Celestin Apprentice 5/Apprentice-Release5.iso/Information/CSMP Digest/volume 3/csmp-digest-v3-001

Text File | 1995-12-31 | 89.1 KB | 2,304 lines | [TEXT/R*ch]

Received-Date: Thu, 10 Mar 1994 19:07:16 +0100
From: [e-mail] (Francois Pottier)
Subject: CSMP Digest, Issue 1
To: [e-mail]
Date: Thu, 10 Mar 94 19:07:11 MET
X-Mailer: ELM [version 2.3 PL11]
Errors-To: [e-mail]
Reply-To: [e-mail]
X-Sequence: 2

C.S.M.P. Digest             Thu, 10 Mar 94       Volume 3 : Issue 1

Today's Topics:

        Async PBRead in a VBL
        C compilers for the MAC - Summary
        Fortran compilers for the Mac
        Getting time zone settings from Map control panel
        Grow-box icon drawn w-o scroll lines
        How to get all fonts ?
        Interacting with a POP Server
        Interrupting an interrupt
        Partial pathnames in FSMake
        PowerPC & CFM query
        Rules for Using This Newsgroup???
        Sending Apple Events from Excel
        Strange launch erroor
        TCL- CDLOGDialog [?]
        Time Manager
        Tip for Think C and Radius monitor users
        Universal Headers (A)
        c source code for simple-classic-old mac arcade type games wanted
        showtext-pascal



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

The digest is a collection of article threads from the internet newsgroup
comp.sys.mac.programmer.  It is designed for people who read c.s.m.p. semi-
regularly and want an archive of the discussions.  If you don't know what a
newsgroup is, you probably don't have access to it.  Ask your systems
administrator(s) for details.  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
nef.ens.fr).  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 digest is officially distributed by two means, by email and ftp.

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-digest Your Name    Adds you to the mailing list
    signoff csmp-digest            Removes you from the list
Once you have subscribed, you will automatically receive each new
issue as it is created.

The official ftp info is //ftp.dartmouth.edu/pub/csmp-digest.
Questions related to the ftp site should be directed to
[e-mail]. Currently no previous volumes of the CSMP
digest are available there.

Also, the digests are available to WAIS users as comp.sys.mac.programmer.src.


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

>From [e-mail] (Matt Mora)
Subject: Async PBRead in a VBL
Date: 23 Feb 1994 09:53:24 -0800
Organization: SRI International, Menlo Park, CA


I'm looking for some advice. I need to poll a device on the
serial port and read the data it spits out. Since I don't know how long the
device might take to respond (usally longer than one tick) I'm assuming
I need to to a Async PBRead since I don't want to tie up the system
waiting for a character. So what the best way to appoach this? Do I use
a few global flags to indicate a read is in progress and time out and
try agian if it doesn't respond the first time?

For example (c style pseudo code follows):

// see if any data awaits and start the read is there is

if (SerGetBuf(refNum,&count) == noErr || !gReadInProgress){
   gReadInProgress = true;
   gTimeout = TickCount() + kTimeOutValue;
     <fill in pb block>
   oe = PBRead(pb,true);
}

  // check for no error and for completion
  if (pb.ioResult==noErr && gReadComp){ //gReadComp  set by completion rtn
    <we got our data>
  }

  // check for time out
  if (gReadInProgress && TickCount() >= gTimeOut && !gReadComp ){
     // No response Try again
     gReadInProgress = false;
     // How do I stop the async read that is pending? KillIO?
  }


Is SerGetBuf appropriate here or is there a low level version I should be
using. The problem that I'm trying to solve is that we were using FSRead
which would hang the system if no characters were waiting. Appearently,
it was possible to be interupted between the call SGB and FSRead and
something else might have gobbled up that character that we were
about to read, so the system would lock up waiting for the nonexistant
char. Is that what you would call a deadly embrace? This interupt coding
is interesting.


Xavier



--
___________________________________________________________
Matthew Xavier Mora                    [e-mail]
SRI International                       [e-mail]
333 Ravenswood Ave                    Menlo Park, CA. 94025

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

>From [e-mail] (Pete Resnick)
Date: Wed, 23 Feb 1994 17:17:37 -0600
Organization: University of Illinois at Urbana-Champaign

In article <2kg56k$[e-mail]>, [e-mail] (Matt Mora)
wrote:

>I'm looking for some advice. I need to poll a device on the
>serial port and read the data it spits out. Since I don't know how long the
>device might take to respond (usally longer than one tick) I'm assuming
>I need to to a Async PBRead since I don't want to tie up the system
>waiting for a character. So what the best way to appoach this? Do I use
>a few global flags to indicate a read is in progress and time out and
>try agian if it doesn't respond the first time?

Is there any reason to wait for a timeout? Why not simply put an
asynchronous 1-byte PBRead on the serial port? Whenever the data comes in,
you will get your completion routine called, and you can do what you need
to do. If the device will be sending a lot of data, you will need to then
do a SerGetBuf to find out how much data is left and then do a PBRead on
the rest of the data. But forget all this stuff about global flags.

If you really do want a timeout, put an asynchronous PBRead on the serial
port (as much data as you want), and then set up a Time Manager task for
the timeout. When the timeout expires, you would call PBKillIO
asynchronously. Remember that KillIO clears all outstanding data too,
which might not be what you want to do. If the read completes, you can
simply remove the TM task.

>Is SerGetBuf appropriate here or is there a low level version I should be
>using.

You should use the low-level PBStatus call with the appropriate csCode for
SerGetBuf (2). You can call this immediately if you want to. If you do,
you will need to have this glue code:

#pragma parameter __D0 PBStatusImmed(__A0)
pascal OSErr PBStatusImmed(ParmBlkPtr paramBlock) = 0xA205;

Otherwise, you can call it asynchronously.

Interrupt programming is wonderful!

pr
--
Pete Resnick        (...so what is a mojo, and why would one be rising?)
Graduate assistant - Philosophy Department, Gregory Hall, UIUC
System manager - Cognitive Science Group, Beckman Institute, UIUC
Internet: [e-mail]

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

>From [e-mail] (Alnoor Ladak)
Subject: C compilers for the MAC - Summary
Date: Tue, 22 Feb 1994 19:58:57 GMT
Organization: ESRI Canada

Hi Netters:

Due to numerous requests, I have compiled a pseudo-summary of the responses
I received for 'C compilers for the Mac'. I have included all the messages
so that you may form your own opinions based on the various viewpoints for
the same software. The addresses of the individual authors are also included.

Hope this helps!

I would also like to once again thank all the authors for their input.
Thanks tons!

AL.


**********************************************************************
From: Jacques Marcoux <esri!cmc.aes.doe.CA!jmarcoux>
**********************************************************************

There is such beast as shareware C compiler for the Mac, however
they are pain in the neck.

IMHO your friend would be better off shelling out a $120.00 or so
to get Think C from Symantec (educationnal discount)
-------------------------------------------------------------------
language:    C
package:    Harvest C
version:    1.3
parts:        compiler, assembler, linker.
author:        ?
how to get:    ftp mac/development/languages/harves* from archive.umich.edu
description:    The parts of the system are all integrated in a unique
        application, which manages a "project" composed by several C
        source files and resource files (which contain data).
ports:        Macintosh
contact:    Eric W. Sink
updated:    1992/05/26

language:    C
package:    C-Interp
version:    ?
parts:        interpreter
author:        ?
how to get:    ftp Mac/Misc/C_Interp.sit from oac2.hsc.uth.tmc.edu
description:    An interpreter for a small subset of C, originally part of a
        communications package.
contact:    ? Chuck Shotton <[e-mail]>
updated:    1993/05/14

language:    C, C++, Objective-C, RTL
package:    GNU CC (gcc)
version:    2.5.5
parts:        compiler, runtime, examples, documentation
        Library listed separately
author:        Richard Stallman <[e-mail]> and others
how to get:    ftp gcc-2.4.5.tar.gz from a GNU archive site
    msdos:    ftp pub/msdos/djgpp/* from omnigate.clarkson.edu
    Macintosh:    ftp mpw-gcc-1.37.1r14 from ?
----- try sumex.aim and/or ftp.apple.com -------------------
description:    A very high quality, very portable compiler for C, C++,
        Objective-C.  The compiler is designed to support multiple
        front-ends and multiple back-ends by translating first
        into RTL (Register Transfer Language) and from there into
        assembly for the target architecture.    Front ends for
        Ada, Pascal, and Fortran are all under development.
conformance:    C: superset of K&R C and ANSI C.
        C++: not exactly cfront 3.0? [could someone tell me which
        version of cfront it is equivalent to, if any?    --ed]
        Objective-C: ?
bugs:        gnu.gcc.bug
restriction:    GNU General Public License
ports:        3b1, a29k, aix385, alpha, altos3068, amix, arm, convex,
        crds, elxsi, fx2800, fx80, genix, hp320, clipper,
        i386-{dos,isc,sco,sysv.3,sysv.4,mach,bsd,linux,windows,OS/2},
        iris,i860, i960, irix4, m68k, m88ksvsv.3, mips-news,
        mot3300, next, ns32k, nws3250-v.4, hp-pa, pc532,
        plexus, pyramid, romp, rs6000, sparc-sunos,
        sparc-solaris2, sparc-sysv.4, spur, sun386, tahoe, tow,
        umpis, vax-vms, vax-bsd, we32k, hitachi-{SH,8300}

**********************************************************************
From: esri!hydra.rose.brandeis.edu!reuber (Kent Reuber)
**********************************************************************

Think C from Symantec is one of the best C compilers around. It isn't free,
but Symantec offers very generous educational discounts. The cost at
Brandeis is about US$60.

**********************************************************************
From: esri!ecn.purdue.edu!jess (Jess M Holle)
**********************************************************************

DO NOT go shareware on this!  On just about any other type of software,
I'd advise you to look at shareware first, but not for a Mac C compiler.
The only free/shareware C compiler for the Mac OS that works to some
degree is Harvest C, but its author has admitted it has lots of bugs and
that he has no time to fix them or upgrade the package.

Actually, there is 1 _reasonable_ option for the C-guru who likes UNIX.
The option is to get MacMint, a freeware UNIX-like pseudo-OS.  It runs
as a process under the Mac OS, but runs programs compiled for it
pre-emptively under it.  If you get gcc, etc, for MacMint, then you can
compile programs for MacMint and run them.  This will only cost you in
hard drive space, download time, setup time, learning curve (unless you
are a UNIX/gcc guru), etc.

I'd recommend either picking up Think C or, for the more serious user,
CodeWarrior (which is still in the development stage).  Think C is only
about $60 educational price.  A "teacher friend" ought to be able to
finagle this price.  Symantec C++ still has some real problems, but Think
C is a rather solid ANSI C compiler with a very easy learning curve, nice
environment, and good debugger.  The $60 would pay off in the first few
hours of debugging, since the best debugger the MacMint stuff provides
is the GNU version of dbx, if that.  Also, Think C automatically figures
out how to "make" your projects, so you save all the time that you would
spend writing make files under MacMint.  Finally, don't forget that
programs compiled under MacMint only run under MacMint, so if he would
like to have others run the software someday, MacMint is probably out.
Think C and Symantec C++ are available from Symantec.

CodeWarrior is similar to Think C, BUT only available on CD, is still in
development (though anyone can buy it now even), and _probably_ not as
well tested.  CodeWarrior's C++ _sounds_ more stable than Symantec's,
and some limitations of the Think environment are not present in the
CodeWarrior environment.  CodeWarrior also has educational prices and
is available from Metrowerks.

>PS. He does not have net access, so please respond to me & I'll

This would make MacMint usage more difficult since all the pieces and
all the support are available almost solely from the net.

The bottom line _to a degree_ is that you get what you pay for.  This is
not true in the extreme, however.  I would not recommend MPW C, which is
the other major C option on the Mac.  It is expensive, hard-to-use, hard
to learn, etc.  After using MPW for quite a while, I'd advise against its
purchase except for C++ development and even in that case with CodeWarrior
out on the market now and Symantec C++ improving.

**********************************************************************
From: "Joseph S. D'Elia" <esri!crl.com!silk>
**********************************************************************

I don't know of any Shareware C programs out there (that doesn't mean
there aren't any) but I have a copy of Think C 5.0 he can buy with all
documentation and manuals.  It would only cost him $50.

I recently purchased Think C 6.0 so I don't need 5.0 anymore.  If nothing
else comes your way, keep it in mind.  It is just sitting on my shelf.

**********************************************************************
From: esri!katya.acs.ilstu.edu!bavogler (Benjamin A. Vogler)
**********************************************************************

You can get Apple's C GCC compiler from ftp.apple.com...

**********************************************************************
From: Chong Richard Ng Kim <esri!cdf.toronto.edu!a260chon>
**********************************************************************

Hi Al!

Don't know of any shareware C compilers for the Mac. In fact, the only Mac C
compilers that I know of are Symantec's THINK C and Apple's MPW C.

If your friend is a member of a user group, tell him to call Symantec Canada.
(They're in the Toronto phone book.) As a UG member, you can get THINK C for
only $79 Canadian (+PST+GST of course!)

If your friend isn't a member of a UG, then might I suggest ClubMac? (I'm a
CMAC member myself.) There's an info line for CMAC at [phone]. If
your friend wants more info then just leave name & address &/or phone # on the
answering machine and someone will be in touch soon thereafter.

Hope this helps out some!

**********************************************************************
From: esri!astro.uu.se!Olof.Morell (Olof Morell)
**********************************************************************

Think C, current version 6.0. Commercial from Symantec corp.
If you can find an used 5.0.x version for not too much it's a bargain...

Good compiler.

**********************************************************************
From: Ben Cottrell <esri!soda.berkeley.edu!benco>
**********************************************************************

Try Harvest C, which can be gotten off of sumex-aim.stanford.edu and numerous
other places.

Tell me if you want more information or need help finding it!

**********************************************************************
From: esri!marie.physik.tu-berlin.de!georg (Georg Schwarz)
**********************************************************************

Harvest C 1.3 is freeware. Try it!


--
______________________________________________________________________________
Alnoor Ladak (email [e-mail]) |Somebody ought to cross ball point pens
ESRI Canada, 49 Gervais Dr.,         |  with coat hangers  so that the pens
Don Mills, Ont. M3C 1Y9              |   will muliply instead of disappear

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

>From [e-mail] (Chandrashekhar S Damle)
Subject: Fortran compilers for the Mac
Date: 22 Feb 1994 18:59:45 GMT
Organization: The Ohio State University


If anybody has any experience with Macintosh Fortran compilers sold
by Language Systems or Absoft, I would appreciate it if they could
relate their experiences to me.

I am thinking of buying a Fortran compiler for our Mac IIci with system 7.1
I have received some information from both of these companies, but
would appreciate it if somebody who has real experience on them could
give me their opinions.

Thank you.

Shekhar Damle
[e-mail]

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

>From [e-mail] (Peter Chang)
Date: 23 Feb 1994 08:37:02 -0500
Organization: Carderock Division, NSWC, Bethesda, MD


In comp.lang.fortran, [e-mail] (Chandrashekhar S Damle) writes:
>If anybody has any experience with Macintosh Fortran compilers sold
>by Language Systems or Absoft, I would appreciate it if they could
>relate their experiences to me.
>
I have been using Language Systems Fortran for the Mac for the past
5 years.  I have no experience with Absoft. Here's a random list of
thoughts concerning LS Fortran running inside MPW:
- you have to learn how to get around inside the MPW shell. At first
a little arcane, but eventually feels like a low power Unix shell. YOu
can customize MPW to the same commands as Unix.  MPW gives you command
line file maniupulation (as opposed to Finder) which is very powerful
and makes up for the deficiencies in Finder.
- the LS Fortran has some very nice extensions, e.g. cycle and leave
statements for do loops.
- the editor makes programming very quick. For instance indenting large
blocks of code makes programming much more organized. I have not used
a statement label (except for formats) in a long time.
- many compiler options for various debugging modes and optimization
levels.
- source code debugger
- I do all my programming in MPW, then port over to various Crays etc.
- My comments are more based on use of MPW. I am very pleased with LS
Fortran, but as stated above cannot compare it with Absoft.

Please email me if you have more specific questions.
------------------------------------------------------------------------------
Peter Asha Chang, III         |"That's exactly the method", the Bellman bold
[e-mail]       |              In a  hasty parenthesis cried,
301.227.1798 (voice)          |"That's exactly the way I have always been told
301.227.3883 (fax)            |   That the capture of Snarks should be tried"
David Taylor Model Basin      |               --Lewis Carroll
Bethesda, MD, USA             |               Alice in Wonderland
-------------------------------------------------------------------------------

-------------------------------------------------------------------------------
Peter Asha Chang, III         |"That's exactly the method", the Bellman bold
[e-mail]       |              In a  hasty parenthesis cried,
301.227.1798 (voice)          |"That's exactly the way I have always been told
301.227.3883 (fax)            |   That the capture of Snarks should be tried"
David Taylor Model Basin      |               --Lewis Carroll
Bethesda, MD, USA             |               Alice in Wonderland
-------------------------------------------------------------------------------


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

>From [e-mail] (Ron Shepard)
Date: Wed, 23 Feb 1994 18:34:51 GMT
Organization: Argonne National Laboratory, Chicago, Illinois

In article <[e-mail]> [e-mail] (Peter Chang) writes:
>
>In comp.lang.fortran, [e-mail] (Chandrashekhar S Damle) writes:
>>If anybody has any experience with Macintosh Fortran compilers sold
>>by Language Systems or Absoft, I would appreciate it if they could
>>relate their experiences to me.
>>
>I have been using Language Systems Fortran for the Mac for the past
>5 years.  I have no experience with Absoft. Here's a random list of
>thoughts concerning LS Fortran running inside MPW:
[...]

I have experimented with various versions of LS Fortran on the Mac too,
but I always seemed to run into some small bug that would stop me from
doing what I needed to do.  I don't mean to pick on LS Fortran in
particular; this comment probably applies to all PC type fortran
compilers.  I just don't know if you can really expect to find
"industrial strength" compilers like those you find on workstations,
minis, supercomputers, etc.

For example, I remember with LS Fortran that list-directed I/O on
character arrays would not work.  Many of my application codes use
this, so when I ran into the bug, it pretty much halted progress.

      character*8 carray(5)
      data carray/ '1', '2', '3', '4', '5' /
      read(5,*) carray
      write(6,*) carray
      end

input file:
 'new1' 'new2' / take the default values for the rest

When executed, the correct output should be:

 new1    new2    3       4       5

But with (on several versions at least) LS Fortran you get:

 new1    new2    /       take    the

Maybe this has been fixed (anyone know?).  It is just something to watch
out for with PC-class compilers.

$.02 -Ron Shepard

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

>From [e-mail] (Fred Walsteijn)
Date: Wed, 23 Feb 1994 22:58:14 GMT
Organization: Physics Department, University of Utrecht, The Netherlands

In <[e-mail]> [e-mail] (Ron Shepard) writes:

>....  I just don't know if you can really expect to find
>"industrial strength" compilers like those you find on workstations,
>minis, supercomputers, etc.
>For example, I remember with LS Fortran that list-directed I/O on
>character arrays would not work.  Many of my application codes use
>this, so when I ran into the bug, it pretty much halted progress.
>      character*8 carray(5)
>      data carray/ '1', '2', '3', '4', '5' /
>      read(5,*) carray
>      write(6,*) carray
>      end
> [stuff deleted]
>Maybe this has been fixed (anyone know?).  It is just something to watch
>out for with PC-class compilers.

Hello Ron,
The latest version of LS Fortran (3.3) works OK:
***input:
'new1' 'new2' / take the default values for the rest
***output:
new1    new2    3       4       5

I agree with your remark on the quality of compilers.
If you limit yourself to the Mac, however, then LS Fortran is a good choice.
I use the MPW Shell for 90% of my Fortran code editing, and use LS Fortran
(sometimes) for some initial tests (and use FTNCHEK in MPW as well).
At home I do all my work with LS Fortran.

_____________________________________________________________________________
Fred Walsteijn                                | Internet: [e-mail]
Institute for Marine and Atmospheric Research | FAX:      [phone]
Utrecht University, The Netherlands           | Phone:    [phone]

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

>From [e-mail] (Matthias Neeracher)
Subject: Getting time zone settings from Map control panel
Date: 20 Feb 1994 21:08:56 GMT
Organization: Swiss Federal Institute of Technology (ETHZ)

I am wondering if there is a way to derive time zone information from
the settings of the Map control panel. I seem to have seen this in
a few programs. On a related topic, can anybody who has used this
information comment on how high a percentage of Mac users configures
their Map control panel correctly?

Respondents will earn my eternal gratitude.

Matthias

-----
Matthias Neeracher                                      [e-mail]
      "Success is what me make of the mess we have made out of things"

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

>From [e-mail] (Bill Hofmann)
Date: Sun, 20 Feb 1994 22:12:43 GMT
Organization: NETCOM On-line Communication Services (408 241-9760 guest)

[e-mail] (Matthias Neeracher) writes:
>I am wondering if there is a way to derive time zone information from
>the settings of the Map control panel. I seem to have seen this in
>a few programs. On a related topic, can anybody who has used this
>information comment on how high a percentage of Mac users configures
>their Map control panel correctly?
ReadLocation reads a rather bizarre data structure which encodes the
current lat/long and the time zone.  Currently undocumented (uh oh) but
important: the union that encodes the time zone info and the "dlsDelta"
uses the high order bit of dlsDelta to tell you that daylight savings
is currently on, at least in System 7 Pro.  See IM VI for more info.
Another piece of info: gmtDelta reflects not the time zone but the delta
from GMT, which may need to be adjusted if DST is on.

Well, I'd guess a small percentage use this feature.  7Pro encourages it,
but since my software requires it, I more or less force the user to set it.

>Respondents will earn my eternal gratitude.
Wow.
>Matthias

>-----
>Matthias Neeracher                                      [e-mail]
>      "Success is what me make of the mess we have made out of things"
--
-Bill Hofmann                    [e-mail]
 Fresh Software and Instructional Design    [phone]

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

>From [e-mail] (Matthias Neeracher)
Date: 22 Feb 94 17:44:14
Organization: Integrated Systems Laboratory, ETH, Zurich

In article <[e-mail]>, [e-mail] (Bill Hofmann) writes:

> [e-mail] (Matthias Neeracher) writes:
>>I am wondering if there is a way to derive time zone information from
>>the settings of the Map control panel.

> ReadLocation reads a rather bizarre data structure which encodes the
> current lat/long and the time zone.

Thank you very much. As soon as I had the routine name, I was able to find what
was documented in the Script Manager chapter of IM VI. This information doesn't
seem to have made it into the Script Manager chapter of NIM:Text (doesn't have
much to do with text, anyway), I wonder where it will end up in the NIM series.

> Well, I'd guess a small percentage use this feature.

This was also the consensus of the e-mail messages I received. Thanks to
Michael Hanson, Scott M. Silver, and Douglas Rowe for their helpful
suggestions.

Matthias

-----
Matthias Neeracher                                  [e-mail]
  "And that's why I am going to turn this world upside down, and make
   of it a fire so *bright* that someone real will notice"
                                -- Vernor Vinge, _Tatja Grimm's World_

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

>From [e-mail] (Tim Shimmin)
Subject: Grow-box icon drawn w-o scroll lines
Date: 22 Feb 1994 13:34:50 +1100
Organization: Zikzak Public Access UNIX, Melbourne Australia

I would like to be able to draw the grow-box icon for
a window but without the scroll-lines being drawn -
which happens when calling DrawGrowIcon.
Could someone please enlighten me as to how it is done.
I have a vertical palette on the left hand side which
I don't want "over-drawn" by a scroll line.

Thanks.

--tim
--
                   Tim Shimmin - [e-mail]
        Zikzak public access UNIX, Melbourne, Australia.

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

>From [e-mail] (Brian Clark)
Date: Mon, 21 Feb 1994 22:25:31 -0600
Organization: Northwestern University

In article <2kbr0o$[e-mail]>, [e-mail] (Tim
Shimmin) wrote:

> I would like to be able to draw the grow-box icon for
> a window but without the scroll-lines being drawn -
> which happens when calling DrawGrowIcon.
> Could someone please enlighten me as to how it is done.
> I have a vertical palette on the left hand side which
> I don't want "over-drawn" by a scroll line.

I believe one way to do this is covered in one of the Macintosh Inside Out
series of books from Addison-Wesley titled "Macintosh Programming Secrets"
by a couple of unknowns named Knaster and Rollins :-). As I recall, the
secret is to save the current clip region, set the new clip region to
include only the icon, call DrawGrowIcon, then restore the original clip
region.

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

>From Carl R. Osterwald <[e-mail]>
Date: Mon, 21 Feb 94 04:39:43 GMT
Organization: National Renewable Energy Lab.

In article <2kbr0o$[e-mail]> Tim Shimmin,
[e-mail] writes:
>I would like to be able to draw the grow-box icon for
>a window but without the scroll-lines being drawn -
>which happens when calling DrawGrowIcon.
>Could someone please enlighten me as to how it is done.

Sure.

1.  Save existing window clipping region
2.  Set clipping region to rectangle containing only GrowIcon
3.  Call DrawGrowIcon
4.  Restore window clipping region

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

>From [e-mail] (Robert S. Mah)
Date: Tue, 22 Feb 1994 00:01:51 -0500
Organization: One Step Beyond

[e-mail] (Tim Shimmin) wrote:

> I would like to be able to draw the grow-box icon for a window but without
> the scroll-lines being drawn - which happens when calling DrawGrowIcon.
> Could someone please enlighten me as to how it is done.

Sure thing.  See the code below.

Cheers,
Rob
________________________________________________________________________
 Robert S. Mah              One Step Beyond              [e-mail]


#define kGrowBoxSize 16

void DrawGrowIconNoLines( WindowPtr wp, Boolean hideLines )
{
    RgnHandle  saveClip;
    GrafPtr    savePort;
    Rect       growBox;

    if( hideLines == false ){
        DrawGrowIcon( wp );
    }else{
        saveClip = NewRgn();
        if( saveClip == NULL )
            return;

        GetPort( &savePort );
        SetPort( wp );

        GetClip( saveClip );

        growBox.right  = wp->portRect.right;
        growBox.bottom = wp->portRect.bottom;
        growBox.left   = growBox.right - kGrowBoxSize;
        growBox.top    = growBox.bottom - kGrowBoxSize;

        ClipRect( &growBox );
        DrawGrowIcon( wp );

        SetClip( saveClip );
        SetPort( savePort );
        DisposeRgn( saveClip );
    }
}

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

>From [e-mail] (Axel Kowald)
Subject: How to get all fonts ?
Date: Mon, 21 Feb 1994 10:20:08 GMT
Organization: Manchester Computing Centre

Hi

What is the proper way to get a list of all fonts which are installed
on a mac ?  I could use GetFontName() but it seems the font numbers can be
scattered around the whole integer range. I can't imagine I have to scan
all integer numbers, so what is the way to do it ?

            Thanks
                                Axel ([e-mail])

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

>From Reid Ellis <[e-mail]>
Date: Tue, 22 Feb 1994 20:06:41 GMT
Organization: Alias Research, Inc., Toronto ON Canada

Axel Kowald <[e-mail]> writes:
|What is the proper way to get a list of all fonts which are
|installed on a mac ?  I could use GetFontName() but it seems the
|font numbers can be scattered around the whole integer range. I
|can't imagine I have to scan all integer numbers, so what is the
|way to do it ?

I can't remember where I saw it, but the documented way of doing
this is to create a menu, call AddResMenu(theMenu, 'FOND') [*not
'FONT'!], loop through all the menu items, copying the item's
string into a list of font names you are creating, and when you
are done, get rid of the menu again.

Intuitive, eh?  :-/

Reid
--
Reid Ellis, Alias Research Inc.
[phone] <[e-mail]>

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

>From [e-mail] (Scott Gruby)
Subject: Interacting with a POP Server
Date: Sat, 19 Feb 1994 23:19:35 -0800
Organization: Harvey Mudd College, Claremont CA

Does anyone out there know how to interact with a POP server via MacTCP?
What I would like to do is simply send a username and password and see if
new mail is there and then close the connection (I don't want to get the
mail.).

a) What commands do I have to send to the POP server?

b) What is the easiest way to do it?

Thanks.

--
Scott Allen Gruby                         ([e-mail])

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

>From [e-mail] (Peter N Lewis)
Date: 21 Feb 1994 11:16:19 +0800
Organization: NCRPDA, Curtin University

[e-mail] (Scott Gruby) writes:

>Does anyone out there know how to interact with a POP server via MacTCP?
>What I would like to do is simply send a username and password and see if
>new mail is there and then close the connection (I don't want to get the
>mail.).

POP, like all good protocols (and surprisingly enough, POP is actually quite
a good protocol) is documented in an RFC (Request for Comment).  They are
stored at munnari.oz.au:/rfc (amoungst other places).  This directory has
zillions of files in it, so listing it is somewhat tedious.  The relevent file
is munnari.oz.au:rfc/POP3.Z.

telnet <ip> pop3
R: +OK ncrpda POP3 Server (Version 1.002) ready.
S: user peter
R: +OK please send PASS command
S: pass *****
R: +OK 0 messages ready for peter in /usr/spool/mail/peter
S: quit
R: +OK whatever

Enjoy,
   Peter.
--
_______________________________________________________________________
Peter N Lewis <[e-mail]>       Ph: [phone]

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

>From [e-mail] (Philip Machanick)
Date: Mon, 21 Feb 1994 12:21:54 +0200
Organization: Computer Science Dept, U of Witwatersrand

In article <[e-mail]>, [e-mail]
(Scott Gruby) wrote:

> Does anyone out there know how to interact with a POP server via MacTCP?
> What I would like to do is simply send a username and password and see if
> new mail is there and then close the connection (I don't want to get the
> mail.).
>
> a) What commands do I have to send to the POP server?
>
> b) What is the easiest way to do it?

Telnet to your mail host to port 110 (or 109 if that doesn't work), tell it
who you are then quit.

For example - replace "concave" by the name of your mail host in what
follows, and of course put in your own user name and password (lines
starting with "+" are output from the server):
--
telnet concave 110
Trying 146.141.15.20...
Connected to concave.cs.wits.ac.za.
Escape character is '^]'.
+OK UCB Pop server (version 1.831beta) at concave.cs.wits.ac.za starting.
user philip
+OK Password required for philip.
pass ########
+OK philip has 0 message(s) (0 octets).
quit
+OK Pop server at concave.cs.wits.ac.za signing off.
Connection closed by foreign host.
--
If you want to write a program to do this, good luck.
--
Philip Machanick                   [e-mail]
Department of Computer Science, University of the Witwatersrand
2050 Wits, South Africa
phone [phone]  fax [phone]

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

>From [e-mail] (Andy Shiekh)
Subject: Interrupting an interrupt
Date: Mon, 21 Feb 1994 12:24:48 +0100
Organization: International Centre for Theoretical Physics

  I recently read in "The Art of Electronics" concerning programming
the 68000 series that:

  In order to prevent pandemonium, the CPU disables interrupts when
it jumps to the handler, and reenables them when it executes RTE.
If you have a long winded handler, you may want to re-enable interrupts
(at higher-priority levels only) within the handler, which you can do
by sending the appropriate byte to the status register.

  Can anyone enlighten me on what the appropriate byte is?


Andy

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

>From [e-mail] (Lawrence D'Oliveiro, Waikato University)
Date: 22 Feb 94 11:32:53 +1300
Organization: University of Waikato, Hamilton, New Zealand

In article <[e-mail]>, [e-mail] (Andy Shiekh) writes:
>   I recently read in "The Art of Electronics" concerning programming
> the 68000 series that:
>
>   In order to prevent pandemonium, the CPU disables interrupts when
> it jumps to the handler, and reenables them when it executes RTE.
> If you have a long winded handler, you may want to re-enable interrupts
> (at higher-priority levels only) within the handler, which you can do
> by sending the appropriate byte to the status register.

I was aware of no such thing. My understanding of interrupt handling on
the 68K family (or on any CPU family, for that matter), is that

a) the priority of the interrupt request is compared against the current
CPU priority, and the interrupt is only granted if its priority is higher, and
b) before entering the interrupt service routine, the CPU priority is raised
to the interrupt priority. This blocks out further interrupts at the same or
a lower priority, until the interrupt service routine has finished.

Thus there is no automatic disabling of _all_ interrupts--just those that
haven't got a higher priority.

The only time that code needs to fiddle the CPU priority is when it is
synchronizing with interrupt code. For example, say you're writing a device
driver for a device that generates interrupts at priority p. When driver
code (for example, invoked as a result of a user call) needs to access data
structures that are also accessed by the interrupt service routine, then you
need to raise the priority to at least p while doing the access (and restore
it to what it was before afterwards), otherwise you could end up with
inconsistent data.

Note: you should never *lower* CPU priority from what it was when your code
was entered. Doing so can destroy the interrupt nesting scheme, break the
synchronization assumptions in other interrupt handlers, and generally cause
havoc.

Lawrence D'Oliveiro                       fone: [phone]
Info & Tech Services Division              fax: [phone]
University of Waikato            electric mail: [e-mail]
Hamilton, New Zealand    37^ 47' 26" S, 175^ 19' 7" E, GMT+13:00

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

>From [e-mail] (Pindar Infotek Ltd)
Subject: Partial pathnames in FSMake
Date: 22 Feb 1994 03:02:35 -0600
Organization: UTexas Mail-to-News Gateway

>From Pindar user Ian McCall

Subject: _Partial pathnames in FSMakeFSSpec
Hello.

I'm having a problem locating files in a directory structure that's
been set up. The structure is as follows:

File server root:Notes:a:b:c:abcxxx

where a, b, and c are single digits, and xxx are also digits.

I've set up an alias resource to 'File server root', which is
correctly being resolved. I'd now like to use the FSSpec I receive
through that resolution plus a partial pathname of
'Notes:a:b:c:abcxxx' to specify a file. Whenever I try however, I'm
constistently getting back a -35 err - no such volume.

I've set the FSSpec of the file to equal fileServerRoot.vRefNum,
fileServerRoot.parID, and 'Notes:a:b:c:abcxxx' for each of the
fields.

What's going wrong? Please reply via email, since I don't actually
get to read this group.

Thanks in advance for any information.



---
Please make 'Attn:<space>Ian McCall' the first line of your reply
(not the subject) to ensure automatic forwarding to the correct
Pindar user.

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

>From [e-mail] (Peter N Lewis)
Date: 23 Feb 1994 11:25:17 +0800
Organization: NCRPDA, Curtin University

[e-mail] (Pindar Infotek Ltd) writes:

>through that resolution plus a partial pathname of
>'Notes:a:b:c:abcxxx' to specify a file. Whenever I try however, I'm
>constistently getting back a -35 err - no such volume.

Thats because 'Notes:a:b:c:abcxxx' is not a partial pathname.  Its a full
pathname.  ':Notes:a:b:c:abcxxx' is what you want.  A pathname is partial
if and only iff it contains no colons, or it starts with a colon.

>What's going wrong? Please reply via email, since I don't actually
>get to read this group.

See my note about "rules for posting" ;-)
   Peter.
--
_______________________________________________________________________
Peter N Lewis <[e-mail]>       Ph: [phone]

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

>From Brian Patrick Arnold <[e-mail]>
Subject: PowerPC & CFM query
Date: Tue, 22 Feb 1994 18:58:45 -0500
Organization: Sponsored account, Engineering and Public Policy, Carnegie Mellon, Pittsburgh, PA


I have some hard questions regarding PowerPC migration that I figure I
ought to start addressing as soon as possible.

I want some of you to help me guestimate the tradeoffs in migrating a
MacApp 2.0 application to PowerPC.  Please reply directly to me and I
will do my best to summarize any helpful replies.

Assumptions:
-----------
o  App uses MacApp 2.0 with "7.0 for MacApp 2.0" modifications
o  App's user interface is written in MPW Object Pascal 3.2
o  App's kernel (50% of code) is written in C, with a tiny bit of
   C++, with no MacApp/PascalObject dependency (except via
   vanilla procedures)
o  App's kernel has all relevant computation-intensive code
o  App has been tested to run on PowerPC architecture with its
   unmodified 68k code

Hard choices:
------------
a. Port the kernel to PowerPC but keep the interface in MC68k
b. Port the kernel and interface code as-is, all for PowerPC
c. Translate interface code to C++, upgrade to MacApp 3.0 and then 3.1

The goal being sought is to change as little code as possible in as
little time as possible.  I have a sneaky suspicion that I can get away
with option (a) in far less time than option (b) or (c) if I knew the
rights answers as to how to do it.

Tradeoffs:
---------
Option (a) requires dipping into code fragments and/or dynamic linking.
How hard is it to turn half of an existing API on its head, compile
part of it for PowerPC and have it be directly accessible with 68k code
via the Code Fragment Manager?  Easy?  Tedious?

Option (b) I assume requires modifying MacApp 2.0 to handle Universal
headers and having a PowerPC Pascal compiler than can handle MacApp
2.0.  How hard can one estimate modifying MacApp 2.0 to be?  Also has
anyone heard whether a PowerPC Pascal compiler will support Pascal
Objects and MacApp 2.0?  Is this simply too far fetched an option?

Option (c) is more well-defined but less appealing to me at least in
the time frame being considered.  This is because translation is
expected to take one month, upgrading to MacApp 3.0 is expected to take
another month, MacApp 3.1 might take another week, and by then a lot of
code has changed, so testing will become a big deal.  This sounds like
too much work, although I'd like to move to MacApp 3.1 at some point in
the distant future (well...).

Anyone willing to rise to this challenge and help me make better
estimates for option (a), or convince me why option (b) or (c) is
better?

- Brian


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

>From [e-mail] (Jon Wdtte)
Date: 23 Feb 1994 07:56:03 GMT
Organization: Royal Institute of Technology, Stockholm, Sweden

In <[e-mail]> Brian Patrick Arnold <[e-mail]> writes:

>o  App uses MacApp 2.0 with "7.0 for MacApp 2.0" modifications

Ouch!

>o  App's user interface is written in MPW Object Pascal 3.2

Aaaahhh... (now has a serious limp and terrible cut above eyes)

>o  App's kernel (50% of code) is written in C, with a tiny bit of
>   C++, with no MacApp/PascalObject dependency (except via
>   vanilla procedures)

Oh, that's better (bandages in place)

>o  App has been tested to run on PowerPC architecture with its
>   unmodified 68k code

Well, that doesn't make transition harder nor easier. If you run
on a 68k Mac, you run on the PowerPC emulation. No surprise there.

>a. Port the kernel to PowerPC but keep the interface in MC68k

This is, technically, a bear, since there is no linker currently
available which can grok 68k and PPC code at the same time.

>b. Port the kernel and interface code as-is, all for PowerPC

Hard to do, sincC++ and MacApp 3.1 is needed.

>c. Translate interface code to C++, upgrade to MacApp 3.0 and then 3.1

I would suggest re-implementing the interface code in MacApp 3.1
using whatever cool interface editors/code generators you can get
your grubby on.

>The goal being sought is to change as little code as possible in as
>little time as possible.  I have a sneaky suspicion that I can get away

I think my suggestion would actually be the fastest. AND you get a
good shot at cleaning up your UI, if that's needed.

--
 -- Jon W{tte, [e-mail], Mac Hacker Deluxe --
  "And now, from the makers of EDLIN, comes: Windows NT!"

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

>From [e-mail] (Zalman Stern)
Date: Thu, 24 Feb 1994 01:51:55 GMT
Organization: Adobe Systems Incorporated

Jon W tte writes
> In <[e-mail]> Brian Patrick Arnold
<[e-mail]> writes:
> >a. Port the kernel to PowerPC but keep the interface in MC68k
>
> This is, technically, a bear, since there is no linker currently
> available which can grok 68k and PPC code at the same time.

Its very easy. Write a PowerPC code fragment which has an entry point that
accepts a table of function pointers and returns (a possibly different)
table of function pointers. Write jacket functions that call through the
table of function pointers on each side. Add initialization code that loads
the code fragment and calls it to exchange function tables. If it takes you
more than an hour to do this given a list of routine names, you are not an
accomplished tools user. (Generating the procinfos for the kernel functions
is harder. There's a tool to do it for you, but I doubt its available.)

This ignores the problem of global variables. If you have lots of them,
think of this as an opportunity to improve your code and remove them. You
can add pointers to the globals to information that gets exchanged during
the initialization handshake. You will have to revise the kernel code to
deal with an extra level of indirection on its globals.

[...]
> >c. Translate interface code to C++, upgrade to MacApp 3.0 and then 3.1
>
> I would suggest re-implementing the interface code in MacApp 3.1
> using whatever cool interface editors/code generators you can get
> your grubby on.
>
> >The goal being sought is to change as little code as possible in as
> >little time as possible.  I have a sneaky suspicion that I can get away
>
> I think my suggestion would actually be the fastest. AND you get a
> good shot at cleaning up your UI, if that's needed.

Brian's solution is almost certainly faster and less intrusive to the code.
(And I have enough direct experience with both MacApp and PowerPC to be very
confident in saying that.) The "goodness" of such an approach depends on how
well the "kernel" matches the performance cross section of your app. (I.e.
does it get you the speed you need.) It also becomes an issue of deferred
maintenance. There will be a day when compiling Object Pascal for a 68K is
about as easy as compiling Algol 68 for a CDC 6600.
--
Zalman Stern           [e-mail]            [phone]
Adobe Systems, 1585 Charleston Rd., POB 7900, Mountain View, CA 94039-7900
"Do right, and risk consequences." Motto of Sam Houston (via Molly Ivins)

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

>From [e-mail]
Subject: Rules for Using This Newsgroup???
Date: 20 Feb 1994 09:44:05 -0500
Organization: Virginia Tech, Blacksburg, VA


>From: David A Lyons <[e-mail]>
>Date: Sat, 19 Feb 1994 06:40:14 GMT
>Organization: Apple Computer, Inc.
>
>(Having the right books handy is a very good thing.  You'll save
>yourself and everybody else a lot of time by getting them and becoming
>familiar with them.)
>
>Dave Lyons, [e-mail]
>Mr Tangent


I posed a question to the newsgroup dealing with UpdateEvents and
Redrawing the contents of a window.  I do not have IM at this time due
to lack of funds.  I do have Think Reference which has been very helpful
and I use this Newsgroup as a source of information.  Everyone has been
very helpful but after this last response I feel like I'm misusing this
newsgroup!  What exactly are the rules for posting questions on this
Newsgroup?  Is there a limit to the quantity of questions one person may
ask or do your questions have to be of a certain complexity?

I have been using the Macintosh C Programming Primers as a source of
information and I was interested in finding out more about UpdateEvents.
The Primer led me to belive that you could redraw the whole window or
you could keep track of exactly what needed to be updated and only
redraw that portion of the window.  The examples in the Primers only
showed redrawing the whole window so I was wondering if others lived
with that approach or if they used more complex code to speed things
up!?  I used the BeginUpdate/redraw/EndUpdate logic and it worked; I was
just trying to make things faster so I asked what others did.

Should I not ask such trivial questions on this Newsgroup?


Stephen L. Greenfield                  Phone:     [phone]
1700 Pratt Drive - Computing Center    Bitnet:    FEATS@VTVM1
Virginia Polytechnic Institute         Internet:  [e-mail]
Blacksburg, Virginia  24060-6361 ,/\              [e-mail]
                                /   `\/`'\
                           ,/\/          /
                         ,/             <_          ._.
                        /                 `\       / /
             ,/\_     _/'                   `-`-\  |/
          ,/'    `_'_'  *  Blacksburg          \_> '
       ,/'                                    \__>_/\
      +----------------------------------------------+

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

>From [e-mail] (Steven Lane)
Date: Sun, 20 Feb 1994 17:10:07 GMT
Organization: University of Chicago

In article <2k7svl$[e-mail]> [e-mail] writes:

>Everyone has been
>very helpful but after this last response I feel like I'm misusing this
>newsgroup!  What exactly are the rules for posting questions on this
>Newsgroup?  Is there a limit to the quantity of questions one person may
>ask or do your questions have to be of a certain complexity?

This being the net, there aren't many rules per se. There are
standards of etiquette; if you transgress these, you will not be
fined, but you may be gently mocked (or not so gently, depending on
the offense). In general, when posting questions to a
technically-oriented group such as this one, the best rule is not to
ask a question until you've tried to figure out the answer yourself.
If you didn't read some obvious piece of documentation which answers
your question, people may be a little grumpy with you. If, on the
other hand, you don't know which obvious piece of your thousands of pages
of docs you should read, or you read it and made no sense of it, post
away. This is a very helpful group, personality-wise -- many rank
beginners (myself included) post questions and receive speedy and
thorough answers. At worst, if your question is somehow inappropriate,
it will just sit by itself in your thread selector, with no responses.

>I have been using the Macintosh C Programming Primers as a source of
>information and I was interested in finding out more about UpdateEvents.
>The Primer led me to belive that you could redraw the whole window or
>you could keep track of exactly what needed to be updated and only
>redraw that portion of the window.  The examples in the Primers only
>showed redrawing the whole window so I was wondering if others lived
>with that approach or if they used more complex code to speed things
>up!?  I used the BeginUpdate/redraw/EndUpdate logic and it worked; I was
>just trying to make things faster so I asked what others did.

I can't tell whether your question has been answered already; I'll
pretend it hasn't. It's actually the Window Manager which keeps track
of which part of the window needs to be refreshed. Look at the
structure of a WindowRecord and you'll see a field called updateRgn.
When you use the Begin/EndUpdate sequence, you are warning the Window
Manager that it should make sure your drawing is clipped so as only to
go to the updateRgn.

[ long .sig deleted ]

You might consider trimming your signature file a little, to save
bandwidth.

--
----
Steve Lane
University of Chicago, Department of History
[e-mail]

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

>From [e-mail] (Edwin H. Watkeys III)
Date: Sun, 20 Feb 1994 18:18:41 GMT
Organization: Drexel University

In article <[e-mail]> [e-mail] writes:
>In article <2k7svl$[e-mail]> [e-mail] writes:
>
>>[ far too considerate poster's comments deleted. ]
>
>This being the net, there aren't many rules per se. There are
>standards of etiquette; if you transgress these, you will not be
>fined, but you may be gently mocked (or not so gently, depending on
>the offense). In general, when posting questions to a
>technically-oriented group such as this one, the best rule is not to
>ask a question until you've tried to figure out the answer yourself.
>If you didn't read some obvious piece of documentation which answers
>your question, people may be a little grumpy with you. If, on the
>other hand, you don't know which obvious piece of your thousands of pages
>of docs you should read, or you read it and made no sense of it, post
>away. This is a very helpful group, personality-wise -- many rank
>beginners (myself included) post questions and receive speedy and
>thorough answers. At worst, if your question is somehow inappropriate,
>it will just sit by itself in your thread selector, with no responses.

Isn't that depressing when it happens to one of your posts? It's
not necessarly because your question is inappropriate: often, there just
isn't anyone out there at the time who knows much of anything about you
question. Of course, it could be because everyone thinks you're an
asshole. I usually assume the former when I see other people's lonely
posts, but the latter when my posts go unanswered... Oh well...

You realy need to think before strting a new thread... How are you going
to appeal to potential readers with your subject line? How will you phrase
your post, so as to get the right people in a responding mood? You want to
go for the big names on csmp -- it'll make you feel somehow more worthy...

Or maybe I need to see a psychologist...

Ed
--
Ed Watkeys              [e-mail]  Drexel University
Philosopher-Programmer  [e-mail]           Distant Software
                        - Use finger for my PGP key. -

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

>From [e-mail] (Jon Wdtte)
Date: 20 Feb 1994 20:29:43 GMT
Organization: Royal Institute of Technology, Stockholm, Sweden

In <[e-mail]> [e-mail] (Edwin H. Watkeys III) writes:

>to appeal to potential readers with your subject line? How will you phrase
>your post, so as to get the right people in a responding mood? You want to
>go for the big names on csmp -- it'll make you feel somehow more worthy...

A __BIG__ potential catch here:

Once a week or so, I get a letter starting:

"I see you answer a lot of questions on comp.sys.mac.programmer, so
I figured you could help me with..."

I usually answer (politely I hope) that I post to the UseNet on my
spare time (lunch breaks or such) for the potential benefit of all
UseNet readers, while individual consulting is what I do for a
living; short contracts are in the $100/hour range.

In my eye, a question answered on the net is worth 10x as much as
one answered in a letter, because there are a lot of lurkers who don't
post much, but read for their education. If, however, the posted
question is of a simple nature; I usually just mail a reply saying
that, yes, PBGetCatInfo CAN be your friend, too :-) and then point
towards the FAQ.

Cheers,

                        / h+
--
 -- Jon W{tte, [e-mail], Mac Hacker Deluxe --

  "From now on I will re-label the EQ on the deck as Fizz and Wobble
   instead of HF and LF."

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

>From [e-mail] (Peter N Lewis)
Date: 21 Feb 1994 11:45:58 +0800
Organization: NCRPDA, Curtin University

[e-mail] writes:

>newsgroup!  What exactly are the rules for posting questions on this
>Newsgroup?  Is there a limit to the quantity of questions one person may
>ask or do your questions have to be of a certain complexity?

There are no rules, which just makes it worse really.  Most people wont
single you out because you've posted too many messages, but they may
well not answer because of what's in your message.  Things to try to
remember:

First, try to answer the question yourself, if you don't spend at least
a few hours trying to figure it out first, then you probably wont even
know the right question to ask.

Next, make the subject appropriate.  Subjects like "HELP!!!" are useless,
subjects need to define the problem so that people interested in that
area, or experts in that area will read your post.

Then be consise but complete.  Make it clear what you are trying to do
and why you are failing.  If you have bugs in your code, make sure
you have found exactly which bit is failing, make sure you are checking
all the error return values, and produce a simple, short piece of code
that demonstrates the error.  No one here can (or will) debug your
application for you - a code snippet sure, but not your application.

Offer to take email responses and to summarize them, but also follow
the thread in news.  Don't say "Email responses, I don't read this group".
If you don't read the group, then you aren't contributing, so why should
anyone help you.  If you can't even take the time to read the group for
a few weeks, forget about posting.  And if you don't normally read the
group, then check vbery carefully all the articles already there and
see if your question comes up.  Also, check if it is in the FAQ (usually
available from rtfm.mit.edu:/pub/usenet).  And do summarize the responses
if there are more than a few (or if there are any in Email that aren't
already posted).  Others have the same questions as you do.

And then put what you learnt back - if someone else posts a question
that you've had answered, answer it!

There are only a few reasons that your post wont get any responses:
1) No one knows.
2) Some people know.
  2.1) but they didn't see your post (bad subject)
  2.2) couldn't figure out what your problem was (bad article)
  2.3) you pissed them off with some comment (insulted someone/something,
       perhaps had some random religous comment in your sig, whatever).
  2.4) you sounded like you had no idea what you were talking about
       (this makes it difficult to answer, since the answer must start
       from basic principles and go from there which is too tedious).

Enjoy,
   Peter.
--
_______________________________________________________________________
Peter N Lewis <[e-mail]>       Ph: [phone]

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

>From [e-mail] (Ari I Halberstadt)
Date: Tue, 22 Feb 1994 06:42:22 GMT
Organization: The World Public Access UNIX, Brookline, MA

In article <2k8h7n$[e-mail]>,
Jon Wdtte <[e-mail]> wrote:
>Once a week or so, I get a letter starting:
>
>"I see you answer a lot of questions on comp.sys.mac.programmer, so
>I figured you could help me with..."

Hi Jon, I see you answer a lot of questions on
comp.sys.mac.programmer, so I figured you could help me with this
simple question that's been bugging me for the last 5 minutes. I've
been trying to figure out precisely how, using only BeginUpdate and
EndUpdate and 3,500 other assorted Toolbox calls, to solve the mystery
of Life the Universe and Everything. So far, no obvious solution has
appeared, but perhaps you've already tackled and solved this problem.
Forgive me if this is a really obvious thing which is right there in
the docs, but I couldn't find it anywhere. :-)
--
Ari Halberstadt    [e-mail]     #include <std/disclaimer.h>
"These beetles were long considered to be very rare because very few
entomologists look for beetles in the mountains, in winter, at night,
during snow storms." -- Purves W. K., et al, "Life: The Science of

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

>From [e-mail] (Hla Tun "Tim" Thein)
Subject: Sending Apple Events from Excel
Date: Tue, 22 Feb 1994 18:07:40 GMT
Organization: NETCOM On-line Communication Services (408 241-9760 guest)

Does anyone know how to send Apple Events from MS Excel to an Apple
Event Savvy application without writing any code resource?

I've looked through the Excel manuals and even the Excel SDK book and
it seems to suggest that some code resource needs to be written.  If
that's the case, would anyone have some code resource that they would
like to share?

Thanks,
Tim

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

>From [e-mail] (Robert S. Mah)
Date: Tue, 22 Feb 1994 16:22:57 -0500
Organization: One Step Beyond

[e-mail] (Hla Tun "Tim" Thein) wrote:

> Does anyone know how to send Apple Events from MS Excel to an Apple
> Event Savvy application without writing any code resource?
>
> I've looked through the Excel manuals and even the Excel SDK book and
> it seems to suggest that some code resource needs to be written.  If

Just write your code and make sure to name the resource correctly.  It's
all pretty straitforward -- the only complications come in "linking" the
external to Excel.

The one thing about using custom AE's from externals -- don't ask for a
reply.  The host application probably won't know what to do with it.

An alternative is to simply use DDE.  Excel supports this and DDE calls
_can_ receive replies.  The DDE calls are documented in the Macro reference
(I think).  And since DDE calls simply go out as high level events, your
other application (if it's yours) can respond without any fuss.

Another alternative is to use an external scripting system such as Frontier
or AppleScript.  This is probably the best way, as it will use the OSA
standard and involve less mucking around inside Excel and speed development
time.

However, be careful about using Excel with DDE as the foundation of an
application.  You should be aware that while Excel is in "edit" mode, it
cannot receive or send or receive any AppleEvents or DDE "events".  I was
once on a team that had to rewrite an Excel 4.0 based application because
of this limitation.  At the time, MS said they had no intention of changing
this, and that to a customer that bought almost a million dollars of MS
software every year.  It may be fixed in 4.1 or maybe not -- I haven't
checked.

Cheers,
Rob
________________________________________________________________________
 Robert S. Mah              One Step Beyond              [e-mail]

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

>From Steven Ritter <[e-mail]>
Date: Wed, 23 Feb 1994 10:40:37 -0500
Organization: Psychology, Carnegie Mellon, Pittsburgh, PA

Excerpts from netnews.comp.sys.mac.programmer: 22-Feb-94 Re: Sending
Apple Events fr.. by Robert S. [e-mail]
> [e-mail] (Hla Tun "Tim" Thein) wrote:
>
> > Does anyone know how to send Apple Events from MS Excel to an Apple
> > Event Savvy application without writing any code resource?
> >
> > I've looked through the Excel manuals and even the Excel SDK book and
> > it seems to suggest that some code resource needs to be written.  If
>
> Just write your code and make sure to name the resource correctly.  It's
> all pretty straitforward -- the only complications come in "linking" the
> external to Excel.
>

This isn't much trouble either. If you're not up to writing this
yourself, ISIS sells a product called S7PX that includes macros to send
appleevents from Excel -- [phone]. I wrote a version for myself
that is more limited than the ISIS one but allows other options that I
need. If you tell me more specifically what you want to do, I might be
able to help.

> The one thing about using custom AE's from externals -- don't ask for a
> reply.  The host application probably won't know what to do with it.
>

I don't have any problem sending and receiving replies (you have to wait
for a reply, not queue it).

> However, be careful about using Excel with DDE as the foundation of an
> application.  You should be aware that while Excel is in "edit" mode, it
> cannot receive or send or receive any AppleEvents or DDE "events".
> I was once on a team that had to rewrite an Excel 4.0 based
application > because of this limitation.  At the time, MS said they had
no intention > of changing this, and that to a customer that bought
almost a million
> dollars of MS software every year.  It may be fixed in 4.1 or maybe
not > -- I haven't checked.
>

Yes, this is really a big, big hassle. It's not so bad if you're waiting
for a reply to your appleevent because SendAE will return "appleevent
not handled". But if you're sending an appleevent that doesn't get a
reply (like SetData), there is no way to tell whether Excel is listening
or not. I now confirm SetData with subsequent GetData calls and retry
the setdata if the result it different.

On a related tack, there is a send mode parameter called "Want Receipt",
which is supposed to give the client a "return receipt" if the
appleevent was received. IM is very vague about what this means ("your
application receives a return receipt as a high-level event") and I
haven't been able to figure out how this works (if it does at all). Does
anyone use this? Would this be a way of telling whether Excel accepted a
SetData event?

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

>From [e-mail] (Mike Cohen)
Date: Wed, 23 Feb 1994 19:04:36 GMT
Organization: ISIS International

[e-mail] (Robert S. Mah) writes:

>[e-mail] (Hla Tun "Tim" Thein) wrote:

>> Does anyone know how to send Apple Events from MS Excel to an Apple
>> Event Savvy application without writing any code resource?
>>
>> I've looked through the Excel manuals and even the Excel SDK book and
>> it seems to suggest that some code resource needs to be written.  If

My company (ISIS International) sells a product called System 7 Pack for
Excel which lets you send AppleEvents, run AppleScript or Frontier scripts,
and launch applications from within Excel macros. A demo is available via
anonymous ftp on netcom.com in pub/isis/S7PX.demo.hqx.

Sorry about the commercial post.
--
Mike Cohen - [e-mail]
NewtonMail: MikeC49506 / ALink: D6734 / AOL: MikeC20


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

>From Jon Little <[e-mail]>
Subject: Strange launch erroor
Date: 21 Feb 1994 22:18:21 GMT
Organization: Dept. of Comp. Sci., Victoria Uni. of Wellington, New Zealand.

Hi netters/hackers

While testing a Think Pascal routine which searches the memory for a
reqhired application, iff the application isn't in memory then the
routine tries to launch the application. This is when I get the strange
error: "Chain and  launch not allowed" in the following code - in the
line marked {*****************}


 PROCEDURE LaunchApp (creator: OSType;
       tofront: boolean);
  VAR
   psn: ProcessSerialNumber;
   targetAddress: AEDesc;
   theEvent, theReply: AppleEvent;
   fileList: AEDescList;
   launchDesc: AEDesc;
   app_fs: FSSpec;
   launchThis: LaunchParamBlockRec;
   oe: OSErr;
   gv: longInt;
   sendmode: AESendMode;
 BEGIN
  IF (Gestalt(gestaltOSAttr, gv) = noErr) & (BTST(gv,
gestaltLaunchControl)) THEN BEGIN
    IF GetPSN(creator, psn) THEN BEGIN
      IF tofront THEN BEGIN
       oe := SetFrontProcess(psn);
       END;
     END
    ELSE BEGIN
      IF MyGetAPPL(creator, app_fs) = noErr THEN BEGIN
       oe := AECreateDesc(typeApplSignature, @creator, sizeof(creator),
targetAddress);
       oe := AECreateAppleEvent(kCoreEventClass, kAEOpenApplication,
targetAddress, kAutoGenerateReturnID, kAnyTransactionID, theEvent);
       oe := AEDisposeDesc(targetAddress);
       launchThis.launchAppSpec := @app_fs;
       PrepareToLaunch(theEvent, tofront, launchDesc, launchThis);
       oe := LaunchApplication(@launchThis);   {*****************}
       oe := AEDisposeDesc(theEvent);
       END;
     END;
   END;
 END;

I am unable to find this error documernted anywhere so I would be
grateful if anyone could tell me how to correct this error

Thanks in advance

Jon

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

>From [e-mail] (Rich Siegel)
Date: Tue, 22 Feb 1994 00:17:47 GMT
Organization: Bare Bones Software

In article <2kbbvd$[e-mail]> Jon Little <[e-mail]> writes:
>Hi netters/hackers
>
>While testing a Think Pascal routine which searches the memory for a
>reqhired application, iff the application isn't in memory then the
>routine tries to launch the application. This is when I get the strange
>error: "Chain and  launch not allowed" in the following code - in the
>line marked {*****************}

THINK Pascal will not allow you to launch other applications from a
program running within the THINK Pascal debugger. (If memory serves,
either version 4.0.1 or 4.0.2 relaxes this restriction so that you can
sub-launch applications under MultiFinder or System 7.)

R.

--
Rich Siegel % [e-mail]    % Principal, Bare Bones Software
--> For information about BBEdit, finger [e-mail] <--

"He then proceeded to give a history of the universe, in real time."

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

>From [e-mail] (Grant Neufeld)
Subject: TCL- CDLOGDialog [?]
Date: Mon, 21 Feb 1994 09:08:31 GMT
Organization: Carleton University

I am trying to use a DLOG resource to display a dialog using
Symantec's THINK Class Library (SC++ 6.0.1).

My problem is that I can't get it to do the automatic recognition of
the default "OK" button. The dialog comes up and can be dragged (I've
selected a WDEF with a title bar), but the OK button isn't hilited and
doesn't do anything when clicked.

My class CAboutBox subclasses CDLOGDirector.

/* My initialize procedure */

void CAboutBox::IAboutBox(CApplication *itsSupervisor)
{
    TRY
    {
        inherited::IDLOGDirector(WIND_ABOUT, itsSupervisor);
        gDecorator->CenterWindow(itsWindow);
        TalkToUser(); /* display the dialog */
    }
    CATCH
    {
        Dispose();
    }
    ENDTRY;
}

/* my display procedure */

void CAboutBox::TalkToUser()
{
    long dismissCommand;

// The following line gets an "Error: access to 'SetDefaultButton' denied"
//    (CDialog)itsWindow->CDialog::SetDefaultButton(cmdOK);

    BeginModalDialog();
    /* Only one button ("OK", DITL item 1) in my about dialog, */
    /* so dismissCommand isn't used */
    dismissCommand = DoModalDialog(cmdOK);
}

/* finally, I added the following to my application class's DoCommand */

    case cmdAbout:
        theAboutBox = new(CAboutBox);
        theAboutBox->IAboutBox(this);
        theAboutBox->Dispose();
        FlushEvents(mDownMask+mUpMask+keyDownMask+keyUpMask
                        +autoKeyMask, 0);
        break;
--
Grant Neufeld - [e-mail] - Ottawa, Ontario, Canada
www http://arpp1.carleton.ca/arpp/Grant/Grant_Neufeld.html
Apple Research Partnership Program, 1305 Dunton Tower, [phone]ext3537
My views are too extreme to be anyone else's

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

>From [e-mail] (Esa Ristil{)
Date: 21 Feb 1994 13:20:59 +0200
Organization: University of Tampere, Finland

Did you define command number to the OK button in DITL resource?
You can do that by adding #1 after OK, so the text in the button
should be 'OK#1' (without quotes).  This works exactly the way
menu commands are initialized in the menu resource.  In DITL
though, you can't put command numbers anywhere else than plain
buttons, no radiobuttons checkboxes or other things.

--
                       Esa Ristil{  *  [e-mail]
                             Homo homini lupus

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

>From [e-mail] (Jon Wdtte)
Date: 21 Feb 1994 13:45:38 GMT
Organization: Royal Institute of Technology, Stockholm, Sweden

In <gneufeld.761821711@superior> [e-mail] (Grant Neufeld) writes:

>    /* Only one button ("OK", DITL item 1) in my about dialog, */

You probably forgot to give the button a command number by
naming it OK#100

Cheers,

                    / h+
--
 -- Jon W{tte, [e-mail], Mac Hacker Deluxe --
  _/~|   Yellow
 / * \_  Shark                      (This signature has won the "Worst ASCII
 ~~~~\/  Software                    Logo of the Year" award)

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

>From Steve Bryan <[e-mail]>
Date: Mon, 21 Feb 1994 20:54:58 GMT
Organization: Sexton Software

In article <gneufeld.761821711@superior> Grant Neufeld,
[e-mail] writes:
>My problem is that I can't get it to do the automatic recognition of
>the default "OK" button. The dialog comes up and can be dragged (I've
>selected a WDEF with a title bar), but the OK button isn't hilited and
>doesn't do anything when clicked.

There are at least two things you need to do. First use ResEdit to change
the text of your default button in the dialog to add a command number,
just like you do for menu items (they share the same parser). I believe
there is a defined constant that should be used by convention for dialogs
(found it in Commands.h: #define cmdOK    100L). So if the text of your
default button was "OK" change it to "OK#100". The other thing you need
to do is send the message SetDefaultCmd to your dialog director informing
it that cmdOK is to be your default command. I don't have my projects
online but that's the idea and it has worked for me.

-Steve Bryan

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

>From Steve Bryan <[e-mail]>
Date: Mon, 21 Feb 1994 21:55:05 GMT
Organization: Sexton Software

In article <[e-mail]> Esa Ristil{, [e-mail] writes:
>should be 'OK#1' (without quotes).  This works exactly the way

That may work but I would use "OK#100" which would associate the button
with cmdOK. I believe that cmdQuit is asigned to command number 1. The
other step that might need to be taken is sending the SetDefaultCmd
message to the dialog director so it knows what command to be the
default. It draws the outline and handles the default behavior after that.

-Steve Bryan

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

>From [e-mail] (Chris Taylor)
Subject: Time Manager
Date: 21 Feb 1994 18:07:23 -0400
Organization: The Party Palace

I've benn doing some stuff with the extended Time Manager.  I'm using some
of the sample code from Inside Macintosh VI (chapter 23).  I'm setting it
up to use the InsXTime proc, and my actual procedure uses the A5 world of
my app, as well as reactivating itself with a delay of 1000 miliseconds (1
second).  This seems to conflict with the WaitNextEvent function because
when I install the task, THINK Pascal stops the program, and tell me that
there's a bus error in the WaitNextEvent line, then it locks up.  It works
fine without installing the Time Manager task, and I have no extension
conflicts.  Also, it's not in my task installation procedure or in the task
itself (I've checked them all).

Can someone please help me here?!?!  I need to use the extended time
manager!

Please e-mail me if you can help.

Christian

+---------------------+----------------------------------+
|  Christian Taylor   | Internet: [e-mail] |
|The Party Palace BBS | CIS     : 71442,1161             |
|   [phone]    | AOL     : Chris1020              |
+---------------------+----------------------------------+

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

>From [e-mail] (Martin Minow)
Date: Tue, 22 Feb 1994 19:57:32 GMT
Organization: Macintosh Developer Services

In article <[e-mail]>,
 [e-mail] (Chris Taylor) wrote:
>
> I've benn doing some stuff with the extended Time Manager.  I'm using some
> of the sample code from Inside Macintosh VI (chapter 23).  I'm setting it
> up to use the InsXTime proc, and my actual procedure uses the A5 world of
> my app, as well as reactivating itself with a delay of 1000 miliseconds (1
> second).

Your Time Manager completion routine is called when the time interval
finishes, and some other process or application may be running..
Consequently, before installing the task, you must save a copy of your
application's A5 "somewhere" where the completion routine can access it
(generally, at a fixed offset from the TMTask. The easist way to do this
is to create an extended task structure, such as the following:

    typedef struct {
        TMTask     tmTask;
        long       tmApplicationA5;
    } TMInfoRecord;
    TMInfoRecord   gTMInfoRecord;

         gTMInfoRecord.tmTask.FOO = BAR; /* Setup TMTaskRecord */
         gTMInfoRecord.tmApplicationA5 = SetCurrentA5();
         InsTime((QElemPtr) &gTMInfoRecord.tmTask);
         ...

Your time manager task callback function would look like this:

    pascal void MyTimeManagerTask(void)
    {
         register TMInfoRecord   *tmInfoPtr;
         long                    oldA5;

         asm {
             move.l a1,tmInfoPtr;
         }
         oldA5 = SetA5(tmInfoPtr->tmApplicationA5;
         ... other stuff ...
         (void) SetA5(oldA5);
    }

Note that you cannot call QuickDraw or any function that may move memory
from your time manager task.

Hope this helps.

Martin Minow
[e-mail]
The above does not represent the position of Apple Computer Inc.

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

>From Dan Crevier <[e-mail]>
Subject: Tip for Think C and Radius monitor users
Date: 21 Feb 1994 01:44:40 -0500
Organization: Harvard University

    I've got a radius monitor, and have always been frustrated that
when using Think C, I can't use the option-click in a window's title
feature to get a list of the header files because RadiusWare makes it so
that an option-click in the title zooms the window, and there seemed to be
no way to turn this "feature" off.
    Well, playing around with Resorcerer, I found that there is a
resource type 'RPRM' (with a TMPL resource to match) that has a field
called "no Option Title zoom", and if you change this to on, then you an
use the option-click feature in Think C.
    Just thought I'd mention it in case someone else finds it useful...

Dan


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

>From [e-mail] (Steve Jovanovic)
Subject: Universal Headers (A)
Date: 22 Feb 1994 03:35:54 -0500
Organization: TopSoft, Inc.

Hi Netters,

Recently, someone asked where to find the Universal Headers,
and the universal QuickTime headers.  The generic MPW
Universal headers (CIncludes) are available on the _develop_
#16 CD.  However, they only include the generic headers--
the System 7.1-specific headers are missing (some of the OSL
headers, QT, etc.)  The good news, though, is that you can
find the missing ones on the CodeWarrior DR1 CD!  I was looking
everywhere for the QuickTime uni's too and finally found them
on the DR1 CD!  Oddly, they're not in the 68K folder/s, but
in the PowerPC folder/s.  I can't recall exactly in which
directory you'll find them, but if you use the Finder's
Find... and do a search for "ImageCompression.h" (it'll
take a while :P ) you'll find them in *that* folder :) as
well as the Speech Manager uni header, OSA uni's, etc.

Also, someone else asked about experiences using universal
headers under TC or SymC++.  That's what we did with FilterTop.
I just took CIncludes from _develop_ 16 and trashed Symantec's headers.
For some reason, the Symantec compilers have a problem with
FixMath.h (specifically, 4 lines).  If you comment them out
you'll be fine.  (The problem is caused by a conditional
compilation directive--e-mail me if you need the details).
It took a couple of days of full-time work to convert
several dozen K lines of source to compile and work properly
under the uni headers--there are a bunch of gotchas...but
it's not hard.  The only real guide that I've come across is
the "Making the Leap to PowerPC" article in _develop_ #16.
(It's also available in DocViewer format on the CodeWarrior
CD).

This is a bit of an aside, but if you're starting on a new
project with the uni headers (this is a good rule to follow
in general regardless of which headers you're using)...when
you declare struct's, etc. name your data members starting
with an "f" or something like that...like this:

typedef struct TopSoft
{
    short    fMembers;
    short    fProducts;
    char    *fCharacters;
}
    *TopSoftPtr,
    **TopSoftHdl;

The reason that might be a good idea is becuase I've
come across a bunch of non-obvious name conflicts that
cause MPW C++ to choke (but only w/ uni headers); it's
important to ensure that you use unique names!  If you
get obscure "illegal combination of type specifiers"
errors from MPW C++ (and in those, MPW doesn't give
line #s!) that might be the cause.

Anyway, good luck--I hope this helps.

steve

+----------------------+---------------------------+----------------------+
| Steve Jovanovic      |  President, TopSoft, Inc. |    2918 Goold St.    |
| Marquette University +---------------------------+   Racine, WI 53404   |
| Computer Engineering | Internet:  [e-mail] |    [phone]    |
+----------------------+---------------------------+----------------------+
| "Blue skies above, sun on your arms, strength in your stride, and hope  |
|  in those squeaky clean eyes..." -- "Teenage Wildlife," David Bowie     |
+-------------------------------------------------------------------------+



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

>From [e-mail] (Apple-O)
Subject: c source code for simple-classic-old mac arcade type games wanted
Date: 21 Feb 94 08:16:52 GMT
Organization: Rutgers Univ., New Brunswick, N.J.

I'm interested in learning how to program games on my mac, using C.

Being new to C (and mac) I'm unsure of graphics techniques, functions, etc.
as well as sound.

If anyone has any simple games source code (a plus if well documented), I'd probably be able to learn something from looking at it, & playing around & modifying things.

Does anyone know any C libraries which include graphics or sound functions?

--
Apollo

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

>From [e-mail] (Ken Long)
Date: Mon, 21 Feb 1994 16:33:04 GMT
Organization: NETCOM On-line Communication Services (408 241-9760 guest)

There are a few C game sorces available - from simple to complex.

The Sprite Animation Toolkit has "Space Invaders" type source, as well as
other forms of animation demos.  There's a Pascal source set for
"HeartQuest" in the package, too.

"NewMaxell" has some buggy B/W "game" source, in C.

On ftp oak.oakland.edu, there's some old C source for a "Go" game.  I'm
working on updating this source and will post upon completion.

The "SpriteWorld" source package has some "game" demo source.

The Arashi 1.0 and 1.1 source is on Stanford and Univ. Michigan ftps.
Full color, direct to screen, graphics, animation, sound, sound
compression, etc.

"Cheese Toast" is a Malestrom-type game source package.  FPU required for
the built application.

I ordered and received the source for "MacMan" (the B/W version) and the
source may still be available. ($15)

There's some Tetris-type sources available.

There is a source project called "Hearts" - a card game - available from
the authors.

Here, on Netcom, there are some ANSI game sources. Poker, tetris,
Monopoly, Snake and others.  Not much help for a beginner, since they are
not "Mac-like" or even for Think C, but game source, none the less.

Then, you can write to authors and ask for source.  I've gotten several
sources - some of which I'm not allowed (nor would I) to distribute - from
authors upon request.  They rule of thumb is:  When you see a program you
want the source for, ASK immediately.  I've gotten some publically
available source, even at a price, only to find other's later attemps to
get the same source failed due to authors changing their mind about
releasing it.

One of the best game source projects I got was from an author who had not
released anything since 1985.  No addresses in his program were current.
So, with a little detective work, and some 555-1212 calls, I was able to
contact him.  A Mac to Mac telecom connection transmitted the source.

Also, you may be able to get source for games authors have given up on.
They may no longer have the time or willingness to complete the
development of some project.  You can work deals to complete the
development of such projects.  I was given source on that basis.  But,
upon reviewing the source, I could see that a complete rewrite, using
modern methods, would have been the best course of action.  As such, the
old code only had historical value, or training value.  And, since the
original game was started, several competitive games have already been
released by others, making it an unfeasable shareware proposition.  This
may not always be the case, though.

Ken

[e-mail]

"It's better to have a Mac and not need it, than to need a Mac and not
have it."


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

>From [e-mail] (todd J. Lipcon)
Subject: showtext-pascal
Date: 20 Feb 1994 11:29:14 -0600
Organization: 4th Dimension BBS

Is there a very easy way to save the contents of THINK Pascal's built-in Text
window to a teachtext file?  I really don't want to have to figure out the
entire TE manager!  The program generates a series of prime numbers and right
now it puts them into the text window using writeln().  If I cant do what I
want, does anyone have any simlpe TE routines?
                                                                                                Thanks for the help!
                                        Scott Lipcon
reply to: [e-mail], or [e-mail]

********************************************************************
System: fourd.com                                Phone: [phone]
Cute quote:  Being a computer means never having to say you're sorry
********************************************************************

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

>From [e-mail] (Robert L Mathews)
Date: Mon, 21 Feb 1994 11:32:46 GMT
Organization: Tiger Technologies

todd J. Lipcon ([e-mail]) wrote:
: Is there a very easy way to save the contents of THINK Pascal's built-in Text
: window to a teachtext file?  I really don't want to have to figure out the
: entire TE manager!  The program generates a series of prime numbers and right
: now it puts them into the text window using writeln().  If I cant do what I
: want, does anyone have any simlpe TE routines?

You're under a misapprehension here: TextEdit has nothing to do whatsoever
with TeachText files (not from the point of view of a program that wants
to create them, anyway). TextEdit is what TeachText uses internally to
display your file, but all your file needs to contain is a series of
characters.

This being the case, simply do the following:

o  Create a new file (data fork) on the disk of type 'TEXT', creator 'ttxt'.
   Use a toolbox call such as FSpCreate or HCreate, depending on whether
   the System 7-only FSSpec calls are available.

o  Open the file using FSpOpenDF or HOpen.

o  Use FSWrite to write the text to the file. Do this for each line you
   want in the file, writing a carriage return (or two) wherever you want
   to have carriage returns.

o  Use FSClose when you're finished.

This should get you going - if this is going into a real program, you'll
want to be sure to check for errors and allow the user to name and choose
the location of the file, using StandardPutFile or one of its more complicated
ancestors or cousins.

--
Robert L Mathews, Tiger Technologies

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

>From [e-mail] (Harry A. Reinke)
Date: Mon, 21 Feb 1994 14:50:18 GMT
Organization: Drexel University, Phila. Pa.

My only suggestion is that you use two procedures, one for text output,
and another to a file.  If you need code for how to output to a file I can
help you there as well.

If you want to look into it on your own ... look at
FSWRITE,FSREAD,STANDARDPUTFILE.

Bye.


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

>From bn9606@beorn (Brian Novack)
Date: 21 Feb 94 17:09:24 GMT
Organization: WU-Wien

In article <[e-mail]> from [20 Feb 94 17:29:14 GMT] you wrote:
 |> Is there a very easy way to save the contents of THINK Pascal's built-in Text
 |> window to a teachtext file?  I really don't want to have to figure out the
 |> entire TE manager!  The program generates a series of prime numbers and right
 |> now it puts them into the text window using writeln().  If I cant do what I
 |> want, does anyone have any simlpe TE routines?
 |>                                                                                                 Thanks for the help!
 |>                                         Scott Lipcon
 |> reply to: [e-mail], or [e-mail]
 |>
 |> ********************************************************************
 |> System: fourd.com                                Phone: [phone]
 |> Cute quote:  Being a computer means never having to say you're sorry
 |> ********************************************************************

Under the Run Options, you can tell it to save the text window to a file, and adjust the
buffer size of the window as well (how many characters it will hold).

---Brian Novack


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

>From [e-mail] (Peter N Lewis)
Date: 22 Feb 1994 12:37:59 +0800
Organization: NCRPDA, Curtin University

[e-mail] (todd J. Lipcon) writes:

>Is there a very easy way to save the contents of THINK Pascal's built-in Text
>window to a teachtext file?  I really don't want to have to figure out the

Well, choose Save as... works for me.  I think it requires at least TP 4,
maybe even 4.0.2.  But it works fine, saves the text window to a TeachText
document.
   Peter.
--
_______________________________________________________________________
Peter N Lewis <[e-mail]>       Ph: [phone]

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

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