Celestin Apprentice 5/Apprentice-Release5.iso/Information/THINK C Digest/1992/92-03

Text File | 1995-12-31 | 81KB | 2,422 lines

Path: ucivax!gateway
From: [e-mail] (David Sugar)
Subject: Sound stuff??
Message-ID: <[e-mail]>
Newsgroups: fa.think-c
Lines: 16
Date: 2 Mar 92 04:29:01 GMT


  Umm, I have a quick question here.  I hope this one isn't too hard..
I just want to write a simple little thing to generate 2 tones.  They
have to be 2 different tones.  And I would like to do this so that
it will put one tone on the left channel and one on the right of the
stereo output.  I was going to use the wave Table Synth routines, but I
can't find a way to make it produce tones on just the left or right, I
can only get it to produce a mono tone..
  I only want this to control some audio hardware that is indirectly
hooked to the computer.  So it doesn't have to be great code or anything.

Thanks

Dave Sugar
[e-mail]











Path: ucivax!gateway
From: [e-mail]
Subject: Re: example code to display PICT files
Message-ID: <[e-mail]>
Newsgroups: fa.think-c
Lines: 28
Date: 2 Mar 92 15:27:51 GMT

I put together something to read and display PICT files using Think C 4.0

I'm afraid my Mac isn't connected to Internet, so I'll have to do this from
memory:

1)  I took the Starter project a base.

2)  I made the thing create a color window for each document instead of a
    black and white one. ( I think all windows have color abilities in
    Think 5.0.1 )

3)  I brought in the CPictFile object thoughtfully provided in the Think
    Class Library.

4)  I changed the default file type to 'PICT'.

5)  I can't remember exactly, but I think I made BuildNewWindow (?) do a
    ReadAll on my data file and stick it into a scroll paine in my color
    window.  (maybe someone can clarify this step)

And then I think I could read and view 32 bit color image files.  ( I later
modified this base project to do some color fractals - sort of. )

I hope this helps as a start.

John

[e-mail]










Path: ucivax!gateway
From: [e-mail] (Daniel Sears)
Subject: Quality is Job 1.01
Message-ID: <[e-mail]>
Newsgroups: fa.think-c
Lines: 5
Date: 4 Mar 92 18:19:33 GMT

I recently bought Think Reference.  I think it's a very good product
with a fair number of minor blemishes.  Then I noticed that my disk
was dated May 1991.  Are there any updates available for this product?

--Dan










Path: ucivax!gateway
From: [e-mail] (Scott McAlister)
Subject: Mac CASE tools?
Message-ID: <9203041958.AA01633@apollo101>
Newsgroups: fa.think-c
Lines: 1
Date: 4 Mar 92 19:55:20 GMT

Are there any CASE tools especially for Mac programming?










Path: ucivax!gateway
From: [e-mail] ("R.G. Hawbaker ")
Subject: Updating Think-C 4 to 5
Message-ID: <[e-mail]>
Newsgroups: fa.think-c
Lines: 5
Date: 5 Mar 92 04:17:48 GMT


Hi, I have the Think C 4 version, and would like to upgrade to 5.  What's the
number that I should call about this?

    Bobert










Path: ucivax!gateway
From: [e-mail]
Subject: Please include me.
Message-ID: <[e-mail]>
Newsgroups: fa.think-c
Lines: 3
Date: 5 Mar 92 05:15:22 GMT

Hi, please include me in your subscription list.

Thanks.










Path: ucivax!gateway
From: roscom![e-mail] (Monty Solomon)
Subject: Re: Updating Think-C 4 to 5
Message-ID: <[e-mail]>
Newsgroups: fa.think-c
Lines: 19
Date: 5 Mar 92 07:12:08 GMT

> From: "R.G. Hawbaker " <[e-mail]>
> Subject: Updating Think-C 4 to 5
> Newsgroups: fa.think-c
> Date: 5 Mar 92 04:17:48 GMT
> To: [e-mail]
>

> Hi, I have the Think C 4 version, and would like to upgrade to 5.  What's the
> number that I should call about this?
>

>     Bobert

1-800-228-4122 x811
The upgrade is $89.00 + $8.00 S/H

---
# Monty Solomon / Roscom / PO Box 2486 / Framingham, MA  01701-0405
# roscom![e-mail]










Path: ucivax!gateway
From: [e-mail]
Subject: remove from list
Message-ID: <[e-mail]>
Newsgroups: fa.think-c
Lines: 7
Date: 5 Mar 92 19:34:34 GMT


Hi

Please remove me from this mailing list.

Thank you.











Path: ucivax!gateway
From: [e-mail]
Subject: Unsubscribe
Message-ID: <[e-mail]>
Newsgroups: fa.think-c
Lines: 2
Date: 6 Mar 92 14:37:32 GMT


Please remove me from the list.  Thanx.










Path: ucivax!gateway
From: [e-mail] (egw)
Subject: Corrupted Project Res Map
Message-ID: <[e-mail]>
Newsgroups: fa.think-c
Lines: 24
Date: 6 Mar 92 16:23:54 GMT

Corrupted Project Res Map
I've been having a recurring problem with corrupted Think C project files.
Occasionally, I try to open a project and Think C replies"

"Resource file 'MyProject.#185#' not found"

When I try opening the project file with Resedit or Resorcerer they give me a
-199 mapReadErr (map inconsistent with operation) from the Resource Manager.

This is a huge project (>250 source files) which takes about an hour to compile
on an FX so this is a major pain.  I've rebuilt the project from scratch (Which
made me wish that project file lists could be saved and restored from text
files.) and still have the problem.  I had been blaming the problem on my code
corrupting the heap somehow which then got written into the project.  However,
the last time it happened my code wasn't even in a runnable state so I was only
editing, compiling, and adding a few files in the project.

I'm using Think C 5.0.2 under Tuned-up System 7 in 32bit mode on a IIfx with
20Mb of RAM.  Lots of extensions (NOW utilities, QKeys, etc.).

Eric
[e-mail]












Path: ucivax!gateway
From: [e-mail] (Rick Holzgrafe)
Subject: Re: Corrupted Project Res Ma
Message-ID: <[e-mail]>
Newsgroups: fa.think-c
Lines: 20
Date: 6 Mar 92 23:47:02 GMT

         Reply to:
                    RE>Corrupted Project Res Map
> I've been having a recurring problem with corrupted Think C project files.
> [...]
> the last time it happened my code wasn't even in a runnable state so I was
only
> editing, compiling, and adding a few files in the project.
> Eric
> [e-mail]

Check the resource forks of the source files you were editing. I have found
that a trashed source file resource fork can wreak havoc in a project file.
Until I found my offending source file and deleted its resource fork, I was
replacing my project file several times a day, and swearing like a sailor.

Rick Holzgrafe   [e-mail] (now)
                 [e-mail] (soon)













Path: ucivax!gateway
From: [e-mail] (Jean-Marie Laborde)
Subject: FSSpec and PBGetFInfo
Message-ID: <[e-mail]>
Newsgroups: fa.think-c
Lines: 42
Date: 9 Mar 92 21:07:15 GMT

I am in trouble with FSSpec recommended records for file starting with
system 7.
Indeed when one obtain after a call to StandardGet(Put)File an answer in a
StandardFileReply variable, How to fill the parameters from a bloc to pass
to PBGetFInfo.
Before system 7 it was possible to have something like
   lebloc.fileParam.ioCompletion    =0L;
                    lebloc.fileParam.ioNamePtr       =reponse.name;
                    lebloc.fileParam.ioVRefNum       =reponse.vRefNum;
                    lebloc.fileParam.ioFDirIndex     =0;

Now using a st_reponse of type StandardFileReply
    lebloc.fileParam.ioCompletion   =0L;
                    lebloc.fileParam.ioNamePtr       =st_reponse.sfFile.name;
                    lebloc.fileParam.ioVRefNum       =st_reponse.sfFile.vRefNum;
                    lebloc.fileParam.ioFDirIndex     =0;
do not work.
(very often    'st_reponse.sfFile.vRefNum' is -1 when
       'reponse.vRefNum'  obtained via the older SFPutFile would be some
thing like
 -32413 !)


(At least most of the time PBGetFInfo return fnfErr (File Not Found).
GetVRefNum via a pathrefNum does not help.

It exist a way to go from old-fashioned file description to FSSpec:
        FSMakeFSSpec(vReNum, dirID, filename, &spec),
what about the converse?

Thank you for helping to anybody!
(By the way who knows (I did not remember what is the special semantic
meaning of the value -1 for a vRefNum?)
Thank you again


Jean-Marie Laborde          [e-mail]
Laboratoire de Structures Discretes et de Didactique - IMAG
Universite Joseph Fourier - CNRS          BP 53x 38041 Grenoble cedex
FRANCE
(33) 76 51 46 10   (sec 76 51 46 16)      (fax [phone])











Path: ucivax!gateway
From: [e-mail] (Raymond Degennaro)
Subject: building an application & MacBugs
Message-ID: <[e-mail]>
Newsgroups: fa.think-c
Lines: 16
Date: 9 Mar 92 21:47:49 GMT

Hi y`all

I have an interesting problem. My program runs fine in the debugger, but when
build an application, it crashes with an unimplemented trap error.  I'm using
the TCL (I've used the updater from the archives) and have no idea where the
problem could be comming from.  Any suggestions?  I don't think that needed
methods were stripped out, because it crashes if I build the app with smart
linking, and it crashes if I don't.

I thought I might get a hint as to what the problem is by using MacBugs.  Well,
when I popped into it, I realized that I hadn't the faintest idea how to use
it.  Are there any instructions around?


Ray (DJ) DeGennaro II
[e-mail]










Path: ucivax!gateway
From: [e-mail] ("Edward J. Huff")
Subject: Re: FSSpec and PBGetFInfo
Message-ID: <[e-mail]>
Newsgroups: fa.think-c
Lines: 78
Date: 10 Mar 92 03:28:10 GMT
X-Envelope-to: [e-mail]

>I am in trouble with FSSpec recommended records for file starting with
>system 7.
>Indeed when one obtain after a call to StandardGet(Put)File an answer in a
>StandardFileReply variable, How to fill the parameters from a bloc to pass
>to PBGetFInfo.
>Before system 7 it was possible to have something like
>   lebloc.fileParam.ioCompletion    =0L;
>                    lebloc.fileParam.ioNamePtr       =reponse.name;

This requires that response exists for at least as long as leblock, since
a pointer to response.name is created here.  If response is an automatic
variable of the current procedure and lebloc is static, this creates a
dangling pointer upon exit from the current procedure.

>                    lebloc.fileParam.ioVRefNum       =reponse.vRefNum;
>                    lebloc.fileParam.ioFDirIndex     =0;
>
>Now using a st_reponse of type StandardFileReply
>    lebloc.fileParam.ioCompletion   =0L;
>                    lebloc.fileParam.ioNamePtr       =st_reponse.sfFile.name;
>                    lebloc.fileParam.ioVRefNum       =st_reponse.sfFile.vRefNum;
>                    lebloc.fileParam.ioFDirIndex     =0;
>do not work.
>(very often    'st_reponse.sfFile.vRefNum' is -1 when
>       'reponse.vRefNum'  obtained via the older SFPutFile would be some
>thing like
> -32413 !)
Notice that sfFile (an FSSpec) contains three fields:  vRefNum, parID, and
name.
The old SFReply record (designed prior to HFS) contains only vRefNum and
fName.
The vRefNum field was modified for HFS to return a working directory ID
number,
something like -32413, rather than a real volume reference number, like -1.
Working directory ID numbers are to be avoided.  I believe they cause such
things as the inability to trash a folder from finder prior to quitting an
application which used the old SF package to select a file in that folder.
(After you quit the application, the folder can be deleted).
I would hope that using the new SF package does not create a working
directory and hence, you should be able to trash a folder without first
quitting from the application which last opened a file from that folder via
the new SF package, but I haven't tried it.
>
>
>(At least most of the time PBGetFInfo return fnfErr (File Not Found).
>GetVRefNum via a pathrefNum does not help.
>
>It exist a way to go from old-fashioned file description to FSSpec:
>        FSMakeFSSpec(vReNum, dirID, filename, &spec),
>what about the converse?
You are supposed to change your PBGetFInfo call to a FSpGetFInfo high level
call (no paramater block).

If you must use parameter blocks, then you must copy the parID field from
the FSSpec into lebloc.fileParam.ioDirID and use PBHGetFInfo instead of
PBGetFInfo.  "PGHGetFInfo is identical to PBGetFInfo except that it accepts
a directory ID in ioDirID."  IV-149.  You might also have to change the
declaration of lebloc if it doesn't contain the ioDirID field.  (Sorry,
I am looking at Inside Macintosh, not at the C header files).
>
>Thank you for helping to anybody!
>(By the way who knows (I did not remember what is the special semantic
>meaning of the value -1 for a vRefNum?)
-1 is the actual volume reference number for the first volume, rather
than a working directory reference number.
>Thank you again
>
>
>Jean-Marie Laborde          [e-mail]
>Laboratoire de Structures Discretes et de Didactique - IMAG
>Universite Joseph Fourier - CNRS          BP 53x 38041 Grenoble cedex
>FRANCE
>(33) 76 51 46 10   (sec 76 51 46 16)      (fax [phone])
--
Edward J. Huff   [e-mail]   [phone]
Keck Laboratory for Biomolecular Imaging
NYU Chemistry Deptartment, 31 Washington Place, New York NY 10003











Path: ucivax!gateway
From: [e-mail] ("Edward J. Huff")
Subject: Re: building an application & MacBugs
Message-ID: <[e-mail]>
Newsgroups: fa.think-c
Lines: 18
Date: 10 Mar 92 03:48:56 GMT
X-Envelope-to: [e-mail]

>I thought I might get a hint as to what the problem is by using MacBugs.  Well,
>when I popped into it, I realized that I hadn't the faintest idea how to use
>it.  Are there any instructions around?
>
>
>Ray (DJ) DeGennaro II
>[e-mail]
The "MacsBug Reference and Debugging Guide for MacsBug version 6.2" is
published by Addison-Wesley, ISBN 0-201-56767-9, $22.95, available in
bookstores.
MacsBug version 6.2 should be available on ftp.apple.com, I got a copy from
there recently.  There is a "help" command if you installed MacsBug and its
preferences file properly.
--
Edward J. Huff   [e-mail]   [phone]
Keck Laboratory for Biomolecular Imaging
NYU Chemistry Deptartment, 31 Washington Place, New York NY 10003











Path: ucivax!gateway
From: [e-mail] (Eric Slosser)
Subject: example code to display PICT files
Message-ID: <[e-mail]>
In-Reply-To: "Haseen I. Alam"'s message of 28 Feb 92 03:17:14 GMT <[e-mail]>
Newsgroups: fa.think-c
Lines: 13
Date: 10 Mar 92 08:28:10 GMT

CopyBits is a good way to blit images on the screen.  Having setup a PixMap,
you'll be able to alter colors and bitmapped images fairly easily.

The TIFF format is the same on all computers, except that everyone uses a
different subset of its fields.

Don't try to learn the internals of a Picture.   Rely only on knowing which
QD calls will get stored in a pict, and that a PICT file is a 512 byte
header followed by the contents of the pict handle.

very brief, so feel free to ask for clarification.
eric slosser











Path: ucivax!gateway
From: [e-mail] (Eric Slosser)
Subject: building an application & MacBugs
Message-ID: <[e-mail]>
In-Reply-To: "Edward J. Huff"'s message of 10 Mar 92 03:48:56 GMT <[e-mail]>
Newsgroups: fa.think-c
Lines: 30
Date: 10 Mar 92 08:38:51 GMT


If you dropped into MacsBug because of an unimplement trap, type "ip".
A disassembly of your current location will result, with the
offending trap marked with a '*'.

If it's something like "TRAP 1F", then check your source to see what
the compiler thought it was trying to do.  If it's got a name like
"RGBForeColor", check to be sure that trap is implemented on your platform.

If the whole listing is filled with instructions like "ORI.B #0,D0",
then you've been naughty and written data on top of your code, or
moved a CODE segment that was referenced as a return address somewhere,
or messed up your stack and return addresses, and have jumped into
"the tall grass".

If none of the above, each of the following, following each with a <return>.
log unimptrapcrash
ip
td
sc
hc
ht
log

Then look for the file named "unimptrapcrash" on your disk, and send
it to some one you don't mind bothering, like anyone who replied to
your original post.

Good luck,
eric slosser










Path: ucivax!gateway
From: [e-mail]
Subject: Re: building an application & MacBugs
Message-ID: <[e-mail]>
In-reply-to: Your message of 10 Mar 92 03:48:56 +0000.
             <[e-mail]>
Newsgroups: fa.think-c
Lines: 6
Date: 10 Mar 92 14:47:33 GMT

And speaking of MacsBugs, does anyone out there know if it is
possible to get it to work (load at startup) on a IIcx with
an Emachines Z21 IQ monitor?

Ian Dowell
[e-mail]










Path: ucivax!gateway
From: [e-mail] (egw)
Subject: Project Corruptions (Solved
Message-ID: <[e-mail]>
Newsgroups: fa.think-c
Lines: 20
Date: 11 Mar 92 16:27:00 GMT

Project Corruptions (Solved)
Last week I posted a message to this group regarding the periodic corruption of
Think C project files.

Symptom: a project file which Think C refused to open (giving various error
messages, "Resource file 'MyProject.#185#' not found", "memory full", etc.).

Disease: the project file was approaching/exceeding 16Mb limit on the size of a
resource fork.

Cure: by precompiling the TCL headers along with the usual MacHeaders stuff we
were able to reduce the project to a "svelte" 8Mb.  Since doing this we've had
no further corruptions.

Thanks to Rick Holzgrafe and Rich Siegel for their help.

Eric
[e-mail]












Path: ucivax!gateway
From: [e-mail] (Jean-Marie Laborde)
Subject: FSSpec and PBGetFInf (solved)
Message-ID: <[e-mail]>
Newsgroups: fa.think-c
Lines: 45
Date: 11 Mar 92 18:37:52 GMT

Thanks to Edward J. Huff but also especially to Francois Menneteau,
I was able to solved the problem of getting old-fashioned file description
from a new FSSpec description and achieved some sort of compatibility
beetween code writen before system 7 and after system 7.
Indeed it suffices to open for a while a working directory that return what
is needed by many old (before system 7) routines:

PROBLEM: to fill fields of a "reponse" variable of type SFReply
gieven a variable "st_reponse" of type StandardFileReply.

SOLUTION:
{
                     WDPBRec leBloc;
                     ParamBlockRec lebloc;
    OSErr    coderreur;

                     leBloc.ioCompletion=0L;
                     leBloc.ioNamePtr        =reponse.fName;   /* here just as a place-holder */

                     leBloc.ioVRefNum        =st_reponse.sfFile.vRefNum;
                     leBloc.ioWDProcID        ='XXXX';        /* the signature of the
application*/
                     leBloc.ioWDDirID        =st_reponse.sfFile.parID;

                     coderreur=PBOpenWD(&leBloc,FALSE);
                     reponse.vRefNum=leBloc.ioVRefNum;
/* reponse.vRefNum are now correct */

                     PtoCstr(st_reponse.sfFile.name);
                     strcpy((char *)reponse.fName,(char *)st_reponse.sfFile.name);
                     CtoPstr(st_reponse.sfFile.name);
                     CtoPstr(reponse.fName);
/* reponse.fName is now correct */
}

I hope that people encountering the problem will understand.
Thank you.


Jean-Marie Laborde          [e-mail]
Laboratoire de Structures Discretes et de Didactique - IMAG
Universite Joseph Fourier - CNRS          BP 53x 38041 Grenoble cedex
FRANCE
(33) 76 51 46 10   (sec 76 51 46 16)      (fax [phone])











Path: ucivax!gateway
From: [e-mail] ("Hisham A. Abboud")
Subject: Sound Manager question
Message-ID: <[e-mail]>
Newsgroups: fa.think-c
Lines: 21
Date: 11 Mar 92 19:33:13 GMT


Dear netters,

I am having a problem using the Sound Manager.  I read a MacRecorder
file, make a 'snd ' resource out of it, create a sound channel, and
try to play it async.  It refuses to play async under system 6.07
(no sound at all), though it works fine under system 7.  Under 6.05,
it refuses to play -sync-!  It always plays async!!

I have the feeling the problem is probably in the way I am creating
my 'snd ' resource (I tried both format 1 and 2).  Does any have a sample
source on how this is done?

Any hints would be appreciated.

                        Hisham.


Hisham A. Abboud, Cedrus Corp.  [Internet: [e-mail]]












Path: ucivax!gateway
From: [e-mail] (John J Chew)
Subject: TC5 disassembler gripes
Message-ID: <[e-mail]>
Newsgroups: fa.think-c
Lines: 68
Date: 11 Mar 92 19:57:16 GMT

Finally got my TC5 upgrade this week and eagerly started converting a
work in progress.  Got through two files of new and improved pickiness
about argument types before I came to a more interesting problem.

I need to store a handle in a code resource.  (Pause for stunned silence.)
It's for an 'atlk' and I don't see any place else to put it.  I put it in
my main() right after the initial two BRA.S's, and referred to it using a
'LEA 4(main), A0', since that looks nice even if the syntax is a bit loose,
and it did work under TC4.

Not under TC5.  'Syntax Error' it helpfully says.  I try 'LEA 4+main, A0'.
This generates the interesting disassembled result 'LEA $0004(A4),A0'.
From which I deduce that... um... everything would be hunky-dory if I
had the address of main in A4, which I don't.

Read the fine manual.  It says (p. 264) the DC.W directive can accept a
function name ... and generate the PC-relative offset of the function ...
as a word.  Okay.  I try 'dc.w main'.  Disassembled, this gives me a zero
word.  Looks like 'dc.w anyotherfunction' does the same thing.  This got
me suspicious, and I start thinking more clearly and realize that there
is no way the disassembler could give me a reasonable value here, since
the function that I referenced could well be in another as yet uncompiled
source file.

Back to the fine manual.  Read the note on page 168 "The Disassemble
command shows all references to global variables and subroutine calls as
being at 0x0000(A5).  This value is merely a place holder for the
disassembler.  When THINK C compiles your program it uses the correct
offset." With a bit of imagination, I see how a confused disassembler
might decide that if globals to be shown as $0000(A5) then 4 + a global
should be $0004(A5) and 4 + a global in a code resource should be
$0004(A4).  Hmm.

In fact, the compiler does generate correct code for both the above
examples.

My gripe is therefore directed at the person(s) responsible for the
disassembler, who IMHO has left a few ragged edges.  At the very least
a comment in the listing indicating what has happened seems appropriate.
And I suppose as long as I am listing wishes for a revised disassembler,
I don't see why globals can't be represented symbolically, and traps too
for that matter.  I would also like to be able to click on a source file
in the project window and say Disassemble, just as I can with Compile.
Storage allocated inline by DC directives should be shown as such and should
certainly not interfere with subsequent disassembly.  It would be very
nice if some semblance of C could be shown in the listing in another column,
though I don't know how feasible that would be when optimisation is
enabled.  Maybe a preprocessor directive to insert comments in the
disassembled output.  Every line should be shown in hex as well as
disassembled form.

I suppose what I really want is the functionality of ResEdit Code Viewer
that I've been using so happily up until now, and that I will no doubt
continue to use for the time being, if only because its magic little
arrows have spoiled me.

Thank you for your attention, and thank you Symantec for what is really
overall an excellent product.

John

P.S. Is there anyone out there who has worked on writing an 'atlk' before?
--
John J. Chew, III                            * [e-mail]
AIS/Berger-Levrault * phone [phone]  * [e-mail]
18, rue des Glacis  * fax   [phone]  * [e-mail]
F-54017 Nancy Cedex * telex BL NANCY 960 212 * [e-mail]
"Les arbres sont souvent en cause dans les accidents de la circulation."










Path: ucivax!gateway
From: [e-mail] (Raymond Degennaro)
Subject: building an application (revisited)
Message-ID: <[e-mail]>
Newsgroups: fa.think-c
Lines: 8
Date: 11 Mar 92 22:55:33 GMT

Here's what I figured-out 'bout my app running fine in the debugger and crashing w/ an unimplemented trap error (appearently) from the finder.  I installed MacBugs, and tried again.  When it dropped into the debugger, the cause was a user break; I left the oopsDebug library in the application.  Any way, after poking around, I found that "new_by_name" was being called by CDLOGDirector::AddOverloadedItem, for an unknown class.

So, what it appears is that the Linker, when building an application, strips out "stuff" even when smart linking is off.  Now I just have to find the "stuff" that's been stripped out, and force a reference (or something, time to RTM) to the "stuff" so it isn't stripped.

Thanks for the help.  I'll post what I find.

Ray DeGennaro
[e-mail]










Path: ucivax!gateway
From: [e-mail] (David the Gris)
Subject: Unsubscribe
Message-ID: <[e-mail]>
Newsgroups: fa.think-c
Lines: 3
Date: 12 Mar 92 01:52:01 GMT

Please remove me from the list.  Thanx.
gris
[e-mail]










Path: ucivax!gateway
From: [e-mail] (John Miller)
Subject: Re:  Sound Manager Question
Message-ID: <[e-mail]>
In-Reply-To: <[e-mail]>; from "Mail Delivery Subsystem" at Mar 12, 92 9:17 am
X-Mailer: ELM [version 2.3 PL11]
Newsgroups: fa.think-c
Lines: 48
Date: 12 Mar 92 15:23:22 GMT

 > Dear netters,
 >
 > I am having a problem using the Sound Manager.  I read a MacRecorder
 > file, make a 'snd ' resource out of it, create a sound channel, and
 > try to play it async.  It refuses to play async under system 6.07
 > (no sound at all), though it works fine under system 7.  Under 6.05,
 > it refuses to play -sync-!  It always plays async!!
 >
 > I have the feeling the problem is probably in the way I am creating
 > my 'snd ' resource (I tried both format 1 and 2).  Does any have a sample
 > source on how this is done?
 >
 > Any hints would be appreciated.
 >
 >                         Hisham.
 >
 >
 > Hisham A. Abboud, Cedrus Corp.  [Internet: [e-mail]]
 >
 >


 The problem is in the documentation for the sound manager.  In Inside Mac
 VI, after going into great (and useful) detail about the different
 structures of both Format 1 and Format 2 sound headers, one expects that
 either of these is somehow accessible to the sampled synth.

 Only later (page 22-44) do you find the description of the "structure of
 the standard sampled sopund header used by the sampled sound
 synthesizer..."

 What follows is an abbreviated version of the other types.  My solution
 was to create the sound file (in MacRecorder or anything else), save it as
 a resource, and then edit it in ResEdit to conform to the header described
 on 22-44.  This involves merely removing the first (I don't remember how
 many) bytes.  Everything else then works fine.

 Good luck.

 John
_____________________________
 John Miller
 Music Theory
 Eastern Illinois University














Path: ucivax!gateway
From: [e-mail] (Rex Bontrager)
Subject: Mac X-Ref book
Message-ID: <[e-mail]>
Newsgroups: fa.think-c
Lines: 10
Date: 12 Mar 92 21:52:52 GMT

I visited my local computer bookstore today and noticed that their X-Ref
book from Apple does not yet include Inside Mac Vol VI.  Can anyone
tell me if or when IM VI will be included in the X-Ref?


Rex Bontrager
Purdue University Computing Center
Bitnet:    rexb@purccvm
Internet:  [e-mail]
Phone:     [phone] ext. 256










Path: ucivax!gateway
From: [e-mail] (John J Chew)
Subject: SystemSevenOrLater?
Message-ID: <[e-mail]>
Newsgroups: fa.think-c
Lines: 26
Date: 13 Mar 92 02:52:29 GMT

Can someone tell me what the correct way to get FindFolder() defined is
when using TC5?  I found two definitions for it in <Folders.h>, presumably
one for glue and one calling the the trap (but why the MOVEQ #$0, D0?),
one of which is chosen depending on the value of SystemSevenOrLater.

In <Types.h>, this macro is set to 0 if it hasn't already been defined,
which makes me think that I should put a #define SystemSevenOrLater 1 in
my Prefix.  Can't see how, but would this break anything?

For the moment, I say:

  #undef SystemSevenOrLater
  #define SystemSevenOrLater 1
  #include <Folders.h>
  #undef SystemSevenOrLater
  #define SystemSevenOrLater 0

and feel silly doing so.

John
--
John J. Chew, III                            * [e-mail]
AIS/Berger-Levrault * phone [phone]  * [e-mail]
18, rue des Glacis  * fax   [phone]  * [e-mail]
F-54017 Nancy Cedex * telex BL NANCY 960 212 * [e-mail]
"Les arbres sont souvent en cause dans les accidents de la circulation."










Path: ucivax!gateway
From: [e-mail]
Subject: Re: Mac X-Ref book
Message-ID: <[e-mail]>
Newsgroups: fa.think-c
Lines: 11
Date: 13 Mar 92 16:09:13 GMT

> I visited my local computer bookstore today and noticed that their X-Ref
> book from Apple does not yet include Inside Mac Vol VI.  Can anyone
> tell me if or when IM VI will be included in the X-Ref?
>

Just last week, I purchased the updated X-Ref from a book store in the
Dallas area.  Inside Mac VI WAS included in the X-Ref.

Richard Beecher
Anchor Beech Software











Path: ucivax!gateway
From: [e-mail]
Subject: Problem with FSpExchangeFiles
Message-ID: <[e-mail]>
Newsgroups: fa.think-c
Lines: 71
Date: 13 Mar 92 17:15:26 GMT

I have implemented a "safe save" strategy in my code as suggested in IM VI
using the FSpExchangeFiles function.  Works like a charm, except that the
temporary file that is created in the process doesn't get deleted.
FSpDelete() returns a file busy error (-47).  The end result is that the
temporary file ends up in the Trash after each restart/reboot.

Incidentally, I can delete the file BEFORE calling FSpExchangeFiles, which
makes me think that FSpExchangeFiles opens the data and/or resource forks,
but does not close them.  Am I right?  If so, what can I do to fix the
problem?  Any help would be much appreciated.

My code is given below (some portions removed for brevity).

Richard Beecher
([e-mail])



/****************** SaveSettings *****************/

void SaveSettings()
{
    short              foundVRefNum;
    long               foundDirID;
    OSErr              err;
    FSSpec             tempFile;
    short              tempRefNum;
    Str255             tempFileName = { "\pFP Scrap" };


    /* Create a temporary file for the safe save */

    err = FindFolder( kOnSystemDisk, kTemporaryFolderType, kCreateFolder,
                   &foundVRefNum, &foundDirID );
    if ( err )
    ErrorString( "\pCould not find/create the Temporary Items Folder.",
stopError );
    FSMakeFSSpec( foundVRefNum, foundDirID, tempFileName, &tempFile );

    FSpCreateResFile( &tempFile, 'trsh', 'trsh', smSystemScript );
    err = ResError();
    if ( err )
    ErrorString( "\pCouldn't create the temporary save file!", stopError );

tempRefNum = FSpOpenResFile( &tempFile, fsRdWrPerm );
    err = ResError();
    if ( err != noErr )
   ErrorString( "\pCouldn't open the temporary save file!", stopError );


    /* Add the resources to the temporary file */

    UseResFile( tempRefNum );
    SaveResources();
    CloseResFile( tempRefNum );


    /* Now, exchange the files */

/* err = FSpDelete( &tempFile );  <---- works fine here */

    err = FSpExchangeFiles( &tempFile, &settingsFile );  /* settingsFile
already exists */
    if ( err )
    ErrorString( "\pCouldn't finish the save operation!", stopError );

    err = FSpDelete( &tempFile );  /* returns a file busy error -47 */
    if ( err )
    ErrorString( "\pCouldn't delete the temporary file!", stopError );
}











Path: ucivax!gateway
From: [e-mail] (Thomas Edward Van Lenten)
Subject: Gray Text
Message-ID: <wdkFXQm00WC=[e-mail]>
Newsgroups: fa.think-c
Lines: 12
Date: 13 Mar 92 21:05:25 GMT

    Can anyone tell me an easy way to draw grey text?  I need to be able
to dim some text in the same manner that a button dims when not allowed.
 I see no easy way to do this, so I'm open to any ideas.

TVL
=========================================================
 Tom Van Lenten                               MacMail II
 [e-mail]                            Special Projects
 [e-mail]          Carnegie Mellon University
 bitnet%"tv0c@andrew"                     Pittsburgh, PA
=========================================================











Path: ucivax!gateway
From: [e-mail] (John J Chew)
Subject: PC-relative addresses
Message-ID: <[e-mail]>
Newsgroups: fa.think-c
Lines: 12
Date: 13 Mar 92 21:30:39 GMT

Is there a good reason why THINK C uses A4-relative addressing to get at
globals in a single-segment code resource?  Why can't it just use PC-relative
addressing as I think it does for string literals, and not require the
manipulation of the A4 register?

John
--
John J. Chew, III                            * [e-mail]
AIS/Berger-Levrault * phone [phone]  * [e-mail]
18, rue des Glacis  * fax   [phone]  * [e-mail]
F-54017 Nancy Cedex * telex BL NANCY 960 212 * [e-mail]
"Les arbres sont souvent en cause dans les accidents de la circulation."










Path: ucivax!gateway
From: [e-mail] (Rich Siegel)
Subject: Re:  PC-relative addresses
Message-ID: <[e-mail]>
Newsgroups: fa.think-c
Lines: 4
Date: 14 Mar 92 03:40:44 GMT

Among other things, globals are not read-only. It's easy to do a
MOVE xxx, y(a4) than it is to do a LEA z(pc), An, MOVE xxx, (An).

R.










Path: ucivax!gateway
From: [e-mail] (Pete Resnick)
Subject: Re: SystemSevenOrLater?
Message-ID: <[e-mail]>
X-Mailer: Eudora [version 1.3b28 MacTCP]
Newsgroups: fa.think-c
Lines: 16
Date: 14 Mar 92 20:24:31 GMT

>In <Types.h>, this macro is set to 0 if it hasn't already been defined,
>which makes me think that I should put a #define SystemSevenOrLater 1 in
>my Prefix.  Can't see how, but would this break anything?

That should be fine unless, as most of us, you use MacHeaders, in which
case you have to rebuild MacHeaders to get it to stick. Then again, if you
are including Folders.h in your source file, then adding it to your prefix
should work fine.

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]











Path: ucivax!gateway
From: [e-mail] (Mark Nagel)
Subject: ARCHIVE: Project Printer FKEY
Message-ID: <[e-mail]>
Newsgroups: fa.think-c
Reply-To: [e-mail]
Lines: 28
Date: 14 Mar 92 23:56:50 GMT


/*****************************************************************************/

From: Phil Shapiro <[e-mail]>
Date: Tue, 10 Mar 92 22:50:28 -0500
Subject: Project Printer FKEY 1.0 for THINK C users

Here's a FKEY for THINK C programmers that produces a list of the
files in C project, including segmentation information. This is a
no-frills hack, but it can come in handy when you need to rebuild a
large project from scratch.

To use it, you first need to install the FKEY in your System file or
in THINK C itself. Once installed, just invoke the FKEY after opening
the project you want to list. The file names, object sizes, and
segmentation information is pasted into a new THINK C edit window.

For more info, see the enclosed ReadMe.

    -phil
----
   Phil Shapiro                                   Software Engineer
   Language Products Group                     Symantec Corporation
           Internet: [e-mail]

/*****************************************************************************/

[saved as: /mac/think-c/symantec/proj-printer-fkey.hqx; 5K]










Path: ucivax!gateway
From: [e-mail] (Mark Nagel)
Subject: ARCHIVE: CTreeView
Message-ID: <[e-mail]>
Newsgroups: fa.think-c
Reply-To: [e-mail]
Lines: 16
Date: 15 Mar 92 00:07:40 GMT


/*****************************************************************************/

From: david wagner <[e-mail]>
Date: Sat, 7 Mar 92 10:09:45 -0500

     I have placed in the /incoming directory a file called
ctreeview.hqx.  This contains a simple class I wrote for people (like me)
who are beginners at programming with TCL, and may have difficulty
visualizing the visual hierarchy and the chain of command.  This class
places a window on the screen which will display these things.
                                                           David Wagner

/*****************************************************************************/

[saved as: /mac/think-c/classes/ctreeview.hqx; 20K]










Path: ucivax!gateway
From: [e-mail] (Francois Menneteau)
Subject: TCL ETS patch question
Message-ID: <[e-mail]>
Newsgroups: fa.think-c
Organization: IMAG Institute, University of Grenoble, France
Lines: 32
Date: 17 Mar 92 11:59:21 GMT


Here is the patch use in the TCL libraries for ExitToShell():

     static long origETS, origLoadSeg;

     static pascal void ETS_Patch( void)
     {
         long realETS;
         long oldA5 = SetCurrentA5();

         if (gApplication)
             gApplication->RemovePatches();
         realETS = origETS;

         SetA5( oldA5);
        CallPascal( realETS);
     }

My question is how can this patch work when it is an other application that
call ExitToShell? (SetCurrentA5() puts CurrentA5 low memory into A5, but
this A5 can be different from my application. So how it is possible to test
for the existence of the gApplication global variable?).





--
Francois Menneteau ()   __|||||__   () "... I had their lives in my hands
================== ()    /O   O\    () their fate their fortune in my visions
[e-mail]       ()    - .|. -    () No one believed in my true prophecy
================== ()     \=^=/     () And now it's too late."  (Iron Maiden)










Path: ucivax!gateway
From: [e-mail] (Ephraim Vishniac)
Subject: Re: TCL ETS patch question
Message-ID: <[e-mail]>
In-Reply-To: Your message of "17 Mar 92 11:59:21 GMT."
             <[e-mail]>
Newsgroups: fa.think-c
Lines: 32
Date: 17 Mar 92 14:06:59 GMT


   From: Francois Menneteau <[e-mail]>
   Date: 17 Mar 92 11:59:21 GMT

   Here is the patch use in the TCL libraries for ExitToShell():

     static long origETS, origLoadSeg;

     static pascal void ETS_Patch( void)
     {
         long realETS;
         long oldA5 = SetCurrentA5();

         if (gApplication)
             gApplication->RemovePatches();
         realETS = origETS;

         SetA5( oldA5);
        CallPascal( realETS);
     }

   My question is how can this patch work when it is an other application
   that call ExitToShell? (SetCurrentA5() puts CurrentA5 low memory into
   A5, but this A5 can be different from my application. So how it is
   possible to test for the existence of the gApplication global
   variable?).

Under Multifinder, an application's patches are part of its
environment. They get swapped out when another application runs. So,
no other application will execute your application's patches. Under
Unifinder, there is no other application. So, you're safe either way.











Path: ucivax!gateway
From: [e-mail] (Phil Shapiro)
Subject: TCL ETS patch question
Message-ID: <[e-mail]>
In-Reply-To: Francois Menneteau's message of 17 Mar 92 11:59:21 GMT <[e-mail]>
Newsgroups: fa.think-c
Lines: 25
Date: 17 Mar 92 14:13:11 GMT

>>>>> On 17 Mar 92 11:59:21 GMT, Francois Menneteau <[e-mail]> said:

 > Here is the patch use in the TCL libraries for ExitToShell():

    [ code for ETS_Patch() removed ]

 > My question is how can this patch work when it is an other
 > application that call ExitToShell? (SetCurrentA5() puts CurrentA5
 > low memory into A5, but this A5 can be different from my
 > application. So how it is possible to test for the existence of the
 > gApplication global variable?).

When more than one application is running (under MultiFinder or System
7), each application has its own table of patched traps. If a TCL
application patches ExitToShell, then that patch is only "visible" on
that application layer. This is accomplished by a patch that's
installed at boot time by the System which collects calls to
SetTrapAddress. These patches are swapped out during any context
switch, minor or major.

    -phil
----
   Phil Shapiro                                   Software Engineer
   Language Products Group                     Symantec Corporation
           Internet: [e-mail]










Path: ucivax!gateway
From: [e-mail] (Igor Livshits)
Subject: Bus errors
Message-ID: <[e-mail]>
Newsgroups: fa.think-c
Lines: 63
Date: 17 Mar 92 15:28:11 GMT

Good morning,

I am writing an object that handles folder and file information.  To test
for errors from Toolbox routines, I call FailOSErr(error).  If error is
anything but noErr, a bus error is triggered as soon as FailOSErr returns
to the main program.

I use far code and data.  I have also recompiled oopsdebug with the far
options on.  Another symptom is that NO error alert is presented when error
!= noErr.

I am still tinkering with the libraries and my code, and I think the
problem is with the debugger calls.  If someone has a clue about this,
please drop me a note.

example:
/*===---------------===*/

long    CIsleDataFile::CreateFolder(Str63 folderName)
Begin
    long    createdDirID;                        //    Directory ID of the folder we will create
    OSErr    error;                                    //    Possible error condition

    if (error= DirCreate(volNum, dirID, folderName, &createdDirID))
    {                                                            //    Oops!
        if (error == dupFNErr)
        {                                                        //    It already exists, what is its directory ID?
            CInfoPBRec    specs;                //    The duplicate file specifications

            specs.dirInfo.ioCompletion= Null;
            specs.dirInfo.ioNamePtr=        folderName;
            specs.dirInfo.ioVRefNum=        volNum;
            specs.dirInfo.ioFDirIndex=    kDefault;
            specs.dirInfo.ioDrDirID=        dirID;

            FailOSErr (PBGetCatInfo(&specs, False) );
            if (specs.dirInfo.ioFlAttrib & kDirTest)
                createdDirID= specs.dirInfo.ioDrDirID;
            else
                FailOSErr(error);                //    The duplicate is a file, not a directory!
        }
        else
            FailOSErr(error);
    }

    return createdDirID;
End

/*===---------------===*/



Many thanks, Igor

<=====-----------------------------------------=====>

 Igor S. Livshits               Education Specialist
 NCSA                                 e: [e-mail]
 University of Illinois         w: [phone]
 Champaign-Urbana campus        h: [phone]

<=====-----------------------------------------=====>











Path: ucivax!gateway
From: [e-mail] (Ian McEachern VE3PFH)
Subject: Debugger Bombs with large arrays
Message-ID: <[e-mail]>
X-Mailer: ELM [version 2.3 PL11]
Newsgroups: fa.think-c
Reply-To: [e-mail]
Lines: 26
Date: 17 Mar 92 22:01:22 GMT

I'm having a problem getting a program developed on a previous version
of think-c running on version 5.0.2.

The debugger (and the program when compiled) bombs when this routine is called:

void foo(long *ptr)

    int   big[8][1500];
    long  bigtoo[1500];


etc.

This did work on the older compilers and now crashes hard every time the
routine is called.  I've tried Far DATA and CODE with the same results ...
tried recompiling all the libraries too, with no luck.

When I reduce the dimesions of the array to 150 it runs (seemingly) normal.

Does anyone have an explanation, solution to my problem?

Thanks

Ian McEachern












Path: ucivax!gateway
From: [e-mail] (Rich Siegel)
Subject: Re:  Debugger Bombs with large arrays
Message-ID: <[e-mail]>
Newsgroups: fa.think-c
Lines: 5
Date: 18 Mar 92 02:22:57 GMT

You're overrunning your stack, plain and simple - you should either use
a global or a dynamically allocated array, or use SetApplLimit to make
more room for the stack.

R.










Path: ucivax!gateway
From: [e-mail] ("Bob (Mr. Grape loves ME!) Miller" )
Subject: Re: Debugger Bombs with large arrays
Message-ID: <9203172349.AA26358@Mr_Grape.convex.com>
In-Reply-To: Your message of 17 Mar 92 22:01:22 +0000.
             <[e-mail]>
Newsgroups: fa.think-c
X-Quote-Of-The: Patriotism is the veneration of real estate above principles.
                                        -- George Jean Nathan
Lines: 15
Date: 18 Mar 92 17:32:05 GMT

> The debugger (and the program when compiled) bombs when this routine
> is called:
>
> void foo(long *ptr)
>
>     int   big[8][1500];
>     long  bigtoo[1500];

That brings up an interesting question.  What is the default stack
size on the Mac, and how can you vary it?  (This is probably a FAQ;
sorry about that.)


                    K<bob>











Path: ucivax!gateway
From: [e-mail]
Subject: thePort (?)
Message-ID: <[e-mail]>
Newsgroups: fa.think-c
Lines: 11
Date: 19 Mar 92 21:00:55 GMT

I apologize for asking this question - I should
remember the answer, but I do not.  When I compile
a program, I get the message "thePort not declared"
(in the line InitGraf(&thePort); )

What am I missing?

If this is too simple a ? for the list, please say so.
Pls reply to me.
Peter Gerhardstein ([e-mail])











Path: ucivax!gateway
From: [e-mail]
Subject: Mailing list for THINK-C
Message-ID: <[e-mail]>
Newsgroups: fa.think-c
Prev-Resent-From: Mark Nagel <[e-mail]>
Prev-Resent-Message-Id: <[e-mail]>
Lines: 11
Prev-Resent-To: cigna!nagel
Date: 19 Mar 92 21:57:32 GMT
Prev-Resent-Date: Thu, 19 Mar 92 11:25:53 PST

Hello,

        I thought there existed a list server for stuff related to THINK-C
and I wanted to subscribe for the same. If a list server does exist, then
please add my name to the mailing list. I am a beginer with regard to
programming on a Mac and I have just started using THINK-C.

                -- Vijaya Sankar

 [e-mail]   (the new INET address)











Path: ucivax!gateway
From: [e-mail] (Craig Miller)
Subject: Re:  Debugger Bombs with large arrays
Message-ID: <[e-mail]>
Newsgroups: fa.think-c
Lines: 50
Date: 19 Mar 92 23:25:14 GMT

<[e-mail]> writes:

> The debugger (and the program when compiled) bombs when this routine
> is called:
>
> void foo(long *ptr)
>
>       int   big[8][1500];
>       long  bigtoo[1500];

<[e-mail]> replies:

> You're overrunning your stack, plain and simple - you should either use
> a global or a dynamically allocated array, or use SetApplLimit to make
> more room for the stack.

<[e-mail]> comments:

> That brings up an interesting question.  What is the default stack
> size on the Mac, and how can you vary it?

Is it really true?  Think C has no automatic runtime check for a
stack overflow? I've been programming professionally with Think C
for two years, assuming that any stack overflow problems would
generate some kind of runtime error message, just like all the other
systems I've programmed in.

A quick check in the Think C User's Manual (for v.4) reveals only
one mention of stack size and/or overflow:  p.450 has a title "Data
segment too big/Stack frame too big" and explains how to dynamically
allocate memory instead of using automatic memory from the stack. I
see no mention of stack size,  changing it or checks for overflow.
I.M. vol.II, chap. 1 (The Memory Manager) says that the initial
application stack size is 8k and can be increased by reducing the
application heap size because they share the same block of memory.
It also mentions the "stack sniffer" which checks for stack overflow
after the fact ONLY IF IT HAPPENS DURING A VERTICAL RETRACE PERIOD
(and generates a system error alert). I.M IV, p.78 documents the
StackSpace function which returns the currently available stack
space.

If this is correct, why doesn't Think C, or even the debugger use the
StackSpace function function for a REAL stack overflow check?
Please verify my statements or certify me a raving reactionary
lunatic.

Craig D. Miller, Computer Programmer       \  [e-mail]
Scientific Reasoning Research Institute     \      tel. [phone]
SRRI, Hasbrouck Lab, Univ. of Massachusetts  \
Amherst, MA  01003                       USA  \










Path: ucivax!gateway
From: [e-mail] ("Kenneth B. Kirksey")
Subject: Update Offer Expires?
Full-Name: Kenneth B. Kirksey
Message-ID: <[e-mail]>
Newsgroups: fa.think-c
Lines: 7
Date: 20 Mar 92 00:48:54 GMT


Well, I finally graduated (hurrah!) and now I've got the money to upgrade
from Think-C 4.0 to 5.0. I pulled out the upgrade card that Symantec sent me,
and it said that the upgrade offer expired December 31, 1991. Bummers! How
strict are they about this? Could I just send in the card and a check anyway?

                                     Ken










Path: ucivax!gateway
From: [e-mail] (Peter Miller)
Subject: Re: Debugger Bombs with large arrays
Message-ID: <[e-mail]>
In-Reply-To: Your message of 19 Mar 92 23:25:14 +0000.
             <[e-mail]>
Newsgroups: fa.think-c
Lines: 52
Date: 20 Mar 92 04:22:51 GMT


Craig Miller <[e-mail]> writes:
> Is it really true?  Think C has no automatic runtime check for a
> stack overflow? I've been programming professionally with Think C
> for two years, assuming that any stack overflow problems would
> generate some kind of runtime error message, just like all the other
> systems I've programmed in.

Whom do you thinks should be doing the checking?
Should the compiler generate stack checking code immediately before each
LINK opcode it generates, to make sure the imminent LINK will not bang into
the stack?  If it did find a problem, what then who and how does it notify,
since the _compiler_ has no idea if the code will be used in a TCL environment,
dumb application, code resource, or burnt into a rom and put into a toy train!

I know I have been spoiled by VM environments - when my program wants more
stack, it goes and grabs it.  This generates page faults, and the operating
system either gives me more memory (because it's a stack opcode) or bombs my
process with a memory violation.  If, in some weird situation, there happened
to be 4GB of page disk and my process just happened to consume 4GB of memory,
and the stack pointer overran the heap, I know of no VM system which would tell
me "memory violation, your stack just trashed your heap".  The UNIX box I am
writing this on certainly has no such error, but then, it doesn't have
4GB of page disk either, so it can never happen :-).

The Mac, even in its more expensive incarnations, still allocates a finite
amount of space to each "process" (I use the term loosely).  There is no
elegant VM environment for each process, and so there is no dynamic allocation
of (virtual) memory for the process's stack.  Your heap starts at one end of
the process' space, and the stack starts at the other.  All of the memory you
are ever going to get has already been given to you, and the OS doesn't much
care if you shoot yourself in the foot.  The VBL check, as you have found, is
mostly useless.

The "VM" available on the more expensive Macs (yes, I am bitter about apple
abandoning we humble Plus users) is mostly hyp: is simply allows you to make
you mac look like its got LOTS of memory to allocate fixed sized chunks in,
rather than just the little amount of pyhsical: process memory is still
allocated the same way.

The answer to "who should check the (imminent) stack overflow?" is,
as various sources have said, "use the Memory Manager".
The Memory Manager checks the stack, and it has a way of reporting failure.
TCL adds even more wrapping, allowing elegant recovery.
What to do about lack of space, heap or stack, is application
and environment specific: there is no point putting up a dialog box asking
them to close windows if the only person who will see it is the little plastic
man in the model train's cab.

Regards
Peter Miller.
(software engineer and C addict)










Path: ucivax!gateway
From: [e-mail] (geoff george)
Subject: Setting stack size
Message-ID: <[e-mail]>
Newsgroups: fa.think-c
Lines: 30
Date: 20 Mar 92 08:16:51 GMT

[e-mail] asks:
 > That brings up an interesting question.  What is the default stack
 > size on the Mac, and how can you vary it?  (This is probably a FAQ;
 > sorry about that.)
 >
 >                                        K<bob>

Here's the code I use, as the very first statement my apps execute:

   #define STACK_SIZE        (48 * 1024L)

   SetApplLimit (CurStackBase - STACK_SIZE);
   InitGraf ((Ptr) & thePort);
   InitFonts ();
   ...

CurStackBase is a low-memory global at 0x908, of type Ptr, defined in
<MemMgr.h> in TC4 and in <LoMem.h> in TC5. SetAppLimit sets where the
end of the heap is. These are documented in the Memory Manager section
of IM I-III. I don't know whether this works under A/UX - anyone?

geoff
--
geoff george    [e-mail]    (my vax)
        or    [e-mail]   (a touch of warmth from GaTech OCS)

"Ordinary f---ing people - I hate 'em. Ordinary person spends his life avoiding
tense situations; repo man spends his life getting INTO tense situations."












Path: ucivax!gateway
From: [e-mail] (Paul Potts)
Subject: THINIK C stack overflow check
Message-ID: <[e-mail]>
Newsgroups: fa.think-c
Lines: 15
Date: 20 Mar 92 14:16:50 GMT


The C language, in general, doesn't provide for things like stack overflow
checks. (Some implementations of C for DOS/Windows support a "stack probe,"
which checks how much stack space is available before calling a function and
placing parameters and variables on the stack). One of the goals of C was to
use only a very small runtime library, or no runtime library at all if no
functions such as standard I/O need to be linked in.

It raises an interesting idea, though, especially for purposes of debugging.
Has anyone written code which does stack checking before calling functions?
It would be more difficult to check stack size before declaring variables,
and this would probably have to be a compiler option, but I imagine that it
could be done without too much pain.

-Paul Potts-










Path: ucivax!gateway
From: [e-mail]
Subject: (none)
Message-ID: <[e-mail]>
Newsgroups: fa.think-c
Lines: 17
Date: 20 Mar 92 15:38:05 GMT

I am using TCL and I am building menus on the fly.
I have been able to add menu items as well. No problkems here.

Now I would like to add command-key options to some of the menu
items that I have built. Can anybody give me an idea how to do this?

I have always done this in ResEDit.

Is there an obsacureobscure method in CBartender that does this?

Any help on this is very much apperciated.

TaThanks

[e-mail]












Path: ucivax!gateway
From: [e-mail] (Francis Taylor)
Subject: THINIK C stack overflow check
Message-ID: <[e-mail]>
In-Reply-To: Paul Potts's message of 20 Mar 92 14:16:50 GMT <[e-mail]>
Newsgroups: fa.think-c
Reply-To: [e-mail]
Lines: 5
Date: 22 Mar 92 08:49:38 GMT

It's my impression (I don't have my Inside Mac with me right now) that
the Mac has a "stack sniffer" that runs during vertical blank.  Of
course, that won't do you very much good, for, as they point out, the
stack can easily run into the heap and back off again without the
sniffer catching it.  It will catch the most egregious cases, though.










Path: ucivax!gateway
From: [e-mail] (Chris McNeil)
Subject: Text edit and multiple wind
Message-ID: <[e-mail]>
Newsgroups: fa.think-c
X-Charset: MACINTOSH
Organization: University of New Brunswick     Fredericton NB Canada
Lines: 13
Date: 22 Mar 92 09:02:12 GMT

Subject:   Text edit and multiple windows
A simple programming question. I have an application that creates multiple
windows and uses text edit to write stuff to them. I the use FindWindow to
get the WindowPtr when I select a window. My question is how do I get the
tePtr for the text that is in the window I select ?

Chris McNeil
Systems Support
University of New Brunswick
Fredericton NB Canada
[e-mail]












Path: ucivax!gateway
From: [e-mail] (Mark Alldritt)
Subject: Adding objects
Message-ID: <9203201719*[e-mail]>
Newsgroups: fa.think-c
Lines: 8
Date: 22 Mar 92 09:34:52 GMT

Hi,

Is there a way in THINK C to add object classes to a program that has already
been linked?  I want to add something like a resource that contains the
implementation of an object class to an existing program.

Thanks in advance.
-Mark










Path: ucivax!gateway
From: [e-mail] (Bob Miller)
Subject: Graf3D?
Message-ID: <9203221938.AA18178@Mr_Grape.convex.com>
Newsgroups: fa.think-c
Lines: 8
Date: 22 Mar 92 19:39:22 GMT

Think C contains a library, Graf3D, and an include file, Graf3D.h.
I've been trying to find some documentation of this library.

Can someone point me toward Graf3D documentation?

Thanks.

                                        K<bob>










Path: ucivax!gateway
From: [e-mail] (WALSHAG)
Subject: Beyond 32k TextEdit
Message-ID: <[e-mail]>
Return-receipt-to: "WALSHAG" <[e-mail]>
Newsgroups: fa.think-c
Lines: 22
Date: 23 Mar 92 15:36:57 GMT

Re:  Using TCL to create text documents greater that 32k

Does anyone have sample code for creating (under TCL) text documents larger
than 32k (the limit on TextEdit)?

Using TCL, you are told to create a subclass of CAbstractText and define your
own buffer.  This doesn't sound too bad, though I would prefer not to re-invent
the wheel.... if someone has already done it
and would be willing to share it, I would appreciate it.

While I'm on the topic of extending TextEdit, has anyone used (or know
of) a commercially available TextEdit extension package for TCL (such as the
"Word Solution Engine", which, to my knowlege does not support TCL or even C)?

Thanks,
Aaron Walsh
Boston College, IPS
([e-mail])














Path: ucivax!gateway
From: [e-mail]
Subject: Re: Beyond 32k TextEdit
Original-From: attmail!ehorvath (Ned Horvath @ CE Software, Inc. )
Lines: 33
Date: 24 Mar 92 03:35:36 GMT
Phone: [phone]
Message-ID: <[e-mail]>
In-Reply-To: your message <internet0831617070> of Mon Mar 23 15:36:57 GMT 1992
>To: internet!ics.uci.edu!think-c
Content-Type: text
Content-Length: 1681
Newsgroups: fa.think-c
Message-Version: 2
Fax-Phone: [phone]
EMail-Version: 2
UA-Message-ID: <MAC-1.3.4A2-618034-ehorvath-3>
UA-Content-ID: <MAC-1.3.4A2-618034-ehorvath-3>
MTS-Message-ID: <ehorvath0840333260>

Aaron Walsh ([e-mail]) asks
> While I'm on the topic of extending TextEdit, has anyone used (or know
> of) a commercially available TextEdit extension package for TCL (such as the
> "Word Solution Engine", which, to my knowlege does not support TCL or even
C)?

DataPak's WS Engine supports C quite nicely, and ships with Think C compatible
interfaces and libraries.  The definitions in the doc are all in pascal, not
unlike the bulk of Inside Macintosh, but the translation to C is generally
straightforward and works just fine, and headers are provided.

It took me 2-3 days to design and implement WS-based replacement classes for
the TCL 1.0 EditText classes (actually I did more: the style commands are
applied to the selection, not the whole buffer as in the (unstyled) TCL 1.0
classes).

I haven't converted those over to TCL 1.1 yet, but expect to soon.  I'll warn
about one gotcha: when scrolling a panorama containing WS text, you have to
scroll, ask WS where it REALLY is, and (usually) set the scroll bar values and
limits a second time: scrolling a WS record is imprecise because the
calculations are deferred.  The structure of the TCL means the scrollbars get
drawn twice, which can be a bit distracting the first time you see it.

WS is also (mostly) Script-Manager compatible.  The only exception I know of
is that, while it handles right-to-left scripts, it does not work with
mixed-direction scripts (e.g. right-to-left Arabic with embedded left-to-right
English).  When I asked DataPak tech support they told me there wasn't enough
demand for the feature to justify the (high) development cost.

Hope that helps.

=Ned Horvath=
[e-mail]










Path: ucivax!gateway
From: [e-mail] (Sigurdur Darri Skulason)
Subject: ShowInit
Message-ID: <[e-mail]>
X-Mailer: ELM [version 2.3 PL11]
Newsgroups: fa.think-c
X-Charset: ASCII
Lines: 6
Date: 24 Mar 92 07:22:36 GMT
X-Char-Esc: 29


Could someone post to me the ShowINIT source code? I recently lost it in a
disk crash, and can't seem to find it on any ftp servers.

  Thanks, Sigurdur.











Path: ucivax!gateway
From: [e-mail] ("Edward J. Huff")
Subject: Re: ShowInit
Message-ID: <[e-mail]>
Newsgroups: fa.think-c
Lines: 218
Date: 24 Mar 92 15:09:28 GMT
X-Envelope-to: [e-mail]

>Could someone post to me the ShowINIT source code? I recently lost it in a
>disk crash, and can't seem to find it on any ftp servers.
>
>  Thanks, Sigurdur.
It appears that a lot of people have never heard of Archie.
You telnet to nis.ans.net, and login as archie, no password:

AIX telnet (nis.ans.net)

IBM AIX Version 3 for RISC System/6000
(C) Copyrights by IBM and by others 1982, 1990.
login: archie
*====================================================================*
|                 Welcome the ANS Archie Server !                    |
|                                                                    |
| *** As of 02/19/92, the default search method is set to "exact".   |
| *** Type "help set search" for more details.                       |
|                                                                    |
| Australian users:   archie.au (139.130.4.6)                        |
| Canadian users:     archie.mcgill.ca     (132.206.2.3)             |
| European users:     archie.funet.fi      (128.214.6.100)           |
|                     archie.doc.ic.ac.uk  (146.169.11.3)            |
| Other U.S. servers:                                                |
|   archie.sura.net      (128.167.254.179)                           |
|   archie.rutgers.edu   (128.6.18.15)                               |
|   archie.unl.edu       (129.93.1.14)                               |
|                                                                    |
| ** 'help' for help                                                 |
| ** Problems, comments etc. to [e-mail]                 |
|                                                                    |
| Client software is available on ftp.ans.net:/pub/archie/clients,   |
| and documentation can be found in /pub/archie/doc on the same      |
| machine.                                                           |
*====================================================================*
# term set to vt100 24 80
archie> prog showinit
# matches / % database searched:    0 /100%

# No matches were found.
archie> set search sub
archie> prog showinit
# matches / % database searched:   31 /
Host earth.rs.itd.umich.edu   (141.211.164.153)
Last updated 12:29 16 Mar 1992

    Location: /mac.bin/system.extensions/init/.AppleDouble
      FILE      rw----r--      7151  Jul 24  1991   ShowInitIcon
    Location: /mac.bin/system.extensions/init
      FILE      rw-r-----         0  Mar  2 21:38   ShowInitIcon
    Location: /mac.bin/system.extensions/init/Beep Shuffle :c4/.AppleDouble
      FILE      rw----r--       937  May 20  1991   CShowINIT.c
    Location: /mac.bin/system.extensions/init/Beep Shuffle :c4
      FILE      rw-r-----      6044  Mar 15  1991   CShowINIT.c

Host nic.funet.fi   (128.214.6.100)
Last updated 05:45 24 Mar 1992

    Location: /pub/mac/source
      FILE      rw-rw-r--     36479  Nov 10  1989   cshowinit.sit

Host garbo.uwasa.fi   (128.214.87.1)
Last updated 11:45 18 Mar 1992

    Location: /mac/source
      FILE      rw-rw-r--     36479  Nov 10  1989   cshowinit.sit

Host ucselx.sdsu.edu   (130.191.1.10)
Last updated 06:08  6 Mar 1992

    Location: /pub/mac/ThinkC
      FILE      rw-r--r--     51629  Oct 18  1990   cshowinit.hqx

Host plains.nodak.edu   (134.129.111.64)
Last updated 12:00  2 Mar 1992

    Location: /pub/mac/pub/programming/src
      FILE      rw-rw-r--     51470  Nov 10  1989   cshowinit.hqx

Host rusmv1.rus.uni-stuttgart.de   (129.69.1.12)
Last updated 15:35  6 Feb 1992

    Location: /soft/mac/source
      FILE      rw-r--r--     16114  Feb 19  1991   showinit.hqx
    Location: /serv2/soft/mac/source
      FILE      rw-r--r--     16114  Feb 19  1991   showinit.hqx

Host wuarchive.wustl.edu   (128.252.135.4)
Last updated 05:43  8 Mar 1992

    Location: /mirrors/macintosh/code
      FILE      rw-r--r--      7593  Aug 30  1988   asm-showinit.sit

Host rigel.acs.oakland.edu   (141.210.10.117)
Last updated 05:48  2 Mar 1992

    Location: /pub/macintosh/code
      FILE      rw-r--r--      7593  Aug 30  1988   asm-showinit.sit

Host procyon.cis.ksu.edu   (129.130.10.80)
Last updated 05:01  2 Mar 1992

    Location: /pub/Macintosh/source/asm
      FILE      rw-r--r--      7593  Aug  1  1991   showinit.shar

Host wuarchive.wustl.edu   (128.252.135.4)
Last updated 05:43  8 Mar 1992

    Location: /mirrors3/rascal.ics.utexas.edu/programming/+_ASM
      FILE      r--r--r--      6718  Jul 31  1988   ShowINIT.asm

Host rascal.ics.utexas.edu   (128.83.138.20)
Last updated 05:22  2 Mar 1992

    Location: /misc/mac/programming/+_ASM
      FILE      rw-rw-r--      6718  Aug  1  1988   ShowINIT.asm

Host wuarchive.wustl.edu   (128.252.135.4)
Last updated 05:43  8 Mar 1992

    Location: /mirrors3/rascal.ics.utexas.edu/programming/+_C-src+bin
      FILE      r--r--r--     36480  Nov 11  1989   CShowINIT_SIT_bin

Host rascal.ics.utexas.edu   (128.83.138.20)
Last updated 05:22  2 Mar 1992

    Location: /misc/mac/programming/+_C-src+bin
      FILE      rw-rw-r--     36480  Nov 12  1989   CShowINIT_SIT_bin

Host wuarchive.wustl.edu   (128.252.135.4)
Last updated 05:43  8 Mar 1992

    Location: /mirrors3/rascal.ics.utexas.edu/programming/+_C-src+bin
      FILE      r--r--r--      1999  Nov 11  1989   CShowINIT_intro

Host rascal.ics.utexas.edu   (128.83.138.20)
Last updated 05:22  2 Mar 1992

    Location: /misc/mac/programming/+_C-src+bin
      FILE      rw-rw-r--      1999  Nov 12  1989   CShowINIT_intro

Host plaza.aarnet.edu.au   (139.130.4.6)
Last updated 18:15 16 Mar 1992

    Location: /micros/mac/info-mac/Old/source
      FILE      rw-r--r--     14799  Jul  1  1991   showinit.hqx.Z

Host toklab.ics.osaka-u.ac.jp   (133.1.12.30)
Last updated 05:33  6 Mar 1992

    Location: /mac/info-mac/info-mac/recent/source
      FILE      rw-r--r--     14799  Jan 17 12:29   showinit.hqx.Z

Host shark.mel.dit.csiro.au   (144.110.16.11)
Last updated 10:50  3 Mar 1992

    Location: /info-mac/Old/source
      FILE      rw-r--r--     14799  Jul  1  1991   showinit.hqx.Z

Host plaza.aarnet.edu.au   (139.130.4.6)
Last updated 18:15 16 Mar 1992

    Location: /micros/mac/info-mac/Old/source/c
      FILE      rw-r--r--     45437  Jul  1  1991   cshowinit.hqx.Z

Host shark.mel.dit.csiro.au   (144.110.16.11)
Last updated 10:50  3 Mar 1992

    Location: /info-mac/Old/source/c
      FILE      rw-r--r--     45437  Jul  1  1991   cshowinit.hqx.Z

Host toklab.ics.osaka-u.ac.jp   (133.1.12.30)
Last updated 05:33  6 Mar 1992

    Location: /mac/info-mac/news
      FILE      rw-r--r--     45828  Mar 15  1990   CShowINITforTC.hqx.Z

Host rigel.acs.oakland.edu   (141.210.10.117)
Last updated 05:48  2 Mar 1992

    Location: /pub/macintosh/code
      FILE      rw-r--r--      7593  Aug 30  1988   asm-showinit.txt

Host gdr.bath.ac.uk   (138.38.32.1)
Last updated 06:25 18 Mar 1992

    Location: /pdsoft/macintosh/code
      FILE      rw-r--r--      7593  Aug 31  1988   showinit.asm

Host kalikka.jyu.fi   (130.234.0.2)
Last updated 05:18 21 Mar 1992

    Location: /maclib/source
      FILE      rw-rw-r--     11392  Feb 25 16:05   ShowInit_1.1b1.sit

Host jyu.fi   (130.234.0.6)
Last updated 05:16 21 Mar 1992

    Location: /maclib/source
      FILE      rw-rw-r--     11392  Feb 25 16:05   ShowInit_1.1b1.sit

Host kalikka.jyu.fi   (130.234.0.2)
Last updated 05:18 21 Mar 1992

    Location: /maclib/source
      FILE      rw-rw-r--       649  Feb 25 16:05   ShowInit_1.1b1.sit.doc

Host jyu.fi   (130.234.0.6)
Last updated 05:16 21 Mar 1992

    Location: /maclib/source
      FILE      rw-rw-r--       649  Feb 25 16:05   ShowInit_1.1b1.sit.doc

archie>
--
Edward J. Huff   [e-mail]   [phone]
Keck Laboratory for Biomolecular Imaging
NYU Chemistry Deptartment, 31 Washington Place, New York NY 10003











Path: ucivax!gateway
From: "ENG::TLEAMING"@gemini.sim.es.com (Taylor Leaming x3836)
Subject: RE: Beyond 32k TextEdit
Message-ID: <[e-mail]>
Newsgroups: fa.think-c
Lines: 33
Date: 24 Mar 92 17:26:27 GMT
X-Vmsmail-To: SIM::SMTP%"[e-mail]"

Aaron,

>Does anyone have sample code for creating (under TCL) text documents larger
>than 32k (the limit on TextEdit)?
>
>Using TCL, you are told to create a subclass of CAbstractText and define your
>own buffer.  This doesn't sound too bad, though I would prefer not to re-invent
>the wheel.... if someone has already done it
>and would be willing to share it, I would appreciate it.
>
>While I'm on the topic of extending TextEdit, has anyone used (or know
>of) a commercially available TextEdit extension package for TCL (such as the
>"Word Solution Engine", which, to my knowlege does not support TCL or even C)?
>
>Thanks,
>Aaron Walsh
>Boston College, IPS
>([e-mail])


I've been looking for the very same thing.  My best solution to date is to use
DataPak's WordEngine.  But, it costs $$.  Please let me know if you find a
better solution (ie: source code).

Taylor

--------------------------------------------------------------------------
Taylor Leaming                                  [e-mail]
Simulation Division
Evans & Sutherland Computer Corp.
SLC, UT                                                       801/582-5847












Path: ucivax!gateway
From: [e-mail] (Trey Campbell)
Subject: Getting along with the Finder
Message-ID: <[e-mail]>
Newsgroups: fa.think-c
Lines: 18
Date: 24 Mar 92 20:04:57 GMT
X-Vmsmail-To: SMTP%"[e-mail]"

Hi, all!

I was wondering if anybody out there knows how to coax the Finder into noticing
changes that have been made to a file via FSpSetFInfo? I've tried using
FlushVol to no avail. Clearly the Finder is caching this new info somehwere,
but how can I get it written to disk? I've seen several programs where new
files and icons show up without having to open and close the window.

A second, sort of related, question: Is there a way to keep the "File you're
opening is locked" dialog from appearing when you open a locked file? I've
written a program to allow the user to change the Finder flags, etc. and this
dialog is sort of redundant since the program displays that info. I wrote a
dialog hook to go with CustomGetFile to record the file's status and unlock the
file when the OK button receives a hit (and it works), but I still get the
dialog. Any suggestions?

Thanks,
Trey Campbell, X-ray Diffraction Lab, Texas A&M University










Path: ucivax!gateway
From: [e-mail] (Trey Campbell)
Subject: Finder and File Updates
Message-ID: <[e-mail]>
Newsgroups: fa.think-c
Lines: 18
Date: 24 Mar 92 21:00:21 GMT
X-Vmsmail-To: SMTP%"[e-mail]"

Hi, all!

I was wondering if anybody out there knows how to coax the Finder into noticing
changes that have been made to a file via FSpSetFInfo? I've tried using
FlushVol to no avail. Clearly the Finder is caching this new info somehwere,
but how can I get it written to disk? I've seen several programs where new
files and icons show up without having to open and close the window.

A second, sort of related, question: Is there a way to keep the "File you're
opening is locked" dialog from appearing when you open a locked file? I've
written a program to allow the user to change the Finder flags, etc. and this
dialog is sort of redundant since the program displays that info. I wrote a
dialog hook to go with CustomGetFile to record the file's status and unlock the
file when the OK button receives a hit (and it works), but I still get the
dialog. Any suggestions?

Thanks,
Trey Campbell, X-ray Diffraction Lab, Texas A&M University










Path: ucivax!gateway
From: [e-mail] ("Hisham A. Abboud")
Subject: gamma tables
Message-ID: <[e-mail]>
Newsgroups: fa.think-c
Lines: 15
Date: 24 Mar 92 21:31:00 GMT


Hi there,

Any gurus out there familiar with changing the gamma table on color
Macs?  Also, would you know how quickly (in msecs) the gamma table can
be switched around?

Thanks.  I'd appreciate any info.

                        Hisham.


Hisham A. Abboud, Cedrus Corp.  [Internet: [e-mail]]












Path: ucivax!gateway
From: [e-mail] (Hal Zabin)
Subject: Modal Dialogs with filterProc
Message-ID: <[e-mail]>
X-Mailer: LeeMail 1.2
Newsgroups: fa.think-c
Lines: 21
Date: 25 Mar 92 14:15:22 GMT

Does anyone have some sample code they would be willing to send me in which one
uses a filter procedure in the modal dialog call? Specifically, I am trying to
make a rather simple dialog box with a field for text entry which can be
scrolled, and if I am not wrong, I must do this by passing ModalDialog a
ProcPtr which can handle this. This is probably a rather simply question, but I
am stumped! I would appreciate any help any of you might be able to offer me.
Thanks, Hal Zabin




Please respond to: [e-mail]

<=====-------------------------------=====>
Dr. Hal B. Zabin
Max-Planck-Institut fr molekulare Genetik
Abteilung Trautner
1000 Berlin 33
Germany
<=====-------------------------------=====>











Path: ucivax!gateway
From: [e-mail] (Stephen Roderick)
Subject: RE: Beyond 32k TextEdit
Message-ID: <[e-mail]>
Newsgroups: fa.think-c
Lines: 9
Date: 26 Mar 92 04:33:44 GMT


If you are interested I am selling WS Engine ($395) with the
virtual memory extension ($175).

I will sell the two together for $295 or a reasonable offer!

Steve Roderick
[e-mail]
[phone]










Path: ucivax!gateway
From: [e-mail] (Hal Zabin)
Subject: Crashes...
Message-ID: <[e-mail]>
X-Mailer: LeeMail 1.2
Newsgroups: fa.think-c
Lines: 21
Date: 26 Mar 92 22:50:39 GMT

I have been working on a program which is frustrating me to no end due to the
following: the program always crashes at a particular point, EXCEPT when I am
using the debugger! I have a sneaking suspicion that this should be telling me
something important about the problem. If anyone has any ideas on what would
lead to such a situation, I would greatly appreciate hearing them! (BTW: I am
using Think-C 5.0.2).

Thanks,
Hal Zabin


Please respond to: [e-mail]

<=====-------------------------------=====>
Dr. Hal B. Zabin
Max-Planck-Institut fr molekulare Genetik
Abteilung Trautner
1000 Berlin 33
Germany
<=====-------------------------------=====>











Path: ucivax!gateway
From: [e-mail] (Andre "Just A Plumber" Srinivasan)
Subject: Re: Crashes...
Message-ID: <[e-mail]>
In-Reply-To: Hal Zabin's message of 26 Mar 92 22:50:39 GMT <[e-mail]>
Newsgroups: fa.think-c
Lines: 21
Date: 27 Mar 92 00:01:54 GMT


) following: the program always crashes at a particular point, EXCEPT when I am
) using the debugger! I have a sneaking suspicion that this should be telling me
) something important about the problem. If anyone has any ideas on what would
) lead to such a situation, I would greatly appreciate hearing them! (BTW: I am
) using Think-C 5.0.2).

you're dereferencing a pointer that works out ok when the debugger is
in memory.  try localizing the problem to a particular function or
even a few lines of code if you can.  then, using either macsbug or
the debugger, check every pointer on each line in the suspect area.

good luck.

                                -andre.

---
Andre Srinivasan  :     "This file provides programmers with information
734 LRDC          :     proving that it really was a hardware problem..."
U. of Pittsburgh  :
[e-mail] :                - MacsBug Ballon Help










Path: ucivax!gateway
From: [e-mail] ("Hisham A. Abboud")
Subject: Re:  Sound Manager Question
Message-ID: <[e-mail]>
Newsgroups: fa.think-c
Lines: 13
Date: 27 Mar 92 02:18:35 GMT


Earlier, I posted a question about playing async sound under
System 6.07.  Thanks to the couple of people who resonded.

While looking for help on AppleLink, I ran into a note by Apple
Apple documenting a bug in 6.07.  Check it out (I could send a copy
to those interested, just ask).

                    Hisham.



Hisham A. Abboud, Cedrus Corp.  [Internet: [e-mail]]










Path: ucivax!gateway
From: [e-mail] (Aric Friesen)
Subject: That header info on a printed page...
Message-ID: <[e-mail]>
Newsgroups: fa.think-c
Lines: 8
Date: 29 Mar 92 02:07:16 GMT


This is probably a really stupid question, but I have yet to figure out
how to stop ThinkC from printing the time, date, and filename at the
tope of each page.  Is there a way to stop it?  I really don't want people
to know that I do most of my programming at 2:30 in the morning.

        Aric Friesen











Path: ucivax!gateway
From: [e-mail] (Igor Livshits)
Subject: Exception handler v. CError
Message-ID: <[e-mail]>
Newsgroups: fa.think-c
Lines: 13
Date: 31 Mar 92 16:31:40 GMT

Hello,

I getting confused by the overlapping utility of the CError object the
Exception handler in TCL 1.1.  Which one should I use and when?  Does
CError rely on the Exception handler?

One pf the problems with the Exception handler that I noticed is that while
the applciation is starting up and creates a new document, the errors are
caught by the Apple Event object and not CApplication.  This results in a
lack of appropriate error alerts.

Thanks, igor











Path: ucivax!gateway
From: [e-mail]
Subject: Re: Beyond 32k TextEdit
Message-ID: <9203311843.AA16744@husc10>
In-Reply-To: <[e-mail]>; from "WALSHAG" at Mar 23, 92 3:36 pm
X-Mailer: ELM [version 2.3 PL11]
Newsgroups: fa.think-c
Lines: 18
Date: 31 Mar 92 18:43:51 GMT

> Does anyone have sample code for creating (under TCL) text documents larger
> than 32k (the limit on TextEdit)?
>
> Using TCL, you are told to create a subclass of CAbstractText and define your
> own buffer.  This doesn't sound too bad, though I would prefer not to
> re-invent the wheel.... if someone has already done it
> and would be willing to share it, I would appreciate it.

I've written just such a subclass of CAbstractText that supports
documents larger than 32k.  It's not Script Manager-compatible or
anything like that, but it works reasonably well.  If anyone would be
interested in testing my code, drop me a message and I'll get back to
you at some point.

Chris Wysocki
[e-mail]