 14:35:45 GMT
Organization: National Instruments Corp.

In article <[e-mail]>, [e-mail] (Matthew T. Russotto) writes:
>
> In article <[e-mail]> [e-mail] (Joachim Ostheimer) writes:
> >
> >
> >With the introduction of the quadra, apple changed
> >the way the mac reacts to a BusError.
> >
> >The Change: when you try to _write_  to a nonexistent
> >address -> nothing happens, no BusError as with
> >any other Mac.
> >For a hardware projekt, I have to make _even_the_Quadra_
> >generate a BusError on illegal writes in NuBus
> >address space.
>
> On the 68030 and 68851, if the memory address is not in the
> translation table (or is marked invalid), a bus error is generated
> whether it is a write or a read.  Apple would not have been able to
> change this-- therefore, either the 68040 changed it, or a bus error
> DOES happen, but the default handler just skips the offending
> instruction.  If the latter, all you have to do is install your own
> bus error.

The 68040 hasn't changed bus error behavior, but the Quadra is a different
beast entirely.  On it, the bus interface will cause an NMI (level 7 interrupt)
when a bus error occurs on a write.  I suggest you contact Mac DTS for more
details on this esoteric subject.

Russell "Mr. Bus Error" Davoli

+++++++++++++++++++++++++++

From: [e-mail] (Kent Sandvik)
Date: 17 Apr 92 23:34:49 GMT
Organization: MacDTS Mongols

In article <[e-mail]>, [e-mail] (Matthew T.
Russotto) writes:
> On the 68030 and 68851, if the memory address is not in the
> translation table (or is marked invalid), a bus error is generated
> whether it is a write or a read.  Apple would not have been able to
> change this-- therefore, either the 68040 changed it, or a bus error
> DOES happen, but the default handler just skips the offending
> instruction.  If the latter, all you have to do is install your own
> bus error.

..our group is just now looking into the issue of installing own
bus handlers, it's doable but skanky, many other managers create temp
patches, and the chain of vectors could be broken due to an interrupt.
So the best way is to turn off interrupts during the bus vector patch,
then again this is not good either...

Cheers,
Kent

+++++++++++++++++++++++++++

From: [e-mail] (Joachim Ostheimer)
Organization: Techn. University of Berlin, Germany
Date: Sat, 18 Apr 1992 01:36:13 GMT


In article <[e-mail]>, [e-mail] (Matthew T.
Russotto) writes:
> On the 68030 and 68851, if the memory address is not in the
> translation table (or is marked invalid), a bus error is generated
> whether it is a write or a read.  Apple would not have been able to....

This is not the only reason for a Bus Error.
When the Address is marked valid in the Translation Table
 and is a Nubus Address
 and the addressed Card does not answer
     ( for Example when there is no Card )
 then also a Bus Error is generated.

 The only exception is the Quadra 700 or 900 where write Accesses
 to an Empty NuBus Slot are ignored and cause no Bus Error.

* * *
 In his article  [e-mail] (Kent Sandvik) writes:
> ..our group is just now looking into the issue of installing own
> bus handlers, it's doable but skanky ...

The Handler even has to check out the CPU Type, 24 or 32 Bit addressing Mode
and VirtualMemory ...

       Joachim  Ostheimer

+++++++++++++++++++++++++++

From: [e-mail] (Matthew T. Russotto)
Date: Mon, 20 Apr 92 03:21:39 GMT
Organization: College of Engineering, University of Maryland, College Park

In article <[e-mail]> [e-mail] (Joachim Ostheimer) writes:
>
>In article <[e-mail]>, [e-mail] (Matthew T.
>Russotto) writes:
>> On the 68030 and 68851, if the memory address is not in the
>> translation table (or is marked invalid), a bus error is generated
>> whether it is a write or a read.  Apple would not have been able to....
>
>This is not the only reason for a Bus Error.
>When the Address is marked valid in the Translation Table
> and is a Nubus Address
> and the addressed Card does not answer
>     ( for Example when there is no Card )
> then also a Bus Error is generated.
>
> The only exception is the Quadra 700 or 900 where write Accesses
> to an Empty NuBus Slot are ignored and cause no Bus Error.

Hmm-- maybe it would be possible to determine which nubus slots were
empty and MARK them invalid in the translation table, if bus error on
write to empty slot is necessary.
- --
Matthew T. Russotto    [e-mail]    [e-mail]
Some news readers expect "Disclaimer:" here.
Just say NO to police searches and seizures.  Make them use force.
(not responsible for bodily harm resulting from following above advice)

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

Organization: University of Illinois at Chicago
Date: Saturday, 18 Apr 1992 23:56:25 CDT
From: John Galidakis <[e-mail]>
Subject: Segment loader question.

 Hello mac gurus.
I was considering purchasing a 680x0 assembler the other day,
and I've decided it would be cheaper (and more fun) to sit down
and write one. (pretty radical, huh?)
 I have a pretty good grasp of what's involved, except one minor
point. I want to be able to create multiple segments. Now:
Specifically, as soon as the assembler encounters a directive
(say {$I segName}) I want The following code to be stored on
the segment of my choice. I know that the segment loader uses
the jump table (CODE id=0) to keep track of what's loaded and
what's not. (Entry containing the routine offset from begginning
of segment, MOVE.W #segid,-(SP), and _LoadSeg trap)
 Question is, is that all I need in the creation of a multisegment
app?. I.E., is the following strategy correct?
Assembler: encounter {$I...} directive.
add one more entry to the jump table with offset=end of segment id=
segName and then append the generated code there.
Repeat, until {$I segName2} is reached, etc.
What about CODE=1 which loads anyway?
Any help is greatly appreciated.
J.Galidakis-programmer/Designed Data Co.
"Things are simple; don't complicate  them..."

+++++++++++++++++++++++++++

From: [e-mail] (Jon W{tte)
Date: 19 Apr 92 14:36:54 GMT
Organization: Royal Institute of Technology, Stockholm, Sweden

.edu> [e-mail] (John Galidakis) writes:

   the jump table (CODE id=0) to keep track of what's loaded and
   what's not. (Entry containing the routine offset from begginning
   of segment, MOVE.W #segid,-(SP), and _LoadSeg trap)

That's the format.

   What about CODE=1 which loads anyway?

Does it ? I thought all the loader did was jump to table entry one
in the jump table. By convention, that always refers to segment one.

Or am I off track ?

- --
"You should meet yourself someday. I'm sure you would hate it."
- - Me: [e-mail]; Jon W{tte (The Diplomat - NOT!)

+++++++++++++++++++++++++++

From: [e-mail] (Rich Siegel)
Organization: GCC Technologies
Date: Sun, 19 Apr 1992 15:18:13 GMT

In article <[e-mail]> John Galidakis <[e-mail]> writes:
> Hello mac gurus.
>I was considering purchasing a 680x0 assembler the other day,
>and I've decided it would be cheaper (and more fun) to sit down
>and write one. (pretty radical, huh?)
> I have a pretty good grasp of what's involved, except one minor
>point. I want to be able to create multiple segments. Now:
>Specifically, as soon as the assembler encounters a directive
>(say {$I segName}) I want The following code to be stored on
>the segment of my choice. I know that the segment loader uses
>the jump table (CODE id=0) to keep track of what's loaded and
>what's not. (Entry containing the routine offset from begginning
>of segment, MOVE.W #segid,-(SP), and _LoadSeg trap)
>"Things are simple; don't complicate  them..."

You're missing the forest for the trees. The creation of multiple
segments and the jump table is the function of the linker, not
the assembler. All the assembler itself really needs to do is to
emit code and relocation information so that the linker can patch
up the proper code offsets at link time. The standard format for
representing this information on the Mac is the MPW .O format,
which is understood and interpreted by most development systems'
linkers, including MPW, THINK Pascal, and (with conversion) THINK
C.

For maximum usefulness, your assembler should emit .O files
as its output. Not only will this broaden its appeal, it will
remove the burden of writing a linker from your shoulders.
Another good goal is to be source-compatible with the MPW
assembler, and to offer a few features that the MPW assembler
doesn't that people just can't live without. That way, people
have a compelling reason to switch, and will be able to do
so easily, without investing a lot of time in porting.

R.


- --
- -----------------------------------------------------------------------
Rich Siegel                              Internet: [e-mail]
Software Engineer, Quickdraw Group
GCC Technologies

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

From: [e-mail] (Brendan T. McNichols)
Subject: Action! information?
Date: Mon, 13 Apr 92 17:01:21 EDT


Hi all,

I have heard of a programming tool called Action! from Expertelligence
Inc.  Action! is a toolkit which allows a programmer to create user
interfaces and build programs visually (ala NeXT Interface Builder).
Does anyone know a telephone number, US Mail address, or email address
for Expertelligence so I can get more information on this software?

Thanks,
Brendan

- --
Brendan T. McNichols, Computer Engineer               [phone]
Psyracuse U. Sychology Dept.             [e-mail] (NeXT)
430 Huntington Hall               [e-mail] (ASCII)
Syracuse, NY 13244                              [e-mail]

+++++++++++++++++++++++++++

From: [e-mail] (Kent Sandvik)
Date: 18 Apr 92 23:47:53 GMT
Organization: MacDTS Mongols

In article <[e-mail]>,
[e-mail] (Brendan T. McNichols) writes:
>
>
> Hi all,
>
> I have heard of a programming tool called Action! from Expertelligence
> Inc.  Action! is a toolkit which allows a programmer to create user
> interfaces and build programs visually (ala NeXT Interface Builder).
> Does anyone know a telephone number, US Mail address, or email address
> for Expertelligence so I can get more information on this software?

This toolkit is for MCL (Macintosh Common Lisp), but if you are interested
in MCL, go ahead. Sorry, I don't have their email address just now.

Cheers,
Kent

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

From: [e-mail] (Danny McCampbell)
Subject: (Q): Does anyone know about 'ictb' rsrcs?
Date: 15 Apr 92 12:21:24 GMT
Organization: University of Tennessee

Hi all.

I am adding color to the buttons in my dialogs and I have just encountered
the 'ictb' resource.  The only editor for it is the hex editor.  Does
anyone know of another editor that can be used?  I cannot really understand
where what goes where.

Also, is there a list of the hex codes for all colors that will tell me
what hex numbers I need to put together to yield certain colors?

Some info would really be appreciated.  Thanks in advance.

Danny

+++++++++++++++++++++++++++

From: [e-mail] (Pete Gontier)
Date: 16 Apr 92 05:56:17 GMT


 DM> From: [e-mail] (Danny McCampbell)

 DM> I am adding color to the buttons in my dialogs and I have just
 DM> encountered the 'ictb' resource. The only editor for it is the
 DM> hex editor. Does anyone know of another editor that can be used?

Resorcerer does a wonderful job of this editing where nothing else does.
Contact info is in this group's FAQ.

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

From: [e-mail] (Alen Shapiro)
Subject: COMBINATORIX...math-type programming question?
Date: 16 Apr 92 20:27:04 GMT
Organization: Crash TimeSharing, El Cajon, CA

ok you budding math-type mac programmers...here's a question...

I need to figure out how many ways n objects may be picked from a set
of m objects. Order is not relevant (ab = ba).

Is the answer (m-n)! (!=factorial)?...this works for the boundary condition
where m=n only if 0! = 1 and my math is too shaky for this.

Any takers
- --alen
[e-mail]

mail me and I'll summarize...book recommendations appreciated.

+++++++++++++++++++++++++++

From: [e-mail] (Matthew Hall)
Date: 17 Apr 92 19:33:46 GMT
Organization: Oberlin College Computer Science

In article <[e-mail]> [e-mail] (Alen Shapiro) writes:

>   ok you budding math-type mac programmers...here's a question...
>
>   I need to figure out how many ways n objects may be picked from a set
>   of m objects. Order is not relevant (ab = ba).
>
>   Is the answer (m-n)! (!=factorial)?...this works for the boundary condition
>   where m=n only if 0! = 1 and my math is too shaky for this.
>

well, lets see.
First pick - m possible choices
Second pick - m-1 possible choices
nth pick m-n+1 choices

algorithm = m*(m-1)*(m-2)*...*(m-n+1)

      = m!
      -----
      (m-n)!  = works for boundary condition.  In fact, it just works.


- -matt hall
- --


- -------------------------------------------------------------------------------
Matt Hall.    [e-mail]  OR  [e-mail]
              (216)-775-5805 (That's a Cleveland Area code. Lucky Me)

"If a man comes up to you and says:
    'A dog just carried away your ear.'
Do you run after the dog, or search first for your ear?" - Moon over Morocco


+++++++++++++++++++++++++++

From: [e-mail] (Steven J. Edwards)
Date: 17 Apr 92 20:04:05 GMT
Organization: Bull HN, Worldwide Information Systems, Billerica, Mass., USA

In article <[e-mail]> [e-mail] (Matthew Hall) writes:
# In article <[e-mail]> [e-mail] (Alen Shapiro) writes:
#
#>    ok you budding math-type mac programmers...here's a question...
#>
#>    I need to figure out how many ways n objects may be picked from a set
#>    of m objects. Order is not relevant (ab = ba).
#>
#>    Is the answer (m-n)! (!=factorial)?...this works for the boundary condition
#>    where m=n only if 0! = 1 and my math is too shaky for this.
#>
#
# well, lets see.
# First pick - m possible choices
# Second pick - m-1 possible choices
# nth pick m-n+1 choices
#
# algorithm = m*(m-1)*(m-2)*...*(m-n+1)
#
#         = m!
#         -----
#         (m-n)!  = works for boundary condition.  In fact, it just works.

    Nope.  Remember, he said that order was unimportant.  The
correct answer is:

    C = n! / ((n - k)! * k!)

For example, the number of different five card poker hands that can
be dealt from a fifty two card deck is:

    C = 52! / (47! * 5!) = 2598960

For thirteen card bridge hands:

    C = 52! / (39! * 13!) = approx 6.35 * 10 ^ 11

    I guess that there just aren't that many card players among
Mac programmers.

 [The above opinions expressed are my own; not necessarily held by others.]
      == Steven J. Edwards           Bull HN Information Systems Inc. ==
      == [phone]              300 Concord Road         MS 820A ==
      == [e-mail]     Billerica, MA 01821          USA ==
"That Government which Governs the Least, Governs Best." -- Thomas Jefferson

+++++++++++++++++++++++++++

From: [e-mail] (Stephen L Chamberlin)
Date: 17 Apr 92 20:11:25 GMT
Organization: Massachusetts Institute of Technology


>>   ok you budding math-type mac programmers...here's a question...
>>
>>   I need to figure out how many ways n objects may be picked from a set
>>   of m objects. Order is not relevant (ab = ba).
>>
>>   Is the answer (m-n)! (!=factorial)?...this works for the boundary condition
>>   where m=n only if 0! = 1 and my math is too shaky for this.
>>
>
>well, lets see.
>First pick - m possible choices
>Second pick - m-1 possible choices
>nth pick m-n+1 choices
>
>algorithm = m*(m-1)*(m-2)*...*(m-n+1)
>
>      = m!
>      -----
>      (m-n)!  = works for boundary condition.  In fact, it just works.
>

This is correct when ordering is important. The original poster wanted to
find the formula when order is _not_ relevant.

The correct formula is "m choose n," or:

     m!
  --------
  n!(m-n)!


Have fun!

+++++++++++++++++++++++++++

From: [e-mail] (Matthew T. Russotto)
Date: Fri, 17 Apr 92 21:25:38 GMT
Organization: College of Engineering, University of Maryland, College Park

In article <[e-mail]> [e-mail] (Matthew Hall) writes:
>In article <[e-mail]> [e-mail] (Alen Shapiro) writes:
>
>>   ok you budding math-type mac programmers...here's a question...
>>
>>   I need to figure out how many ways n objects may be picked from a set
>>   of m objects. Order is not relevant (ab = ba).
>>
>>   Is the answer (m-n)! (!=factorial)?...this works for the boundary condition
>>   where m=n only if 0! = 1 and my math is too shaky for this.
>>
>
>well, lets see.
>First pick - m possible choices
>Second pick - m-1 possible choices
>nth pick m-n+1 choices
>
>algorithm = m*(m-1)*(m-2)*...*(m-n+1)
>
>      = m!
>      -----
>      (m-n)!  = works for boundary condition.  In fact, it just works.

Nope, doesn't work.  Remember, order doesn't matter.  It is

   m!
- --------
n!(m-n)!


- --
Matthew T. Russotto    [e-mail]    [e-mail]
Some news readers expect "Disclaimer:" here.
Just say NO to police searches and seizures.  Make them use force.
(not responsible for bodily harm resulting from following above advice)

+++++++++++++++++++++++++++

From: [e-mail] (Brad Hards)
Organization: Organisation! Not likely around here
Date: Fri, 17 Apr 1992 11:39:48 GMT

In article <[e-mail]>, [e-mail] (Alen Shapiro) writes:
> I need to figure out how many ways n objects may be picked from a set
> of m objects. Order is not relevant (ab = ba).
>
> Is the answer (m-n)! (!=factorial)?...this works for the boundary condition
> where m=n only if 0! = 1 and my math is too shaky for this.

Nope. The answer (I hope!) is m!/(n)!(m-n)! , where ! is the factorial
operator i.e 5!=5.4.3.2.1, and 1! = 0! = 1.
It is wprth noting that when you want to calculate it, it is possible to
reduce the computational overhead by noting that m!/(m-n)! can be simplified
to m.(m-1).(m-2)...(n+1), and also that the if you want to choose n items from
a set of m, that is identical to chooing (m-n) from the same set of m. If any
of this doesn't make sense ( I can't do the proper notation in ascii ) let
me know, and I'll do this again with a better word processor.

As for books, any high school probability book will cover this -> if it
sounds like "Introduction to Probability" it will do this.

Brad Hards


+++++++++++++++++++++++++++

From: [e-mail] (The Weasel)
Date: 20 Apr 92 16:18:24 GMT
Organization: DSG, Stanford University, CA 94305, USA

In article <[e-mail]> [e-mail] (Alen Shapir
o) writes:

>   ok you budding math-type mac programmers...here's a question...
>
>   I need to figure out how many ways n objects may be picked from a set
>   of m objects. Order is not relevant (ab = ba).
>
>   Is the answer (m-n)! (!=factorial)?...this works for the boundary condition
>   where m=n only if 0! = 1 and my math is too shaky for this.
>

What you want is the combination function. It is (n!)/(m! * (n-m)!)

Peter

- --
Peter Chang                        |  "I'm sure I would be much more upset
E-Mail: [e-mail] |   if I weren't under such heavy sedation."
Snail Mail: PO Box 9603            |                    David St. Hubbins
            Stanford, CA  94309    |

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

From: [e-mail] (Spencer W. Thomas)
Subject: Programming Apple OneScanner?
Organization: University of Michigan HSITN
Date: Fri, 17 Apr 1992 22:01:50 GMT

This may be the second posting of this question.  If so, I apologize
- -- our news feed has been flaky.

I am looking for programming information on the Apple OneScanner.  I
need to write a special-purpose scanning application.  Does anybody
know where I can get this info?  (If the answer is APDA, please feel
free to direct a kick at my posterior.)

Thanks.

- --
=Spencer W. Thomas         HSITN, U of Michigan, Ann Arbor, MI 48109
[e-mail]    313-747-2778

+++++++++++++++++++++++++++

From: [e-mail] (Spencer W. Thomas)
Date: 13 Apr 92 16:42:24 GMT
Organization: University of Michigan HSITN

Does anyone know where I can get information on programming the Apple
OneScanner?  Would APDA have it (I'm not a member, but others here
are)?

Please respond by mail.

- --
=Spencer W. Thomas         HSITN, U of Michigan, Ann Arbor, MI 48109
[e-mail]    313-747-2778

+++++++++++++++++++++++++++

From: [e-mail] (Brian Bechtel)
Date: 18 Apr 92 14:08:21 GMT
Organization: Apple Computer Inc., Cupertino, CA

[e-mail] (Spencer W. Thomas) writes:

>Does anyone know where I can get information on programming the Apple
>OneScanner?  Would APDA have it (I'm not a member, but others here
>are)?

R0138LLA       Programmer's Guide to Apple Scanners $25.00

The APDA price lists are available for anonymous ftp on ftp.apple.com,
in the directory /dts/apda/price-lists.  Also on that server is a
variety of other information about APDA, in /dts/apda.

You can also send a message to [e-mail].  By the way,
APDA hasn't had a membership fee for a while now...

- --Brian Bechtel     [e-mail]     "My opinion, not Apple's"

+++++++++++++++++++++++++++

From: [e-mail]
Date: 19 Apr 92 02:48:36 GMT
Organization: Adept Solutions

In article <[e-mail]> [e-mail]
(Spencer W. Thomas) writes:
> This may be the second posting of this question.  If so, I apologize
> -- our news feed has been flaky.
>
> I am looking for programming information on the Apple OneScanner.  I
> need to write a special-purpose scanning application.  Does anybody
> know where I can get this info?  (If the answer is APDA, please feel
> free to direct a kick at my posterior.)
>
The answer _IS_ APDA. I'll leave the kicking to yourself. ;)
- --
Roy Lovejoy      | internet:  [e-mail]
Head RGB Guy     | AppleLink: adept
Adept Solutions  | CIS:       72447,1447
.................| dual certified developer: NeXT & Apple ;)

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

From: [e-mail] (Herb Poppe)
Subject: Serial Port Programming
Date: 15 Apr 92 16:12:49 GMT
Organization: National Center for Atmospheric Research

In article <[e-mail]> [e-mail] (Bradley
Green ) writes:

> I'm trying to run a distributed processing network, using a Mac as the
> master controller, with dedicated microcontrollers at the oother nodes
> of the network.  What I want to do is to be able to run the Mac serial
> port at its maximum asynchronous speed, which I believe is 230Kbaud, but
> I haven't found any real info on how to program the serial port to do
> this.  Note that I don't want to use AppleTalk protocols, but just to
> send data byte-by-byte.

The Mac's serial driver (software) is limited to a maximum speed of
57.6KB. According to the conventional wisdom, this is the maximum speed
that Apple could achieve with an interrupt driven driver on the original
Mac. Note that the 230KB AppleTalk drivers are not event driven. It is
probably the case that Macs like the IIsi, etc. could support higher
interrupt driven baud rates, but the current serial driver has no
mechanism for specifying a baud rate higher than 57.6KB.

You will have to write you own driver if you want to run at higher baud
rates. This means you will have to get the technical details on the Zilog
SCC.

Suppose you decide to live with the 57.6KB limit and use the Apple serial
driver. Depending on how you wire your network, there may be another
consideration that may not be handled by the current serial driver. Are
you using RS-485 transcievers like the SN75176 with your microcontrollers;
that is, is your network a single twisted pair? If so, then only one
RS-485 driver on the network can be enabled at any one time. This includes
the driver in the Mac. I do not know if the Mac serial driver always
enables the driver before each transmission and then disables it at the
end of transmission, or whether it enables the driver at open and disables
it at close, or whether it is always enabled. You can get around this
problem by using two twisted pair and separate RS-485 drivers and
receivers with your micros. The Mac's driver would be wired to all the
micros' receivers and the micros' drivers would all be wired to the Mac's
receiver. Then it would not be an issue if the Mac's driver were always
enabled.

At some point in the future, I plan to implement a network like the one
you mention. For me, the 57.6KB limit is not a problem, but I will need to
know how the Mac serial driver handles enabling/disabling the RS-422
driver. If anyone can answer that question, I'd like to know.

If you decide you need to write your own driver, I seem to recall that at
one time Apple made the source code of a (maybe not THE) serial driver
available. If I remember correctly, does anyone remember how to get this
code?

Herb Poppe             [e-mail]
NCAR                      [phone]
1850 Table Mesa Dr.
Boulder, CO  80307-3000

+++++++++++++++++++++++++++

From: [e-mail] (Jack Brindle)
Organization: Motorola, Mobile Data Division - Seattle, WA
Date: Wed, 15 Apr 1992 21:01:01 GMT

In article <[e-mail]> [e-mail] (Herb Poppe) writes:
>In article <[e-mail]> [e-mail] (Bradley
>Green ) writes:
>
>> I'm trying to run a distributed processing network, using a Mac as the
>> master controller, with dedicated microcontrollers at the oother nodes
>> of the network.  What I want to do is to be able to run the Mac serial
>> port at its maximum asynchronous speed, which I believe is 230Kbaud, but
>> I haven't found any real info on how to program the serial port to do
>> this.  Note that I don't want to use AppleTalk protocols, but just to
>> send data byte-by-byte.
>
>The Mac's serial driver (software) is limited to a maximum speed of
>57.6KB. According to the conventional wisdom, this is the maximum speed
>that Apple could achieve with an interrupt driven driver on the original

The 57.6KB limit has far more to do with limitations imposed by the
hardware designers at Zilog who did the 8530 than the software folks
at Apple. Quite simply, the limitation is imposed by the use of the
internal SCC baud rate generation clock. Take a good look at the SCC
technical manual; it will explain why. You will also find out that the
SCC will go faster than 56KB for asynchronous format if you supply an
external clock.

>Mac. Note that the 230KB AppleTalk drivers are not event driven. It is

They also run quite different. The incoming LocalTalk frame causes an
interrupt when the address matches the one programmed into the SCC. The
remaining bytes of the frame are then read during that first interrupt
service response. So, it's one interrupt per packet.

>probably the case that Macs like the IIsi, etc. could support higher
>interrupt driven baud rates, but the current serial driver has no
>mechanism for specifying a baud rate higher than 57.6KB.

Only if the clocking rate of the SCC is changed, or external clocking is
supplied. The SCC has to run faster, not the CPU. Again, this is simply
not primarily  a software issue!

>
>You will have to write you own driver if you want to run at higher baud
>rates. This means you will have to get the technical details on the Zilog
>SCC.

... And run in synchronous mode. Writing a synchronous driver for the
SCC can be quite tricky. There _is_ some *magic* that has to be applied.
Zilog's technical notes make it a lot easier these days, but there still
is a pretty good learning curve.

>
>Suppose you decide to live with the 57.6KB limit and use the Apple serial
>driver. Depending on how you wire your network, there may be another
>consideration that may not be handled by the current serial driver. Are
>you using RS-485 transcievers like the SN75176 with your microcontrollers;
>that is, is your network a single twisted pair? If so, then only one
>RS-485 driver on the network can be enabled at any one time. This includes
>the driver in the Mac. I do not know if the Mac serial driver always
>enables the driver before each transmission and then disables it at the
>end of transmission, or whether it enables the driver at open and disables
>it at close, or whether it is always enabled. You can get around this

In async mode, the SCC always drives the output line.

>problem by using two twisted pair and separate RS-485 drivers and
>receivers with your micros. The Mac's driver would be wired to all the
>micros' receivers and the micros' drivers would all be wired to the Mac's
>receiver. Then it would not be an issue if the Mac's driver were always
>enabled.

The output ports on the Mac are controlled by the SCC's RTS signal. This
signal would need to be controlled to avoid problems. By the way this is
not simply a problem of blocking other transmissions. Directly connecting
the outputs of two drivers that are in opposite states can cause a very
large current flow through both devices, and will probably end up damaging
one or both devices.  The other way to resolve the situation is to do what
Apple did with LocalTalk, and provide some form of isolation for each Mac.
The LocalTalk interface box effectively provides DC isolation for each Mac,
while passing the rapidly changing data signals onto the main bus. Possibly
the best way to do this in this situation is to use an RS442 to open-collector
interface. This is also the method used for the ADB. You can then simply
tie the outputs of each open-collector buffer together on the network bus.
You would also need a pull-up resistor to make the network voltage level
sit high when no-one is driving it. I would probably opt for an actual
discrete transistor for this purpose (say a 2N4124, available at Radio Shack)
rather than use an IC. You only need one per node, and the transistor has
higher current capabilities. This would be useful for driving longer cable
lengths.

>
>At some point in the future, I plan to implement a network like the one
>you mention. For me, the 57.6KB limit is not a problem, but I will need to
>know how the Mac serial driver handles enabling/disabling the RS-422
>driver. If anyone can answer that question, I'd like to know.

See above. The RTS line does it.

>
>If you decide you need to write your own driver, I seem to recall that at
>one time Apple made the source code of a (maybe not THE) serial driver
>available. If I remember correctly, does anyone remember how to get this
>code?

If it wasn't THE driver, it sure was a good fake. I remember it quite well.
I have no idea where you could get it. I have a feeling that Apple's
driver will suit the original poster's purpose quite well. All that is
needed is a bit of cleverness on the hardware side to make things play
together. It's not hard to do, and is a great opportunity to learn things.

By the way, you might wish to look at an MC14469 Addressable Asynchronous
Receiver/Transmitter. It is a really neat device that allows you to set
an address for each AART, and send data on a common bus to each device.
It will respond only to data sent to its address. It's only catch is that
it runs rather slow, max data rate is 9600 baud. Still, the addressing
concepts and usage might provide some good ideas.

Jack Brindle
ham radio: wa4fib
internet: [e-mail]


+++++++++++++++++++++++++++

From: [e-mail] (Bradley Green )
Date: 16 Apr 92 02:39:05 GMT
Organization: University of Pennsylvania

Back to the original poster (me), I guess.  The conclusion I have drawn from
this debate is that trying to run the serial interface at over 57,600 baud is
more trouble than its worth; I'd rather write my software to keep traffic
low.

However, the discussion of the RS-422 interface has left me confused.  I
thought that this standard was just a 'fast' RS-232 with most of the
control pins removed; you can still hook up an RS-232 to the Mac as
long as it runs under 19200 baud.  However, the hardware discussions I've
seen have led me to realize that this may not be the case.  What are
the differences between the RS-232 and the RS-422?  What is the easiest
manner in which to implement the latter (422) in hardware?

- --
*----------------------------------------------------------------------------*
* Bradley S. Green                University of Pennsylvania   *
* [e-mail]            Dept. Of Systems Engineering *
*----------------------------------------------------------------------------*

+++++++++++++++++++++++++++

From: [e-mail] (Bill Coleman)
Date: 16 Apr 92 11:00:34 GMT
Organization: Hayes Microcomputer Products, Norcross, GA

In article <[e-mail]>, [e-mail] (Herb Poppe) writes:
>
> The Mac's serial driver (software) is limited to a maximum speed of
> 57.6KB. According to the conventional wisdom, this is the maximum speed
> that Apple could achieve with an interrupt driven driver on the original
> Mac.

Wrong. This is the maximum speed you get when using the internal bit
rate generator the smallest programmable divisor (zero, which causes a
divide by 4) with the internally supplied clock to the Z8530.
ie: 3.672 MHz / 4 = 57375 * 16 (which is close enough to 57600)
A 16x clock is necessary to properly operate asynchronous receivers, since
the receiver doesn't know when the bit transitions will take place (ie
it is not synchronized to the clock)

> Note that the 230KB AppleTalk drivers are not event driven.

Perhaps you mean "interrupt driven." In that case, you are wrong again.
230.4 Kbps LocalTalk *IS* interrupt driven.

The difference here is that LocalTalk is synchronous, not asynchronous,
and uses FM0 encoding. FM0 requires a 4x clock, not 16x, so the speed of
LocalTalk is 230.4 = 4 * 57600.

> It is
> probably the case that Macs like the IIsi, etc. could support higher
> interrupt driven baud rates, but the current serial driver has no
> mechanism for specifying a baud rate higher than 57.6KB.

It is a hardware limitation. If you were able to externally clock the
serial chip, you could run asynchronous rates up to 250 Kbps. (Highest
speed of the clock input is 4 MHz.) Unfortunately, the serial driver has
no provisions for reprogramming the clock inputs, so you'd have to
go directly to the chip.

> You will have to write you own driver if you want to run at higher baud
> rates. This means you will have to get the technical details on the Zilog
> SCC.

Something you should have done to start with. <grin>

> I do not know if the Mac serial driver always
> enables the driver before each transmission and then disables it at the
> end of transmission, or whether it enables the driver at open and disables
> it at close, or whether it is always enabled.

With the serial driver, the "drivers" are always enabled.

> If you decide you need to write your own driver, I seem to recall that at
> one time Apple made the source code of a (maybe not THE) serial driver
> available. If I remember correctly, does anyone remember how to get this
> code?

Apple did make the code available to some developers, back about 1985. This
is no longer the case, primarily because of the changes in the Mac hardware.

Writing your own driver is not recommended, because you will bump right into
compatibility problems with the IIfx and Quadras, and potentially won't work
at all with future Macs.

There's a lot of stuff the serial driver could do, but doesn't have the
interface calls to do. Here's to hoping Apple fixes this in the future.

- --
Bill Coleman, AA4LR                ! CIS: 76067,2327    AppleLink: D1958
Principal Software Engineer        ! Packet Radio: AA4LR @ W4QO
Hayes Microcomputer Products, Inc. ! UUCP: uunet!hayes!bcoleman
POB 105203 Atlanta, GA 30348 USA   ! Internet: [e-mail]
Disclaimer: "My employer doesn't pay me to have opinions."
Quote: "The same light shines on vineyards that makes deserts." -Steve Hackett.


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

End of C.S.M.P. Digest
**********************