MacTech 1 to 12/MacTech-vol-1-12.toast/Reference/the cmsp digests ('94-'97)/csmp digest Vol 4 No 012

Text File | 1996-10-27 | 165KB | 4,749 lines

C.S.M.P. Digest             Thu, 17 Oct 96       Volume 4 : Issue 12

Today's Topics:

        *** Jasik Debugger Student Special $99 ***
        68k based library in a PPC project
        ADB Info Requests
        Applescript HELP!
        C++ Classes to do spreadsheet?
        Calling SAT from code resource plug-in
        Changing colors in a PICT
        Cross-compiling Visual C++ w- MFC to the Mac
        GetSharedLibrary ()?  Documentation on loading shared libraries?
        Help! Icons not showing up!
        How do you call math routines in a code resource
        How to create MacOS 6-7 dual-boot system?
        How to do Unix-like scripting on a Mac?
        Inside Macintosh's Venn Diagrammer: has anyone done it?
        Looking for ictb, aete editor, NOT resourcerer
        Memory Loss with CustomGetFile
        Problems with SetDialogDefaultItem
        Re^2: MacOS 8 suggestion (Re: Disk space efficiency)
        Setting dialog default buttons
        Stack size in BOA
        StandardGetFile doesn't display only 'MooV's
        Where is Resedit in Info-Mac
        Windows and Mac programming
        [ANN ]MWDebug-MacOS 1.6.3 is available for downloading
        [ANN] Licorice for Game Sprocket Developers
        [Q] Drawing Lines?
        [Q] Switching the Sound Input source
        atan2: PPC vs. 68k
        password dialogs-- how?
        problems with XCMD using OT



The Comp.Sys.Mac.Programmer Digest is moderated by Mark Aiken
([e-mail]).

The digest is a collection of article threads from the internet
newsgroups comp.sys.mac.programmer.help, csmp.tools, csmp.misc and
csmp.games. It is designed for people who read news 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. If you don't have access to news, you
may still be able to post messages to the group by using a mail server
like anon.penet.fi (mail [e-mail] for more information).

Each issue of the digest contains one or more sets of articles (called
threads), with each set corresponding to a 'discussion' of a particular
subject.  The articles are not edited; all articles included in this digest
are in their original posted form (as received by our news server at
ee.mcgill.ca).  Article threads are not added to the digest until the last
article added to the thread is at least two weeks old (this is to ensure that
the thread is dead before adding it to the digest).  Article threads that
consist of only one message are generally not included in the digest.

The digests can be obtained by email, ftp or through the World Wide Web.

If you want to receive the digest by mail, send email to
[e-mail] with no subject and one of the following commands
as body:

    help                        Sends you a summary of commands
    subscribe csmp                      Adds you to the mailing list
    unsubscribe csmp                    Removes you from the list

Once you have subscribed, you will automatically receive each new
issue as it is created.

Back issues are available by ftp from Info-Mac mirror sites in the
per/csmp subdirectory, e.g.

  ftp://sumex-aim.stanford.edu/info-mac/per/csmp/

The contents of all back issues can be searched by accessing the
following URL, courtesy of Andrew Barry ([e-mail]):

    http://marvin.stattech.com.au/search.html

They can also be searched through the following URLs, thanks to
Tim Tuck ([e-mail]):

    http://wais.sensei.com.au/searchform.html
    wais://wais.sensei.com.au:210/csmp?

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

From [e-mail] (Steve Jasik)
Subject: *** Jasik Debugger Student Special $99 ***
Date: Wed, 02 Oct 1996 00:48:58 -0700
Organization: Jasik Designs

Jasik Designs announces a special Fall sale for University and College students
(undergraduate and graduate students).

For the months of October and November,
full time college students may purchase a Full copy of the 'Jasik Debugger'
from Jasik Designs for the special sale price of $99 .

For more info on the Jasik Debugger visit my WEB page at http://www.jasik.com

To order your copy, you -must- submit a copy of your student id card
showing current registration or some other proof.

You may e-mail your order to Jasik Designs at : [e-mail]

It should include:

a) A 'jpeg' file readable by Photoshop which contains a scanned in copy
   of your Student ID showing current registration.

b) your mailing address, e-mail address ( [e-mail] )
   and your phone number.

c) A credit card number, expiration date (Visa or MasterCard)
   and name of the cardholder.
  Those who are paranoid about sending their credit card numbers via
  e-mail should contact me about security measures.

Steve Jasik


Note:
 The Jasik Debugger runs on most Macintoshs, and is NOT compatable
 with Virtual Memory or Ram Doubler.
 Address questions to : [e-mail]

- --------------------------------------------------------
Steve Jasik             Author of 'The Debugger & MacNosy'
343 Trenton Way         http://www.jasik.com
Menlo Park CA 94025     415-322-1386  Voice Only

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

From Manuel Veloso <[e-mail]>
Date: 3 Oct 1996 10:39:21 GMT
Organization: Active Paper, Inc.

In article <macnosy-0210960049380001@204.119.88.14> Steve Jasik,
[e-mail] writes:
>Jasik Designs announces a special Fall sale for University and College students
>(undergraduate and graduate students).
>
>For the months of October and November,
>full time college students may purchase a Full copy of the 'Jasik Debugger'
>from Jasik Designs for the special sale price of $99 .

And, if you're a student who's into Mac development, Jasik's is the
most powerful debugging tool, period. It's worth it at list, and it's
an insane deal at $99. Heck, even sticking it on your resume'll be
a massive plus, especially if you know how to use it (and you'll
have to...they'll ask).

Oh, std disclaimer for plugs applies.

"when the shit hits the fan, use The Debugger to clean it off"
- ---------------------------------------
Manny Veloso          Digital Plumber
Active Paper, Inc.    http://www.apix.com
- ---------------------------------------
"bandwidth overload in progress"

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

From Manuel Veloso <[e-mail]>
Date: 3 Oct 1996 10:39:21 GMT
Organization: Active Paper, Inc.

In article <macnosy-0210960049380001@204.119.88.14> Steve Jasik,
[e-mail] writes:
>Jasik Designs announces a special Fall sale for University and College students
>(undergraduate and graduate students).
>
>For the months of October and November,
>full time college students may purchase a Full copy of the 'Jasik Debugger'
>from Jasik Designs for the special sale price of $99 .

And, if you're a student who's into Mac development, Jasik's is the
most powerful debugging tool, period. It's worth it at list, and it's
an insane deal at $99. Heck, even sticking it on your resume'll be
a massive plus, especially if you know how to use it (and you'll
have to...they'll ask).

Oh, std disclaimer for plugs applies.

"when the shit hits the fan, use The Debugger to clean it off"
- ---------------------------------------
Manny Veloso          Digital Plumber
Active Paper, Inc.    http://www.apix.com
- ---------------------------------------
"bandwidth overload in progress"

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

From [e-mail] (Lionel Attuly)
Subject: 68k based library in a PPC project
Date: Mon, 30 Sep 1996 11:16:53 +0100
Organization: none

When I try to add a 68k based library to my PPC project, the linker
complains with the message "Not a Power PC library". Is there a way to use
the library ?

I'm using CW 9.

Thanks.

_____________________________________________
Lionel Attuly    France    [e-mail]

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

From [e-mail] (Jeff Miller)
Date: Mon, 30 Sep 1996 10:08:46 -0700
Organization: Network TeleSystems, Inc.

In article <[e-mail]>,
[e-mail] (Lionel Attuly) wrote:

> When I try to add a 68k based library to my PPC project, the linker
> complains with the message "Not a Power PC library". Is there a way to use
> the library ?

No, you can't link it in directly.  You'll need to build the library into
a 68K code resource, which you can then load and call from your PPC code
using the Mixed Mode Manager.

Jeff

- ----------------------------------------------------------
 Jeff Miller  |  [e-mail]  |  Network TeleSystems, Inc.
- ----------------------------------------------------------

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

From [e-mail] (MW Ron)
Date: Wed, 02 Oct 1996 16:14:39 -0400
Organization: Metrowerks

In article <[e-mail]>,
[e-mail] (Lionel Attuly) wrote:

>When I try to add a 68k based library to my PPC project, the linker
>complains with the message "Not a Power PC library". Is there a way to use
>the library ?

No,  If this is a third party library you would need to have them rebuild
the library for PowerPC,  If this is just a libray that you have in a 68k
project like MacOS.Lib you would use Interface.Lib for PowerPC instead.
Or the proper ANSI library.

Ron

--
METROWERKS                   Ron Liechty
"Software at Work"    [e-mail]
http://www.metrowerks.com/about/people/rogues.html#mwron

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

From [e-mail] (David Phillip Oster)
Date: 3 Oct 1996 20:27:11 GMT
Organization: Digital Arts & Sciences Corp.

In article <[e-mail]>,
[e-mail] (MW Ron) wrote:

>In article <[e-mail]>,
>[e-mail] (Lionel Attuly) wrote:

>>When I try to add a 68k based library to my PPC project, the linker
>>complains with the message "Not a Power PC library". Is there a way to use
>>the library ?

You can use the library, but it is tedious. Define a new procedure
that takes a integer "selector" argument, and uses a "switch" statement
to dispatch to each routine in the library. Don't give the selector
routine too many parameters, but it is OK to pass it a pointer to a
parameter record. You may need ot add
initialization and sutdown routines. Link this with the library,
and make of it an executabe code resource. Don't forget that executable
code resources have to initialize, setup, and restore register A4,
and that callbacks that the library hands out need to have appropriate
setupa4, restorea4 calls.

Then, from your powerpc code, you can load the above excutable resource
into memory, lock it, and call it with CallUniversalProcPtr().

My powerpc code calls 68000 adobe photoshop plug-ins just fine using
this technique.

--
I used to say, "At least Congress doesn't make death worse every year."
Then I had to probate an estate.

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

From [e-mail] (Ed Anson)
Date: Thu, 03 Oct 1996 20:56:41 -0400
Organization: Tulip Software

In article <DavidO-0310961330200001@206.14.165.117>, [e-mail]
(David Phillip Oster) wrote:

 and make of it an executabe code resource. Don't forget that executable
> code resources have to initialize, setup, and restore register A4,
> and that callbacks that the library hands out need to have appropriate
> setupa4, restorea4 calls.

If you are creating a code resource from scratch, using A4 based globals
is common practice. However, when calling an existing library, it is
probable that it is compiled to run in an A5 world.

Consequently, you have one of two cases:

1. You are lucky enough to find that the library uses no globals, and no
special setup is required.

2. If the library uses A5 globals, you will have to create and maintain a
separate A5 world for it. There is a Tech Note (old numbering #256) that
describes in detail how to do this.

- --------------------
Ed Anson
Tulip Software
Andover, MA 01810
U.S.A.              <http://www.tiac.net/users/tulip/home.html>

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

From [e-mail] (Ed Anson)
Date: Thu, 03 Oct 1996 20:56:41 -0400
Organization: Tulip Software

In article <DavidO-0310961330200001@206.14.165.117>, [e-mail]
(David Phillip Oster) wrote:

 and make of it an executabe code resource. Don't forget that executable
> code resources have to initialize, setup, and restore register A4,
> and that callbacks that the library hands out need to have appropriate
> setupa4, restorea4 calls.

If you are creating a code resource from scratch, using A4 based globals
is common practice. However, when calling an existing library, it is
probable that it is compiled to run in an A5 world.

Consequently, you have one of two cases:

1. You are lucky enough to find that the library uses no globals, and no
special setup is required.

2. If the library uses A5 globals, you will have to create and maintain a
separate A5 world for it. There is a Tech Note (old numbering #256) that
describes in detail how to do this.

- --------------------
Ed Anson
Tulip Software
Andover, MA 01810
U.S.A.              <http://www.tiac.net/users/tulip/home.html>

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

From [e-mail] (Lane Roathe)
Subject: ADB Info Requests
Date: Thu, 03 Oct 1996 04:31:11 -0800
Organization: Ideas From the Deep, Ltd.

1) Is the ADB Manager native? If not, will it be any time soon?
  (ie, does it make a difference if I write a 68K ADB driver?)

2) Does anyone have, or can anyone point me to, C source to an ADB driver?
  (I have 68k source, but would like to be able to build a FAT driver)

If you could CC replies to my email account <[e-mail]> I would
appriciate it, as I miss articles on the newsgroups.

Thanks!

_________________________________________________________________
Lane Roathe, Authorized Representative, Ideas From the Deep. Ltd.
Distribution in any form by Microsoft or Prodigy is PROHIBITED!!!
-->Unsolicited bulk email will be proofread @ $500/K, 2K minimum!
Views expressed are my own, not IFD's! Content (c) Copyright 1996
<http://www.ifd.com/lane>                    <http://www.ifd.com>

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

From [e-mail] (TheBug)
Date: Thu, 03 Oct 1996 15:40:31 +0100
Organization: privat

In article <lroathe-0310960431110001@news>, [e-mail] (Lane Roathe) wrote:

> 1) Is the ADB Manager native? If not, will it be any time soon?
>   (ie, does it make a difference if I write a 68K ADB driver?)

ADB Manager is not native and it will not become native until the new I/O
architecture that was planned for System 8 is rolled in. Writing a PPC
native driver will reduce your performance compared to a 68K driver. It
adds two code switches as the ADB Manager is 68K and you have to switch to
PPC on entry of the driver and back to 68K on exit.

> 2) Does anyone have, or can anyone point me to, C source to an ADB driver?
>   (I have 68k source, but would like to be able to build a FAT driver)

There is no such thing as C code for ADB drivers the calling interface is
Assembly. The least thing you need is some glue code. And stay away from a
fat driver, it slows you down.

--
*******************************************************************
Guido Kˆrber -            Programmer and hardware hacker
[e-mail]    Specialised in mistreating the ADB
fax: x49-30-773 81 36     Ask me about:
                           Flightstick Pro, Jetstick
                           MacEnjoy, MacEnjoy Style

Opinions expressed herein are mine unless expressly stated
otherwise. Similarities with living or undead persons are
coincidence and not intended - really!   ;-)
Best use before: (see date printed on backside of message)
*******************************************************************

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

From [e-mail] (TheBug)
Date: Thu, 03 Oct 1996 15:40:31 +0100
Organization: privat

In article <lroathe-0310960431110001@news>, [e-mail] (Lane Roathe) wrote:

> 1) Is the ADB Manager native? If not, will it be any time soon?
>   (ie, does it make a difference if I write a 68K ADB driver?)

ADB Manager is not native and it will not become native until the new I/O
architecture that was planned for System 8 is rolled in. Writing a PPC
native driver will reduce your performance compared to a 68K driver. It
adds two code switches as the ADB Manager is 68K and you have to switch to
PPC on entry of the driver and back to 68K on exit.

> 2) Does anyone have, or can anyone point me to, C source to an ADB driver?
>   (I have 68k source, but would like to be able to build a FAT driver)

There is no such thing as C code for ADB drivers the calling interface is
Assembly. The least thing you need is some glue code. And stay away from a
fat driver, it slows you down.

--
*******************************************************************
Guido Kˆrber -            Programmer and hardware hacker
[e-mail]    Specialised in mistreating the ADB
fax: x49-30-773 81 36     Ask me about:
                           Flightstick Pro, Jetstick
                           MacEnjoy, MacEnjoy Style

Opinions expressed herein are mine unless expressly stated
otherwise. Similarities with living or undead persons are
coincidence and not intended - really!   ;-)
Best use before: (see date printed on backside of message)
*******************************************************************

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

From [e-mail]
Subject: Applescript HELP!
Date: 22 Sep 1996 23:09:48 GMT
Organization: a2i network

I am puzzled on how to write an applescript to dragndrop convert/change
creator of a bunch of text files to say, Tex Edit Plus format. Anyone
know how?

I am new to apple script and it amazes me how complex yet powerful it is.

Email please; [e-mail]



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

From [e-mail] (Matt Bishop)
Date: Sat, 28 Sep 1996 11:58:44 -0500
Organization: Zippo

In article <524gvt$[e-mail]>, [e-mail] wrote:

> I am puzzled on how to write an applescript to dragndrop convert/change
> creator of a bunch of text files to say, Tex Edit Plus format. Anyone
> know how?
>
> I am new to apple script and it amazes me how complex yet powerful it is.
>
> Email please; [e-mail]

AppleScript (AS) itself won't do it; you use AS to talk to another
application like ClarisWorks or BBEdit (I think it is scriptable.  Not
sure.). If you want to go to Tex Edit Plus, check that program for an AS
dictionary (it is in the 'File' menu) for commands to open/import and
save.

Then, you take each filename, tell TEP to open the filename and then save
it as another name.  It should save it in its own format.

voila!  For help, get Danny Goodman's AppleScript Handbook.  It is
invaluable for scripting.

Have fun!

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

From [e-mail] (Jim Correia)
Date: Sat, 28 Sep 1996 16:27:29 -0500
Organization: University of Massachusetts Dartmouth

In article <[e-mail]>,
[e-mail] (Matt Bishop) wrote:

>In article <524gvt$[e-mail]>, [e-mail] wrote:
>
>> I am puzzled on how to write an applescript to dragndrop convert/change
>> creator of a bunch of text files to say, Tex Edit Plus format. Anyone
>> know how?
>>
>> I am new to apple script and it amazes me how complex yet powerful it is.
>>
>> Email please; [e-mail]
>
>AppleScript (AS) itself won't do it; you use AS to talk to another

Right.

>application like ClarisWorks or BBEdit (I think it is scriptable.  Not
>sure.). If you want to go to Tex Edit Plus, check that program for an AS
>dictionary (it is in the 'File' menu) for commands to open/import and
>save.
>
>Then, you take each filename, tell TEP to open the filename and then save
>it as another name.  It should save it in its own format.

While this is one approach, this is not one I'd highly recommend.  If you
have a text file, and only want to change it so that it will open in
another text editor when double clicked, there are other, more efficient
ways to do it.

Your first option is TextChanger.  It is available at
<http://www.barebones.com>.  It will iterate through a folder tree.  It is
very fast.  Extremely fast compared to the same applescript.  If you insist
on doing it in applescript, you do it by talking to the scriptable finder
and change the creator code of each file.  The example below does just
that, but doesn't iterate through folders.  That is left as an excercise
for the reader.

on open theFileList
     repeat with theFile in theFileList
          tell application "Finder"
               if the file type of theFile is "TEXT" -- only work on text files
                    set the creator type of theFile to "R*ch" -- BBEdit's
               end if
          end tell
     end repeat
end open

Jim

--
Jim Correia
Software Developer
SimCalc Project     <http://tango.mth.umassd.edu/>
[e-mail] <http://tango.mth.umassd.edu/correia/>

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

From Forrest Johnson <[e-mail]>
Date: 4 Oct 1996 04:10:22 GMT
Organization: Fred's Fill Dirt & Croissants

In article <524gvt$[e-mail]>, [e-mail] wrote:

> I am puzzled on how to write an applescript to dragndrop convert/change
> creator of a bunch of text files to say, Tex Edit Plus format. Anyone
> know how?

Use the On Open function to make the script drag'n'drop.  Within the On
Open, loop through the list of files dropped on the script and change the
creator with something like:

On Open filelist

repeat with droppedfile in filelist

tell application "Finder"
        set creator type of alias droppedfile to "MOSS"
end tell

end repeat

end open

(This code changes the creator to Netscape; you'll have to replace the
MOSS with whatever creator string you need.  You might also have to play
around with the "alias" keyword depending on how you parsed the filename.
 The code might not be exact since I don't have my Applescript books
handy and I'm writing from memory.)

When you save the script, save it as an application and check the stay
open and splashscreen boxes to your preference.

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

From "Thomas L. Ferrell" <[e-mail]>
Date: 4 Oct 1996 05:24:14 GMT
Organization: Oak Ridge National Lab

By the way, there's a nice little app for this called Tex-Edit Drop Box
that's just been posted to the INFO-MAC.
tom



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

From "Thomas L. Ferrell" <[e-mail]>
Date: 4 Oct 1996 05:24:14 GMT
Organization: Oak Ridge National Lab

By the way, there's a nice little app for this called Tex-Edit Drop Box
that's just been posted to the INFO-MAC.
tom



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

From [e-mail] (Eli Meir)
Subject: C++ Classes to do spreadsheet?
Date: 24 Sep 1996 16:51:32 GMT
Organization: Dept. of Zoology, Univ. of Washington

I am using TCL in an educational program, and am looking for a way to
implement a simple spreadsheet in the program.  Does anyone know of a
third party or shareware or pub. domain set of classes that I could use
to do this.  The spreadsheet just needs to do very simple arithmatic
(ie- add and multiply and maybe raise to the power of).  Any
suggestions would be appreciated.  Thanks in advance,

Eli Meir
Dept of Zoology, Univ of Washington
[e-mail]

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

From [e-mail] (Matt Bishop)
Date: Fri, 27 Sep 1996 14:52:02 -0500
Organization: Zippo

In article <5293ik$[e-mail]>,
[e-mail] (Eli Meir) wrote:

> I am using TCL in an educational program, and am looking for a way to
> implement a simple spreadsheet in the program.  Does anyone know of a
> third party or shareware or pub. domain set of classes that I could use
> to do this.  The spreadsheet just needs to do very simple arithmatic
> (ie- add and multiply and maybe raise to the power of).  Any
> suggestions would be appreciated.  Thanks in advance,
>
> Eli Meir
> Dept of Zoology, Univ of Washington
> [e-mail]


Metrowerks PowerPlant has a 'table' class (including a 'tableviewer' class
to display it in a window/navigate the info) that might help...post your
question in comp.sys.mac.programmer.codewarrior to see if anyone has
written one for it.

I know Adrenaline Software has an OpenDoc spreadsheet part that works
really well if you want to go that route.

Just an idea,

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

From [e-mail] (David Dunham)
Date: 4 Oct 1996 17:26:10 GMT
Organization: Pensee Corporation

In article <[e-mail]>,
[e-mail] (Matt Bishop) wrote:

> > I am using TCL in an educational program, and am looking for a way to
> > implement a simple spreadsheet in the program.  Does anyone know of a
> > third party or shareware or pub. domain set of classes that I could use
> > to do this.  The spreadsheet just needs to do very simple arithmatic
> > (ie- add and multiply and maybe raise to the power of).

> Metrowerks PowerPlant has a 'table' class (including a 'tableviewer' class
> to display it in a window/navigate the info) that might help...post your
> question in comp.sys.mac.programmer.codewarrior to see if anyone has
> written one for it.

TCL has table classes too, and I suspect they're more robust (much of
PowerPlant's are in the "In Progress" folder). But no table view class
solves the problem of calculation...

David Dunham   Pensee Corporation   [e-mail]
Voice/Fax: [phone]             http://www.pensee.com/dunham/
    "I say we should listen to the customers and give them what they want."
    "What they want is better products for free." --Scott Adams

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

From [e-mail] (Matthew Clay)
Date: 4 Oct 1996 20:13:33 GMT
Organization: The University of Iowa

From article <[e-mail]>, by [e-mail] (David Dunham):
> In article <[e-mail]>,
> [e-mail] (Matt Bishop) wrote:
>
>> > I am using TCL in an educational program, and am looking for a way to
>> > implement a simple spreadsheet in the program.  Does anyone know of a
>> > third party or shareware or pub. domain set of classes that I could use
>> > to do this.  The spreadsheet just needs to do very simple arithmatic
>> > (ie- add and multiply and maybe raise to the power of).
>
>> Metrowerks PowerPlant has a 'table' class (including a 'tableviewer' class
>> to display it in a window/navigate the info) that might help...post your
>> question in comp.sys.mac.programmer.codewarrior to see if anyone has
>> written one for it.
>
> TCL has table classes too, and I suspect they're more robust (much of
> PowerPlant's are in the "In Progress" folder). But no table view class
> solves the problem of calculation...

They are robust. There's actually a simple, but fairly well-featured
spreadsheet interface in one of the demo projects. It supports a nice
gray background around the edges, resizable rows and columns, etc.
Note it is only an interface, so you must provide the computational
backend. I think the original poster wants not only an interface, but
the computational part as well.

-mc



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

From [e-mail] (Mark Chrisman)
Subject: Calling SAT from code resource plug-in
Date: Mon, 30 Sep 1996 00:31:26 -0700
Organization: Inreach's InterNetNews Site

Suppose I want to make the levels of my game plug-ins (code resources).
Is there a good way to do this?  They would need to call the SAT routines.
Is that even possible for a code resource?

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

From [e-mail] (Ingemar Ragnemalm)
Date: 3 Oct 1996 08:42:11 GMT
Organization: (none)

[e-mail] (Mark Chrisman) writes:

>Suppose I want to make the levels of my game plug-ins (code resources).
>Is there a good way to do this?  They would need to call the SAT routines.
>Is that even possible for a code resource?

For some yes, for some no. Let's take the no first.

Linking the SAT lib into the plug-in is hard or impossible. SAT wants to
have a bunch of globals available. That could be worked around by passing
a pointer to gSAT to all calls that needs it, but that is not the way it
works today.

Plug-ins like screen savers are basically out of the question. However, should
a screen saver allocate all the memory needed for double-buffered animation?

Plug-ins to a program of your own is another matter. In such a case, you
have control over the interface between the code resource and the main
program, so you can pass the addresses to the SAT calls you need. Make a
record with a bunch of procedure pointers, and perhaps a pointer to gSAT
as well, and pass a pointer to that record to the plug-in. Then the plug-in
can make any of these calls. SAT is linked with the main program, so it has
all the globals it needs.

So in your case, it seems to me that the answer is YES!

--
- -
Ingemar Ragnemalm, PhD
Image processing, Mac shareware games
E-mail address: [e-mail] or [e-mail]

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

From [e-mail] (Erik Ackerman)
Subject: Changing colors in a PICT
Date: Fri, 4 Oct 1996 16:34:15 GMT
Organization: Analysts International Corporation

Ok, I feel foolish.  This ought to be really easy, but so far no dice:

I have a bunch of PICTs in resources.  I put them together to show
different things.  Depending on the state of these things, I want to color
them differently.  So in the PICTs there are areas that are colored with
"false" colors (i.e. cyan) that usually will be changed to one color (i.e.
gray), but sometimes will be changed to another (i.e. red).  I tried using
the following routine to effect this change, but no dice.  I have come to
the (perhaps obvious) conclusion that GetPictInfo() creates a color table
rather than returning a handle to an existing one, and therefore it is
unreasonable to try to effect changes through the returned table.  Any
suggestions?

void ChangePictColors(  PicHandle inPict,
                        const RGBColor inFromColor,
                        const RGBColor inToColor) {
   PictInfo   thePictInfo;

   short anErr = GetPictInfo(inPict, &thePictInfo, 0x1f, 256, 0, 0);
   if(anErr != noErr)
      return;

   CTabHandle  theColorTable = thePictInfo.theColorTable;
   ColorSpec   *theCurrentEntry;
   for(short i = 0; i < theColorTable[0]->ctSize; i++) {
      theCurrentEntry = &(theColorTable[0]->ctTable[i]);
      if(   theCurrentEntry->rgb.red == inFromColor.red &&
            theCurrentEntry->rgb.green == inFromColor.green &&
            theCurrentEntry->rbg.blue == inFromColor.blue) {
         theCurrentEntry->rgb = inToColor;
         CTabChanged(theColorTable);
   }
}

--
Life's not complete 'til your heart's skipped a beat. - Prefab Sprout

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

From [e-mail] (Erik Ackerman)
Date: Fri, 4 Oct 1996 17:31:55 GMT
Organization: Analysts International Corporation

In article <[e-mail]>,
[e-mail] (Erik Ackerman) wrote:

>Ok, I feel foolish.  This ought to be really easy, but so far no dice:
>
>I have a bunch of PICTs in resources.  I put them together to show
>different things.  Depending on the state of these things, I want to color
>them differently.  So in the PICTs there are areas that are colored with
>"false" colors (i.e. cyan) that usually will be changed to one color (i.e.
>gray), but sometimes will be changed to another (i.e. red).  I tried using
>the following routine to effect this change, but no dice.  I have come to
>the (perhaps obvious) conclusion that GetPictInfo() creates a color table
>rather than returning a handle to an existing one, and therefore it is
>unreasonable to try to effect changes through the returned table.  Any
>suggestions?
>
>void ChangePictColors(  PicHandle inPict,
>                        const RGBColor inFromColor,
>                        const RGBColor inToColor) {
>   PictInfo   thePictInfo;
>
>   short anErr = GetPictInfo(inPict, &thePictInfo, 0x1f, 256, 0, 0);
>   if(anErr != noErr)
>      return;
>
>   CTabHandle  theColorTable = thePictInfo.theColorTable;
>   ColorSpec   *theCurrentEntry;
>   for(short i = 0; i < theColorTable[0]->ctSize; i++) {
>      theCurrentEntry = &(theColorTable[0]->ctTable[i]);
>      if(   theCurrentEntry->rgb.red == inFromColor.red &&
>            theCurrentEntry->rgb.green == inFromColor.green &&
>            theCurrentEntry->rbg.blue == inFromColor.blue) {
>         theCurrentEntry->rgb = inToColor;
>         CTabChanged(theColorTable);
>   }
>}
>
>--
>Life's not complete 'til your heart's skipped a beat. - Prefab Sprout

Ok, fixed it.  I do roughly the same thing, but in a BitsProc with the
individual PixMaps instead of to the PICT as a whole.  Works great.  Sorry
for the waste of bandwidth.

Erik

--
Life's not complete 'til your heart's skipped a beat. - Prefab Sprout

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

From [e-mail] ( nick.c @MT )
Date: Fri, 04 Oct 1996 17:48:52 -0800
Organization: MacTech Magazine



[e-mail] (Erik Ackerman) wrote:

>Ok, fixed it.  I do roughly the same thing, but in a BitsProc with the
>individual PixMaps instead of to the PICT as a whole.  Works great.  Sorry
>for the waste of bandwidth.


   FYI, there was an article in _develop_ (around issue 20, give or
     take an issue) that dealt with swapping out color tables for
     picts.  As I recall it was a method for taking up less disk
     space by using a single color table for multiple picts--sorry
     I don't remember the details too well and don't have the issue
     handy.  Anyway, it might be worth digging up that article
     just as a general reference for what you're working on.  Luck,

____Nicholas C. DeMello, Ph.D.________________________________________

   Online from MacTech Magazine, for Mac OS Programmers and Developers
     http://www.MacTech.com/
                                       _/   _/  _/  _/_/_/   _/   _/
  Chemistry: [e-mail]       _/_/ _/  _/  _/   _/  _/_/_/
    MacTech: [e-mail]      _/ _/_/  _/  _/       _/ _/
       http://www.chem.ucla.edu/~nick/   _/  _/   _/_/_/  _/   _/

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

From [e-mail] (William Walker)
Subject: Cross-compiling Visual C++ w- MFC to the Mac
Date: Tue, 01 Oct 1996 15:57:36 -0400
Organization: Section of Medical Informatics, U. of Pittsburgh


I have a graphics animation and sound intensive Windows application that
works fine under Windows and I am creating working Macintosh 680x0 and PPC
applications using the Visual C++ Cross Compiler.  However, the Mac
applications are slow (esp. the 680x0).

I am trying to find out two things:

1) Is the lower performance typical of the Visual C++ Cross Compiler (and
the associated Windows API emulation library)?  If so, is there anything
simple I can do or should avoid to get better performance?

2) With the libraries and source code supplied with the Visual C++ Cross
Compiler, is it possible to compile the Windows code using a Mac-based
compiler (CodeWarrior?) and get better performance?

Anyone's experience here is greatly appreciated.  Please no flames, I am
in a bind and I have inherited this situation--I would not have chosen to
develop a cross platform application using this method.

Peace.

--Willy
_____________________________________________
Willy Walker                    [e-mail]    In Pittsburgh we have
Image Engine Project                             two seasons, how many
University of Pittsburgh, PA     412/692-4454    you got?        -WDVE

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

From [e-mail] (Brad Pettit)
Date: Tue, 01 Oct 1996 14:22:58 -0700
Organization: Apple Computer

In article <[e-mail]>, [e-mail]
(William Walker) wrote:

> I have a graphics animation and sound intensive Windows application that
> works fine under Windows and I am creating working Macintosh 680x0 and PPC
> applications using the Visual C++ Cross Compiler.  However, the Mac
> applications are slow (esp. the 680x0).
>
>
> Anyone's experience here is greatly appreciated.  Please no flames, I am
> in a bind and I have inherited this situation--I would not have chosen to
> develop a cross platform application using this method.
>
> Peace.
>
> --Willy

Willy,

Check out http://www.willows.com/ and their Willows Toolkit (Formerly TWIN API).

--Brad

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

From [e-mail] (MW Ron)
Date: Tue, 01 Oct 1996 22:11:25 -0400
Organization: Metrowerks

In article <[e-mail]>, [e-mail]
(William Walker) wrote:

>I have a graphics animation and sound intensive Windows application that
>works fine under Windows and I am creating working Macintosh 680x0 and PPC
>applications using the Visual C++ Cross Compiler.  However, the Mac
>applications are slow (esp. the 680x0).
>
>I am trying to find out two things:
>
>1) Is the lower performance typical of the Visual C++ Cross Compiler (and
>the associated Windows API emulation library)?  If so, is there anything
>simple I can do or should avoid to get better performance?

I believe that thise is the problem, maybe someone else can help you with
optimization techniques

>2) With the libraries and source code supplied with the Visual C++ Cross
>Compiler, is it possible to compile the Windows code using a Mac-based
>compiler (CodeWarrior?) and get better performance?

Sorry  Metrowerks CodeWarrior can not build Macintosh applications with
the MFC for Mac.  CodeWarrior can use MFC for Windows and generate Windows
code but not the other way around.  I believe that the MFC For Mac files
are only compiliable under Win32 and Visual C++.


Ron

--
METROWERKS                   Ron Liechty
"Software at Work"    [e-mail]
http://www.metrowerks.com/about/people/rogues.html#mwron

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

From "Brian Mowatt" <[e-mail]>
Date: 3 Oct 1996 08:39:04 GMT
Organization: Crocodile Clips Ltd.

William Walker <[e-mail]> wrote in article
<[e-mail]>...
>
> I have a graphics animation and sound intensive Windows application that
> works fine under Windows and I am creating working Macintosh 680x0 and
PPC
> applications using the Visual C++ Cross Compiler.  However, the Mac
> applications are slow (esp. the 680x0).
>
> I am trying to find out two things:
>
etc.
I have the same problem..
As far as i'm aware though with the cross compiler you can make calls to
the Mac API more directly, bypassing the MFC layer altogether. Ive already
done a little for mac sound that is more advanced than the cross compiler
would offer. Therefore If you want to learn the mac equivalent graphics
calls
(e.g. copybits <=> bitblt etc) and use them then in key places I would
imagine that speed would be improved. I'm probably going to give it a go
soon anyway, as I'm in exactly the same boat!
Hope thats of some help.

[e-mail]


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

From [e-mail] (Frank Kim)
Subject: GetSharedLibrary ()?  Documentation on loading shared libraries?
Date: 3 Oct 1996 12:12:53 GMT
Organization: MIT Laboratory for Computer Science, Cambridge MA

Hi,

The Targeting MacOS manual mentions the function GetSharedLibrary() but
I have been unable to find any documentation in the Inside Mac series or
CodeWarrior stuff.

Basically I am looking for documentation on Mac equivalents to these
commands on other platforms:

Windows: LoadLibrary, FreeLibrary, GetProcAddress
Unix:    dlopen, dlclose

Any help would be most appreciated!

--
Frank Kim    [e-mail]
        http://cag-www.lcs.mit.edu/~frankkim/


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

From [e-mail] (Adam Winer)
Date: Thu, 03 Oct 1996 12:05:28 -0800
Organization: Oracle Corporation

In article <530ak5$[e-mail]>,
[e-mail] (Frank Kim) wrote:

> Hi,
>
> The Targeting MacOS manual mentions the function GetSharedLibrary() but
> I have been unable to find any documentation in the Inside Mac series or
> CodeWarrior stuff.
>
> Basically I am looking for documentation on Mac equivalents to these
> commands on other platforms:
>
> Windows: LoadLibrary, FreeLibrary, GetProcAddress
> Unix:    dlopen, dlclose

The documentation is in Inside Mac:PowerPC System Software.
Despite the "PowerPC" moniker, it also applies to CFM-68K
code, though not to standard 68K applications.

All the docs can be found at:
<http://devworld.apple.com/dev/techsupport/insidemac/PPCSoftware/PPCSoftware-
2.html>

-- Adam Winer
[e-mail]

--
Adam Winer
[e-mail]
Technical Staff, Oracle Corp.

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

From [e-mail] (Adam Winer)
Date: Thu, 03 Oct 1996 12:05:28 -0800
Organization: Oracle Corporation

In article <530ak5$[e-mail]>,
[e-mail] (Frank Kim) wrote:

> Hi,
>
> The Targeting MacOS manual mentions the function GetSharedLibrary() but
> I have been unable to find any documentation in the Inside Mac series or
> CodeWarrior stuff.
>
> Basically I am looking for documentation on Mac equivalents to these
> commands on other platforms:
>
> Windows: LoadLibrary, FreeLibrary, GetProcAddress
> Unix:    dlopen, dlclose

The documentation is in Inside Mac:PowerPC System Software.
Despite the "PowerPC" moniker, it also applies to CFM-68K
code, though not to standard 68K applications.

All the docs can be found at:
<http://devworld.apple.com/dev/techsupport/insidemac/PPCSoftware/PPCSoftware-
2.html>

-- Adam Winer
[e-mail]

--
Adam Winer
[e-mail]
Technical Staff, Oracle Corp.

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

From [e-mail] (Chris Weiss)
Subject: Help! Icons not showing up!
Date: Wed, 25 Sep 1996 01:46:18 GMT
Organization: CCnet Communications (510-988-7140 guest)

We're trying to re-master several old products on one large CD with
some new files added and are running into a weird problem with icons.
Several of the files are displaying the incorrect icons (either icons
of the creator program or incorrect color-depth) on most systems.
We've tried rebuilding the desktop and have verified that the
resources have the correct icons in them (using ResEdit). We have one
machine that displays the icons correctly, but we can't use it to
build the masters. We've run into this problem in the past, but nobody
remembers how we solved it and our Mac gurus are out of town.

A few questions - What should the IDs of the icons be? Most of the
files are set to 128, but a few have -65xxx. Is there a procedure to
select the default icon that is used (from the ICL4, ICL8, etc
resources). What happens if a file only has an ICN# resource and no
ICL4, ICL8, etc?

I know the solution is something simple, but I'm really not familiar
with Mac procedures so I'm sort of lost on where to even start
looking.

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

From [e-mail] (Brian Kendall)
Date: Wed, 25 Sep 1996 17:30:51 -0400
Organization: Programmer

In article <[e-mail]>, [e-mail] (Chris Weiss) wrote:

> We're trying to re-master several old products on one large CD with
> some new files added and are running into a weird problem with icons.
> Several of the files are displaying the incorrect icons (either icons
> of the creator program or incorrect color-depth) on most systems.
> We've tried rebuilding the desktop and have verified that the
> resources have the correct icons in them (using ResEdit). We have one
> machine that displays the icons correctly, but we can't use it to
> build the masters. We've run into this problem in the past, but nobody
> remembers how we solved it and our Mac gurus are out of town.
>
> A few questions - What should the IDs of the icons be? Most of the
> files are set to 128, but a few have -65xxx. Is there a procedure to
> select the default icon that is used (from the ICL4, ICL8, etc
> resources). What happens if a file only has an ICN# resource and no
> ICL4, ICL8, etc?
>
> I know the solution is something simple, but I'm really not familiar
> with Mac procedures so I'm sort of lost on where to even start
> looking.

The easiest way to set the icon of something is to get the info for it,
select the icon picture in the top left hand corner, and paste in a
picture. (That's good for folders and documents. Applications should use a
BNDL.)

This creates an icl8, icl4, icl#, ics8, ics4, and an ics# all with an ID
of -16455 (or your basic icon family). This also turns on the 'use custom
icon' setting to the file or folder. It's probably the easiest way to set
up an icon.
If you did this to a folder, an invisible file called 'icon' is placed in
the folder to set everything up.

Hope this helps.

Brian K.

ãããããããããããããããããããããããããããããããããããããããããããããããããããããããããããããããããã
"If you take cranberries and stew them like apple sauce, it tastes
much more like prunes then rhubarb does." ã Groucho Marx
ãããããããããããããããããããããããããããããããããããããããããããããããããããããããããããããããããã

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

From Russ Hendy <[e-mail]>
Date: Thu, 26 Sep 1996 11:26:09 +0000
Organization: tui interactive media

Chris -

Your default custom icon ID is -16455.

If you make an icon, add ICN#, icl8, icl4, ics#, ics8 ics4 icons all
with -16455 ID's. The computer reading the icons will choose icons
depending on the colour attributes set on that machine, and whether it's
looking at small (ics) or large (ics) icons.


That's the easiest solution, although BNDL resources are probably the
best way to go.


(By the way, when you save your resource fork with the icons in, first
change the file attributes - in ResEdit it's FILE|Get Info For
<yourfile> - so that 'Inited' is false, and 'Use Custom Icon' is true.
When you restart, you should see your new icons, although a desktop
database rebuild can be a good move.)

Regards,
Russ.

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

From "Aidan Cully" <[e-mail]>
Date: 3 Oct 96 02:19:48 -0400
Organization: INTAC Access Corporation - An Internet Service Provider

>In article <[e-mail]>, [e-mail] (Chris Weiss)
>wrote:
>
>> We're trying to re-master several old products on one large CD with
>> some new files added and are running into a weird problem with icons.
>> Several of the files are displaying the incorrect icons (either icons
>> of the creator program or incorrect color-depth) on most systems.
>> We've tried rebuilding the desktop and have verified that the
>> resources have the correct icons in them (using ResEdit). We have one
>> machine that displays the icons correctly, but we can't use it to
>> build the masters. We've run into this problem in the past, but nobody
>> remembers how we solved it and our Mac gurus are out of town.
>>
>> A few questions - What should the IDs of the icons be? Most of the
>> files are set to 128, but a few have -65xxx. Is there a procedure to
>> select the default icon that is used (from the ICL4, ICL8, etc
>> resources). What happens if a file only has an ICN# resource and no
>> ICL4, ICL8, etc?
>>
>> I know the solution is something simple, but I'm really not familiar
>> with Mac procedures so I'm sort of lost on where to even start
>> looking.
>
>The easiest way to set the icon of something is to get the info for it,
>select the icon picture in the top left hand corner, and paste in a
>picture. (That's good for folders and documents. Applications should use a
>BNDL.)
>
>This creates an icl8, icl4, icl#, ics8, ics4, and an ics# all with an ID
>of -16455 (or your basic icon family). This also turns on the 'use custom
>icon' setting to the file or folder. It's probably the easiest way to set
>up an icon.
>If you did this to a folder, an invisible file called 'icon' is placed in
>the folder to set everything up.
>
>Hope this helps.
>
>Brian K.
>
>??????????????????????????????????????????????????????????????????
>"If you take cranberries and stew them like apple sauce, it tastes
>much more like prunes then rhubarb does." ? Groucho Marx
>??????????????????????????????????????????????????????????????????

It might also be the case that you have to use ResEdit to get the catalog
info for your files and clear the INITed field, and rebuild the desktop
afterwards.

HTH
--Aidan



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

From [e-mail] (Kent Gardner)
Subject: How do you call math routines in a code resource
Date: 30 Sep 1996 16:26:14 GMT
Organization: UCLA

I am using the Symantec C++ compiler to write a code resource that
requires calls to math routines. Math routines (e.g., SIN) are not
part of the ANSI-A4 library. Any suggestions on how to make the call.
I assume I'll need to save an restore registers... Is there an
example out there on how to do this. Thanks in advance.

Kent Gardner

Pharmacology/Crump Institute, UCLA
B2-086 CHS 694817
Los Angeles, California 90095
e-mail: [e-mail]

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

From [e-mail] (Andrew Cunningham)
Date: Mon, 30 Sep 1996 18:42:32 -0700
Organization: Vibro-Acoustic Sciences Inc

In article <kgardner-3009960929290001@149.142.171.200>,
[e-mail] (Kent Gardner) wrote:

> I am using the Symantec C++ compiler to write a code resource that
> requires calls to math routines. Math routines (e.g., SIN) are not
> part of the ANSI-A4 library. Any suggestions on how to make the call.
> I assume I'll need to save an restore registers... Is there an
> example out there on how to do this. Thanks in advance.
>
> Kent Gardner


Kent,
Easy solution - just add the library sources you need to the math library
right into your code resource project! They just left out that stuff to
keep the ANSI-A4 libary small.

Andrew

--
Andrew Cunningham
Vibro-Acoustic Sciences Inc
Ph:  [phone]
Fax: [phone]
e-mail: [e-mail]

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

From jgarden@192.219.29.90
Date: 3 Oct 96 15:02:49 -0300
Organization: isis, Incorporated

>I am using the Symantec C++ compiler to write a code resource that
>requires calls to math routines. Math routines (e.g., SIN) are not
>part of the ANSI-A4 library. Any suggestions on how to make the call.
>I assume I'll need to save an restore registers... Is there an
>example out there on how to do this. Thanks in advance.
>
>Kent Gardner
>

I haven't tried it myself, but you might try just including the source for
the math routines - ie. math.c or whatever.

JG@PI

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

From [e-mail] (Paul C. Pharr)
Date: Thu, 03 Oct 1996 19:06:35 -0400
Organization: Graphsoft, Inc.

In article <kgardner-3009960929290001@149.142.171.200>,
[e-mail] (Kent Gardner) wrote:

>I am using the Symantec C++ compiler to write a code resource that
>requires calls to math routines. Math routines (e.g., SIN) are not
>part of the ANSI-A4 library. Any suggestions on how to make the call.
>I assume I'll need to save an restore registers... Is there an
>example out there on how to do this. Thanks in advance.
>

Although I use Metrowerks compilers these days, I recall that simply
adding math.c to the ANSI/A4 project & rebuilding does the trick. If you
need to use sprintf with floating point numbers, you also need to set the
prefix for that project to enable floating point support.

Paul C. Pharr

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

From [e-mail] (Paul C. Pharr)
Date: Thu, 03 Oct 1996 19:06:35 -0400
Organization: Graphsoft, Inc.

In article <kgardner-3009960929290001@149.142.171.200>,
[e-mail] (Kent Gardner) wrote:

>I am using the Symantec C++ compiler to write a code resource that
>requires calls to math routines. Math routines (e.g., SIN) are not
>part of the ANSI-A4 library. Any suggestions on how to make the call.
>I assume I'll need to save an restore registers... Is there an
>example out there on how to do this. Thanks in advance.
>

Although I use Metrowerks compilers these days, I recall that simply
adding math.c to the ANSI/A4 project & rebuilding does the trick. If you
need to use sprintf with floating point numbers, you also need to set the
prefix for that project to enable floating point support.

Paul C. Pharr

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

From [e-mail] (Tim DeBenedictis)
Subject: How to create MacOS 6-7 dual-boot system?
Date: Thu, 26 Sep 1996 21:46:50 GMT
Organization: NETCOM On-line Communication Services (408 261-4700 guest)

I got a hold of an ancient Mac IIci in order to test some software with
very old Mac System Software.  I want to install System 6.0.8, 7.0.1,
7.1.3, and 7.5.5 all on this machine; in order to do so, I need to divide
the hard drive into (at least) 4 partitions, each containing the
requisite Mac system software.

When I used Apple's HD SC Setup program to custom-partition the drive,
however, I ran into a problem: once I created a Mac volume on the drive,
the HD SC Setup program would no longer present me with the option of
creating another one!  This was true with the version of HD SC Setup
included with -every- System Software version mentioned above.  Is there
any way to get Apple HD SC Setup to do what I want?  Or do I have to use
a third-party HD formatting utility like FWB or Silverlining?

Alternatively, is it possible to create a multiple-System configuration
without partitioning the hard drive?  Somewhere I remeber seeing a
control panel (?) called System Picker (I think) which would let the user
pick an "active" system folder on a single drive with multiple System
folders.

Would simply renaming the System Folders I don't want to use something
other than "System Folder" work?  I haven't tried this yet but I don't
think it will be this easy.

Thanks, and please send e-mail as well as posting,

-Tim DeBenedictis
[e-mail]


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

From [e-mail] (Nevin ":-]" Liber)
Date: Sat, 28 Sep 1996 11:32:23 -0700
Organization: University of Arizona CS Department, Tucson "It's too dang hot!" Arizona

In article <[e-mail]>, [e-mail] (Tim
DeBenedictis) wrote:

> When I used Apple's HD SC Setup program to custom-partition the drive,
> however, I ran into a problem: once I created a Mac volume on the drive,
> the HD SC Setup program would no longer present me with the option of
> creating another one!  This was true with the version of HD SC Setup
> included with -every- System Software version mentioned above.  Is there
> any way to get Apple HD SC Setup to do what I want?

Make sure you have a copy of the latest Apple HD SC Setup (7.3.5); I don't
believe that the previous versions allow multiple partitions.

> Alternatively, is it possible to create a multiple-System configuration
> without partitioning the hard drive?

Yes.  What makes a system folder "blessed" is having both "Finder" and
"System" in it.  You can unbless a folder by moving one of these out of the
system folder (I usually put the Finder in a subfolder), and bless a
different one by putting the appropriate Finder and System files in the
same folder.  I know that there are some utilities out there that do this,
but I haven't used any of them in years.

> Would simply renaming the System Folders I don't want to use something
> other than "System Folder" work?

You can rename system folders to whatever you want; it doesn't matter.
--
        Nevin ":-)" Liber       [e-mail]    [phone]
                                http://www.cs.arizona.edu/people/nevin/

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

From [e-mail] (Chaz Larson)
Date: Thu, 03 Oct 1996 22:41:57 -0500
Organization: Mystery Men

In article <[e-mail]>, [e-mail] (Tim
DeBenedictis) wrote:

> Or do I have to use
> a third-party HD formatting utility like FWB or Silverlining?

Yes, so far as I know.

> Alternatively, is it possible to create a multiple-System configuration
> without partitioning the hard drive?  Somewhere I remeber seeing a
> control panel (?) called System Picker (I think) which would let the user
> pick an "active" system folder on a single drive with multiple System
> folders.

oooh.  Don't do this.  In theory, it will work, but the method is fraught
with peril.

> Would simply renaming the System Folders I don't want to use something
> other than "System Folder" work?  I haven't tried this yet but I don't
> think it will be this easy.

Say you have four system folders on the disk.  To make System Folder A
active, you'd make sure that it was the only one in which System and
FInder are at the same level, then open and close it, then reboot.

But you really don't want to do that.  Partition the drive, or better yet
get a zip drive and use four zip carts for your four Systems.  If you
regularly change between System 6, 7, and 7.5, you'll be rebuilding your
desktop with every boot.  Yuck.

chazl
10.03.96
- ------

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

From [e-mail] (Chaz Larson)
Date: Thu, 03 Oct 1996 22:41:57 -0500
Organization: Mystery Men

In article <[e-mail]>, [e-mail] (Tim
DeBenedictis) wrote:

> Or do I have to use
> a third-party HD formatting utility like FWB or Silverlining?

Yes, so far as I know.

> Alternatively, is it possible to create a multiple-System configuration
> without partitioning the hard drive?  Somewhere I remeber seeing a
> control panel (?) called System Picker (I think) which would let the user
> pick an "active" system folder on a single drive with multiple System
> folders.

oooh.  Don't do this.  In theory, it will work, but the method is fraught
with peril.

> Would simply renaming the System Folders I don't want to use something
> other than "System Folder" work?  I haven't tried this yet but I don't
> think it will be this easy.

Say you have four system folders on the disk.  To make System Folder A
active, you'd make sure that it was the only one in which System and
FInder are at the same level, then open and close it, then reboot.

But you really don't want to do that.  Partition the drive, or better yet
get a zip drive and use four zip carts for your four Systems.  If you
regularly change between System 6, 7, and 7.5, you'll be rebuilding your
desktop with every boot.  Yuck.

chazl
10.03.96
- ------

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

From [e-mail] (Bill Vaughan)
Subject: How to do Unix-like scripting on a Mac?
Date: 24 Sep 1996 21:49:44 GMT
Organization: Sun Microsystems Inc.

I need to be able to do Unix-like scripting on the Mac to set environment variables and run tools/apps without going through the GUI. (No, I'm not one of those Unix/DOS Neanderthals who walks around grunting, "Me, I REAL man, I use command line interface, GUI for sissies.") I need to set up a multi-platform test harness. I have considered AppleScript, but the problem is that I will be working with some tools and apps which are not AppleScript Aware. Any suggestions would be greatly appreciated.

Thanks,

Bill



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

From [e-mail] (Robert Dalgleish)
Date: Wed, 25 Sep 1996 09:11:42 -0600
Organization: Industrial Strength Software

In article <529l1o$[e-mail]>, [e-mail] wrote:

>>I need to be able to do Unix-like scripting on the Mac to set environment
>variables and run tools/apps without going through the GUI. (No, I'm not
>one of those Unix/DOS Neanderthals who walks around grunting, "Me, I REAL
>man, I use command line interface, GUI for sissies.") I need to set up a
>multi-platform test harness. I have considered AppleScript, but the
>problem is that I will be working with some tools and apps which are not
>AppleScript Aware. Any suggestions would be greatly
>appreciated.

The best multiplatform test harness would be Perl.  The MacPerl implementation
is currently at Perl 5.001, with an advanced beta test version at 5.002. It is
very competent and fast.   The MacPerl FAQ can be found at
http://www.perl.com/CPAN/doc/FAQs/mac/MacPerlFAQ.html

To help with the cross-platform stuff, you will have to have a plug-in
architecture for things such as process-spawning and piping.  If you look
at the test harness that Matthias created for testing MacPerl, you may
see what I mean.

--
Bob Dalgleish   [e-mail]
Web Page        http://www.sasknet.com/~dalgl/

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

From [e-mail] (Ray Johnson)
Date: 25 Sep 1996 16:44:24 GMT
Organization: Sun Microsystems Inc., Mountain View, CA

Robert Dalgleish ([e-mail]) wrote:
: In article <529l1o$[e-mail]>, [e-mail] wrote:

: >>I need to be able to do Unix-like scripting on the Mac to set environment
: >variables and run tools/apps without going through the GUI. (No, I'm not
: >one of those Unix/DOS Neanderthals who walks around grunting, "Me, I REAL
: >man, I use command line interface, GUI for sissies.") I need to set up a
: >multi-platform test harness. I have considered AppleScript, but the
: >problem is that I will be working with some tools and apps which are not
: >AppleScript Aware. Any suggestions would be greatly
: >appreciated.

: The best multiplatform test harness would be Perl.  The MacPerl implementation
: is currently at Perl 5.001, with an advanced beta test version at 5.002. It is
: very competent and fast.   The MacPerl FAQ can be found at
: http://www.perl.com/CPAN/doc/FAQs/mac/MacPerlFAQ.html

You should also consider the Tcl scripting language whoch has been available
on the Mac for some time & is also very stable.  You can check out the
home page for the project at http://www.sunlabs.com/research/tcl/

There is a package called swig that will parse your C header files and
generate Tcl commands for all of your C API's.  It is a perfect tool
for testing.  Much better than Perl.  In fact, Apple has used Tcl for
testing parts of the Copland kernal - or so I hear through the grape vine...

Ray



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

From [e-mail] (Robert Dalgleish)
Date: Thu, 26 Sep 1996 08:21:49 -0600
Organization: Industrial Strength Software

In article <52bnh8$[e-mail]>, [e-mail] (Ray
Johnson) wrote:
>You should also consider the Tcl scripting language whoch has been available
>on the Mac for some time & is also very stable.  You can check out the
>home page for the project at http://www.sunlabs.com/research/tcl/
>
>There is a package called swig that will parse your C header files and
>generate Tcl commands for all of your C API's.  It is a perfect tool
>for testing.  Much better than Perl.

I cannot let that go unchallenged.  Perl has several facilities to
translate C header files to Perl files.  Perl's capabilities at pattern
matching make it a better platform for testing, especially against
systems that support only minimal pattern matching.  TCL is not "Much better"
than Perl.

--
Bob Dalgleish   [e-mail]
Web Page        http://www.sasknet.com/~dalgl/

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

From [e-mail] (Matthias Neeracher)
Date: 26 Sep 1996 18:39:33 GMT
Organization: Integrated Systems Laboratory, ETH, Zurich

In article <[e-mail]>, [e-mail] (Robert Dalgleish) writes:
> In article <52bnh8$[e-mail]>, [e-mail] (Ray
> Johnson) wrote:
>> You should also consider the Tcl scripting language whoch has been available
>> on the Mac for some time & is also very stable.  You can check out the
>> home page for the project at http://www.sunlabs.com/research/tcl/
>>
>> There is a package called swig that will parse your C header files and
>> generate Tcl commands for all of your C API's.  It is a perfect tool
>> for testing.  Much better than Perl.

> I cannot let that go unchallenged.  Perl has several facilities to
> translate C header files to Perl files.

Right. In particular, SWIG also generates Perl5, so any header which can be
translated by SWIG to Tcl should automatically also work in Perl5.

> Perl's capabilities at pattern matching make it a better platform for
> testing, especially against systems that support only minimal pattern
> matching.

Agreed, but Tcl has a simpler syntax and is much easier to embed in another
application.

> TCL is not "Much better" than Perl.

I think Ray meant his comment only in the context of testing, for which I
would tend to agree with him (Tcl is also probably better for Tk on Un*x
platforms, and definitely so on the Mac).

Matthias

- ---
Matthias Neeracher   <[e-mail]>   http://www.iis.ee.ethz.ch/~neeri
   "Computer languages follow the extended Sapir-Whorf hypothesis, which
    states that because the Eskimos had 900 words for snow, they decided
    to move north." -- Dave Griffith <[e-mail]>


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

From [e-mail] (Roger Smith)
Date: Tue, 24 Sep 1996 21:51:07 -0500
Organization: Base Online

In article <529l1o$[e-mail]>, [e-mail] wrote:

> I need to be able to do Unix-like scripting on the Mac to set
environment variables and run tools/apps without going through the GUI.
(No, I'm not one of those Unix/DOS Neanderthals who walks around grunting,
"Me, I REAL man, I use command line interface, GUI for sissies.") I need
to set up a multi-platform test harness. I have considered AppleScript,
but the problem is that I will be working with some tools and apps which
are not AppleScript Aware. Any suggestions would be greatly appreciated.
>

Depending on what you want to accomplish, you can use the MPW shell or you
can use the Apple VU tool (or you can use them in tandem) see either this
or last months's MacTech mag for a review on VU -- virtual user).

Roger Smith

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

From [e-mail] (Glenn Andreas)
Date: 27 Sep 1996 14:45:37 GMT
Organization: GAndreas Software

In article <[e-mail]>,
[e-mail] (Robert Dalgleish) wrote:

> In article <52bnh8$[e-mail]>, [e-mail] (Ray
> Johnson) wrote:
...
> >There is a package called swig that will parse your C header files and
> >generate Tcl commands for all of your C API's.  It is a perfect tool
> >for testing.  Much better than Perl.
>
(while we're at it...)

Python is also supported by swig, as is, for that matter, Perl 5 (and to
some extent, Perl 4).  "Much better" depends on what you need it for:

Perl:  Very good regular expressions, but "interesting" syntax
Python: Very good object model, easy to extend, easy to embed, but
structure determined by indentation bothers some people
TCL: Imagine infix scheme written as a macro processor, with all the power
and pitfalls that intails.

Personally, I use Python (since it has pretty good mac os support,
including apple events and a fair chunk of the toolbox).

--
Glenn Andreas                                 Author of Macintosh games:
[e-mail]                               Theldrow 2.3
http://www.skypoint.com/members/gandreas            Blobbo 1.0.2
ftp://ftp.skypoint.com/pub/members/g/gandreas
Unsolicited bulk email will be proofread for a US$500/k, min $1000

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

From [e-mail] (Ray Johnson)
Date: 2 Oct 1996 15:55:25 GMT
Organization: Sun Microsystems Inc., Mountain View, CA

Glenn Andreas ([e-mail]) wrote:
: Python is also supported by swig, as is, for that matter, Perl 5 (and to
: some extent, Perl 4).  "Much better" depends on what you need it for:

: Perl:  Very good regular expressions, but "interesting" syntax
: Python: Very good object model, easy to extend, easy to embed, but
: structure determined by indentation bothers some people
: TCL: Imagine infix scheme written as a macro processor, with all the power
: and pitfalls that intails.

I guess I should have mentioned that swig was also available for python
and perl.  It's a great package and should cerainly be checked out no
matter which of the above three languages you use.

Last time I checked Python was not that easy to embed.  Perhaps it has
been improved?  Tcl is trivial to embed.  It took me only two weeks
to write the Tcl/Tk plugin for Netscape.  (Mind you it's still rather
buggy, but that's due to problems with the Netscape API...)

Tcl's other advantage is that it is very easy to extend.  The API for
creating new commands is simple.  In fact, you can create new control
structures as an extension.

Of course, Tcl's greatest advantage is Tk.  Tk is a cross platform
windowing toolkit available for UNIX, Windows & Mac.  Unfortunantly,
the Mac version currently use the Motif look & feel (I'm working
full time to make it native look & feel.)  In fact, both Python
and Perl use Tk as a toolkit for it's environments!

I would like to point out one thing, however.  Perl and Python are
both very cool languages.  I look at languages as tools.  The more
the better - use the best tool for the job.  The intent of my original
message was not to start a flame war - but rather to encourage
developers to look at all the available tools before deciding what
to use.

Ray



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

From Mary Smith <[e-mail]>
Date: Thu, 03 Oct 1996 15:58:16 -0400
Organization: Voicenet - Internet Access - [phone]

Hi.

I'm responding from a friend's account to someone's post regarding
how to do Unix scripting on the Mac.

One answer is to use a C Shell. Some time ago I wrote one called
UltraShell, from scatch in fact, for commercial sale. I was selling it
for $40 at the time, but the version that I still have on hand is
somewhat buggy and I am willing to sell it at $20 plus shipping,
until I can put out another version. Note, this shell does not
support AppleScript, and it prefers System 7.1 more than 7.5.
However it incorporate about 25 common Unix commands e.g. tar, du,
ps, find, etc; and the executable is only about 100 kB.

If anyone is interested in UltraShell, just send an email to
this account. I'll send you a broshure.

Cheers,

Zack T. Smith

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

From [e-mail] (Steve Byan)
Date: Fri, 04 Oct 1996 12:32:11 -0400
Organization: Hitachi Computer Products, Inc.

Use the MPW (Macintosh Programmer's Workbench) shell. The literals are
different (alas) but the syntax is quite similar.  It's the best
command-line interface for the Mac, and I wish I had it on my HP
workstation too.

Regards,
-Steve

--
Steve Byan                                  internet: [e-mail]
Hitachi Computer Products (America), Inc.
1601 Trapelo Road                           phone: [phone]
Waltham, MA 02154                           FAX: [phone]

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

From [e-mail] (Steve Byan)
Date: Fri, 04 Oct 1996 12:32:11 -0400
Organization: Hitachi Computer Products, Inc.

Use the MPW (Macintosh Programmer's Workbench) shell. The literals are
different (alas) but the syntax is quite similar.  It's the best
command-line interface for the Mac, and I wish I had it on my HP
workstation too.

Regards,
-Steve

--
Steve Byan                                  internet: [e-mail]
Hitachi Computer Products (America), Inc.
1601 Trapelo Road                           phone: [phone]
Waltham, MA 02154                           FAX: [phone]

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

From Zoila Deleg-Bricker <[e-mail]>
Subject: Inside Macintosh's Venn Diagrammer: has anyone done it?
Date: 2 Oct 1996 14:38:38 GMT
Organization: Rensselaer Polytechnic Institute/School of Management & Technology

Hi:

I would like to be instructed by the Venn Diagrammer application that is
described in the IM Overview volume:

            "The remainder of this book illustrates how to write a Macintosh
            application by gradually dissecting the source code of a very
            simple sample application, called Venn Diagrammer. This
            application allows the user to use Venn diagrams as a method of
            determining whether a given syllogism is valid (that is, whether
            the conclusion must be true if both premises are true). This
            section briefly describes the operation of the Venn Diagrammer
            application."

It's far beyond the present state of my programming skills to create this
application, yet I would like to see how it works sooner rather than the much
later that it would take for me to develop the necessary knowledge. Does
anyone have the Venn Diagrammer that they created, by working through IM
Overview, somewhere sitting in a Syquest cartridge that they wouldn't mind
binhexing and emailing me? Do you know someone else who created it?

Thanks very much

Dale Bricker
[e-mail]

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

From [e-mail] (Matt Bishop)
Date: Wed, 02 Oct 1996 15:19:18 -0500
Organization: Zippo

In article <52tupe$[e-mail]>, Zoila Deleg-Bricker
<[e-mail]> wrote:

> Hi:
>
> I would like to be instructed by the Venn Diagrammer application that is
> described in the IM Overview volume:

I made the mistake of buying IM Overview, too.  If you still have the
receipt, take it back and buy "Programming Starter Kit for Macintosh" by
Jim Trudeau.  Not only does it clearly explain how the Macintosh OS works,
it teaches you how to program it, something Overview does not do.  It also
teaches in C as opposed to Pascal.

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

From Zoila Deleg-Bricker <[e-mail]>
Date: 3 Oct 1996 03:55:51 GMT
Organization: Rensselaer Polytechnic Institute/School of Management & Technology

In article <[e-mail]> Matt Bishop,
[e-mail] writes:
>ubject: Re: Inside Macintosh's Venn Diagrammer: has anyone done it?
>From: Matt Bishop, [e-mail]
>Date: Wed, 02 Oct 1996 15:19:18 -0500
>>In article <52tupe$[e-mail]>, Zoila Deleg-Bricker
><[e-mail]> wrote:
>
>> Hi:
>>
>> I would like to be instructed by the Venn Diagrammer application that is
>> described in the IM Overview volume:
>
>I made the mistake of buying IM Overview, too.  If you still have the
>receipt, take it back and buy "Programming Starter Kit for Macintosh" by
>Jim Trudeau.  Not only does it clearly explain how the Macintosh OS works,
>it teaches you how to program it, something Overview does not do.  It also
>teaches in C as opposed to Pascal.

Whoops! When I said "I would like to be instructed by the Venn Diagrammer
application....", I mean instructed in syllogistic reasoning, not programming.
I already own Programming Starter Kit for Macintosh and would never think of
solely relying on IM Overview to learn how to program for the Mac.

So, again......does anyone have a copy of Venn Diagrammer? Pascal programmers?


Zoila

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

From [e-mail] (Matt Bishop)
Subject: Looking for ictb, aete editor, NOT resourcerer
Date: Sat, 21 Sep 1996 13:57:30 -0500
Organization: Bishop Shareware

I am searching for a resource editor that can edit ictb and aete
resources, but so far the only thing I have come up with is Resourcerer.
It is a great program, but it costs so damn much!  Some of us have to pay
for these utilities ourselves.

If you have anything, please let me know.

-Matt

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

From [e-mail] (Steve Wilson)
Date: 21 Sep 1996 20:24:15 GMT
Organization: Emergent Behavior

In article <[e-mail]>, [e-mail]
(Matt Bishop) wrote:

> I am searching for a resource editor that can edit ictb and aete
> resources, but so far the only thing I have come up with is Resourcerer.
> It is a great program, but it costs so damn much!  Some of us have to pay
> for these utilities ourselves.

I know.  I had to pay for it myself too, but it is worth every penny.
Actually, knowing what I know now, I'd pay double the price for it.
Resourcerer is soooooo much better than ResEdit.  The tools for editing
PICT resources are awsome.  Couldn't live without them.

Steve Wilson
Emergent Behavior
[phone]
[e-mail]
http://www.emer.com

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

From [e-mail] (Steve Wilson)
Date: 21 Sep 1996 20:24:15 GMT
Organization: Emergent Behavior

Reposting article removed by rogue canceller.

In article <[e-mail]>, [e-mail]
(Matt Bishop) wrote:

> I am searching for a resource editor that can edit ictb and aete
> resources, but so far the only thing I have come up with is Resourcerer.
> It is a great program, but it costs so damn much!  Some of us have to pay
> for these utilities ourselves.

I know.  I had to pay for it myself too, but it is worth every penny.
Actually, knowing what I know now, I'd pay double the price for it.
Resourcerer is soooooo much better than ResEdit.  The tools for editing
PICT resources are awsome.  Couldn't live without them.

Steve Wilson
Emergent Behavior
[phone]
[e-mail]
http://www.emer.com

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

From [e-mail] (Mike Tilstra)
Date: Thu, 26 Sep 1996 10:11:37 -0500
Organization: TaDPoL

In article <[e-mail]>, [e-mail]
(Matt Bishop) wrote:

>Reposting article removed by rogue canceller.
>
>I am searching for a resource editor that can edit ictb and aete
>resources, but so far the only thing I have come up with is Resourcerer.
>It is a great program, but it costs so damn much!  Some of us have to pay
>for these utilities ourselves.
>
>If you have anything, please let me know.
>
>-Matt

There are templates for ResEdit for atleast the aete resource.  I'm not
sure about ictb though.

--
Mike Tilstra------TaDPoL-------Limiter Task
     mailto:[e-mail]
http://www.tc.umn.edu/nlhome/g019/tils0007/
    For the first time in your memory,
    you're generating your own light.

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

From [e-mail] (Don Arbow)
Date: Sat, 28 Sep 1996 11:37:17 -0700
Organization: Wolfe Internet Access, L.L.C.

In article <[e-mail]>,
[e-mail] (Matt Bishop) wrote:

:  Reposting article removed by rogue canceller.
:
:  I am searching for a resource editor that can edit ictb and aete
:  resources, but so far the only thing I have come up with is Resourcerer.
:  It is a great program, but it costs so damn much!  Some of us have to pay
:  for these utilities ourselves.
:
:  If you have anything, please let me know.
:
:  -Matt

Resorcerer has a educational discount.

There's also an 'aete' editor out there called EightyRez.  It's only about
$40.  See their web page at http://www.mv.com/biz/condes/EightyRez.html.

Don

--
- ------------------------------------------------------------------------
    \ | /    Don Arbow, Partner, CTO |  [e-mail]
  -- EDO --  EveryDay Objects, Inc.  |  http://www.pla-net.net/edo
    / | \    Seattle, WA             |
- ------------------------------------------------------------------------
  My Prographing web page:   http://www.wolfe.net/~donarb/Dataflow.html
        "The fix is only temporary, unless it works" - Red Green
- ------------------------------------------------------------------------

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

From [e-mail] (Steve Herman)
Date: 3 Oct 1996 16:14:43 GMT
Organization: CSC

In article <[e-mail]>,
[e-mail] (Don Arbow) wrote:

> In article <[e-mail]>,
> [e-mail] (Matt Bishop) wrote:
>
> :  Reposting article removed by rogue canceller.
> :
> :  I am searching for a resource editor that can edit ictb and aete
> :  resources, but so far the only thing I have come up with is Resourcerer.
> :  It is a great program, but it costs so damn much!  Some of us have to pay
> :  for these utilities ourselves.
> :
> :  If you have anything, please let me know.
> :
> :  -Matt
>
> Resorcerer has a educational discount.
>
> There's also an 'aete' editor out there called EightyRez.  It's only about
> $40.  See their web page at http://www.mv.com/biz/condes/EightyRez.html.
>
> Don
>

There's also a HyperCard stack for creating and modifying 'aete's which (I
believe) was created by Apple.  It comes with products such as the
AppleScript development kit and ETO, but I'm not sure if it's freely
available or not.

It's called "Aete Editor Stack" or something like that.

Steve

- --------------------------------------------------
- Steve Herman - PrISMS
- Computer Sciences Corporation
- Marshall Space Flight Center
- Huntsville, AL
- --------------------------------------------------

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

From [e-mail] (Michael Grueter)
Subject: Memory Loss with CustomGetFile
Date: Wed, 02 Oct 1996 15:29:48 -0800
Organization: Misty City Software

Hi,

Every time I call the toolbox function "CustomGetFile", with a custom
dialog hook and file filter, I loose approximately 288 bytes of memory. If
I make the same call but pass nil for the dialog hook and file filter
parameters, I do not loose any bytes. I can't figure out why I'm loosing
the bytes! I've included my calling sequence and my dialog hook and file
filter functions (which is mostly just taken from Inside Mac) below. Note
that if the code for "A" file type is used, the bytes are lost. If the
code for "B" file type is used, the bytes are not lost. If anyone can help
me make sense of this, I would greatly appreciate it.

- Mike

Code to follow:
- ------------------------------------------

var
   custFilterUPP: FileFilterYDUPP;
   dlogHookUPP: DlgHookYDUPP;

...

if fileType0 = kAFileType then
   begin {"A" File Type}
{causes bytes to be lost}
      dlogHookUPP := NewDlgHookYDProc(DlgHookYDProcPtr(@MyFilterDlgHook));
      custFilterUPP := NewFileFilterYDProc(FileFilterYDProcPtr(@MyFileFilter));
      dlogID := kCustomGetDialog;
   end {"A" File Type}
else
   begin {"B" File Type}
{does not cause bytes to be lost}
      dlogHookUPP := nil;
      custFilterUPP := nil;
      dlogID := 0;
   end; {"B" File Type}

CustomGetFile(custFilterUPP, numTypes, @theTypeList, theSFReply, dlogID,
dlogTopLeft, dlogHookUPP, nil, nil, nil, nil);
if theSFReply.sfGood then
   ; {user didn't cancel}

DisposeRoutineDescriptor(dlogHookUPP);
DisposeRoutineDescriptor(custFilterUPP);


{--------------------------------------------------------------}
   function MyFileFilter (pb: CInfoPBPtr; myDataPtr: Ptr): Boolean;
   begin {MyFileFilter}
      MyFileFilter := true;   {default: don't show the file (note reverse
sense)}

      if gCurrentType = kAllFilesChoice then
         MyFileFilter := false

      else if gCurrentType = kDosWinFileChoice then
         begin {kDosWinFileChoice}
            if (pb^.ioFlFndrInfo.fdType = kTextType) |
(pb^.ioFlFndrInfo.fdType = kDOSWinType) then
               MyFileFilter := false;
         end {kDosWinFileChoice}

      else if pb^.ioFlFndrInfo.fdType = kMacFileType then {default to
kMacFileChoice}
         MyFileFilter := false;
   end; {MyFileFilter}

{--------------------------------------------------------------}
   function MyFilterDlgHook (theItem: integer; theDialog: DialogPtr;
myDataPtr: Ptr): integer;
      var
         pickedType: integer; {menu item selected}

      const
         kFileTypePopup = 10; {item number of File Format pop-up menu}

   begin {MyFilterDlgHook}
      MyFilterDlgHook := theItem;   {by default, return the item passed in}
      if GetWRefCon(WindowPtr(theDialog)) <> longint(sfMainDialogRefCon) then
         Exit(MyFilterDlgHook);  {this function is only for main dialog}

   {Do processing of pseudo-items and your own additional item.}
      case theItem of
         sfHookFirstCall: {pseudo-item: first time function called}
            begin {sfHookFirstCall}
               gCurrentType := kMacFileChoice;
               SetControlValue(ControlHandle(GetItemHandle(theDialog,
kFileTypePopup)), gCurrentType);
               MyFilterDlgHook := sfHookNullEvent;
            end; {sfHookFirstCall}

         kFileTypePopup:   {user selected File Type pop-up menu}
            begin {kFileTypePopup}
               pickedType :=
GetControlValue(ControlHandle(GetItemHandle(theDialog, kFileTypePopup)));
               if pickedType <> gCurrentType then
                  begin
                     gCurrentType := pickedType;
                     MyFilterDlgHook := sfHookRebuildList;
                  end;
            end; {kFileTypePopup}

         otherwise
            ;  {ignore all other items}
      end; {case theItem}
   end; {MyFilterDlgHook}

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

From [e-mail]
Date: Fri, 04 Oct 1996 19:03:57 -0700
Organization: (none)

In article <[e-mail]>,
[e-mail] (Michael Grueter) wrote:

> Hi,
>
> Every time I call the toolbox function "CustomGetFile", with a custom
> dialog hook and file filter, I loose approximately 288 bytes of memory.

This could be one of the known problems with system software.  You don't
mention which system software version you are running, but 7.5 and 7.5.1
(at least) had  standardfile leaks.  7.5.3 fixed all of them, as far as I
remember.  Look at the various system software technotes at
<http://devworld.apple.com/dev/technotes.shtml> for details.

--
(Pointers to other Mac programming web sites at
<http://devworld.apple.com/dev/geeks.html>)

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

From [e-mail]
Date: Fri, 04 Oct 1996 19:03:57 -0700
Organization: (none)

In article <[e-mail]>,
[e-mail] (Michael Grueter) wrote:

> Hi,
>
> Every time I call the toolbox function "CustomGetFile", with a custom
> dialog hook and file filter, I loose approximately 288 bytes of memory.

This could be one of the known problems with system software.  You don't
mention which system software version you are running, but 7.5 and 7.5.1
(at least) had  standardfile leaks.  7.5.3 fixed all of them, as far as I
remember.  Look at the various system software technotes at
<http://devworld.apple.com/dev/technotes.shtml> for details.

--
(Pointers to other Mac programming web sites at
<http://devworld.apple.com/dev/geeks.html>)

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

From Josh Birnabaum <[e-mail]>
Subject: Problems with SetDialogDefaultItem
Date: Wed, 02 Oct 1996 23:03:32 -0500
Organization: FGBNY

Hello,

I'm currently writing a cdev that has two main buttons in it to control
it's functionality. I'm trying to declare one of them as the default
button and draw the thick banding around it.
The furthest I get is the button being greyed out with the banding
around it greyed out as well. Ie, it's useless.  This is only after I
hit the other of the 2 buttons and then go back and hit the first target
button.
Below is the code called upon hitting the button I want as the default
banded button.

case BUTTON_ONE_ITEM:
     GetDItem( the_dialog, BUTTON_ONE_ITEM, &itemType,
                                                &itemHandle, &itemRect
);
     SetDialogDefaultItem( the_dialog, BUTTON_ONE_ITEM );
     /* code to handle this items functionality here */
     SysBeep(1);
     break;

If you could e-mail me back with any suggestions etc I would be most
greatful.

Thank You

Josh Birnbaum
[e-mail]

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

From [e-mail]
Date: Fri, 04 Oct 1996 19:07:28 -0700
Organization: (none)

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

> Hello,
>
> I'm currently writing a cdev that has two main buttons in it to control
> it's functionality. I'm trying to declare one of them as the default
> button and draw the thick banding around it.

Look at the technote TB 37, "Pending Update Perils" at
<http://devworld.apple.com/dev/technotes/tb/tb_37.html>.  You must call the
standard filter proc in order for SetDialogDefaultItem to work.

--
(Pointers to other Mac programming web sites at
<http://devworld.apple.com/dev/geeks.html>)

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

From [e-mail] (Jan-Hinnerk Reichert)
Subject: Re^2: MacOS 8 suggestion (Re: Disk space efficiency)
Date: 2 Oct 96 21:16:05 +0200
Organization: (none)

In der Nachricht vom 24.Sep 96. 11:28 schrieb Laurent Pointal
([e-mail]) unter dem Betreff "Re: MacOS 8 suggestion
(Re: Disk space efficiency)":

> Daniel Nash wrote:
> [STUFF DELETED]
> > Well, a nice feature to have in MacOS 8 (I know it's probably a
> > bit late
> > for changes now) would be to have the ability to mount volumes as
> > a folder
> > on another volume, as in most UNIXes.  That way, you could set up
> > your
> > partition, and then place them all into one desktop 'drive' icon.
> > I hope
> > someone form Apple is listening.
> >
> >  - Daniel Nash
>
> ? Cant an alias of a partition (volume) do the same thing ?
> The difference: its just mounted when accessed and not a boot time/
> mount command.

Yes, you could do this. But you still would have all these volumes
lying on the Desktop. I have got 5 HD-Icons on my Desktop (14'') from
only one HD (2GB) :-(((

Of course, you could place the Icons on the same place of the Desktop,
so you can see/access only the topmost. But that doesn't look like a
MAC-like solution ;-)

- --------------------------------------------------------------------
 "Few other griefs amid the ill chances    **************************
  of this world have more bitterness       *                        *
  and shame for a man's heart than         *  Jan-Hinnerk Reichert  *
  to behold the love of a lady so fair     *   [e-mail]    *
  and brave that cannot be returned."      *                        *
                  Aragorn son of Arathorn  **************************
- --------------------------------------------------------------------

- MacZPoint 1.92 -


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

From David Thomas Mcwherter <[e-mail]>
Date: Fri,  4 Oct 1996 09:36:10 -0400
Organization: Carnegie Mellon, Pittsburgh, PA

>>>
stuff regarding the mounting of volumes in MacOS8 clipped for brevity
<<<<

I'd like to see MacOS 8 be able to make your 5-6 HD icons all look like
one...support the virtual addition of numerous drives.  It's similar
to what has been described as making a volume appear as a folder, but
slightly different.  In this manner, any folder on a virtual hard drive
could have its contents stored on perhaps 4 or more different drives.
This is available through RAID setups, but I'd like to see it built
into the OS and available for any type of HD.  This would make nightmares
if one of your hard drives went down, however (you'd lose data "randomly")

-David McWherter


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

From Josh Birnabaum <[e-mail]>
Subject: Setting dialog default buttons
Date: Thu, 03 Oct 1996 20:10:32 -0500
Organization: FGBNY

Hello,

Does anyone have any code to declare a button item in a cdev as the
default button with the thick banding around it??
I am using the following code that I think is correct to an extent as
after I hit another button in the same cdev, the target button becomes
greyed out with the banding around it but only then:

case BUTTON_ONE_ITEM:
     SetDialogDefaultItem( the_dialog, BUTTON_ONE_ITEM );
     /* buttons functionality here */
     SysBeep(1);
     break;

Any help would be appreciated. I can be reached at the e-mail address
that follows.

Thank You

Josh Birnbaum
[e-mail]

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

From [e-mail](KevDog) (Kevin Whitty)
Date: Fri, 04 Oct 1996 15:14:16 +0300
Organization: ≈bo Akademi Fˆrbr‰nningskemi

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

> Hello,
>
> Does anyone have any code to declare a button item in a cdev as the
> default button with the thick banding around it??
> I am using the following code that I think is correct to an extent as
> after I hit another button in the same cdev, the target button becomes
> greyed out with the banding around it but only then:
>
> case BUTTON_ONE_ITEM:
>      SetDialogDefaultItem( the_dialog, BUTTON_ONE_ITEM );
>      /* buttons functionality here */
>      SysBeep(1);
>      break;

OK.  I just had a similar question regarding SDDI(), but I've since
figured out the answer myself.  If it's a modal dialog, make sure you're
calling the standard filter procedure (what is it now?...
GetStdDialogProc, or StdDialogProc, or something like that) since SDDI()
requires it.  If it's a modeless dialog, make sure you call the filter
proc on null events, too.  Or maybe it doesn't matter for only a default
button, but you have to make sure and call it if you are using
SetDialogTracksCursor().  Not to get off subject, but if you call SDTC()
in a modeless, then you also have to call DialogSelect() if you want the
carrat to blink.

Hope I didn't confuse you.  :)

KevDog
[e-mail]

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

From [e-mail](KevDog) (Kevin Whitty)
Date: Fri, 04 Oct 1996 15:14:16 +0300
Organization: ≈bo Akademi Fˆrbr‰nningskemi

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

> Hello,
>
> Does anyone have any code to declare a button item in a cdev as the
> default button with the thick banding around it??
> I am using the following code that I think is correct to an extent as
> after I hit another button in the same cdev, the target button becomes
> greyed out with the banding around it but only then:
>
> case BUTTON_ONE_ITEM:
>      SetDialogDefaultItem( the_dialog, BUTTON_ONE_ITEM );
>      /* buttons functionality here */
>      SysBeep(1);
>      break;

OK.  I just had a similar question regarding SDDI(), but I've since
figured out the answer myself.  If it's a modal dialog, make sure you're
calling the standard filter procedure (what is it now?...
GetStdDialogProc, or StdDialogProc, or something like that) since SDDI()
requires it.  If it's a modeless dialog, make sure you call the filter
proc on null events, too.  Or maybe it doesn't matter for only a default
button, but you have to make sure and call it if you are using
SetDialogTracksCursor().  Not to get off subject, but if you call SDTC()
in a modeless, then you also have to call DialogSelect() if you want the
carrat to blink.

Hope I didn't confuse you.  :)

KevDog
[e-mail]

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

From [e-mail] (Jeremy Roussak)
Subject: Stack size in BOA
Date: 1 Oct 1996 06:21:24 GMT
Organization: Queen Mary & Westfield College, London, England

Is there really no way of increasing the stack size of a
background-only application from the default 2k to less than 24k?
TechNote PS2 implies not, but this seems rather curious.

Jeremy

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

From [e-mail] (Brian Stern)
Date: 2 Oct 1996 00:20:05 GMT
Organization: The University of Texas at Austin, Austin, Texas

In article <52qd94$[e-mail]>, [e-mail] (Jeremy
Roussak) wrote:

<Is there really no way of increasing the stack size of a
<background-only application from the default 2k to less than 24k?
<TechNote PS2 implies not, but this seems rather curious.
<
<Jeremy

Never say never.  The limit is LMGetMinStack() and this can be set to
whatever you like.


void
TfbaApp::InitToolBox( short moreMasters, long stackSize )
{
   if ( stackSize > 0 )
   {
      // It's not possible to set the stack to be smaller than
      // DfltStack.  So we adjust it first, set the
      // stacksize and then restore it.
      long  saveDefaultStack = ::LMGetMinStack();
      ::LMSetMinStack( stackSize );

      // Increase the stack size by lowering the heap limit.
       ::SetApplLimit( (Ptr) ( (unsigned long) GetApplLimit() - stackSize ) );
      ::LMSetMinStack( saveDefaultStack );
   }

   ::MaxApplZone();

   for ( int i = 0; i < moreMasters; ++i )
      ::MoreMasters();

   // This is the only manager that should be called
   // from an fba
   ::InitGraf( &qd.thePort );

}

____________________________________________________________________
Brian  Stern  {8-{)}                                 BeBox Hairy Guy                  [e-mail]                          [e-mail]
INIT Writing FAQ etc. at <ftp://ftp.pbcomputing.com//Guests/BrianS/>

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

From [e-mail] (James W. Walker)
Date: Tue, 01 Oct 1996 20:52:45 -0700
Organization: Nisus Software, Inc.

In article <52qd94$[e-mail]>, [e-mail] (Jeremy
Roussak) wrote:

>Is there really no way of increasing the stack size of a
>background-only application from the default 2k to less than 24k?
>TechNote PS2 implies not, but this seems rather curious.

I don't know what the Tech Note says, but it's possible.  However, you have
to temporarily change the low-memory global DefaultStack when you call
SetApplLimit.  It doesn't want to let you change the stack to something
less than the default.
--
  --  Jim Walker

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

From [e-mail] (Jump Long)
Date: 4 Oct 1996 01:02:40 -0400
Organization: America Online, Inc. (1-800-827-6364)

In article <52qd94$[e-mail]>, [e-mail] (Jeremy
Roussak) wrote:

>Is there really no way of increasing the stack size of a
>background-only application from the default 2k to less than 24k?
>TechNote PS2 implies not, but this seems rather curious.

Personally, I'd grow it to 24K and live with a little extra room because
there are lots of things that "borrow" (IMO) too much stack space at
interrupt time. If they borrow too much from your FBA, then it'll crash
and guess who will be blamed? Right... your FBA.

- Jim Luther (who learned that lesson by experience)

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

From [e-mail] (Jump Long)
Date: 4 Oct 1996 01:02:40 -0400
Organization: America Online, Inc. (1-800-827-6364)

In article <52qd94$[e-mail]>, [e-mail] (Jeremy
Roussak) wrote:

>Is there really no way of increasing the stack size of a
>background-only application from the default 2k to less than 24k?
>TechNote PS2 implies not, but this seems rather curious.

Personally, I'd grow it to 24K and live with a little extra room because
there are lots of things that "borrow" (IMO) too much stack space at
interrupt time. If they borrow too much from your FBA, then it'll crash
and guess who will be blamed? Right... your FBA.

- Jim Luther (who learned that lesson by experience)

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

From [e-mail] (Shay Telfer)
Subject: StandardGetFile doesn't display only 'MooV's
Date: Thu, 26 Sep 1996 12:10:46 +0800
Organization: Department of Computer Science, The University of Western Australia

Hi...

I've got this very simple C program:

void main(void)
{
   SFTypeList typeList = {'MooV'};
   StandardFileReply theReply;

   Initialize();

   StandardGetFile(NULL,1,typeList,&theReply);

}

The Initialize() function is the same one in SillyBalls.c (the default C
code for a CodeWarrior project).

When the StandardGetFile dialog is displayed, it shows *any* file, not
just files of type 'MooV'. This apparently trivial problem is driving me
nuts. I'm using CW10, Sys 7.5.5 on a 7200/120.

Any suggestions/solutions? Am I missing anything blindingly obvious?

If you can get the code above to work and only show you 'MooV' movies,
then please tell me!

Thanks in advance,
Shay
PS- I don't normally work in C, they make me do it ;(

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

From [e-mail] (Mike Bitz)
Date: Thu, 26 Sep 1996 22:32:49 -0600
Organization: Best Internet Communications

In article <[e-mail]>,
[e-mail] (Shay Telfer) wrote:

>    SFTypeList typeList = {'MooV'};
>
> When the StandardGetFile dialog is displayed, it shows *any* file, not
> just files of type 'MooV'.



StandardFile.h defines the list as:
     typedef OSType SFTypeList[4];

This line of code will solve your problem:
     SFTypeList     typeList = {MovieFileType, 0, 0, 0};


-Mike

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

From [e-mail] (Shay Telfer)
Date: Mon, 30 Sep 1996 22:54:40 +0800
Organization: Department of Computer Science, The University of Western Australia

In article <[e-mail]>, [e-mail] (Mike
Bitz) wrote:

>In article <[e-mail]>,
>[e-mail] (Shay Telfer) wrote:
>
>>    SFTypeList typeList = {'MooV'};
>>
>> When the StandardGetFile dialog is displayed, it shows *any* file, not
>> just files of type 'MooV'.
>
>
>
>StandardFile.h defines the list as:
>     typedef OSType SFTypeList[4];
>
>This line of code will solve your problem:
>     SFTypeList     typeList = {MovieFileType, 0, 0, 0};

...But as it turns out, that wasn't what was causing the problem. It
actually was Macintosh Easy Open assuming that because my app wanted to
open MooV's, it could also open text documents and just about anything
else.

The solution is to turn MEO off, or to write a filterproc instead of just
passing in the type list.

Thanks for your help,
Shay

--
To get random signatures put text files into a folder called ≥Random Signatures≤ into your Preferences folder.

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

From [e-mail] (Jan-Hinnerk Reichert)
Date: 3 Oct 96 03:57:40 +0200
Organization: (none)

In der Nachricht vom 26.Sep 96. 04:10 schrieb Shay Telfer
([e-mail]) unter dem Betreff "StandardGetFile doesn't
display only 'MooV's":

> I've got this very simple C program:
>
> void main(void)
> {
>    SFTypeList typeList = {'MooV'};
>    StandardFileReply theReply;
>
>    Initialize();
>
>    StandardGetFile(NULL,1,typeList,&theReply);
>
> }
>
> The Initialize() function is the same one in SillyBalls.c (the
> default C code for a CodeWarrior project).
>
> When the StandardGetFile dialog is displayed, it shows *any* file,
> not just files of type 'MooV'. This apparently trivial problem is
> driving me nuts. I'm using CW10, Sys 7.5.5 on a 7200/120.

I can't see any mistakes. But perhaps it could help using

  typeList[0]='MooV'

to initialize typeList.

I know there's no reason why it should work better. But IMHO it's
worth trying ;-)

- --------------------------------------------------------------------
 "Few other griefs amid the ill chances    **************************
  of this world have more bitterness       *                        *
  and shame for a man's heart than         *  Jan-Hinnerk Reichert  *
  to behold the love of a lady so fair     *   [e-mail]    *
  and brave that cannot be returned."      *                        *
                  Aragorn son of Arathorn  **************************
- --------------------------------------------------------------------

- MacZPoint 1.92 -


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

From [e-mail] (Jan-Hinnerk Reichert)
Date: 3 Oct 96 03:57:40 +0200
Organization: (none)

In der Nachricht vom 26.Sep 96. 04:10 schrieb Shay Telfer
([e-mail]) unter dem Betreff "StandardGetFile doesn't
display only 'MooV's":

> I've got this very simple C program:
>
> void main(void)
> {
>    SFTypeList typeList = {'MooV'};
>    StandardFileReply theReply;
>
>    Initialize();
>
>    StandardGetFile(NULL,1,typeList,&theReply);
>
> }
>
> The Initialize() function is the same one in SillyBalls.c (the
> default C code for a CodeWarrior project).
>
> When the StandardGetFile dialog is displayed, it shows *any* file,
> not just files of type 'MooV'. This apparently trivial problem is
> driving me nuts. I'm using CW10, Sys 7.5.5 on a 7200/120.

I can't see any mistakes. But perhaps it could help using

  typeList[0]='MooV'

to initialize typeList.

I know there's no reason why it should work better. But IMHO it's
worth trying ;-)

- --------------------------------------------------------------------
 "Few other griefs amid the ill chances    **************************
  of this world have more bitterness       *                        *
  and shame for a man's heart than         *  Jan-Hinnerk Reichert  *
  to behold the love of a lady so fair     *   [e-mail]    *
  and brave that cannot be returned."      *                        *
                  Aragorn son of Arathorn  **************************
- --------------------------------------------------------------------

- MacZPoint 1.92 -


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

From [e-mail]
Subject: Where is Resedit in Info-Mac
Date: Tue, 01 Oct 1996 20:11:48 -0500
Organization: Northwestern University, Evanston, IL, US

I am looking for this program and can't find it. If anyone knows of the
ftp site where I can find this, it would be very helpful.

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

From [e-mail] (Christopher L Tate)
Date: Wed, 2 Oct 1996 03:36:33 GMT
Organization: The World Public Access UNIX, Brookline, MA

[e-mail] wrote:
:
: I am looking for this program and can't find it. If anyone knows of the
: ftp site where I can find this, it would be very helpful.

You can't get it from any of the public archives; Apple maintain control
over it.  You *can* get it, I believe, from the Apple Developer FTP sites.
The FTP site is listed in the comp.sys.mac.games FAQ; I'll put the URL
here for you:

ftp://ftp.support.apple.com/apple_sw_updates/US/Macintosh/Utilities/

That's the directory whre you can find ResEdit 2.1.3, the most recent
actual release.

--
Christopher Tate               <*>   [e-mail]
http://world.std.com/~ctate/   <*>   Finger for PGP public key
"Priest, it is a cold day for your swim, but so it must be."

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

From [e-mail] (Vince Salupo)
Date: Wed, 02 Oct 1996 11:50:55 -0600
Organization: Eli Lilly and Company

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

>I am looking for this program and can't find it. If anyone knows of the
>ftp site where I can find this, it would be very helpful.

Its not at INFO-MAC

Try any of the following:

<http://www.support.apple.com/pub/apple_sw_updates/US/Macintosh/Utilities/ResEdit_2.1.3.hqx>
<ftp://ftp.unsw.edu.au/pub/mac/apple-sw/util/resedit-213.hqx>
<ftp://ftp.inka.de/pub/comp/Mac/essential/resedit-213.hqx>

Vincent P Salupo
Eli Lilly and Company
[e-mail]
As usual: All opinions are mine and do not reflect the views of my
          Company, Managers, co-workers, spouse, children, dog
          etc....

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

From [e-mail] ( nick.c @MT )
Date: Fri, 04 Oct 1996 17:52:56 -0800
Organization: MacTech Magazine



[e-mail] (Christopher L Tate) wrote:

>You can't get it from any of the public archives; Apple maintain control
>over it.  You *can* get it, I believe, from the Apple Developer FTP sites.
>The FTP site is listed in the comp.sys.mac.games FAQ; I'll put the URL
>here for you:
>
>ftp://ftp.support.apple.com/apple_sw_updates/US/Macintosh/Utilities/


   It's also usually on the CD of whatever compiler you're using
     (CodeWarrior or Symantec C++)--look for a "from Apple" or
     similiarly named folder.

____Nicholas C. DeMello, Ph.D.________________________________________

   Online from MacTech Magazine, for Mac OS Programmers and Developers
     http://www.MacTech.com/
                                       _/   _/  _/  _/_/_/   _/   _/
  Chemistry: [e-mail]       _/_/ _/  _/  _/   _/  _/_/_/
    MacTech: [e-mail]      _/ _/_/  _/  _/       _/ _/
       http://www.chem.ucla.edu/~nick/   _/  _/   _/_/_/  _/   _/

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

From [e-mail] (Lawrence Walters)
Subject: Windows and Mac programming
Date: Thu, 19 Sep 1996 01:58:59 GMT
Organization: Customer of Access One Pty Ltd, Melbourne, Australia

Hi guys,

I am on a team who is writing both Windows and Mac programs.

Does anyone know of a class library or something that will allow me to
use the same source code on both platforms.


Regards

Lawrence Walters


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

From [e-mail] (B.J. Hellstrom)
Date: Thu, 19 Sep 1996 08:41:41 -0500
Organization: jaguNET Access Services

In article <[e-mail]>,
[e-mail] (Lawrence Walters) wrote:

>Hi guys,
>
>I am on a team who is writing both Windows and Mac programs.
>
>Does anyone know of a class library or something that will allow me to
>use the same source code on both platforms.

MS has a cross-platform Visual C++ SDK.  It's pricey tho' (>$1000),
probably has a nasty overhead and will make your Mac look like a Wintel.

- -----------
B.J.Hellstrom
[e-mail]

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

From [e-mail] (Mark Hartman)
Date: 20 Sep 1996 07:29:01 -0700
Organization: Mark Hartman Computer Solutions

In article <[e-mail]>,
[e-mail] (B.J. Hellstrom) wrote:

>In article <[e-mail]>,
>[e-mail] (Lawrence Walters) wrote:
>
>>Hi guys,
>>
>>I am on a team who is writing both Windows and Mac programs.
>>
>>Does anyone know of a class library or something that will allow me to
>>use the same source code on both platforms.
>
>MS has a cross-platform Visual C++ SDK.  It's pricey tho' (>$1000),
>probably has a nasty overhead and will make your Mac look like a Wintel.

Won't PowerPlant do this?
======================================================================
Mark Hartman Computer Solutions - specializing in all things Macintosh
   C  C++   4th Dimension   Networking   System design/architecture
tel [phone] -+- fax [phone] -+- e-mail [e-mail]
======================================================================
        A virus for Windows '95?  Isn't that a bit redundant?

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

From Steve Makohin <[e-mail]>
Date: 21 Sep 1996 19:16:08 GMT
Organization: Water's Edge Software

In article <[e-mail]> Lawrence Walters,
[e-mail] writes:
>I am on a team who is writing both Windows and Mac programs.
>
>Does anyone know of a class library or something that will allow me to
>use the same source code on both platforms.

Give XVT Software Inc. a call at [phone] or fax them at (303)
443-0969. They're pricey, but they work on Mac, Windows, DOS and Unix
(libs for each compiler sold separately).

XVT Software Inc.
4900 Pearl East Circle
Boulder CO
80301 USA

-Steve Makohin
 Water's Edge Software
(Makers of Tools Plus & SuperCDEFs)
 http://www.interlog.com/~wateredg

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

From [e-mail] (vlod kalicun)
Date: Mon, 23 Sep 1996 12:52:04 +0000
Organization: I.I.S.C.

In article <[e-mail]>, [e-mail]
(Mark Hartman) wrote:

>In article <[e-mail]>,
>[e-mail] (B.J. Hellstrom) wrote:
>
>>In article <[e-mail]>,
>>[e-mail] (Lawrence Walters) wrote:
>>
>>>Hi guys,
>>>
>>>I am on a team who is writing both Windows and Mac programs.
>>>
>>>Does anyone know of a class library or something that will allow me to
>>>use the same source code on both platforms.
>>
>>MS has a cross-platform Visual C++ SDK.  It's pricey tho' (>$1000),
>>probably has a nasty overhead and will make your Mac look like a Wintel.
>
>Won't PowerPlant do this?
>======================================================================
>Mark Hartman Computer Solutions - specializing in all things Macintosh

What so PowerPlant allows you to create code that can be compiled on
both platforms? You just set the platform of your choice?

-vlod

*********************************************************************
*        Vlod Kalicun            | Wingz/HyperScript Tools Dev.Team *
*"A disaster waiting to happen!" | Investment Inteligience Sys Corp.*
*       [e-mail]           | London, England. www.wingz.com   *
*********************************************************************

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

From [e-mail] (Kent Johnson)
Date: Mon, 23 Sep 1996 14:45:57 GMT
Organization: Transparent Language, Inc.

In article <[e-mail]>, [e-mail]
(vlod kalicun) wrote:

> In article <[e-mail]>, [e-mail]
> (Mark Hartman) wrote:
>
> >
> >>In article <[e-mail]>,
> >>[e-mail] (Lawrence Walters) wrote:
> >>>
> >>>I am on a team who is writing both Windows and Mac programs.
> >>>
> >>>Does anyone know of a class library or something that will allow me to
> >>>use the same source code on both platforms.
> >>
> >
>
> What so PowerPlant allows you to create code that can be compiled on
> both platforms? You just set the platform of your choice?
>
No. PowerPlant (the application framework) is Mac-only.

CodeWarrior (the development environment) can build Windows applications
using MFC.

CodeWarrior does not include a portable application framework.

Kent Johnson                                [e-mail]
Transparent Language, Inc.                http://www.transparent.com

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

From Steve Makohin <[e-mail]>
Date: 21 Sep 1996 19:16:08 GMT
Organization: Water's Edge Software

Reposting article removed by rogue canceller.

In article <[e-mail]> Lawrence Walters,
[e-mail] writes:
>I am on a team who is writing both Windows and Mac programs.
>
>Does anyone know of a class library or something that will allow me to
>use the same source code on both platforms.

Give XVT Software Inc. a call at [phone] or fax them at (303)
443-0969. They're pricey, but they work on Mac, Windows, DOS and Unix
(libs for each compiler sold separately).

XVT Software Inc.
4900 Pearl East Circle
Boulder CO
80301 USA

-Steve Makohin
 Water's Edge Software
(Makers of Tools Plus & SuperCDEFs)
 http://www.interlog.com/~wateredg

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

From [e-mail] (Greg Chapman)
Date: Mon, 23 Sep 1996 11:30:57 -0500
Organization: Robarts Research Institute

In article <[e-mail]>, [e-mail]
(Mark Hartman) wrote:

> Won't PowerPlant do this?

Because PowerPlant is too dependent on low level Mac technologies...
Window Mgr, Sound Mgr etc etc, where Windows doesn't have anything similar
in nature.

--
Greg Chapman
Mac Developer - Robarts Research Institute
Advanced Imaging Research Group (AIRG)
- -
"You! Out of the gene pool!"

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

From [e-mail] (Francis Meyvis)
Date: 24 Sep 1996 10:16:12 GMT
Organization: University of Ghent, Belgium

Why don't you try amulet : free X, mac and windows class lib
<http://www.cs.cmu.edu/afs/cs/project/amulet/www/amulet-home.html>
I'd like to hear from you're experience ...

Lawrence Walters ([e-mail]) wrote:
: Hi guys,

: I am on a team who is writing both Windows and Mac programs.

: Does anyone know of a class library or something that will allow me to
: use the same source code on both platforms.


: Regards

: Lawrence Walters


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

From [e-mail] (Eli Meir)
Date: 24 Sep 1996 16:48:08 GMT
Organization: Dept. of Zoology, Univ. of Washington

There is also another framework called Zinc that claims cross-platform
support, another one from Neuron Data (800-876-4915) (which I think is
cross-platform), and one from Rogue Wave (which claimed on the phone,
but not in their brochure, to do Mac along with Windoze and Unix -
800-487-3217) and I think there are a couple more.  You might get a
catalog from The Programmers Supershop, which seems to list a few
different cross-platform tools (mostly "cross-platform" to them means
across the different flavors of windoze, unfortunetly).
1-800-421-8006.  Almost all of these are very expensive (like 10's of
thousands of dollars plus royalties), and they don't always give you
code that you just flip a switch and it compiles on another platform -
usually, there is some work involved which they say they will guide you
through.  Even the MS cross-platform library only claims 90% of code
will be compatible

I have actually just switched to Java for my next program, to get
automatic cross-platform support (at the expense of speed, making the
user-interface exactly how I want it, and accessing some platform
dependent functions - its a reasonable trade-off for me, and besides, I
find writing in Java quite a bit faster than writing in C++).  For my
last program, I hired someone to do the conversion.  It's taken quite a
while, but he wasn't familiar with Windows to begin with.  I suspect if
he had been familiar, and if I had isolated the user interface bits a
little more, it wouldn't have been too bad.

Good luck, this is a hard problem.

Eli Meir
Dept of Zoology, Univ of Washington
[e-mail]

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

From [e-mail] (Francois Pottier)
Date: 24 Sep 1996 18:00:10 GMT
Organization: INRIA Rocquencourt, BP 105, 78153 Le Chesnay Cedex, France

In article <5293c8$[e-mail]>,
Eli Meir <[e-mail]> wrote:

>I have actually just switched to Java for my next program, to get
>automatic cross-platform support (at the expense of speed, making the
>user-interface exactly how I want it, and accessing some platform
>dependent functions

Can you make Mac Toolbox or OS calls from Java? Can you build
stand-alone applications? Which compiler are you using?

I'm curious... I thought Java only had a few libraries (awt and
such) which weren't enough to do what we do today in C++. Am I
wrong?

--
Francois Pottier
[e-mail]
http://pauillac.inria.fr/~fpottier/

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

From [e-mail] (Carl Gustafson)
Date: Mon, 23 Sep 1996 11:49:53 -0400
Organization: Imaging and Computer Vision Center, Drexel University

In article <[e-mail]>,
[e-mail] (Lawrence Walters) wrote:

> I am on a team who is writing both Windows and Mac programs.

My sympathy.

> Does anyone know of a class library or something that will allow me to
> use the same source code on both platforms.

There's probably half-a-dozen frameworks that promise this, as well as X,
SunOS, Motif, and WhoKnowsWhatElse. Not having a copy of Dr. Dobbs in the
office (where they all advertise) I can't give you specifics.

Of course, you'll get the least common denominator for your product.

--
Carl Gustafson
carl.gustafson at ece.drexel.edu
(busily trying to avoid spammers)
Imaging and Computer Vision Center
Drexel University, Philadelphia, Penna
- ----------------------------------------------------------
I don't speak for Drexel, and Drexel doesn't listen to me...

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

From [e-mail] (vlod kalicun)
Date: Wed, 25 Sep 1996 13:09:50 +0000
Organization: I.I.S.C.

In article <5297ja$[e-mail]>, [e-mail]
(Francois Pottier) wrote:

>In article <5293c8$[e-mail]>,
>Eli Meir <[e-mail]> wrote:
>
>>I have actually just switched to Java for my next program, to get
>>automatic cross-platform support (at the expense of speed, making the
>>user-interface exactly how I want it, and accessing some platform
>>dependent functions
>
>Can you make Mac Toolbox or OS calls from Java? Can you build
>stand-alone applications? Which compiler are you using?
>
>I'm curious... I thought Java only had a few libraries (awt and
>such) which weren't enough to do what we do today in C++. Am I
>wrong?
>
>--
>Francois Pottier
>[e-mail]
>http://pauillac.inria.fr/~fpottier/

Thats my general impression of java. The speed problem will
eventually go away when JIT (Just-In-Time) compilers start to
really take off and when they get introduced into netscape.

Java is cool though, but has to deal with the lowest common
denominator. ie I want my application to support applescript,
drag-and-drop etc.

-vlod

*********************************************************************
*        Vlod Kalicun            | Wingz/HyperScript Tools Dev.Team *
*"A disaster waiting to happen!" | Investment Inteligience Sys Corp.*
*       [e-mail]           | London, England. www.wingz.com   *
*********************************************************************

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

From [e-mail] (Roger Smith)
Date: Mon, 23 Sep 1996 09:32:57 -0500
Organization: Base Online

> >>MS has a cross-platform Visual C++ SDK.  It's pricey tho' (>$1000),
> >>probably has a nasty overhead and will make your Mac look like a Wintel.
> >


Check again. Microsoft Announced at WWDC that they had reduced the price
to $195 US. My company orderded and received a copy of the MS cross
compiler about 3 weeks ago (at the $195 price).

Roger Smith

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

From [e-mail] (msoori)
Date: Thu, 26 Sep 1996 23:46:13 GMT
Organization: Bio-Rad Laboratories

Dont fall in to hole of trying to get 100% cross-platform capability,
because there is no such thing.

Use a good class library to build your interface and share as much of your
non-interface code as possible.  This way you can get the most out of the
platform capabilities and the look and feel, yet get a good deal of code
shared across the platforms.  We use MacApp (though I actually preffer
PowerPlant) on the Mac and Visual C++ on the windoze to build our
interfaces, and standard c++ libs (stl at the moment) to implement
cross-platform algos.

It seems like most people that I come into contact with falls into this
cross-platform hole and program with one hand tied behind their back for
the rest the program's life cycle and eventually have to re-write it again
as the OSes changes and matures.  If you depend on some body else's libs
to make your app cross-platform, for one thing most of them take the least
common dinominator approach and ignore technologies dependent on the
platform, and for another you are at the mercy of that company.  If they
decide to stop support for one platform and you wanted it - tough luck.

In article <R.521eto$[e-mail]>, Steve Makohin
<[e-mail]> wrote:

> Reposting article removed by rogue canceller.
>
> In article <[e-mail]> Lawrence Walters,
> [e-mail] writes:
> >I am on a team who is writing both Windows and Mac programs.
> >
> >Does anyone know of a class library or something that will allow me to
> >use the same source code on both platforms.
>
> Give XVT Software Inc. a call at [phone] or fax them at (303)
> 443-0969. They're pricey, but they work on Mac, Windows, DOS and Unix
> (libs for each compiler sold separately).
>
> XVT Software Inc.
> 4900 Pearl East Circle
> Boulder CO
> 80301 USA
>
> -Steve Makohin
>  Water's Edge Software
> (Makers of Tools Plus & SuperCDEFs)
>  http://www.interlog.com/~wateredg

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

From [e-mail]
Date: Mon, 30 Sep 1996 06:33:26 GMT
Organization: dex

[e-mail] (Carl Gustafson) wrote:

>In article <[e-mail]>,
>[e-mail] (Lawrence Walters) wrote:

>> I am on a team who is writing both Windows and Mac programs.

>My sympathy.

>> Does anyone know of a class library or something that will allow me to
>> use the same source code on both platforms.

>There's probably half-a-dozen frameworks that promise this, as well as X,
>SunOS, Motif, and WhoKnowsWhatElse. Not having a copy of Dr. Dobbs in the
>office (where they all advertise) I can't give you specifics.

Bewing new to the world of Macs, can you plwease tell me what Dr Dobbs
is as it sounds like it might be useful.

thanks
Dee
>Of course, you'll get the least common denominator for your product.

>--
>Carl Gustafson
>carl.gustafson at ece.drexel.edu
>(busily trying to avoid spammers)
>Imaging and Computer Vision Center
>Drexel University, Philadelphia, Penna
>------------------------------------------------------------
>I don't speak for Drexel, and Drexel doesn't listen to me...



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

From [e-mail] (Carl Gustafson)
Date: Thu, 03 Oct 1996 10:08:49 -0400
Organization: Imaging and Computer Vision Center, Drexel University

In article <52mtfr$[e-mail]>, [e-mail] wrote:

>
> Bewing new to the world of Macs, can you plwease tell me what Dr Dobbs
> is as it sounds like it might be useful.
>

Properly titled: Dr Dobbs Journal.

It is a programmer rag primarily aimed at Windoze droids. Provides shallow
coverage of trendy topics, with the occasionally useful chunk thrown in.
Michael Swain's column is occasional entertaining. (All MAO, of course.)

Don't know if it is available over the counter (or even under) in the UK.

--
Carl Gustafson
carl.gustafson at ece.drexel.edu
(busily trying to avoid spammers)
Imaging and Computer Vision Center
Drexel University, Philadelphia, Penna
- ----------------------------------------------------------
I don't speak for Drexel, and Drexel doesn't listen to me...

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

From [e-mail] (MW Ron)
Subject: [ANN ]MWDebug-MacOS 1.6.3 is available for downloading
Date: Fri, 04 Oct 1996 23:47:35 -0400
Organization: Metrowerks

MWDebug/MacOS 1.6.3 is available for downloading

now or soon from the following URLS

http://www.metrowerks.com/tools/software/updates/current
or
ftp://ftp.metrowerks.com/pub/updates/CW10/
European Mirror Site
ftp://ftp.promo.de/pub/Metrowerks/mirror/updates/CW10/

============================
MW Debug/MacOS Release Notes
============================

Version: MWDebug/MacOS 1.6.3
Date:    10/2/96
Author:  Mike Lockwood
============================

This is patch release to the CW10 Metrowerks debugger.  It fixes some bugs
that have been discovered after
CW10 had been released.

NOTE:  There are still some known bugs in the CW10 debugger that are not
fixed in this release.  We will have
another patch release soon to fix some more bugs.  Since some of these
bugs are crashing bugs that prevent
the debugger from working on certain system configurations, we have
decided to release this now rather than
waiting for other bugs to be fixed.


===================================
Important Installation Instructions
===================================

- In order for Java debugging to work, you must use the copy of Metrowerks
Java from the CW10 CD.
The debugger is not compatible with earlier versions of Metrowerks Java.

- On 68K machines, MWDebug/MacOS 1.6.3 requires 68K-CFM.  You must install the
" CFM-68K Runtime Enabler" in order to run the debugger on 68K machines.
If you use the CW10 Tools
installer, this extension should be installed automatically.  Otherwise,
you can get a copy from:
"CW10 Gold Tools:CodeWarrior
MPW:Interfaces&Libraries:RuntimeLibraries:CFM-68K Additions:"

- The "Debugger Plugins" folder must be in the same directory as the
MWDebug application.
Or optionally, an alias to the "Debugger Plugins" folder can be placed in
the same directory
as the debugger application instead.

- In order to debug MP applications on multiprocessor PowerMacs, you may
have to edit the
'sysz' resource in MetroNub.  If you get a message just after your machine
boots into the
Finder that says there was not enough memory to load the MP library, edit
the 'sysz' resource
in MetroNub and change the value to a large number like 1,300,000.  This
will reserve extra
memory in the System heap so MetroNub can connect to the MP Library at
boot time.  The actual
amount of memory may vary, depending on the amount of RAM and number of
processors in your machine.


==========================
Bugs Fixed in This Version
==========================

- Fixed a bug that caused a crash when launching the debugger application
on some 68K machines running
system software earlier than 7.5.

- The debugger application no longer crashes when launched on a 68K
machine without the
" CFM-68K Runtime Enabler" installed.  Now the debugger uses an alert to
inform the user that
" CFM-68K Runtime Enabler" needs to be installed.

- Fixed a crash that occured on on 68K machines that have the Drag Manager
installed but did not have
NuDragLib.slb in the Extensions folder.

- Fixed a crash that occured when displaying a source file that does not
have an 'MPSR' resource, and
the debugger is unable to read default preferences from the IDE's
preferences file.

- Fixed a problem that prevented the debugger from launching on PowerMacs
with an old version of MathLib
(for example, 1st generation PowerMacs running System 7.5).

- Fixed a crash that occured when displaying 80 bit floating point values
on PowerPC machines, and
when attempting to open the 68K FPU Registers window on a 68K machine
(which is possible, if you have
SoftwareFPU installed).

- When debugging 68K code with the "Catch C++ Exceptions" preference on,
the debugger now stops at
the call to __throw rather than on the first instruction of __throw.

- The debugger now stops at catch when debugging with the "Catch C++
Exceptions" preference on.

- Instead of having a separate log window for each task or thread, there
is now only one log window
per process.

- The popup menu (to the left of the source code, next to the breakpoint
markers) for choosing
between the different expansions of a template is working again.

- The debugger is now able to launch control panel applications (that is,
applications with file
type 'APPC' instead of 'APPL').

- Fixed a problem reading the syntax coloring preferences from the IDE's
preferences file.

- Fixed a problem launching Java applets when their html file contains
quotes embedded in
a pathname.

- Fixed a drawing problem that occured while editing variable values.

- Fixed a crash that sometimes occured the second time you ran an applet
without quitting
the debugger inbetween.

--
METROWERKS                   Ron Liechty
"Software at Work"    [e-mail]
http://www.metrowerks.com/about/people/rogues.html#mwron

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

From Jeremy Vineyard <[e-mail]>
Subject: [ANN] Licorice for Game Sprocket Developers
Date: Fri, 04 Oct 1996 21:22:06 +0000
Organization: Viperware(tm)

Viperware(tm) is proud to announce Licorice, a new series of web pages
designed specifically for developers using Apple's Game Sprockets.  The
URL for Licorice is:

"http://www.viperware.com/licorice"

Licorice features FAQ's for each of the Game Sprockets.  Questions and
answers have been taken from the past 6 months of the
Mac-Games-Developer mailing list.

If you have a question concerning Game Sprockets, you should check
Licorice first to see if your question is answered in a FAQ.  Hopefully
this will reduce noise on the newsgroups and mailing lists, and will
make using these services more enjoyable.

** Also take a look at Butterscotch, the #1 source of information for
game developers: http://www.viperware.com/butterscotch **

+-----------------------+--------------------------+
| Jeremy Vineyard       | http://www.viperware.com |
| Viperware(tm)         |                          |
| [e-mail] |                          |
+-----------------------+--------------------------+

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

From [e-mail] (eric lundberg)
Subject: [Q] Drawing Lines?
Date: Tue, 24 Sep 1996 14:26:51 -0800
Organization: UCSD && Varcon Systems Inc.

Hi, before I fire away with the questions here's a little backround on
what I'm doing and why.  I am _trying_ to make a line drawing tool like
you would see in any paint program.  With the line (as the user drags the
mouse around) drawing over what is currently in the window, and leaving it
as it was as the line moves past it.
   I didn't want my program to take a lot of memory so what I thought I
would do is have one offscreen where I stored the contents of the window.
Then for each iteration of the user draging the mouse, I would draw a line
into the window with the standard MoveTo LineTo, followed by a call to a
routine I wrote that uses a bresenham line algorithm to follow that same
line, but copy those specific pixels out of my offscreen and into the
window.  I actually got
that bresenham routine working with the direct access to screen and access
to the offscreen.  HOWEVER, it doesn't follow the same line that lineTo
follows!!  Or I guess more to the point, it makes the same line that
lineTo does, but it is offset down that line.  Here's an example:
o
 oo
   oo
     o
      oo
This is my line above here

oo
  oo
    o
     oo
       oo
This is Quickdraw version of that same line...

They both follow the same pattern (2,2,1) but they start in different places in
that pattern.  My two guesses thus far (for those of you familer with
bresenhams alg.) are that either quickdraw doesn't set it's error_term to
zero and instead sets it to some function of the x and y differences.  Or
that quickdraw isn't drawing the line from the start point to the finish
point, but is drawing it from somewhere farther up the line to the finish
point and clipping till it actually gets to the start point.
Unfortunately neither of them sounds too plausable, which I guess really
just says I don't know whats going on. :)  My bresenham code looks just
about the same as a million other versions of the algorithm I've seen in
the last couple of days so I really don't think it's that. Anyone have any
ideas one why I have that difference problem with the lineto?

My current solution is to not use LineTo and to use my bresenham line to
draw.  However in order to use more than 256 color mode I am just
calculating the pixel offsets and calling SetCPixel (ugly ugly I know)
with a RGBColor.  If someone could explain to me how I can convert a
RGBColor to the appropriate values for different color depths (8bit 16bit
24bit etc) I could just do that and draw the line to memory myself and it
would be plenty fast.

So if anyone has any thoughts on this I'd love to hear them.  Even if they
are on the order of "Silly, why didn't you just use the XXXXToolbox call
to do that for you?" :) Thanks in advance.

-e

BTW - Just so you know I did my homework, I did scan throught the c.s.m.p
digest archives from about 92-96, visted about 8 bresenham sites, went
through most of the snippet code at devworld, and scaned through the ever
fun 'imaging with quickdraw.'

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

From [e-mail] (Broc Seib)
Date: 26 Sep 1996 16:51:58 GMT
Organization: Purdue University

In article <[e-mail]>,
[e-mail] (eric lundberg) wrote:

> Hi, before I fire away with the questions here's a little backround on
> what I'm doing and why.  I am _trying_ to make a line drawing tool like
> you would see in any paint program.  With the line (as the user drags the
> mouse around) drawing over what is currently in the window, and leaving it
> as it was as the line moves past it.

The most common way of doing this is by using the XOR PenMode.

While you drag your line's endpoint, you are in "modal" operation.
You show your "modal" change to the image by xor'ing your line
with the original image. You return the image to it's original
state by performing the same XOR again, e.g.,

[1]   orig_image ^ your_image = modal_image
[2]  modal_image ^ your_image =  orig_image

You should only do [2] when a MouseMoved event occurs, else you'll
get unecessary redraw/flicker. (Of course you loop and apply [1] again
until the button is released...)

When you release the mouse button, then update your image for real.

[...]

> I actually got
> that bresenham routine working with the direct access to screen and access
> to the offscreen.  HOWEVER, it doesn't follow the same line that lineTo
> follows!!  Or I guess more to the point, it makes the same line that
> lineTo does, but it is offset down that line.  Here's an example:
> o
>  oo
>    oo
>      o
>       oo
> This is my line above here
>
> oo
>   oo
>     o
>      oo
>        oo
> This is Quickdraw version of that same line...
>
> They both follow the same pattern (2,2,1) but they start in different places
> in that pattern.
[...]
> Anyone have any
> ideas one why I have that difference problem with the lineto?

[...]

On this point, remember that there are an infinite number of *real*
lines that could produce one of a finite number of *pixel* lines
between two pixels on your screen. e.g.,
      |
   ---*---+---+---+---+---+---+---+---+---
      | a |   |   |   |   |   |   |   |
   ---+---+---+---+---+---+---+---+---+--
      |   |   |   |   |   |   |   |   |
    --+---+---+---+---+---+---+---+---+-
      |   |   |   |   |   |   |   |   |
   ---+---+---+---+---+---+---+---+---+--
      |   |   |   |   |   |   |   |   |
    --+---+---+---+---+---+---+---+---+-
      |   |   |   |   |   |   |   |   |
   ---+---+---+---+---+---+---+---*---+--
      |   |   |   |   |   |   |   | b |
    --+---+---+---+---+---+---+---+---#-
      |   |   |   |   |   |   |   |   |

Think in a real number coord system for a moment instead of discrete
pixel cells. I put two asterisks at the upper left corner of the two
pixel cells we want to connect.

If you draw a *real* line between these two asterisks and apply the
Bresenham algorithm using that real line, you'll a particular result.

Now, move the asterisks to the centers of the cells, or opposite
corners, or *anywhere* within the cell. You'll get slight variations
of your previous result.

For example from above, realline(a*,b*) is different from realline(a*,b#).
They *may* produce different pixel-rep lines w/ Bresenham alg.

I cannot say what Quickdraw is doing; I'm just giving you my theory
of what might be happening. You might be able to simply trick your
line algorithm to offset the endpoints of your line by half a
pixel (or such) and actually get the same results as Quickdraw...
This would actually be an interesting exercise to do so without using
any floating point math... :-)

I would guess that Apple uses the Bresenham algorithm in some fashion.

:-)
_______________________________________________________________________
  Broc Seib ([e-mail])  |  Purdue University Computing Center
  Network Systems Programmer    |  Instructional Computing Division
_______________________________________________________________________
  Broc Seib ([e-mail])  |  Purdue University Computing Center
  Network Systems Programmer    |  Instructional Computing Division

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

From "Aidan Cully" <[e-mail]>
Date: 3 Oct 96 01:57:13 -0400
Organization: INTAC Access Corporation - An Internet Service Provider

>
>o
> oo
>   oo
>     o
>      oo
>This is my line above here
>
>oo
>  oo
>    o
>     oo
>       oo
>This is Quickdraw version of that same line...
>
>They both follow the same pattern (2,2,1) but they start in different
places
>in
>that pattern.  My two guesses thus far (for those of you familer with
>bresenhams alg.) are that either quickdraw doesn't set it's error_term to
>zero and instead sets it to some function of the x and y differences.  Or
>that quickdraw isn't drawing the line from the start point to the finish
>point, but is drawing it from somewhere farther up the line to the finish
>point and clipping till it actually gets to the start point.
>Unfortunately neither of them sounds too plausable, which I guess really
>just says I don't know whats going on. :)  My bresenham code looks just
>about the same as a million other versions of the algorithm I've seen in
>the last couple of days so I really don't think it's that. Anyone have any
>ideas one why I have that difference problem with the lineto?

As you can probably see, your algorithm goes from (0,0) to (5,8), while
quickdraw's goes from (0,0) to (5,9).  I imagine the problem you are having
comes from that.  Is your line inclusive?





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

From [e-mail] (Reuven J. Sherwin)
Subject: [Q] Switching the Sound Input source
Date: Thu, 03 Oct 1996 07:00:45 +0200
Organization: Scitex Multimedia Publishing

Hi there,

I'm looking for source code example that demonstrate how to use the Sound
Input API for storing the current Sound Input device, then switching it to
the Microphone, and then back again.

Any sample would be of much help.

Thanks,
   Reuven Sherwin

--
Reuven J. Sherwin               email: [e-mail]
Pitagno Multimedia Ltd.
a Scitex company

"Everything ought to turned upside down occasionally; it lets in air and light." - Betty Sorenson, "Star Beast", R.A.H.

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

From [e-mail] (Craig Emery)
Date: Thu, 03 Oct 1996 12:31:00 +0100
Organization: Epic Miltimedia Group

In article <reuven_sherwin-0310960701170001@149.115.2.66>,
[e-mail] (Reuven J. Sherwin) wrote:

>Hi there,
>
>I'm looking for source code example that demonstrate how to use the Sound
>Input API for storing the current Sound Input device, then switching it to
>the Microphone, and then back again.
>
>Any sample would be of much help.
>
>Thanks,
>   Reuven Sherwin
>
>--
>Reuven J. Sherwin               email: [e-mail]
>Pitagno Multimedia Ltd.
>a Scitex company
>
>"Everything ought to turned upside down occasionally; it lets in air and
light." - Betty Sorenson, "Star Beast", R.A.H.

    This is a **very** simple SIOUX application that shows how I've been
doing what you want. Put it in a project and link against standard
Macintosh and ANSI libraries and enjoy. I've compiled this both in 68k and
PPC and it seems to work fine.

    I've a feeling that on occasions the selection you've made isn't kept
after a restart! I haven't figured this out yet!?! :-)

Enjoy.

Craig Emery.

// main.cpp starts here
// includes

#ifndef __SOUNDINPUT__
#include <SoundInput.h>
#endif // !__SOUNDINPUT__

#if !defined(_IOSTREAM_)
#include <iostream>
#endif // !defined(_IOSTREAM_)

#if !defined(_STDLIB)
#include <stdlib.h>
#endif // !defined(_STDLIB)

// proto-types
Boolean InitToolBox();
Size _P2CStr (const StringPtr psSrc, char *lpszDest);

// constants
const Boolean gbToolBoxInited = InitToolBox ();

// globals

// code

// function to initialise the MacOS ToolBox
// returns a flag to be stored in a global so that other functions that
// may be run during startup (such as constructors for static class data)
// can check that this has been run first
Boolean InitToolBox()
{
    InitGraf((Ptr) &qd.thePort);
    InitWindows();
    InitDialogs(nil);
    InitMenus();
    InitCursor ();              // make the cursor the standard arrow and
set cursor show / hide count to 0

    return (true);
};

Size _P2CStr (const StringPtr psSrc, char *lpszDest)
{
    Size szSize;
    BlockMove ((void *) &psSrc[1], (void *) lpszDest, szSize = (Size)
*((char *) psSrc));
    lpszDest[szSize] = '\0';
    return (szSize);
};

// This application will present the user with a list of the
// sound input "sources" in the default device and thgen let the user
// pick which one to use. After setting it, the application quits
void main ()
{
    // make the SIOUX window appear
    // crude but effective
    cout << endl << flush;

    OSErr theErr = noErr;
    Handle hDeviceIconHandle;
    Str255 psDeviceName;
    char lpszDeviceName[255];
    int nCount;
    StringPtr *psInputs = NULL;
    short nsInputs = 1;
    Handle hNames = NULL;
    long lInRefNum;

    // while I don't get siBadSoundInDevice I can ask for info about input
devices
    for (nCount = 1; siBadSoundInDevice != (theErr = SPBGetIndexedDevice
(nCount, psDeviceName, &hDeviceIconHandle)); nCount++)
    {
        _P2CStr (psDeviceName, lpszDeviceName);
        cout << "Name of device number " << nCount << " is " <<
lpszDeviceName << endl;

        if (noErr == SPBOpenDevice (psDeviceName, siWritePermission,
&lInRefNum))
        {
            if (noErr == SPBGetDeviceInfo (lInRefNum, siInputSourceNames,
(char *) &hNames))
            {
                short nCurrInput;
                if (noErr == SPBGetDeviceInfo (lInRefNum, siInputSource,
(Ptr) &nCurrInput))
                {
                    HLock (hNames);

                    nsInputs = **((short **) hNames);
                    psInputs = new StringPtr[nsInputs];
                    cout << "\t" << nsInputs << " input sources are
attached to this device" << endl;

                    char lpszName[255];
                    short nsInput;
                    Size szOffset;
                    StringPtr psName = (StringPtr) (((short *) *hNames) + 1);
                    for (nsInput = 0; nsInput < nsInputs; nsInput++)
                    {
                        psInputs[nsInput] = psName;
                        szOffset = _P2CStr (psName, lpszName);
                        cout << "\t\tinput source number " << nsInput + 1
<< " is called " << lpszName;
                        if ((nsInput + 1) == nCurrInput)
                        {
                            cout << " (currently selected)";
                        };
                        cout << endl;
                        psName += szOffset + 1;
                    };
                };
            };
        };
    };

    if (1 != nsInputs)
    {
        cout << endl << "Enter a number between 1 and " << nsInputs
            << " to change the input source (enter zero to leave it
unchanged)" << endl
            << " : ";
        char lpszMyInput[255], *lpszEnd = NULL;
        cin >> lpszMyInput;
        long lWantInput;
        if ((0 == (lWantInput = strtol (lpszMyInput, &lpszEnd, 0))) &&
(*lpszEnd != '\0'))
        {
            cout << "you didn't enter a number" << endl;
        }
        else if ((0 < lWantInput) && (lWantInput <= nsInputs))
        {
            _P2CStr (psInputs[lWantInput - 1], lpszMyInput);
            cout << "Changing current input source to " << lpszMyInput << endl;
            short nWantInput = lWantInput;
            if (noErr != (theErr = SPBSetDeviceInfo (lInRefNum,
siInputSource, (Ptr) &nWantInput)))
            {
                cout << "Failed to change input source (error number " <<
theErr << ")" << endl;
            };
        };

        delete[] psInputs;
    };

    SPBCloseDevice (lInRefNum);

    if (NULL != hNames)
        DisposeHandle (hNames);

    ExitToShell ();
};

// main.cpp ends here

 -----------------------------------------------------------------------
| Norman Craig Emery B.Sc.,                            [e-mail] |
| Team Leader / Senior CDi Programmer &  [e-mail] |
| Macintosh Programming Coordinator, Voice: +44 (0) (1273) 320637       |
| The Epic Multimedia Group,                 Fax: +44 (0) (1273) 821567 |
|               http://www.epic.co.uk/                                  |
|     VPS House, 52 Old Steine, Brighton, East Sussex, BN1 1NH, U.K.    |
 -----------------------------------------------------------------------

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

From "Helge Baumann" <[e-mail]>
Subject: atan2: PPC vs. 68k
Date: 27 Sep 96 12:12:40 GMT
Organization: C. v. Ossietzky Universitaet Oldenburg - Fachbereich Informatik

Hi folks,

I got a problem, that drives me mad:
In a little C-program I got the line

  c = atan2 (a, b); // a, b, c are double!

and I wantto compile for both PPC and 68k with my CW 9.

The first call for this line is with a = b = 0, so it will compute
(mathematical) arctan (0/0), which schould be a correct value of 0.
(I think, that's what the atan2 (y,x) is for in opposite to atan (x)...)

When I compile it for PowerPC, all works fine.
When I compile for 68020, it says: NAN(040), which should be
not a number.

So, who's going wrong: me, my computer, his interpreter, my compiler,
my library, the OS, Apple, Motorola, anything else?

Thanks for any hints
Bye
Helge
--
 Helge Baumann        ** "Ich bin dagegen,
 Groninger Strafle 28  **  Computer zu vermenschlichen.
 D-26129 Oldenburg    **
 0441-776331          **  Sie mˆgen es nicht!"


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

From Carl R. Osterwald <[e-mail]>
Date: 27 Sep 1996 15:15:23 GMT
Organization: National Renewable Energy Laboratory

In article <52ggs7$[e-mail]> Helge Baumann,
[e-mail] writes:

>I got a problem, that drives me mad:
>In a little C-program I got the line
>
>  c = atan2 (a, b); // a, b, c are double!
>
>and I wantto compile for both PPC and 68k with my CW 9.

>The first call for this line is with a = b = 0, so it will compute
>(mathematical) arctan (0/0), which schould be a correct value of 0.
>(I think, that's what the atan2 (y,x) is for in opposite to atan (x)...)

My first thought is that atan2(0,0) should be an error, because there
is no information on what the angle is, i.e. x cannot be zero:

atan2(10,0) = pi/2
atan2(-13,0) = -pi/2
atan2(0,0) = ?

>When I compile it for PowerPC, all works fine.
>When I compile for 68020, it says: NAN(040), which should be
>not a number.

I think you meant NaN(4), which is invalid division (usually division by
zero).

>So, who's going wrong: me, my computer, his interpreter, my compiler,
>my library, the OS, Apple, Motorola, anything else?

Another point to consider is that SANE (the Standard Apple Numerics
Environment used in all 68k Macs) does not have atan2(y,x) built in.
The single-srgument atan(x) is part of SANE, however.  It is listed in
fp.h in the universal headers, which means it is now available for both
cases.  It looks like there is a difference between 68k and PPC in how
it is implemented.

If indeed the PPC implementation is incorrect, there might be a
difference between Apple's MathLib and Motorola's LibMoto, which now
ships with 7.5.5.  Both implementations should be checked.

I'll try compiling the test example myself later today.

Hope this helps.

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

From Carl R. Osterwald <[e-mail]>
Date: 3 Oct 1996 20:41:44 GMT
Organization: National Renewable Energy Laboratory

In article <52gr2b$[e-mail]> Carl R. Osterwald,
[e-mail] writes:

>I'll try compiling the test example myself later today.

Well, today wasn't today, but I did try a few things out using the
following CodeWarrior10 SOUIX code:

#include <stdio.h>
#include <fp.h>

void calc (double_t a, double_t b);
void calc (double_t a, double_t b)
    {
        double_t    c = atan2(a, b);
        printf("atan2(%lf, %lf) = %lf\n", a, b, c);
    }

void divide (double_t a, double_t b);
void divide (double_t a, double_t b)
    {
        double_t    c = a / b;
        printf("%lf / %lf = %lf\n", a, b, c);
    }

void main(void)
{
    printf ("PPC results:\n\n");

    calc(0.0, 0.0);
    calc(10.0, 0.0);
    calc(-10.0, 0.0);
    divide(0.0, 0.0);
    divide(10.0, 0.0);
    printf( "sqrt(-1.0) = %lf\n", sqrt(-1.0) );
}

Here are the results I obtained on an 8500/180 with SoftwareFPU
installed, CW10 default ANSI project settings:

68k results:

atan2(0.000000, 0.000000) = NAN(004)
atan2(10.000000, 0.000000) = 1.570796
atan2(-10.000000, 0.000000) = -1.570796
0.000000 / 0.000000 = NAN(004)
10.000000 / 0.000000 = INF
sqrt(-1.0) = NAN(255)

PPC results:

atan2(0.000000, 0.000000) = 0.000000
atan2(10.000000, 0.000000) = 1.570796
atan2(-10.000000, 0.000000) = -1.570796
0.000000 / 0.000000 = NAN(000)
10.000000 / 0.000000 = INF
sqrt(-1.0) = NAN(001)

Here are the NaN codes defined in the Apple Numerics Manual,
2nd. Ed., pg. 41:

Name      Code    Meaning
NANSQRT      1  Invalid square root, such as sqrt(-1)
NANADD       2  Invalid addition, such as (+INF) - (+INF)
NANDIV       4  Invalid division, such as 0/0
NANMUL       8  Invalid multiplication, such as 0 x INF
NANREM       9  Invalid remainder or mod such as X rem 0
NANASCBIN   17  Attempt to convert invalid ASCII string
NANCOMP     20  Result of converting comp NaN to floating
NANZERO     21  Attempt to create a NaN with a zero code
NANTRIG     33  Invalid argument to trig routine
NANINVTRIG  34  Invalid argument to inverse trig routine
NANLOG      36  Invalid argument to log routine
NANPOWER    37  Invalid argument to x^i or x^y routine
NANFINAN    38  Invalid argument to financial routine
           255  NaN created by 68882 FPU

So the short answer is, they are both wrong.  The 68k gave NaN(4) for
atan2(0, 0), and the PPC returned 0.0.  I believe atan2(0, 0) should
return NaN(33).

Both returned the correct answers for legitimate atan2 arguments, so my
supposition that the 68k was simply dividing the two arguments and
calling atan was wrong.  Also, both divided 10/0 correctly and
obtained infinity.

For the invalid division test, 0/0, the 68k app returned the correct
NaN code, but the PPC returned NaN(0), which isn't even defined (see the
meaning for NaN(21).

Finally, sqrt(-1) returned the correct code for the PPC, and returned
the FPU code for the 68k case.  This is odd, however, because the 68k
returned the specific NaN code for 0/0, but the generic FPU NaN for
sqrt(-1).

I guess the bottom line is that NaNs aren't very consistent, and should
be used with caution.

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

From Carl R. Osterwald <[e-mail]>
Date: 3 Oct 1996 20:41:44 GMT
Organization: National Renewable Energy Laboratory

In article <52gr2b$[e-mail]> Carl R. Osterwald,
[e-mail] writes:

>I'll try compiling the test example myself later today.

Well, today wasn't today, but I did try a few things out using the
following CodeWarrior10 SOUIX code:

#include <stdio.h>
#include <fp.h>

void calc (double_t a, double_t b);
void calc (double_t a, double_t b)
    {
        double_t    c = atan2(a, b);
        printf("atan2(%lf, %lf) = %lf\n", a, b, c);
    }

void divide (double_t a, double_t b);
void divide (double_t a, double_t b)
    {
        double_t    c = a / b;
        printf("%lf / %lf = %lf\n", a, b, c);
    }

void main(void)
{
    printf ("PPC results:\n\n");

    calc(0.0, 0.0);
    calc(10.0, 0.0);
    calc(-10.0, 0.0);
    divide(0.0, 0.0);
    divide(10.0, 0.0);
    printf( "sqrt(-1.0) = %lf\n", sqrt(-1.0) );
}

Here are the results I obtained on an 8500/180 with SoftwareFPU
installed, CW10 default ANSI project settings:

68k results:

atan2(0.000000, 0.000000) = NAN(004)
atan2(10.000000, 0.000000) = 1.570796
atan2(-10.000000, 0.000000) = -1.570796
0.000000 / 0.000000 = NAN(004)
10.000000 / 0.000000 = INF
sqrt(-1.0) = NAN(255)

PPC results:

atan2(0.000000, 0.000000) = 0.000000
atan2(10.000000, 0.000000) = 1.570796
atan2(-10.000000, 0.000000) = -1.570796
0.000000 / 0.000000 = NAN(000)
10.000000 / 0.000000 = INF
sqrt(-1.0) = NAN(001)

Here are the NaN codes defined in the Apple Numerics Manual,
2nd. Ed., pg. 41:

Name      Code    Meaning
NANSQRT      1  Invalid square root, such as sqrt(-1)
NANADD       2  Invalid addition, such as (+INF) - (+INF)
NANDIV       4  Invalid division, such as 0/0
NANMUL       8  Invalid multiplication, such as 0 x INF
NANREM       9  Invalid remainder or mod such as X rem 0
NANASCBIN   17  Attempt to convert invalid ASCII string
NANCOMP     20  Result of converting comp NaN to floating
NANZERO     21  Attempt to create a NaN with a zero code
NANTRIG     33  Invalid argument to trig routine
NANINVTRIG  34  Invalid argument to inverse trig routine
NANLOG      36  Invalid argument to log routine
NANPOWER    37  Invalid argument to x^i or x^y routine
NANFINAN    38  Invalid argument to financial routine
           255  NaN created by 68882 FPU

So the short answer is, they are both wrong.  The 68k gave NaN(4) for
atan2(0, 0), and the PPC returned 0.0.  I believe atan2(0, 0) should
return NaN(33).

Both returned the correct answers for legitimate atan2 arguments, so my
supposition that the 68k was simply dividing the two arguments and
calling atan was wrong.  Also, both divided 10/0 correctly and
obtained infinity.

For the invalid division test, 0/0, the 68k app returned the correct
NaN code, but the PPC returned NaN(0), which isn't even defined (see the
meaning for NaN(21).

Finally, sqrt(-1) returned the correct code for the PPC, and returned
the FPU code for the 68k case.  This is odd, however, because the 68k
returned the specific NaN code for 0/0, but the generic FPU NaN for
sqrt(-1).

I guess the bottom line is that NaNs aren't very consistent, and should
be used with caution.

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

From [e-mail] (Ed Olson)
Subject: password dialogs-- how?
Date: Thu, 26 Sep 1996 19:11:16 -0400
Organization: MIT

i've been trying to implement a password dialog (show bullets instead of
letters) by intercepting key-down events, and the approach seems to work
for the most part. However (and not suprisingly) this is failing to
catch/intercept the user selecting a range of text and deleting it.

where can i get some source that can show me how, or an outline of how to
go about doing this?

Ed Olson
MIT MacDev
[e-mail]

an email response is preferred :)

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

From [e-mail]
Date: Thu, 26 Sep 1996 19:26:45 -0700
Organization: (none)

In article <[e-mail]>, [e-mail] (Ed
Olson) wrote:

> i've been trying to implement a password dialog [...]
> where can i get some source that can show me how, or an outline
> of how to go about doing this?

The alt.sources.mac archive at <http://www.AmbrosiaSW.com/alt.sources.mac/>
has examples.  Also, there is a PowerPlant class on any info mac archive,
such as
<http://hyperarchive.lcs.mit.edu/HyperArchive/dev/lib/
   l-password-field-cpp-pp.hqx>
which demonstrates some techniques.

--
(Pointers to other Mac programming web sites at
<http://devworld.apple.com/dev/geeks.html>)

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

From "Tedd F. Sperling" <[e-mail]>
Date: Fri, 27 Sep 1996 09:48:34 +0000
Organization: Sperling Corporation

Ed Olson wrote:
>
> i've been trying to implement a password dialog (show bullets instead of
> letters) by intercepting key-down events, and the approach seems to work
> for the most part. However (and not suprisingly) this is failing to
> catch/intercept the user selecting a range of text and deleting it.
>
> where can i get some source that can show me how, or an outline of how to
> go about doing this?
>
> Ed Olson
> MIT MacDev
> [e-mail]
>
> an email response is preferred :)

Hi:

Try installing a bullet font in the resources of your program and then
changing the font type before asking the user for a password.

tedd

--
   ____
  |[  ]| mac programmer                             tedd f. sperling
  |[__]|                               mailto: [e-mail]
  |___-|                http://www.sojourn.com/~sperling/resume.html
- -

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

From [e-mail] (LOTTSIM)
Date: 29 Sep 1996 18:24:57 -0400
Organization: America Online, Inc. (1-800-827-6364)

I had to do this once---it isn't that difficult, either.

What you need to do, first, is create a dialog filter.

Then, you examine each keystroke.  If it is normal (i.e. not return,
enter, delete, or tab), then intercept it, and change the character (to be
passed on to the edit field) to a bullet.  At the same time, you want to
have a char * or string to which you can add the actual characters that
the user inputs.

Then, when the user hits return or enter, simply take the finished correct
string and check it against any other string.

I can probably re-write it, if you need me to.

Alex Rampell

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

From [e-mail] (Gregory Lo)
Date: 3 Oct 1996 19:50:23 GMT
Organization: Queens University

I saw a post a while ago with an interestingly different approach, one
that I never would have thought of:

Install custom Textwidth and Texthook procs in the TextEdit record so that
the width of every character in the password field is the width of a
bullet, and every character drawn in the field by TE is the same bullet
character.  This would save from having to do double bookkeeping on the
entered text.  You would still have to disable Cut and Copy, but that's
something you normally have to implement yourself anyway.

Of course, I haven't actually tried this out yet...

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

From [e-mail] (Gregory Lo)
Date: 3 Oct 1996 19:50:23 GMT
Organization: Queens University

I saw a post a while ago with an interestingly different approach, one
that I never would have thought of:

Install custom Textwidth and Texthook procs in the TextEdit record so that
the width of every character in the password field is the width of a
bullet, and every character drawn in the field by TE is the same bullet
character.  This would save from having to do double bookkeeping on the
entered text.  You would still have to disable Cut and Copy, but that's
something you normally have to implement yourself anyway.

Of course, I haven't actually tried this out yet...

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

From [e-mail] (Pratap Reddy P.)
Subject: problems with XCMD using OT
Date: Tue, 01 Oct 1996 16:03:02 +0530
Organization: Apple Developement India


Hi all,

This is Pratap, sending this mail from India. I am writing an XCMD whcih
interacts with OT streams. I have few problems while doing this. Can
somebody answer to me if you know about any of these problems.

I will brief here about my XCMD (PPC, C++) :

* Writing this using CW

* it opens OT and creates stream (with 3 modules) on opening Hypercard and
returns control back to Hypercard (HC). It also asks HC to lock the
resource using XWHasInterruptCode.

* gets serial port names from OT and gives back to HC, when user clicks on
some button to get this information.

* when user clicks on "check syntax" button of HC, XCMD uses the endpoint
which was already open, issues some ioctls to modules and returns results
back to HC.

* While quitting the hypercard, unbounds and closes the provider and tries
to close opentransport (!!). Also tells the hypercard to unlock the
resource.

Here are my problems :

1.
If i try to send an event to XCMD from a module when XCMD is not in
memory, my system crashes into Macsbug. If I lock the XCMD resource
explicitly it is not crashing. I guess, XWHasInterruptCode is suppossed to
lock the resource!!. Why is it not doing that??

2.
One of my code warrior global is becoming NULL (endpoint that I opened
initially). If I play around with the stack invoking XCMD many time, this
happens. :-((

3.
CloseOpentransport crashes into Macsbug sometimes with "PowerPC unmapped
exception at TOTDeathWatcher::delete" and sometimes with "illegal
instruction at ..(~TOTDeathwatcherFUI)". I can see the assembly code as
dc.l 0x00000000 (??) and sometimes as dc.l 0x02020002.
If I remove CloseOpentransport, looks like everything is fine.

4.
Do i need to issue XWAllowreentrancy from XCMD??

5.
Can't use MW source level debugger more than once (crashes into Macsbug)


Thank you

Pratap
email : [e-mail]
applelink : pratap

PS: Please keep a CC to me <[e-mail]>

--
Pratap P. Reddy
Apple development India
email : [e-mail]
phone - (91) - 80 - 2259495

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

From [e-mail] (Antonio Moura Martins)
Date: Fri, 04 Oct 1996 13:33:51 +0200
Organization: INRA - Lab. Intelligence Artificielle

In article (Dans l'article) <pratap-0110961603060001@17.85.20.112>,
[e-mail] wrote (Ècrivait)†:

> This is Pratap, sending this mail from India. I am writing an XCMD whcih
> interacts with OT streams. I have few problems while doing this. Can
> somebody answer to me if you know about any of these problems.

.....


> 2.
> One of my code warrior global is becoming NULL (endpoint that I opened
> initially). If I play around with the stack invoking XCMD many time, this
> happens. :-((
>

Global?? I think you can't use globals with XCMDs.
you can store them in a HC global.

This also means you cannot use some library routines.

Hope it helps.

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

From [e-mail] (Antonio Moura Martins)
Date: Fri, 04 Oct 1996 13:33:51 +0200
Organization: INRA - Lab. Intelligence Artificielle

In article (Dans l'article) <pratap-0110961603060001@17.85.20.112>,
[e-mail] wrote (Ècrivait)†:

> This is Pratap, sending this mail from India. I am writing an XCMD whcih
> interacts with OT streams. I have few problems while doing this. Can
> somebody answer to me if you know about any of these problems.

.....


> 2.
> One of my code warrior global is becoming NULL (endpoint that I opened
> initially). If I play around with the stack invoking XCMD many time, this
> happens. :-((
>

Global?? I think you can't use globals with XCMDs.
you can store them in a HC global.

This also means you cannot use some library routines.

Hope it helps.

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

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