Nautilus 1992 July/Nautilus-3-8/Nautilus-3-8.bin/Tools & Utilities/Techy Stuff/Doco ƒ/CSMP ƒ/CSMP-V1-056.TXT

Text File | 1992-06-03 | 50.1 KB | 1,424 lines

C.S.M.P. Digest             Tue, 21 Apr 92       Volume 1 : Issue 56

Today's Topics:

    Page flipping
    How to write serial port handling code?
    Mounting AppleShare volumes?
    Object Question
    QuickDraw Regions
    QuickTime Development Question.
    think-c warning : "code overflow", wazzit mean?
    Naming a Print Monitor job
    Seeking Apple File Exchange translator coding guide?
    Sending AppleEvents..
    A/UX 2.0.1 problem
    SndPlayFromDisk buffer size?
    Think C 4.0: debugger crashes on LC
    Apple Event Max Size Query


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

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

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

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

Send administrative mail to [e-mail].

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

From: [e-mail] (Michael A. Kelly)
Subject: Page flipping
Organization: University of Oregon Computer and Information Sciences Dept.
Date: Tue, 10 Mar 1992 10:26:16 GMT


OK, I've been working on this for the last hour or so, and I can't seem to
get any positive results.  Have any of you gotten page flipping to work on
a Mac II?

The code below doesn't cause a crash, but it also doesn't seem
to affect the screen.  A window the size of the GrayRgn is opened and filled
with black, an offscreen PixMap is created with NewScreenBuffer, the main
device's PixMap and the offscreen PixMap are switched, and then a small loop
draws a red square in the middle of the screen.  The problem is that the
red square doesn't show up if I switch the PixMaps.  It _does_ show up if
I don't switch the PixMaps.

The drawing loop draws directly to the screen, using the routine found in
the FAQ.

So, anyone have any ideas?  When and if I get it working, I'll post a tutorial.

Here's the code:

void main()
{

    GDHandle        theDevice;
    short           x, y;
    Ptr             tempBaseAddr;
    PixMapHandle    thePixMap;
    PixMapHandle    tempPixMap;
    OSErr           theErr;

    /* This routine opens a big black window */

    Initialize();

    theDevice = GetMainDevice();

    LockPixels( (*theDevice)->gdPMap );

    theErr = NewScreenBuffer( &(*theDevice)->gdRect, false, &theDevice,
                              &thePixMap );
    (*thePixMap)->baseAddr = StripAddress( (*thePixMap)->baseAddr );

    tempPixMap = (*theDevice)->gdPMap;
    (*theDevice)->gdPMap = thePixMap;
    thePixMap = tempPixMap;

    UnlockPixels( (*theDevice)->gdPMap );

    for ( x = 304; x < 336; ++ x ) {
        for ( y = 224; y < 256; ++ y ) {
            SetPixel( x, y, 0x00000023, (*theDevice)->gdPMap );
        }
    }

    while ( ! Button() ) {
    }

    while ( Button() ) {
    }

    LockPixels( (*theDevice)->gdPMap );

    tempPixMap = (*theDevice)->gdPMap;
    (*theDevice)->gdPMap = thePixMap;
    thePixMap = tempPixMap;

    UnlockPixels( (*theDevice)->gdPMap );

    DisposeScreenBuffer( thePixMap );

    /* This closes the window, etc. */

    ExitCleanup();

}


Thanks for _any_ reply!!

Mike.
- --
_____________________________________________________________________________
Michael A. Kelly                                         University of Oregon
[e-mail]                             Computer Science Department
_____________________________________________________________________________

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

From: [e-mail] (Ben Haller)
Date: 20 Mar 1992 01:35:44 GMT
Organization: Stick Software

In article <[e-mail]>
   [e-mail] (Michael A. Kelly) writes:
>OK, I've been working on this for the last hour or so, and I can't seem to
>get any positive results.  Have any of you gotten page flipping to work on
>a Mac II?
  I believe that in general page flipping is not possible.  If anyone
knows better, I am *extremely* interested.  The documentation for the
INIT "MaxApplZoom" has a very informative discussion of video cards
and drivers and their capabilities, and the gist seems to be that
some boards support vertical panning, which could be used to do
page flipping if there is enough memory on the card, but that there is
no standard way to do this on all cards or anything nice like that.

- -Ben Haller ([e-mail])

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

From: [e-mail] (Dermot Bradley)
Subject: How to write serial port handling code?
Date: 15 Mar 92 19:04:14 GMT
Organization: Extended Bulletin Board Service

Hi,

we are hoping to write code to handle a modem attached to the serial port but
we don't know where to begin. I've heard mention of the Communication Toolbox
but we need to start ASAP and the CTB is supposed to take quite some time to
get into.

Any help would be much appreciated,

Dermot Bradley

[e-mail]  <----- Preferred email address
[e-mail]

- --
   The opinions expressed are not necessarily those of the University of
     North Carolina at Chapel Hill, the Campus Office for Information
        Technology, or the Experimental Bulletin Board Service.
           internet:  bbs.oit.unc.edu or 152.2.22.80

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

From: [e-mail] (Glenn Austin)
Date: Sun, 15 Mar 92 17:18:13 PST
Organization: The Pit Lane


In article <[e-mail]> (comp.sys.mac.programmer), [e-mail] (Dermot Bradley) writes:
> we are hoping to write code to handle a modem attached to the serial port but
> we don't know where to begin. I've heard mention of the Communication Toolbox
> but we need to start ASAP and the CTB is supposed to take quite some time to
> get into.

Not really -- most of the so-called "problems" associated with the CTB have
to do with adapting existing serial code to work with the CTB.  In reality,
using the CTB means that you don't have to do the hard work.  All you have
to do is open the connection and monitor it.

===============================================================================
| Glenn L. Austin                | "Turn too soon, run out of room.           |
| Macintosh Wizard and           |    Turn too late, much better fate."       |
| Auto Racing Driver             |   -- Jim Russell Racing School Instructors |
| Usenet:  [e-mail] or [e-mail]    |
===============================================================================

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

From: [e-mail] (Nick Rothwell)
Date: 16 Mar 92 14:43:49 GMT
Organization: Friends of the Salter Duck

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

   I've heard mention of the Communication Toolbox
   but we need to start ASAP and the CTB is supposed to take quite some time to
   get into.

I wrote some code using the Comms Toolbox to drive a modem, just as a learning
exercise. Took me one (1) day.

Nick.
- --
Nick Rothwell, LFCS, Edinburgh | "We know what to get *you* for
             [e-mail] |  Christmas: a double lobotomy and
Mentation Consultancy Services |  ten rolls of rubber wallpaper."
   [e-mail] |

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

From: [e-mail] (Bryan Socha)
Date: 17 Mar 92 19:04:00 GMT
Organization: FidoNet node 1:273/921 - Unseen Realit, Philadelphia PA


 # we are hoping to write code to handle a modem attached to the
 # serial port but
 # we don't know where to begin. I've heard mention of the
 # Communication Toolbox
 # but we need to start ASAP and the CTB is supposed to take
 # quite some time to
 # get into.

I am doing the same thing..  basically there are several books, ctb is mainly
a system 7 thing so you should look at the system 7 books and apple has a
specific communications tookbox book...

- --
=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=
 Bryan Socha - Internet: [e-mail]

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

From: [e-mail] (Charles McClary)
Subject: Mounting AppleShare volumes?
Organization: Indiana University
Date: Fri, 13 Mar 92 19:51:45 GMT

I would like to mount an AppleShare volume from an application.  I have tried
using the PBVolumeMount as explained in Vol. VI with no success.  I noticed,
on the net, that other people were having trouble with bugs in PBVolumeMount.

My question is how did people mount AppleShare volume for System 6.0.x and
System 7.0.x?

Charles McClary
[e-mail]

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

From: [e-mail] (David Castell)
Date: 16 Mar 92 14:49:07 GMT
Organization: Computer Systems Group

In article <[e-mail]>, [e-mail] (Charles McClary) writes:
>
> I would like to mount an AppleShare volume from an application.  I have tried
> using the PBVolumeMount as explained in Vol. VI with no success.  I noticed,
> on the net, that other people were having trouble with bugs in PBVolumeMount.
>
> My question is how did people mount AppleShare volume for System 6.0.x and
> System 7.0.x?
>
> Charles McClary
> [e-mail]
>
I have been able to get PBVolumeMount to work without many problems (except
for volumes that require volume passwords because PBVolumeMount ignores the
volume password field).  Supposedly AppleShare 3.0 drivers will implement a
PBVolumeMount that will fix this problem and it can be used with System 6 or
System 7.

There is another way to mount volumes in System 6 and System 7.  The MPW
'choose' tool uses this method for System 6, but it is not very pleasant.
The method basically involves getting a special resource from the AppleShare
drivers that is called the 'AppleShare Mounter'.  Using assembler you can
set up the registers and call into this resource, which contains the code
to do all of the work.  The catch is that the calls and how the registers
are set up are currently undocumented by Apple.  They say they might release
this information, but I wouldn't hold my breath.  For our software, we just
hacked the code to see how to do it, but like I said, it is not nice code
to look at.

Dave Castell

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

From: [e-mail] (James)
Date: 19 Mar 92 21:12:17 GMT
Organization: University of Maine System

Hello.
I'm writing for a friend that wanted to access files on a server
from within 4-D.
We want this to be as transparent as possible.  Meaning that
the enduser does not have to share to the server that is going
to be used.  The application will be able to retrieve files
from the server in the background and it would appear as
if the files were on the end users computer..

I've heard about PMVolumeMount but haven't used it.  Any suggestions
as to what we should use will be greatly appreciated.

Thanks
Jim Gray
=-=-=-=-=-=--=-=-=-=
James Gray                           Bitnet   :  IO92142@Maine
University of Maine                  Internet :  [e-mail]

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

From: [e-mail]
Subject: Object Question
Date: 14 Mar 92 20:19:29 GMT
Organization: Drexel University Guerrilla Networking Project

I'm familiar with the basics of OOP, but I can't figure out how to do what
I want (I'm using THINK Pascal).

I've declared a class...

  ThreadErr = (none, done, died);

  CThread = object

    fName: string;
    fThreadID: integer;
    fPriority: integer;

    function Init: ThreadErr;
    function DoIdle: ThreadErr;
    function Kill: ThreadErr;
    function Status: ThreadErr;
    function SetPriority (thePriorty: integer): integer;
    function GetPriority: integer;
    function GetName: string;
    function GetID: integer;
    function SetID (theID: integer): integer;
   end;

...which is going to be used to represent threads (obviously). Now, the code
for these things is going to reside in files (as code resources, I imagine).
All of the code will be common except Init, DoIdle, Kill, SetPriority, and
Status. How would I go about loading these routines and attaching them to
the object? It seems a bit stupid to define sub-classes like CLoginThread,
etc, since putting objects in data files seems very confusing. Any kind of
comments on this situation would be appreciated.

Thanks,

- ---------
Ed Watkeys                         "...if you wish to strive for peace of
phlpa!caligula![e-mail]   soul and pleasure, then believe; if
Drexel University (Comp Sci)        you wish to be a devotee of truth,
Guerrilla Networking Project        then inquire...." -- Nietzsche

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

From: [e-mail] (Jim Spencer)
Date: 17 Mar 92 03:54:43 GMT


 ew> From: [e-mail]
 ew> Newsgroups: comp.sys.mac.programmer
 ew> Organization: Drexel University Guerrilla Networking Project
 ew> Reply-To: caligula!edw (Edwin Howell Watkeys III)
 ew> Message-ID: <[e-mail]>
 ew>
 ew> I'm familiar with the basics of OOP, but I can't figure out how to do
 ew> what
 ew> I want (I'm using THINK Pascal).
 ew>
 ew> I've declared a class...
 ew>
 ew>   ThreadErr = (none, done, died);
 ew>
 ew>   CThread = object
 ew>
 ew>     fName: string;
 ew>     fThreadID: integer;
 ew>     fPriority: integer;
 ew>
 ew>     function Init: ThreadErr;
 ew>     function DoIdle: ThreadErr;
 ew>     function Kill: ThreadErr;
 ew>     function Status: ThreadErr;
 ew>     function SetPriority (thePriorty: integer): integer;
 ew>     function GetPriority: integer;
 ew>     function GetName: string;
 ew>     function GetID: integer;
 ew>     function SetID (theID: integer): integer;
 ew>    end;
 ew>
 ew>...which is going to be used to represent threads (obviously). Now, the
code
 ew>for these things is going to reside in files (as code resources, I
imagine).
 ew>All of the code will be common except Init, DoIdle, Kill, SetPriority, and
 ew>Status. How would I go about loading these routines and attaching them to
 ew>the object? It seems a bit stupid to define sub-classes like CLoginThread,
 ew>etc, since putting objects in data files seems very confusing. Any kind
 ew>of comments on this situation would be appreciated.

It appears that you misunderstand how Object Pascal works.  The linker takes
care of calling methods pretty much the same way it does for normal procedure
calls.  All that changes is that there is a level of indirection to the call.

You are mixing up the concept of the "class" which you are defining above and
the "object" itself which is an instance of a class.  The class is nothing
more than the definition of what an object will be once one is created.  In
your program, you actually create the objects themselves in RAM, not on disk,
using NEW.  One of the advantages of subclassing your CThread is that you
don't have to rewrite the code that is common.  You simply OVERRIDE the
methods that change.  The other advantage is that you can create the objects
you need, for various types of threads in your case, and assign them to a
variable of type CThread.  You can then call say DoIdle or Kill for that
variable without having to worry about having the right code being executed.

I recommend you pick up a copy of "Object Oriented Programming Power for
Think Pascal Programmers" by Chuck Sphar, published by Microsoft Press,
ISBN 1-55615-348-1.  The issues you raise are really beyond an answer here.

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

From: Lund-Larsen/Eine
Subject: QuickDraw Regions
Organization: Scandinavian Customized Prosthesis
Date: Mon, 16 Mar 92 18:10:13 GMT



We are working with QuickDraw regions and wonder if there are any
possibilities of reading the region definition data.
We are especially interested in getting access to points on the outline of the
region.

Any help is appreciated.

Thanks

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

From: [e-mail] (Vince Vann)
Date: 18 Mar 92 22:41:35 GMT
Organization: University of Miami Medical School

Lund-Larsen/Eine writes:

>We are working with QuickDraw regions and wonder if there are any
>possibilities of reading the region definition data.
>We are especially interested in getting access to points on the outline of the
>region.

>Any help is appreciated.

I once wrote a routine to parse through quickdraw region definitions.
I no longer have the code, but fortunately I remember how quickdraw
regions are encoded.

                          *** WARNING ***

All of the information in this description is based on my own
interpretation.  It is not derived from any documented sources.
I have attempted to summarize what I remember about quickdraw regions
definitions, but you should verify the information for yourself by
parsing through several simple regions.

Quickdraw Region Definitions
- ----------------------------
Basically, a region is encoded line by line from top to bottom as sets
of "pixel runs".  Redundancies are eliminated by only encoding information
about those pixels which change from one horizonatal scan line to the
next.  If you are familiar with scan line conversion algorithms or run
length encoding compression algorithms, the idea is very similar.

The actual region data consists of a list of signed integer pairs
that essentially encode horizontal runs of pixels that are to be
turned ON (included within the region) or turned OFF (excluded from the
region).  The runs are encoded by scan line from top to bottom.  The
format for each line is as follows:

     <scan line> <start,stop> [... <start,stop>] <0x7fff>
     <scan line> <start,stop> [... <start,stop>] <0x7fff>
     ......
     ......
     <scan line> <start,stop> [... <start,stop>] <0x7fff>
     <0x7fff>

The <start,stop> integer pairs indicate runs of pixels that must be
toggled either ON/OFF depending on their current state.  Initially, a
region is considered empty and all pixels are OFF.  All runs for the
first scan line will therefore serve to turn ON pixels so that they will
be included in the region.  On subsequent scan lines, pixel runs may
toggle pixels either ON or OFF depending on their current state.  When
there is no change from one scan line to another because the two lines
share exactly the same pattern, the second scan line is not encoded.
This avoids redundancies in the region definition and saves a tremendous
amount of space.

Example 1
- ---------
Here is a simple example.  Say you want to encode a rectangle with
X,Y coordinates 40,10,120,50.  The region would look something like
the one below.  An 'x' indicates a pixel within the region, a '-'
or '|' indicates part of the region boundary.

              40                      120
            10 -------------------------
               |xxxxxxxxxxxxxxxxxxxxxxx|
               |xxxxxxxxxxxxxxxxxxxxxxx|
               |xxxxxxxxxxxxxxxxxxxxxxA|B
               |xxxxxxxxxxxxxxxxxxxxxxx|
               |xxxxxxxxxxxCxxxxxxxxxxx|
            50 -------------------------
                           D

Since the boundaries of a region lie BETWEEN PIXELS like grid lines,
this region includes pixels 40-119 horizontally and 10-49 vertically.
Horizontally, the right boundary occurs at grid line 120.  Therefore,
pixel 119 (A) is in the region, but pixel 120 (B) is not.  Same thing
vertically, pixel 49 (C) is within the region but pixel 50 (D) is not.

The rectangular region in this example would be encoded as follows:

             10 - 40 120 0x7fff
             50 - 40 120 0x7fff
             0x7fff

Initially, the region is EMPTY and contains no pixels.  The first number
[10] means that the top most extent of the region is at quickdraw
coordinate 10.  The next number [40] means that pixels should be
included on this line starting with pixel 40 until the end of the run
[120] is reached.  Thus, pixel 119 gets included in the region but pixel
120 is left out.

The hexadecimal number <0x7fff> terminates encoding on line 10.  Now,
lines 11-49 are identical to line 10 so they are all left out of the
region definition.  Pixels 40-119 will be included within the region
on each of these scan lines.

The next number [50] indicates that at line 50 a change occurs in the
encoding pattern.  The next two numbers [40] [120] represent a run of
pixels which must be toggled either ON/OFF.  Since pixels 40-119 were
previously ON in scan line 49, they will be turned OFF.  The <0x7fff>
terminates encoding for line 50.  This effectively ends the definition
of our rectangular region which is marked by the final <0x7fff>.

Example 2
- ---------
What kind of region do you get if you change the definition above by
inserting two extra lines as follows:

             10 - 40 120 0x7fff
        >>>> 25 - 60 100 0x7fff   /* Excludes pixels 60-99 on */
                                  /* scan lines 25 through 34 */
        >>>> 35 - 60 100 0x7fff   /* Includes pixels 60-99 again */
             50 - 40 120 0x7fff
             0x7fff

Did you guess it?  It's the same rectangle with a hole in it...

               40    60       100     120
             10-------------------------
               |xxxxxxxxxxxxxxxxxxxxxxx|
             25|xxxxxx----------xxxxxxx|
               |xxxxxx|oooooooo|xxxxxxx|
             35|xxxxxx----------xxxxxxx|
               |xxxxxxxxxxxxxxxxxxxxxxx|
             50-------------------------

The pixels marked 'x' are part of the region, the pixels marked 'o' are
not part of the region.  Looks like a rectangular DONUT!!!

I hope all of this makes sense.  If you follow it, then you see that
quickdraw region encoding is very efficient and simple.

Parsing Through a Region Definition
- -----------------------------------
To parse through a region, just remember that 'rgnSize' gives you the
size of the region data in BYTES.  (rgnSize / 2) gives you the number of
integers listed in the region definition.  This includes all of the scan
line terminators <0x7fff> and the final <0x7fff> terminator as well.
Just parse through the integer array by scan line until you reach the
final <0x7fff> teminator.

Let me know if this helps, and good luck!

Vincent Vann
([e-mail])
- --

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

From: sam@staypuffed (Sam Griffith)
Date: 19 Mar 92 04:35:47 GMT
Organization: Pencom Software

In article <[e-mail]>
[e-mail] (Vince Vann) writes:
|> Lund-Larsen/Eine writes:
|>
|> >We are working with QuickDraw regions and wonder if there are any
|> >possibilities of reading the region definition data.
|> >We are especially interested in getting access to points on the outline of
the
|> >region.
|>
|> >Any help is appreciated.
|>
|> I once wrote a routine to parse through quickdraw region definitions.
|> I no longer have the code, but fortunately I remember how quickdraw
|> regions are encoded.
|>

Middle chopped for brevity.....

|> Let me know if this helps, and good luck!
|>
|> Vincent Vann
|> ([e-mail])
|> --



There was also a article in an old MacTutor mag. if I remember right.  You
might check there.

Sam Griffith Jr.
Pencom Software
[e-mail]

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

From: [e-mail] (Donald T. Major)
Date: 19 Mar 92 17:45:56 GMT
Organization: SAS Institute Inc.

In article <[e-mail]>, sam@staypuffed (Sam Griffith) writes:
|> In article <[e-mail]>
|> [e-mail] (Vince Vann) writes:
|> |> Lund-Larsen/Eine writes:
|> |>
|> |> >We are working with QuickDraw regions and wonder if there are any
|> |> >possibilities of reading the region definition data.
|> |> >We are especially interested in getting access to points on the outline of
|> the
|> |> >region.
|> |>
|> |> >Any help is appreciated.
|> |>
|> |> I once wrote a routine to parse through quickdraw region definitions.
|> |> I no longer have the code, but fortunately I remember how quickdraw
|> |> regions are encoded.
|> |>
|>
|> Middle chopped for brevity.....
|>
|> |> Let me know if this helps, and good luck!
|> |>
|> |> Vincent Vann
|> |> ([e-mail])
|> |> --
|>
|>
|>
|> There was also a article in an old MacTutor mag. if I remember right.  You
|> might check there.
|>
|> Sam Griffith Jr.
|> Pencom Software
|> [e-mail]

Also note that the first example, using a solid rectangle, is not quite
accurate--there is NO data after the rgnBBox field in this case, since
the rectangle of rgnBBox is sufficient to describe the structure.  The
size field similarly reflects this, saying that the total structure
size is 10 bytes.  Additional "run" information is appended ONLY if a
non-rectangular (solid) region is to be represented.

                                                         ..
                                                        dtm

- --
Donald Major       SAS Institute Inc.  "Cicely, let's fling something!"
[e-mail] SAS Campus Drive                 - Northern Exposure
[phone]     Cary, NC 27513-2414

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

From: [e-mail] (Captain Kludge)
Subject: QuickTime Development Question.
Organization: Purdue University Engineering Computer Network
Date: Mon, 16 Mar 1992 20:07:18 GMT

Does anyone know where I can find the THINK Pascal interface
files for QuickTime?  Or has someone converted the MPW pascal
interface files to THINK Pascal?

Thanks,

- -John

{ [e-mail] }


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

From: [e-mail] (James)
Date: 19 Mar 92 21:15:33 GMT
Organization: University of Maine System

Hi.  I'd like any info on the QT Pascal question also.
Could you forward?
Thanks.
Jim Gray
=-=-=-=-=-=--=-=-=-=
James Gray                           Bitnet   :  IO92142@Maine
University of Maine                  Internet :  [e-mail]

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

From: [e-mail] (Mike Smithwick)
Subject: think-c warning : "code overflow", wazzit mean?
Organization: SF-Bay Public-Access Unix
Date: Tue, 17 Mar 1992 18:21:01 GMT

[]

I want to staticly define an array of about 28K in size (a trig lookup table),
but Think gives me the warning of "Code Overflow", which is not listed anywhere
in the manual. What does code-overflow mean? (the module is less then the
32K segment size).

mike

- --
"There is no problem too big that can't be solved with high explosives"-Rush

Mike Smithwick - ames!zorch!mike


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

From: [e-mail] (Eric W. Sink)
Date: 17 Mar 92 20:48:57 GMT
Organization: University of Illinois at Urbana-Champaign

In <[e-mail]> [e-mail] (Mike Smithwick) writes:

>I want to staticly define an array of about 28K in size (a trig lookup table),
>but Think gives me the warning of "Code Overflow", which is not listed anywhere
>in the manual. What does code-overflow mean? (the module is less then the
>32K segment size).

Look in Appendix C.  The error message means that you have tried to put more
than 32k of code or data in a single file.

- --
Eric W. Sink,  Spatial Analysis and Systems Team
USACERL, P.O. Box 9005, Champaign, IL 61826-9005
1-800-USA-CERL x449,   [e-mail]

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

From: [e-mail] (Tim W Smith)
Date: 19 Mar 92 05:42:31 GMT
Organization: The Portal System (TM)

>I want to staticly define an array of about 28K in size (a trig lookup table),
>but Think gives me the warning of "Code Overflow", which is not listed anywhere
>in the manual. What does code-overflow mean? (the module is less then the
>32K segment size).

It is too listed in the manual, right where you would expect to find it in
the alphabetical listing of error messages and their meanings.

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

From: [e-mail] (Michael F. Kamprath)
Subject: Naming a Print Monitor job
Date: Wed, 18 Mar 92 01:27:28 EST
Organization: The University of Michigan

I have a program (in THINK C 4.whatever) that can only open one file at a time.
The program has four windows open at a time for any one file.  Now, my
print routine compiles data from all four windows to produce a single
print document (the program is a simple CAD program, and it prints plans).

Here is my problem:  How do I get the document name in the Print Monitor
window to be the same as the current filename?  I have notice that the
name that does show up to be either "Unspecified" or the name of the current
top window (whose name is not dependent on the filename).  Is their
anyway to set a document name in the print record?  I only have IM 1-3, and
never mentions naming a document to be printed.

Help is greatly appreciated.

Michael F. Kamprath
[e-mail]


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

From: [e-mail] (Jamie R. McCarthy)
Organization: Kalamazoo College
Date: Wed, 18 Mar 1992 13:17:42 GMT

[e-mail] (Michael F. Kamprath) writes:
>The program has four windows open at a time for any one file.  Now, my
>print routine compiles data from all four windows to produce a single
>print document
>
>Here is my problem:  How do I get the document name in the Print Monitor
>window to be the same as the current filename?  I have notice that the
>name that does show up to be either "Unspecified" or the name of the current
>top window (whose name is not dependent on the filename).

I understand that the Print Manager grabs the name of the topmost window
for the name of the print job.  If you turn background printing off and
have two people print at the same time, the one who's waiting will get a
status report for the other, something like "User: Fred; document:
NameOfFred'sTopmostWindow; page 1 of 6."  I would imagine the
PrintMonitor uses the same trick.

The standard thing to do (I hear) is to create an offscreen window,
give it the name you want, and bring it to the front before calling the
Print Manager.  Kill it afterwards, of course.  If you have a status
window ("Now printing page x of y"), give it the appropriate name.
- --
 Jamie McCarthy     Internet: [e-mail]     AppleLink: j.mccarthy
 Kzoo randomly kills all my mail;  if I don't acknowledge, try resending.
 If enough people use "#include <stddisclaimer.h>", it will _become_ the
 standard disclaimer, and all the lawyers will go into infinite recursion.

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

From: [e-mail] (Philip Brandenberger)
Organization: Swarthmore College
Date: Thu, 19 Mar 1992 20:24:07 GMT

The important thing here is that the window be in front (topmost) when
PrValidate is called.  This is when the windowname is stolen by the
print manager.  A tech note on the subject suggests a dialog (with a
dbprocbox) which informs the user that printing is in progress while also
having the name of the document you are printing (which isn't shown 'cause
a dbprocbox doesn't have a titlebar).

The alternate solution is to have either a) and offscreen window as suggested
above, or as the tech note suggests a window under the menu bar.

Note that printmonitor is irrelevant to the question; the document name
is taken when the document is spooled by the application.  The username
is taken when printmonitor prints the spool file, as is the app name.
These last two seem to be stupid oversights in the printing/printmonitor
code, IMHO.  [Why not when you're recording the printer name and the
document name also record the user name and the application name so
Printmonitor gets it all right?]

To use an overused phrase, "Cheers,"

Phil


- --
Philip J. Brandenberger
Swarthmore College, but I don't speak for it, usually against it.
[e-mail]

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

From: [e-mail] (VIEN cam thanh)
Subject: Seeking Apple File Exchange translator coding guide?
Date: 18 Mar 92 16:32:39 GMT
Organization: Concordia University, Montreal, Quebec

is there documentation which will allow me to write translator
for the Apple File Exchange?


Thanks in advance




Cam

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

From: [e-mail] (Brian Bechtel)
Date: 19 Mar 92 15:56:25 GMT
Organization: Apple Computer Inc., Cupertino, CA

[e-mail] (VIEN cam thanh) writes:

>is there documentation which will allow me to write translator
>for the Apple File Exchange?

Apple File Exchange Technical Reference Package v1.1, $30.00 US, from
APDA.  Part number M7051.

[e-mail]
800-282-2732 USA
800-637-0029 Canada
408-562-3910
408-562-3971 Fax

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

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

From: [e-mail] (Mark A. Steele)
Subject: Sending AppleEvents..
Date: 18 Mar 92 18:08:03 GMT
Organization: National Geophyical Data Center, Boulder, Colorado

I am trying to send an Apple Event to another application to tell it
to open a specific document. Looking at inside Mac VI I began working
backwards. First I create the Apple Event to send, then I created
a list and added the FSSpec of the file to that list. I added this
list to the apple event and sent it away. Creating and sending generated
no errors. The receiving program (using MacApp 3.0) gets the event but
is unable to get the FSSpec out of the event.
My feeling is that the way I am sending the event (or creating it) is
incorrect. Could somebody please brief out how I should create the
event once I have the FSSpec that I colletec from StandardGetFile();

- -Mark
[e-mail]

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

From: [e-mail] (Eric Berdahl)
Date: 19 Mar 92 22:53:55 GMT
Organization: Taligent, Inc.

In article <[e-mail]>, [e-mail] (Mark A. Steele) writes:
> [Description of sending an Open Document AppleEvent by packaging up an FSSpec deleted]

The Open Document AppleEvent does not take an FSSpec for its direct parameter.  Instead, you
should take the FSSpec you create, make an alias (NewAlias or NewAliasMinimal), and package the
- --ALIAS-- into the AppleEvent.  Then it should work like a charm.

Eric
- --
Internet: [e-mail]
AppleLink: BERDAHL
MaBell: [phone]


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

From: [e-mail] (Steve Dorner)
Subject: A/UX 2.0.1 problem
Date: 18 Mar 92 19:01:02 GMT
Organization: University of Illinois at Urbana-Champaign


I have a little program (MPW C 3.2):

  #include "itall.h"
  main()
  {
    long resp;
    if (!Gestalt(gestaltOSAttr,&resp) && resp&(1L<<gestaltLaunchControl))
    {
      ProcessSerialNumber me, him;
      GetCurrentProcess(&me);
      GetFrontProcess(&him);
    }
  }

This little program crashes (bus error) under A/UX 2.0.1, when it gets to the
"GetFrontProcess" call.

Am I doing something wrong, or is this an A/UX bug?  A little birdie tells
me it's probably an A/UX 2.0.1 bug that's fixed in A/UX 3.0, but I'm still
curious.  I was unable to find any warnings about A/UX and the process
manager in IM or the technotes (at least the ones I have).  Besides,
I thought Gestalt would tell me all I needed to know?

- --
Steve Dorner, U of Illinois Computing Services Office
Internet: [e-mail]  UUCP: uunet!uiucuxc!uiuc.edu!s-dorner

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

From: [e-mail] (JohnC)
Organization: The Royal Society for Putting Things on Top of Other Things
Date: Wed, 18 Mar 1992 21:27:00 GMT

In article <[e-mail]> [e-mail] (Steve Dorner) writes:
>
>I have a little program (MPW C 3.2):
>
>  #include "itall.h"
>  main()
>  {
>    long resp;
>    if (!Gestalt(gestaltOSAttr,&resp) && resp&(1L<<gestaltLaunchControl))
>    {
>      ProcessSerialNumber me, him;
>      GetCurrentProcess(&me);
>      GetFrontProcess(&him);
>    }
>  }
>
>This little program crashes (bus error) under A/UX 2.0.1, when it gets to the
>"GetFrontProcess" call.

Apple has stated that A/UX versions prior to 3.0 don't support any Mac OS
features specific to System 7.  I have no clue as to why Gestalt lied in
this case.  Maybe it is being confused by the A/UX implementation of the
early proto-Process Manager inside Multifinder?


- --
John Cavallino                  |  EMail: [e-mail]
University of Chicago Hospitals |         [e-mail]
Office of Facilities Management | USMail: 5841 S. Maryland Ave, MC 0953
B0 f++ c+ g+ k s+(+) e+ h- pv   |         Chicago, IL  60637

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

From: [e-mail] (Jon W{tte)
Date: 19 Mar 92 12:12:59 GMT
Organization: Royal Institute of Technology, Stockholm, Sweden

> [e-mail] (Steve Dorner) writes:

   I have a little program (MPW C 3.2):

       {
     ProcessSerialNumber me, him;
     GetCurrentProcess(&me);
     GetFrontProcess(&him);
       }

   This little program crashes (bus error) under A/UX 2.0.1, when it gets
   to the "GetFrontProcess" call.

That's not surprising. GetFrontProcess looks like a system-7-process-
manager call. A/UX 2.0.1 only has MultiFinder 6.1...

Maybe you test for the wrong gestalt bit, or gestalt returns erraneous
results ?

- --
This signature is placed into the Public Domain by Jon W{tte ([e-mail])
                     - The worlds only romantic cynic -

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

From: [e-mail] (Jochen Meyer)
Subject: SndPlayFromDisk buffer size?
Date: 19 Mar 92 16:07:12 GMT
Organization: University of Oldenburg, Germany

I am writing an application which is intended to play sampled sounds with
as little memory usage a possible. The play-from-disk routines seemed to
be the solution. Yet they require some buffer in memory.

My question: Which size should this buffer be?

In Inside Macintosh, Vol. 6, there is just some sample code which has a
buffer of 16k, but this seems to be too little for big sounds!?

It might be worth mentioning that the sounds to be played back are
1:6 compressed.

Thanks for answers

Jochen

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

From: [e-mail] (Lawrence D'Oliveiro, Waikato University)
Date: 20 Mar 92 09:25:58 +1200
Organization: University of Waikato, Hamilton, New Zealand

In article <[e-mail]>, [e-mail] (Jochen Meyer) writes:
> I am writing an application which is intended to play sampled sounds with
> as little memory usage a possible. The play-from-disk routines seemed to
> be the solution. Yet they require some buffer in memory.
>
> My question: Which size should this buffer be?
>
> In Inside Macintosh, Vol. 6, there is just some sample code which has a
> buffer of 16k, but this seems to be too little for big sounds!?

It doesn't matter about the size of the sound. The Sound Manager will
just read a bufferful at a time off disk. Bigger buffers => fewer reads,
but more memory consumed. The tradeoff is largely up to you.

> It might be worth mentioning that the sounds to be played back are
> 1:6 compressed.

I have successfully played both compressed and uncompressed sounds of
several tens of seconds duration (i e lots bigger than would fit in my
buffers), using SndStartFilePlay. You just call it the way it's documented,
and let it figure out all the low-level details.

Lawrence D'Oliveiro                       fone: [phone]
Computer Services Dept                     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] (rob oldham)
Subject: Think C 4.0: debugger crashes on LC
Organization: Dept. Computer Science, University of Western Australia.
Date: Fri, 20 Mar 1992 03:40:18 GMT

  I am running ThinkC 4.0 on my LC running system 7. For some reason
the debugger refuses to work - it hangs the machine every time. It
was also doing the same thing when I was running system 6.x when I
originally got the LC.

  I can't quite afford an upgrade to ThinkC 5.0 or whatever, but
would really like a quick fix if it exists.

Thanks a heap,

Rob.


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

From: [e-mail] (Ernest J Hill)
Organization: Massachusetts Institute of Technology
Date: Fri, 20 Mar 1992 05:00:34 GMT

This really should be in the FAQ list by now.  Think C 4.0 is NOT 100% Sys 7
compatible. In particular, the debugger is hopeless.  There are free updates
downloadable from (I think) sumex, and certainly from other sites (that's how
I got mine, by ftp.) The upgrade also fixes some problems in the ANSI scanf()
functions, and some TCL stuff too. The last 4.0 upgrade is numbered 4.0.5.

                        --Foss Hill







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

From: [e-mail] (Nicholas Jackiw)
Subject: Apple Event Max Size Query
Date: 24 Feb 92 18:09:33 GMT
Organization: Whole Earth 'Lectronic Link, Sausalito, CA


I'm trying to send large amounts of data across the network with
Apple Events.  The data contents are microphone-recorded sound,
so I have little ability to predetermine their length.  It seems
that my client app fails to send the event if its too large (bigger
than 64K, empirically); AESend returns error -92 (ddpLenErr).  So
apparently I need to break up the sound into multiple events and
collate them on the server.

My question is, how can the application determine the maximum size?
Is 64K a constant enforced by the AEMgr, or does it depend on some
lower-level transport that might vary?  Or is it configuration
dependent, like how much room the system heap has available in which
to grow?

I'd like to send as large a unit as possible,  under the assumption
that there is lower-level chunkification going on anyway.  I can't
find any discussion of maximum message size in the AEMgr or High
Level Event Mgr documentation of IM#6. I can post code if it would
clarify, but it's a simple AESend()...

As always, thanks...



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

From: [e-mail] (Ed Lai)
Subject:  Apple Event Max Size Query
Date: 24 Feb 92 20:10:41 GMT
Organization: Apple Computer Inc., Cupertino, CA

In article <[e-mail]> [e-mail] (Nicholas Jackiw) writes:
>
>I'm trying to send large amounts of data across the network with
>Apple Events.  The data contents are microphone-recorded sound,
>so I have little ability to predetermine their length.  It seems
>that my client app fails to send the event if its too large (bigger
>than 64K, empirically); AESend returns error -92 (ddpLenErr).  So
>apparently I need to break up the sound into multiple events and
>collate them on the server.
>
>My question is, how can the application determine the maximum size?
>Is 64K a constant enforced by the AEMgr, or does it depend on some
>lower-level transport that might vary?  Or is it configuration
>dependent, like how much room the system heap has available in which
>to grow?
>
>I'd like to send as large a unit as possible,  under the assumption
>that there is lower-level chunkification going on anyway.  I can't
>find any discussion of maximum message size in the AEMgr or High
>Level Event Mgr documentation of IM#6. I can post code if it would
>clarify, but it's a simple AESend()...
>
>As always, thanks...

There is a 64K limit due to the High Level Event Manager.

/* Disclaimer: All statments and opinions expressed are my own */
/* Edmund K. Lai                                               */
/* Apple Computer, MS37-UP                                     */
/* 20525 Mariani Ave,                                          */
/* Cupertino, CA 95014                                         */
/* [phone]                                               */
zW@h9cOi




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

From: [e-mail] (Paul Potts)
Subject:  Apple Event Max Size Query
Date: 25 Feb 92 20:18:33 GMT
Organization: Instructional Technology Laboratory, University of Michigan

In article <[e-mail]> [e-mail] (Nicholas Jackiw) writes:
>
>I'm trying to send large amounts of data across the network with
>Apple Events.  The data contents are microphone-recorded sound,
>so I have little ability to predetermine their length.  It seems
>that my client app fails to send the event if its too large (bigger
>than 64K, empirically);

>From my conversation with Jon Pugh I have heard that AppleEvents cannot
be larger than 64K. This is apparently a bug (or missing feature) and
they are indeed supposed to carry more data.

I'd be interesting in hearing about what you're doing. I have written
code which sends commands to play sounds over the network, and the next
step I was considering was sending digitized sounds (adding data chunks
to the AE).

-Paul-


--
         -Paul [e-mail]-
"Wrong is wrong, even if it helps you." - Popeye



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

From: [e-mail] (Keith Rollin)
Subject:  Apple Event Max Size Query
Date: 2 Mar 92 22:01:43 GMT
Organization: Apple Computer Inc., Cupertino, CA

In article <[e-mail]> [e-mail] (Paul Potts) writes:
>In article <[e-mail]> [e-mail] (Nicholas Jackiw) writes:
>>
>>I'm trying to send large amounts of data across the network with
>>Apple Events.  The data contents are microphone-recorded sound,
>>so I have little ability to predetermine their length.  It seems
>>that my client app fails to send the event if its too large (bigger
>>than 64K, empirically);
>
>From my conversation with Jon Pugh I have heard that AppleEvents cannot
>be larger than 64K. This is apparently a bug (or missing feature) and
>they are indeed supposed to carry more data.

Paul,

I think that you might have misunderstood your conversation with Jon.
Yes, its true that you cannot send more than 64K of data with Apple
events, but there is nothing in the Apple event Manager, per se, that
makes this limitation. Nor is it a bug. Rather, the Apple event Manager
currently uses the Hight Level Event Manager for its transport
mechanism. It is the High Level Event Manager that imposes the 64K
limit, which is in accordance with the design of the HLEventMgr. If the
transport layer were to support more than 64K, then so would the Apple
event Manager. However, there is no bug or missing feature; it's just
a matter of the weakest link in the chain.

--
- ----------------------------------------------------------------------------
Keith Rollin           ---            <Taligent .signature under construction>
Disclaimer: Pretty soon, I really _won't_ be speaking for Apple...



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

From: [e-mail] (Jon Pugh)
Date: 5 Mar 92 20:51:10 GMT
Organization: Apple Co.

In article <[e-mail]>, [e-mail] (Paul Potts) writes:
>
> From my conversation with Jon Pugh I have heard that AppleEvents cannot
> be larger than 64K. This is apparently a bug (or missing feature) and
> they are indeed supposed to carry more data.
>

The High Level Event Manager is the one imposing the 64K restriction for all
events and a lot of us want it fixed.

Jon

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

From: [e-mail] (Paul Potts)
Date: 5 Mar 92 16:45:36 GMT
Organization: Instructional Technology Laboratory, University of Michigan

In article <[e-mail]> [e-mail] (Keith Rollin) writes:
>In article <[e-mail]> [e-mail] (Paul Potts) writes:
>
>Paul,
>
>I think that you might have misunderstood your conversation with Jon.

I did misunderstand. My impression was that the bottleneck was in the AE
Manager. (I haven't read all of I-M 6 yet, so I haven't gotten to the
innards of the High-Level Event Manager...)

Jon also said, I believe, that it would probably be fixed, but
he couldn't say when.

>Yes, its true that you cannot send more than 64K of data with Apple
>events, but there is nothing in the Apple event Manager, per se, that
>makes this limitation. Nor is it a bug.

Well, missing feature then?

>Rather, the Apple event Manager
>currently uses the Hight Level Event Manager for its transport
>mechanism. It is the High Level Event Manager that imposes the 64K
>limit, which is in accordance with the design of the HLEventMgr.

As Gag Halfrunt, Zaphod Beeblebrox's brain-care
specialist says, "Twue but Pwobably Iwwevewant..."  this statement is
somewhat tautological...

I didn't mean to insult the system software... I like System 7. Do you know
if there are plans for new transport mechanisms for the AE Mgr that might
get around this limitation?

(If you can't talk about it, point to it...)
- --
         -Paul [e-mail]-
"A man must be a little mad if he does not want to be even more stupid."
- -Montaigne

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

From: [e-mail] (Jon Pugh)
Date: 19 Mar 92 21:25:36 GMT
Organization: Apple Co.

In article <[e-mail]>, [e-mail] (Paul Potts) writes:
> >Yes, its true that you cannot send more than 64K of data with Apple
> >events, but there is nothing in the Apple event Manager, per se, that
> >makes this limitation. Nor is it a bug.
>
> Well, missing feature then?
>
> >Rather, the Apple event Manager
> >currently uses the Hight Level Event Manager for its transport
> >mechanism. It is the High Level Event Manager that imposes the 64K
> >limit, which is in accordance with the design of the HLEventMgr.
>
> As Gag Halfrunt, Zaphod Beeblebrox's brain-care
> specialist says, "Twue but Pwobably Iwwevewant..."  this statement is
> somewhat tautological...
>
> I didn't mean to insult the system software... I like System 7. Do you know
> if there are plans for new transport mechanisms for the AE Mgr that might
> get around this limitation?

First off, it is a "design limitation".

Second, the transport mechanism is completely seperate. AE runs on ADSP which
also has no 64K limit.  The problem lies in the event manager.  You use an
event, you get this limitation.  We could make AE run on IP and it would
still be limited to 64K.

Personally, I think the EPPC needs a minor kick to the head. ;)

Jon

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

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