PD ROM 1/PD ROM Volume I - Macintosh Software from BMUG (1988).iso/Electronic Messages/USEnet Digests/USEnet Vol. 4/USEnet 4.13

Text File | 1988-04-08 | 30.4 KB | 817 lines | [TEXT/ttxt]

 1-Feb-88 10:02:52-PST,31860;000000000000
Return-Path: <[e-mail]>
Received: from RELAY.CS.NET by SUMEX-AIM.Stanford.EDU with TCP; Mon, 1 Feb 88 10:01:50 PST
Received: from relay2.cs.net by RELAY.CS.NET id ac23115; 1 Feb 88 10:40 EST
Received: from relay.cs.net by RELAY.CS.NET id aa24718; 1 Feb 88 10:25 EST
Received: from sdr.slb.com by RELAY.CS.NET id aa24701; 1 Feb 88 10:22 EST
Date: Mon, 1 Feb 88 09:32 EDT
From: Jeffrey Shulman <[e-mail]>
Subject: Usenet Mac Digest V4 #13
To: [e-mail], [e-mail]
X-VMS-To: in%"[e-mail]",in%"[e-mail]"

Date: Mon 1 Feb 88 09:32:36-GMT
From: Jeff Shulman <SHULMAN@SDR>
Subject: Usenet Mac Digest V4 #13
To: Usenet-List: ;
Message-ID: <570706356.0.SHULMAN@SDR>
Mail-System-Version: <VAX-MM(218)+TOPSLIB(129)@SDR>

Usenet Mac Digest     Sunday, January 31, 1988       Volume 4 : Issue 13

Today's Topics:
     Menus on the title bar.
     JClock vs. MenuClock
     coral allegro common lisp
     3rd party 800K drives
     Merriam-Webster Thesaurus and Multifinder
     NuBus single-chip interface
     Re: TOPS info
     Color Copybits Question
     Math processing
     Re: Macintosh Prolog Recommendations
     List Magager dividers
     Asynchronous SysBeep() on Mac II
     Re: Sorting resources and a question about the list manager
     Re: Are Desktop files a good idea?
     Re: Re: C vs. Pascal for the mac, vs C for MSDOS
     Re: (LSC) Deep Dark Secrets Wanted!
     Re: LaserWriter IISC spec.'s...
     Re: Menus on the title bar.
     Re: Macintosh Prolog Recommendations (2 messages)
     Questions about File/List Managers
     SAS/GRAPH on Mac

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

From: [e-mail] (Mahboud Zabetian)
Subject: Menus on the title bar.
Date: 22 Jan 88 01:00:43 GMT
Organization: Advanced Technology, Princeton University

Does anyone know what the best way is to make a popup menu when you
option-click on the title bar of a window?  I seem to recall having seen
it before.

One way to do that is to fool the mac into thinking that the click is in
the menu bar, and having the menu manager handle the event.  But I don't
want to  do it that way since each window will have its own unique
menu.(plus the standard ones in the menu bar).

Any help will be appreciated.
--
Mahboud Zabetian                [e-mail]
183 Little Hall                     [phone]
Princeton University, Princeton, NJ 08544        [phone]
****** Anyone need a soon-to-graduate hardware/software engineer? ********


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

From: [e-mail] (Mahboud Zabetian)
Subject: JClock vs. MenuClock
Date: 22 Jan 88 01:08:27 GMT
Organization: Advanced Technology, Princeton University

I had used JClock for over a year and had never had anyproblems with it,
until MultiFinder came out and JClock would write over MF's small icon.

Now I use Menu clock which just displays the hour and minutes(as opposed
to JClock's Hour minute and seconds), but I am not satisfied.

With JClock, you knew the time no matter what you were doing.  With
MenuClock the clock stops during certain operations such as mousedown
and downloads.  I don't mean that it would lose time, but that it would
remain unchanged till said operations ended and then would display the
correct time again.

With JClock, when that happened, it was safe to assume that the system
had crashed, but with menuclock, you never know.

Anybody know of a MF compatible JClock?  Thank you.

P.S.  Also, you could get rid of JClock if you clicked on it, but you
can't with MenuClock.
--
Mahboud Zabetian                [e-mail]
183 Little Hall                     [phone]
Princeton University, Princeton, NJ 08544        [phone]
****** Anyone need a soon-to-graduate hardware/software engineer? ********


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

From: [e-mail] (Henry Kautz's mailnews program)
Subject: coral allegro common lisp
Date: 21 Jan 88 19:35:12 GMT
Organization: AT&T Bell Laboratories, Murray Hill

Coral Allegro Common Lisp appears pretty nifty, but the dialog designer
can be hazardous to your Mac II's health.  Using with a SuperMac
Spectrum color board, strange things happened -- random shifts in color.
 Then, POW, the system crashed.  Now it won't boot.  Not from a floppy,
not from the hard disk, period.  No control-shift-etc magic seems to
help.  Maybe parameter RAM is screwed up, but since it won't even boot
from a floppy (it just turns itself off after about 1 second) there is
no way to do the command-shift-option-control panel trick.

So, back to the dealer, who will undoubtably insist we buy a new
motherboard.  Any other suggestions?
--
---- Henry Kautz
:uucp:    allegra!kautz
:arpa: or :csnet: [e-mail]
:old style csnet: [e-mail]
:mail:  AT&T Bell Laboratories
    Room 3C-402A
    600 Mountain Ave.
    Murray Hill, N.J. 07974
:office phone: [phone]


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

From: [e-mail] (benjamin Waldman)
Subject: 3rd party 800K drives
Date: 22 Jan 88 04:14:09 GMT
Organization: Aiken Computation Lab Harvard, Cambridge, MA

A friend of mine wants to buy and 800K external drive, and he's seen
some 3rd party ones that are cheaper than the $300 we have to pay for
Apple drives (university consortium prices).

Since I don't know much about them, I though I'd ask for advice. He's
seen 2 drives - 1)Ehman 800 K external drive, by Ehman engineering
($190) and 2) Phaser 800, by Warp 9 Engineering ($180)

Does anyone know anything about these drives, or can anyone suggest any
other 3rd party drives, and/or a good place to buy one from (Does
MacConnection sell hardware too)?
--
Thanks in advance,
Ben Waldman
[e-mail]


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

From: [e-mail] (Karl C. Kelley)
Subject: Merriam-Webster Thesaurus and Multifinder
Date: 21 Jan 88 17:46:45 GMT

Word Thesaurus published by Target Software, version 1.0v of 2/13/87 is
not functioning under Multifinder.  Gives an error message indicating
you have to use it under Word, but in fact even if you are in Word you
get the same message.  The solution was to go to the ordinary finder.

Response I got when contacting Target Software was that they offer tech
support of the existing product, but that the creator is no longer
updating this program.  I'm a little ambivalent about whether that is
bad or good. The product is "reasonable" for an inexpensive add-on, but
its vocabulary is somewhat limited.

I would be interested to see if anyone knows of the current plans of the
creators, Deneba Systems, Inc.  They are listed as the copyright holder
in the "About Thesaurus" menu.


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

From: [e-mail] (Lawrence A. Dziegielewski)
Subject: NuBus single-chip interface
Date: 21 Jan 88 13:38:33 GMT
Organization: E.I. DuPont Co. Engineering Physics Lab


Does anyone have the name,  address,  and phone number for the company
that was recently written up in MacWeek that has the single-chip NuBus
interface?  Please reply by email.  Thanks in advance for the info.
--
        Lawrence A. Dziegielewski       |       E.I. Dupont Co.
        {uunet!dgis!psuvax1}!eplrx7!lad |       Engineering Physics Lab
        Cash-We-Serve 76127,104         |       Wilmington, Delaware 19898
        MABELL:  [phone]         |       Mail Stop: E357-318


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

From: [e-mail] (Clayton Cramer)
Subject: Re: TOPS info
Date: 21 Jan 88 16:31:53 GMT
Organization: Optilink Corporation, Petaluma, CA

Our system here is:

    Sun 3/280 file server
    many Sun 3/50s
    one Sun 3/60
    a dozen PCs
    a dozen or so Macs
    an Apple LaserWriter Plus
    a Kinetics FastPath box bridging AppleTalk net to Ethernet

The LaserWriter Plus and Macs are on Appletalk network using TOPS to
access the Sun 3/280 file server.  The PCs are on Ethernet, using Sun's
PC-NFS to talk to the Sun 3/280, which also provides the networked
printer support.

We have everything working pretty well now, though our print spooler
will occasionally fail with an very non-descriptive error message, and
require human intervention.  I would strongly encourage anyone
considering a combination like this to reconsider why they are taking
such an approach.

Our particular Kinetics FastPath occasionally fails in a manner that
requires us to reinitialize its software -- yet passes diagnostics. When
this happens, the AppleTalk part stops talking to Ethernet.

There are quirks of trying to use PC software to talk to the LaserWriter
over this torturous path, and it appears that the combination of TOPS
and PC-NFS create problems that preclude using the solutions you would
use for TOPS or PC-NFS alone.

The Appletalk part of our setup works fine -- the PC-NFS part works sort
of OK (though you don't want to use the DOS APPEND command if you are
using PC-NFS), but the combination of all these systems has taken up a
lot of time to get working.  Even now, most Mac users work on hard disks
instead of trusting the network to stay up long enough to complete
editing a document.  (That includes me -- and I'm responsible for
maintaining it).

Clayton E. Cramer


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

From: [e-mail] (Greg Corson)
Subject: Color Copybits Question
Date: 21 Jan 88 22:01:59 GMT
Organization: Math. Dept., Univ. of Notre Dame

If you have a pixmap that is 1 bit deep and copy it to another pixmap
which is half the size and several bits deep will color copybits average
groups of four pixels in the source image to produce one of four grey
scales in the destination?  Here's an example with a 2x2 source array:

Source =  1 1
          0 1

copied to a 1x1 pixmap will you get a black pixel or a 75% grey one?
--
Greg Corson
19141 Summers Drive
South Bend, IN 46637
[phone] (weekdays till 6 PM eastern)
{pur-ee,rutgers,uunet}!iuvax!ndmath!milo



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

From: [e-mail] (Ali Ozer)
Subject: Math processing
Date: 21 Jan 88 21:42:09 GMT
Organization: Stanford University Computer Science Department

--- Seeing all these messages about equation formatters, etc, I thought
I'd mention a program that I hope will soon appear on your dealer's
shelves. The program is "Milo," written by Ron Avitzur, a physics
student at Stanford. He describes the program as a "math processor" ---
just like a "word processor," except it processes equations.

The program does a lot --- It solves equations,  simplifies expressions,
reduces terms, processes matrices, evaluates summations and products,
etc... The goal of the program is to turn the Mac into a scratch pad for
math work --- Normally you can sit in front of your favorite word
processor and just type, even if you do not have a clear idea of what
you want to write. This is easily possible because most word processors
are easy to use and do not interfere with the thinking process. Milo
attempts to do that for math processing --- To give you a environment
where you can type/solve/simply equations, without having to worry about
formatting the equations as you type.

The program provides many menu options, with keyboard equivalents for
all. It'll probably take some time for beginners to get used to, but,
the program is extremely fast once you're used to it. (Ron can type up a
page full of physics equations, all beautifully solved, in under a
minute...) This is very important --- Most easy-to-use programs do not
get faster as you get more accustomed to them... Milo might prove a bit
more difficult to get started in than most Mac programs (mainly due to
the dozens of options), but it does get much faster as you get used to
it...

Milo also provides graphing capabilities, with any level of zooming into
any area of the graph --- this allows you to solve for roots of
equations visually and also allows you to check results obtained through
other methods.

And, of course, you can include text (using multiple fonts/styles) in
your document, as well as all the equations, solutions, and the graphs.

In my opinion, the program is perfect for students doing problem sets
for science and engineering classes. The program will not do your
problem sets for you, but instead, let you get through those tedious and
messy equations without losing terms or minus signs. And it's ease of
use allows you to spend almost no extra time  on problems of
typesetting, formatting, etc.

I think TeX is still a better choice for those doing papers, articles,
or dissertations --- mainly because TeX provides true typesetting with
enough power to satisfy any sort of document processing. But, TeX
doesn't solve equations, just typesets them, so even with TeX, there's
plenty of use for Milo.

The program is available from Sumex, I believe --- in <info-mac>. It's a
demo version that pretty much lets you do most everything, and it also
provides info on how to get in touch with Ron, the author.

Ron took off considerable time from Stanford to work on Milo, and is
currently in the final stages of writing documentation and sorting
through his options for distributing the program. With luck, you'll see
Milo on the dealer shelves within a few months. If you want to contact
him through email (and I think he loves getting mail about Milo!), try
sending mail to [e-mail].

Finally I should mention that I am not associated with Milo in any way,
except that I've seen it and am impressed by it. (I haven't even used
the program for an extended period of time --- mainly because I don't
use Macs... 8-) But, if you have any use for equation processing, and
you use a Macintosh, definitely get your hands on the demo version and
try it out.

Ali Ozer, [e-mail]


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

From: [e-mail] (Mike Newton)
Subject: Re: Macintosh Prolog Recommendations
Date: 22 Jan 88 05:31:32 GMT
Organization: California Institute of Technology


Some comments regarding the recent inquiry regarding Prolog for the Mac:

First, despite the posters comments, CProlog is NOT in the public
domain. Though the academic license is real cheap, it is needed.  I
believe an industry license is big $$.  (Note: I maintain an 'alternate'
version of CProlog that compiles under more systems, like UTS,  than the
original.) I have NOT yet tried to compile it under Mac OS (please dont
ask), though it seems to work under some other (un-named) OS.

I dont know if Quintus or BIM have a Mac version yet, of if they ever
will.  I'm pretty sure that at least BIM was thinking of it.

If Turbo Prolog for the Mac ever comes out _DONT BUY IT_.  It is trash.

A commercial Mac Prolog that I highly recommend is by ALS.  I believe
they are about ready to ship the Mac version.  The PC version has been
out a while.  (Note: i have ties to ALS, so this is biased.  However, it
really seems nice!).  Their phone number is: 315 471 3900.

- mike
--
[e-mail]    {ucbvax!cithep,amdahl}!cit-vax!newton
Caltech 256-80            818-356-6771 (afternoons,nights)
Pasadena CA 91125        Beach Bums Anonymous, Pasadena President

    "Reality is a lie that hasn't been found out yet..."


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

From: [e-mail] (Gregory Dudek)
Subject: List Magager dividers
Date: 21 Jan 88 15:42:12 GMT
Organization: CSRI, University of Toronto

While people are discussing the List manager, is there any easy way to
get dividers between list items rather that defining a custom list?  I
would have though lists with lines between items would be an obvious
thing to include as an option. My apologies if I'm missing something
incredibly obvious.

  If it does require a custom LDEF, and anybody has such a beast
floating around (in source form), please feel free let me see it.
Thanks,
  Greg Dudek
--
Dept. of Computer Science (vision group)    University of Toronto
Reasonable mailers:  [e-mail]
Other UUCP: {uunet,ihnp4,decvax,linus,pyramid,
        dalcs,watmath,garfield,ubc-vision,calgary}!utai!dudek
ARPA: [e-mail]
DELPHI: GDUDEK


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

From: [e-mail] (Dave Platt)
Subject: Asynchronous SysBeep() on Mac II
Date: 22 Jan 88 06:41:33 GMT
Organization: Coherent Thought Inc., Palo Alto CA

I've heard a rumor that someone has written a little INIT which will
convert the SysBeep() call on the Mac II from a synchronous call to the
Sound Manager into an asynchronous call, so that the II doesn't lock up
while the chosen sound is playing.  Can anybody out there point me to a
place where such an INIT (or any system modification of similar effect)
can be found, or [if you've got it] perhaps mail me a BinHex?

advTHANKSance!

--

Dave Platt
  UUCP:    ...!{ames,sun,uunet}!coherent!dplatt
  Internet: coherent![e-mail], [e-mail], [e-mail]


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

From: [e-mail] (John J. Chew III)
Subject: Re: Sorting resources and a question about the list manager
Date: 22 Jan 88 06:43:56 GMT
Organization: The Poslfit Committee

A nicer way to sort things as you put them into a list is to use the
LSearch() function:

  /* for each cell you want to add */
  theCell.h=theCell.v=0;
  LSearch(dataPtr, dataLen, compareProc, &theCell, 1);
  LAddRow(1, theCell.v, 1);
  LSetCell(dataPtr, dataLen, theCell.1);

... where compareProc for a simple alpha sort would be:

pascal int compareProc(ap, bp, al, bl) Byte *ap, *bp; int al, bl; {
  return IUMagString(ap, bp, al, bl)<=0;
  }

However, to answer the original question of how to sort resources, the
best way is to sort them before you create them and to create them in
sorted order.  The second best way is to decide what order they should
be in, create a new resource fork and create copies of them in sorted
order there.  If you are determined to sort existing copies of
resources, read the caveat in IM IV-16 and then:

  - obtain handles to all the resources to be sorted
  - call RsrcMapEntry on each to get the resource reference offset
  - sort the resource references (12 bytes located at the
    above-obtained offsets from TopMapHndl, assuming your
    resources are in the most-recently opened resource fork,
    which they had better be)
  - call ChangedResource() on one of your resource handles.

jjc
--
john j. chew (v3.0)                       [e-mail]
[phone] (300/1200 bps)            [e-mail]
{cbosgd,decvax,mnetor,utai,utcsri,{allegra,linus}!utzoo}!utgpu!poslfit
"Script-G for open, sub-delta for durchschnitt"


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

From: [e-mail] (Steven Martin)
Subject: Re: Are Desktop files a good idea?
Date: 22 Jan 88 01:18:01 GMT
Organization: TI Computer Science Center, Dallas

My solution is to "open" the trash can.  You then can resize its window
and move it to the lower right corner (or anywhere else). Steve Martin
         USENET: {ctvax,im4u,texsun,rice}!ti-csl!martin
                        PHONE: [phone] If you think of
something, it exists somewhere.


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

From: [e-mail] (Ted Johnson)
Subject: Re: Re: C vs. Pascal for the mac, vs C for MSDOS
Date: 20 Jan 88 16:52:36 GMT
Organization: HP Design Tech Center - Santa Clara, CA


I talked to a THINK representative at MacWorld last Sat. about a
debugger for LSC and he said:

    (1)he couldn't give me a date for when the debugger
           would be out, but it would definitely be "before
           it gets cold again (in Massachusetts)".

    (2)the debugger will be in LSC 3.0.  This will be a
           MAJOR upgrade, so they will probably charge for
           it.  He couldn't give me any $ estimates though.

    (3)they're trying to make the debugger in LSC look
           similiar to the one in LSP.  From what I've heard
           of LSP, this seems to be a good idea.



    -Ted


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

From: [e-mail] (Oliver Steele)
Subject: Re: (LSC) Deep Dark Secrets Wanted!
Date: 22 Jan 88 02:44:33 GMT
Organization: University Of North Carolina, Chapel Hill

[e-mail] (Ted Johnson) writes:
>
>Would someone please share the DEEP DARK programming secrets on how to:
>
>    (1)black out the ENTIRE screen (including JCLOCK)?  I have seen several
>           screen savers that do this...  What is the magical ROM routine
>           that I have to call?

There are experts on this on the net, but (deep breath) here goes:
    GrafPort myPort;
    OpenPort(&myPort);    /*does a SetPort too*/
    EraseRect(&myPort.portRect);

This is OK for a stand-alone, MultiFinder-animous application.  If you
want something that will trigger update events when it goes away, try
    Rect bounds;
    bounds = WMgrPort->portRect;
    myWindow = NewWindow(0,&bounds,"\pI may appear on the top screen",
        false,0,-1,false,0);
    ShowWindow(myWindow);
    SetPort(myWindow);
    RectRgn(myWindow.port.visRgn, &bounds);
    EraseRect(&bounds);
    ...
    RectRgn(((WindowPeek)myWindow).strucRgn, &bounds);
    DisposeWindow(myWindow);
    DrawMenuBar();
    /* or something like this.  I'm not at home, so I can't
       try this out */

Remember that if you do some drawing after you black out the screen, you
need to erase the four corners yourself.

Neither of these keeps JCLOCK from doing its thing.  JCLOCK is run off
of the VBL interrupt, and doesn't write into a window, so you need to
  o Disable VBL interrupts.  Nasty.
  o Change ScreenBits to a buffer area.  Nasty.
  o Flip to the alternate screen buffer.  Only works on a Plus and
earlier.
  o Change all the colors to black.  Only works on a II (or later? :-)

>    (2)make a DA live forever (i.e., remain active across launching
>           different applications from the non-multi Finder), unless
>           you explicitly close it by clicking on it's close box.
>       I assume this means patching one or more of the ROM routines....

Set the DA up to launch a VBL task.  I think da-insect at sumex has C
source for this.

>    (3)Write an INIT?

    main(){SysBeep(7);} Set the code type to 'INIT', any number.  Build a
file called Beeper.  User ResEdit or FileMaster or some such to change
the type of Beeper to 'INIT'.  Put Beeper in your System Folder. Set
your sound level to something besides 0.  Reboot.  You get the idea.

"The interrupt system, memory manager, and ROM-based I/O drivers have
already been initialized by the ROM boot code.  The boot code
initializes the Event Manager, the Font Manager, the Resource Manager,
and the file system.  (Althought the toolbox is initialized at this
point, MacsBug does not use the Toolbox.)"  This is from the docs for
MacsBug, and tells the state of the world when MacsBug is loaded.  Since
INITs are loaded at about the same time, you're probably safe assuming
you have use of the above managers from an INIT.  Just remember not to
use any A5-relative locations, such as thePort (in InitGraf(&thePort)),
but I think the LSC documentation talks about this.
--
------------------------------------------------------------------------------
Oliver Steele                  ...!{decvax,ihnp4}!mcnc!unc!steele
                               [e-mail]

"What is L-marked?  Theta-marked.  I mean, the mind
 boggles at the idea of a VP being theta-marked!"    -- Gary Hendrick


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

From: [e-mail] (Eric K. Olson)
Subject: Re: LaserWriter IISC spec.'s...
Date: 22 Jan 88 18:01:53 GMT
Organization: Lexington Software Design

The press release I read implied that the LWIISC did the drawing in the
Mac and downloaded the page bitmap to the LWIISC.  The SCSI port was
touted as keeping the throughput reasonable even though the LWIISC
didn't have any onboard processing.

Regardless of how they are doing that, I have a different question:  Is
the LWIISC capable of printing bitmaps greater than 1 bit deep
[PixMaps]?  I know that the GCC PLP is treated as a 1-bit deep grafport,
and so it cuts off multi-bit images at 50% gray and prints that [the
same thing you'd get if you looked at the PixMap with the display set to
2 gray levels].  The LW and LW+ do the same, but you can get around it
by using the PostScript "Image" operator [in Encapsulated Postscript if
you need transferability to other applications].

Since the printer engine is inherently a 1-bit device, the LWIISC would
have to dither the image (like PS image operator does) to get it to
print.

BTW, so that I can handle all the various cases correctly, how do I tell
what KIND of printer I am outputting to [if Imagewriter, I have to
dither by myself, if LW or LW+, I use PS "image" operator, if GCC PLP, I
dither, if LWIISC, I either dither or print a PICT].  Anybody know?

Thanks!

-Eric
--
Eric K. Olson     [e-mail]     harvard!endor!olson     D0760
   (Name)                (ArpaNet)                 (UseNet)        (AppleLink)


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

From: [e-mail] (David W. Berry)
Subject: Re: Menus on the title bar.
Date: 22 Jan 88 20:05:07 GMT
Organization: Apple Computer Inc., Cupertino, USA

    Presuming your an application and not a desk accessory you should have
something like the following:

    switch(event.what)
    {
    ....
    case mouseDown:
        switch(FindWindow(&event.where, &window))
        {
        ...
        case    inDrag:

Change what normally comes after inDrag, to test the modifiers bits and
call PopUpMenu (a new trap, test for it's existence as in some tech note
or other) if the option key is down, otherwise handle inDrag as normal.

    Remember that option-click in the title bar isn't something defined by
the UI guidelines, so you will have to prominently document it (ie.,
don't bury it in the manual, nobody reads it anyway) or people won't
know whatever feature you implement that way exists.

    If you're a DA it gets a lot harder because the system handles the
clicks in your drag region (ie., the title bar)
--
    David W. Berry
    [e-mail]                   dwb@Delphi
    [e-mail]                   [e-mail]
Disclaimer: Apple doesn't even know I have an opinion and certainly
    wouldn't want if they did.


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

From: [e-mail] (David W. Berry)
Subject: Re: Macintosh Prolog Recommendations
Date: 22 Jan 88 20:16:14 GMT
Organization: Apple Computer Inc., Cupertino, USA

    I just talked to some folks at Quintus and "it's not on the schedule"
It would, however, probably be simple to talk them into doing a port to
A/UX, providing you wanted to pay the n thousand dollar price they ask.
--
    David W. Berry
    [e-mail]                   dwb@Delphi
    [e-mail]                   [e-mail]
Disclaimer: Apple doesn't even know I have an opinion and certainly
    wouldn't want if they did.


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

From: [e-mail] (Michael Deering)
Subject: Re: Macintosh Prolog Recommendations
Date: 22 Jan 88 21:16:44 GMT
Organization: SPAR - Schlumberger Palo Alto Research

Other MAC-II prologs to consider include LPA macprolog, and AAIS prolog
(Advanced AI Systems).  Many large programs run in AAIS prolog on a MAC
II the same speed as Quintus compiled on a SUN-3/50, while still
maintaining a complete debugging environment.  I believe LPA has full
toolbox access, AAIS certainly does.  Both of these prologs have been
shiping commercially for over a year.  I not yet seen a non-beta version
of ALS prolog, but I have not looked recently (2 months). (Note: I have
ties to AAIS, so this biased.  However, it is really nice!)

- MD


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

From: [e-mail] (Andrew Vignaux)
Subject: Questions about File/List Managers
Date: 18 Jan 88 22:54:56 GMT
Organization: Comp Sci, Victoria Univ, Wellington, New Zealand

I have a few questions about the "File Manager" and "List Manager"
sections of IM4.

            File Manager

IM4 says that the directory ID of the root of a volume has dirID==2 with
the name of the directory the same as the name of the volume.  However,
when playing around, I found the parent dirID of the root seems to have
a dirID==1.  This means I can write a "climb_directory_tree" routine
like:

    if (/* I am running HFS */)
    dir = 0L;
    while (dir != 1L) {
        di.ioNamePtr = (StringPtr) name;
        di.ioVRefNum = vol;
        di.ioFDirIndex = -1;
        di.ioDrDirID = dir;

        PBGetCatInfo (&di, False);

        /* do something -- like prepending the name onto a path */

        dir = di.ioDrParID;
    }

where "vol" can be a volume ref number or a WD num (e.g.
sf_reply.vRefNum). This means I don't need to do a separate HGetVInfo or
muck around with special cases with VRefNums.

Is the dirID==1 mentioned in IM4-"File Manager"?  Can I use it safely if
it is not mentioned (I am only using it as the stopping condition)?  Is
there anything wrong with using GetCatInfo() like this?


            List Manager

I have had a few problems convincing the list manager to do what I want.

I would like to set up a window (with grow, zoom, go-away) which
contains a single (1 column wide) list -- I do not want a horizontal
scroll bar.  Is there some marvelous combination of flags (what does
hasGrow do?) that I can pass to NewList(), to give me the vertical
scroll bar with the list either
    * using the horizontal scroll bar area (preferred!)
    * not using the horizontal scroll bar area and NOT UPDATING IT!

Secondly, when I grow the window, the List Manager trys to scroll the
list so that no undefined cells are shown.  However because the update
is redrawn inside BeginUpdate() / EndUpdate() the screen is not redrawn
correctly.

Finally, I am passing a string, which is stored in the cell's data, to
DrawString() inside my LDEF.  Do I need to lock the "cells" handle or
will DrawString() invalidate its argument by moving or purging handles.
--
    - Andrew

I've certainly never come across any irreversible mathematics involving sofas.
------------------------------------------------------------------------------
Domain address: [e-mail]   Path address: ...!uunet!vuwcomp!andrew


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

From: [e-mail] (Robert Annechiarico)
Subject: SAS/GRAPH on Mac
Date: 22 Jan 88 22:35:36 GMT
Organization: Univ. of Rochester Computing Center

I'm looking for information on capturing SAS/GRAPH output on the Mac,
manipulating it slightly (mostly add or change text, and occasionally
reposition some of the elements of the graph), and outputting it to the
LaserWriter.  A dream would be to output it to an HP7475.

Is anyone out there doing such a thing?  It's tempting because SAS/GRAPH
gives me 90% of what I want in a typical graph in about 5 minutes; the
remaining 10% can take hours of work.

I think that the route might be to log into the SAS machine with
VersaTerm Pro, copy the graph, and paste it into MacDraw or Cricket
Draw.  But I'd like to know before I buy the software...

Any experience with this?  Or ideas?  Suggestions for a better approach?

Thanks.
--
Eric Carleen
University of Rochester Medical Center
(716)-275-5391
Bitnet: heartedc@uorhbv
UUCP: seismo!rochester!ur-tut!mpip

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

End of Usenet Mac Digest
************************
-------