Tricks of the Mac Game Programming Gurus/TricksOfTheMacGameProgrammingGurus.iso/Information/CSMP Digest/volume 1/csmp-v1-190.txt

Text File | 1994-12-08 | 41.8 KB | 1,119 lines | [TEXT/R*ch]

C.S.M.P. Digest             Tue, 20 Oct 92       Volume 1 : Issue 190

Today's Topics:

    Does THINK C have glue for FindFolder?
    How do I see if modifiers are down?
    Editor for mini icons?
    Prolog for Mac
    ATrap to Determe my Mac name?
    AUX and memory
    How do I sleep w/o using any CPU time
    Naming my print job: how?
    alias pathname



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

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

Each issue of the digest contains one or more sets of articles (called
threads), with each set corresponding to a 'discussion' of a particular
subject.  The articles are not edited; all articles included in this digest
are in their original posted form (as received by our news server at
cs.uoregon.edu).  Article threads are not added to the digest 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 digest).  Article threads that
consist of only one message are generally not included in the digest.

The entire digest is available for anonymous ftp from ftp.cs.uoregon.edu
[128.223.8.8] in the directory /pub/mac/csmp-digest.  Be sure to read the
file /pub/mac/csmp-digest/README before downloading any files.  The most
recent issues are available from sumex-aim.stanford.edu [36.44.0.6] in the
directory /info-mac/digest/csmp.  If you don't have ftp capability, the sumex
archive has a mail server; send a message with the text '$MACarch help' (no
quotes) to [e-mail] for more information.

The digest is 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 issue as it is created.  Sorry, back issues
are not available through the mailing list.

Send administrative mail to [e-mail].


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

From: [e-mail] (Pondscum Czar)
Subject: Does THINK C have glue for FindFolder?
Organization: Yale CS Mail/News Gateway
Date: Sun, 2 Aug 1992 19:57:27 GMT


Hi... I am trying not to use SysEnvirons to find the system folder (as,
really, SysEnvirons has been replaced w/ Sys7).  However, I need to make
sure this App works under System 6 as well... Anyway, I was poking
through the header files and found this in Folders.h:

- --snip here--
#if SystemSevenOrLater
pascal OSErr FindFolder(short vRefNum,OSType folderType,Boolean
createFolder,
    short *foundVRefNum,long *foundDirID)
    = {0x7000,0xA823};
#else
pascal OSErr FindFolder(short vRefNum,OSType folderType,Boolean
createFolder,
    short *foundVRefNum,long *foundDirID);
#endif
- --snip here--

Does the #else clause mean that THINK C provides glue for this function
like Gestalt?  In other words, can I use this under System 6 to find the
Active System Folder, or do I just stick with SysEnvirons...

Not a big deal, but I guess you could say I am a stickler for details...
:-)

- --Doug
- --
The new Congressmen say they're going to turn the government around.  I
hope I don't get run over again.

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

From: [e-mail] (Mike Dodd)
Date: Mon, 3 Aug 1992 05:15:22 GMT
Organization: Apple Computer Inc.

In article <[e-mail]>, [e-mail] (Pondscum Czar) writes:
>
>
> Hi... I am trying not to use SysEnvirons to find the system folder (as,
> really, SysEnvirons has been replaced w/ Sys7).  However, I need to make
> sure this App works under System 6 as well... Anyway, I was poking
> through the header files and found this in Folders.h:
>
> --snip here--
> #if SystemSevenOrLater
> pascal OSErr FindFolder(short vRefNum,OSType folderType,Boolean
> createFolder,
>     short *foundVRefNum,long *foundDirID)
>     = {0x7000,0xA823};
> #else
> pascal OSErr FindFolder(short vRefNum,OSType folderType,Boolean
> createFolder,
>     short *foundVRefNum,long *foundDirID);
> #endif
> --snip here--
>
> Does the #else clause mean that THINK C provides glue for this function
> like Gestalt?  In other words, can I use this under System 6 to find the
> Active System Folder, or do I just stick with SysEnvirons...
>

yes, you can use it fine. if it doesn't define inline assembly (the
"= {0x7000..." part), as the second definition doesn't, then that means
that it includes glue for it (probably in MacTraps or MacTraps2). and, if
you use something that doesn't have inline assembly or glue, then your
program will refuse to link, so at least you'll know that way.

- -Mike Dodd-
Apple Computer
** What I say is my opinion. Not Apple's. **


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

From: [e-mail] (PEGG, DAVID J)
Subject: How do I see if modifiers are down?
Date: 31 Jul 92 18:45:00 GMT
Organization: University of Tennessee Computing Center

I want to put a feature in my data analysis program that looks to see if Shift
and Option are down (without any other keys).  If they are, then it gets the
mouse position and tells you where it is in the graph's coordinates.  I'm using
GetKeys to see if the keys are pressed.  But I can't seem to get my keyMap to
show anything for any modifier keys (as well as some keys that really surprise
me, like 't' and 'y').  Also, no number keys show up, on the number pad or on
the regular keyboard.

I'm using Think C 5.0.2 with System 7, but programming using mostly stuff from
IM volumes I & II.  I know it has to be possible to do this, since the Key Caps
DA recognizes modifier keys, even when nothing else is pressed.

Thanks in advance for your help,
      Mike

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

From: [e-mail] (John Wallace)
Date: 1 Aug 92 01:27:17 GMT
Organization: TRW Data Systems Center, Redondo Beach, CA

In article <[e-mail]> [e-mail] (PEGG, DAVID J) writes:
>I want to put a feature in my data analysis program that looks to see if Shift
>and Option are down (without any other keys).  If they are, then it gets the
>mouse position and tells you where it is in the graph's coordinates.  I'm using
>GetKeys to see if the keys are pressed.  But I can't seem to get my keyMap to
>show anything for any modifier keys (as well as some keys that really surprise
>me, like 't' and 'y').  Also, no number keys show up, on the number pad or on
>the regular keyboard.
>
>Thanks in advance for your help,
>      Mike


Rather than using the keymap directly, you may want to use
GetNextEvent/WaitNextEvent to grab the next event and check the
event's modifier fields for the shift/option/control keys and the
where field for the mouse location (in global coordinates).

This has several advantages in that you don't have to muck with the
keyboard resources to verify that the user hasn't re-mapped the key
since it is possible for me to swap any of the keys on my keyboard.
Also all of the info you want is available in one simple call.

If you don't want to actually remove any events from the event queue
then pass in 0 for the event mask.

Cheers!
John
- -----
John Wallace     [e-mail]

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

From: [e-mail] (Peter N Lewis)
Organization: NCRPDA, Curtin University
Date: Mon, 3 Aug 1992 01:45:35 GMT

In article <[e-mail]>, [e-mail] (John
Wallace) wrote:

> Rather than using the keymap directly, you may want to use
> GetNextEvent/WaitNextEvent to grab the next event and check the
> event's modifier fields for the shift/option/control keys and the
> where field for the mouse location (in global coordinates).

In fact, OSEventAvail is even better, because it doesn't remove any events
from the queue, and it won't try to switch you out or any foolishness:

dummy := OSEventAvail(0, event);

You can use this to get the mouse position as well, especially if you are a
background only app and want to avoid calling GetMouse which explodes (why
get the mouse position in a backgroun only app?  To test for idleness of
course)

Have fun all,
   Peter.

_______________________________________________________________________
Peter N Lewis, NCRPDA, Curtin University       [e-mail]
GPO Box U1987, Perth WA 6001, AUSTRALIA             FAX: [phone]

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

From: [e-mail] (John Purlia)
Subject: Editor for mini icons?
Date: 30 Jul 92 23:38:26 GMT
Organization: Qualcomm, Inc.

Does anyone know of any utilities that allows for the creation of color
icm# resources?  I'd like to be able to edit mini icons (12 x 12) in the
same manner I edit larger icon families in ResEdit.

Thanks!

...........................................................................
John Purlia          : My brain; not my company's brain.  My brain says...
[e-mail] :      "Just about any movie could be made better
AppleLink:  AM0470   :       if one of the characters were a vampire."
...........................................................................

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

From: [e-mail] (Chris DiGiano)
Date: 3 Aug 92 20:59:30 GMT

[e-mail] (John Purlia) writes:

>Does anyone know of any utilities that allows for the creation of color
>icm# resources?  I'd like to be able to edit mini icons (12 x 12) in the
>same manner I edit larger icon families in ResEdit.

I was looking for the same thing about a month ago.  It was suggested
to me by Tim Dierks at Apple Computer to use the ResEdit 16x16 editor
and then extract the left 12x12 pixels for the mini icon.  I wrote a
utility program which takes an ICS#, ics4, or ics8 icon and converts
it to the appropriate icm icon, by dropping off the last four rows of
pixels and getting rid of the left-most four pixels for each row of
the color icons.  When I drew the mini icons on the screen the results
were a bit strange:

1. Eventhough I used a 12x12 destination rectangle, the icm was
   squashed horizontally.

2. The color icons had a few pixels in the correct place with the
   correct color, but most of it was garbled and had colors which were
   not part of the original ics icons.

I changed my utility program to copy the entire row of 16 colored
pixels, not just the first 12.  This seemed to fix the second
problem--the colors were right and most pixels were in the right
place.  But now just like their black and white cousins, the icon was
compressed along the horizontal axis.

When I converted the B&W small icon below

X
 X
  X
   X
    X
     X
      X
       X
    X
     X
      X
       X

It got a mini icon which looked like this on the screen:

X
 X
 X
   X
    X
    X
      X
       X
       X
     X
         X
       X

That's all the progress I made before giving up and just using small
icons in my application instead.  Let me know if you have any more luck.

- ---
Chris DiGiano  [phone], fax: 0458         [e-mail].{ca,bitnet}
Dynamic Graphics Project, Dept. of Computer Sci.          [e-mail]
University of Toronto, Canada   M5S 1A4         ...!uunet!dgp.toronto.edu!digi

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

From: [e-mail] (John Purlia)
Date: 3 Aug 92 22:19:29 GMT
Organization: Qualcomm, Inc.

In article <[e-mail]>,
[e-mail] (Chris DiGiano) wrote:
>
> [e-mail] (John Purlia) writes:
>
> >Does anyone know of any utilities that allows for the creation of color
> >icm# resources?  I'd like to be able to edit mini icons (12 x 12) in the
> >same manner I edit larger icon families in ResEdit.
>
> I was looking for the same thing about a month ago.  It was suggested
> to me by Tim Dierks at Apple Computer to use the ResEdit 16x16 editor
> and then extract the left 12x12 pixels for the mini icon.  I wrote a
> utility program which takes an ICS#, ics4, or ics8 icon and converts
> it to the appropriate icm icon, by dropping off the last four rows of
> pixels and getting rid of the left-most four pixels for each row of
> the color icons.  When I drew the mini icons on the screen the results
> were a bit strange:

Actually, I came up with a compromise solution and basically handled my
icons the same way the system does.  I created my 12x12 icons using
ResEdit's 16x16 editor using the middle 12x12 pixels, then plotted the
small icons with the destination rectangle overlapping my intended region
by two pixels on each side.  Take a look at the SICN resources in the
system used to plot icons in the Finder.  These icons are 16x16 but only
use the middle 12x12 pixels.  I did the same thing, only using an icon
family and it works MUCH better than plotting a full 16x16 icon within a
12x12 destination Rect.
...........................................................................
John Purlia          : My brain; not my company's brain.  My brain says...
[e-mail] :      "Just about any movie could be made better
AppleLink:  AM0470   :       if one of the characters were a vampire."
...........................................................................

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

From: [e-mail] (Karl Zerangue)
Subject: Prolog for Mac
Organization: gte
Date: Mon, 27 Jul 1992 21:57:47 GMT

Does anyone know of a Prolog interpreter for the Macintosh?

Thanks in advance!

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

From: [e-mail] (Per Mildner)
Date: 28 Jul 92 17:32:55 GMT
Organization: Computing Science Dept.,Uppsala University, Sweden


We have ported our Prolog interpreter/compiler to the Mac. It was
available on info-mac but has been removed. It is still available on
some mirror sites. It is also available by anonymous ftp to
ftp.csd.uu.se (directory /pub/Tricia/Mac/) where there is one extra
version for use with 68000 processors (e.g., Classics).


It is a WAM based vanilla prolog with no fancy Mac specific stuff
except ADSP stream support. The ADSP support have been used to
implement fancy user interfaces for expert systems using SuperCard.

Any questions should go to [e-mail] (This includes
requests for the source code).
Regards,
- --
Per Mildner            [e-mail]
Computing Science Dept.        tel: [phone]
Uppsala University, Sweden    fax: [phone]

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

From: [e-mail] (Robert Marti)
Date: 3 Aug 92 16:16:43 GMT
Organization: Dept. Informatik, Swiss Federal Institute of Technology (ETH), Zurich, CH

In article <[e-mail]>, [e-mail]
(Tony Wingo) writes:
|>
|> In article <[e-mail]>, [e-mail]
|> (Karl Zerangue) wrote:
|> >
|> > Does anyone know of a Prolog interpreter for the Macintosh?
|>
|> Check out Open Prolog, available from
|>    grattan.cd.tcd.ie
|> Last I checked it was in the directory tce/mac.


Make that grattan.cs.tcd.ie .
                  ^^

Also, Open Prolog is now in the directory languages/open-prolog.
The curent version is 1.0d35.

- --
Robert Marti                    |  Phone:    [phone]
Informationssysteme             |  FAX:      [phone]
ETH-Zentrum                     |  E-Mail:   [e-mail]
CH-8092 Zurich, Switzerland     |

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

From: [e-mail] (Lawrence D. Hare)
Subject: ATrap to Determe my Mac name?
Date: 30 Jul 92 17:57:45 GMT
Organization: Control Data Systems

Well - I give up. Assuming that the name my Mac has (System 6 & 7) is the
same
as in a Unix /etc/hosts file, which it jolly well better be, I need to
figure out
what the name of the Mac I am running on is.

I have looked, and looked, and looked.... Surely there is a dead simple
call!

Anyone know what it is and where it is documented?
- --
Lawrence D. Hare       Control Data - Silicon Valley Operations
Consultant             Voice: [phone] - C/N [234] 4339
[e-mail]        Mail: SVLa60  FAX: [phone]

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

From: [e-mail] (Michael Peirce)
Date: 31 Jul 92 05:33:05 GMT
Organization: Peirce Software


In article <[e-mail]> (comp.sys.mac.programmer), [e-mail] (Lawrence D. Hare) writes:
> Well - I give up. Assuming that the name my Mac has (System 6 & 7) is the
> same
> as in a Unix /etc/hosts file, which it jolly well better be, I need to
> figure out
> what the name of the Mac I am running on is.
>
> I have looked, and looked, and looked.... Surely there is a dead simple
> call!
>
> Anyone know what it is and where it is documented?

This comes up ever month or so, but here it is again...

Under System 7 there are two name for a Mac.  The User Name (Choose
name under pre-System 7) is found in the STR resource with ID = -16096
in the System File.  The Machine name is found in the STR resource with
ID = -16416.

You can fetch either of these by calling GetString() since the System
File is open in your application.

I don't know how these names relate to the name found in a Unix system
- - I don't do Unix :-)

- --  Michael Peirce      --   [e-mail]
- --  Peirce Software     --   Suite 301, 719 Hibiscus Place
- --                      --   San Jose, California USA 95117
- --  Makers of...        --   voice: [phone] fax: [phone]
- --            SMOOTHIE  --   AppleLink: peirce & America Online: AFC Peirce

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

From: [e-mail] (Mark Lanett)
Date: 31 Jul 92 07:07:28 GMT
Organization: Apple Computer Inc., Cupertino, CA

[e-mail] (Michael Peirce) writes:


>In article <[e-mail]> (comp.sys.mac.programmer), [e-mail] (Lawrence D. Hare) writes:
>> Well - I give up. Assuming that the name my Mac has (System 6 & 7) is the
>> same
>> as in a Unix /etc/hosts file, which it jolly well better be, I need to
>> figure out
>> what the name of the Mac I am running on is.

[about the Chooser name]

>I don't know how these names relate to the name found in a Unix system
>- I don't do Unix :-)

They don't relate. The Chooser name is only used by AppleTalk, not MacTCP.
For that, you need to make MacTCP calls. Alas, I do sockets, not MacTCP,
and I doubt I could extract meaningful code out of the socket library I
use.
- --
Have a bajillion brilliant Jobsian lithium licks.
Mark Lanett, NOT speaking for anyone. Personal opinion only.

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

From: [e-mail] (David Peterson)
Date: 3 Aug 92 17:49:49 GMT
Organization: University of Southern California, Los Angeles, CA


There is no trap as far as I can tell, but if your name resolution
stuff is working correctly just call AddrToName() with the ip number
of your Mac (which you can get from GetMyIPAddr()).

Look in the MacTCP docs for details. GetMyIPAddr is on pg 87,
AddrToName on 75.

- -dave.

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

From: [e-mail] (Jeremy Roussak)
Subject: AUX and memory
Date: 3 Aug 92 19:42:28 GMT
Organization: Computer Science Dept, QMW, University of London

I've written an INIT, called Apollo. To send AEvents, Apollo
uses a small faceless background app called Eagle. All Eagle
does it to open a PPC port and wait for messages from Apollo,
which it converts into AEvents and sends.

Under MacOS, Eagle works perfectly in a 10k partition. Under
AUX, however, it fails with insufficient memory (-108). If the
partition is increased to 15k, it works nicely.

Why should this be?

Jeremy

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

From: [e-mail] (Keith Rollin)
Organization: Taligent
Date: Tue, 4 Aug 1992 01:59:04 GMT

In article <[e-mail]>, [e-mail] (Jeremy
Roussak) writes:
>
> I've written an INIT, called Apollo. To send AEvents, Apollo
> uses a small faceless background app called Eagle. All Eagle
> does it to open a PPC port and wait for messages from Apollo,
> which it converts into AEvents and sends.
>
> Under MacOS, Eagle works perfectly in a 10k partition. Under
> AUX, however, it fails with insufficient memory (-108). If the
> partition is increased to 15k, it works nicely.
>
> Why should this be?

I think that applications are given a larger stack under A/UX. This is just a
vague recollection, and, as a card-carrying Mac Bigot, I certainly don't have an
A/UX system on which to check this out, so you'll have to verify this for
yourself.

- --
Keith Rollin
Phantom Programmer
Taligent, Inc.


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

From: [e-mail] (David L. Crow)
Subject: How do I sleep w/o using any CPU time
Date: 11 Jul 1992 15:17:45 -0500
Organization: UTexas Mail-to-News Gateway

    I am writing an application to check something every so often.  If
  something exists, I will then put up an error window.  Everything is
  done except that I don't know the best way to sleep until the next
  check and use as little system resources (specifically CPU) as
  possible.  In UNIX, I would just sleep and the process would be
  put to sleep.

    I looked in the UNIX compatibility source in THINK C, and they
  implemented a sleep as "Delay".  When I tried Delay, I couldn't
  get control back from the app during the delay.

    I looked in the UUPC source, and they did it by getting the tick
  count and then going through a while loop until so many ticks had
  gone by.  While this works, it seems like it would not be very
  efficient (read MultiFinder friendly) even though I don't have
  any problems with UUPC and the MultiFinder.

    Basically, I guess that I just want to know the best way to put
  my application to sleep for a given amount of time.

    Thanks for any help you can provide!

  David L. Crow    UUCP:  cs.utexas.edu!rpp386!spicewood!crow
               Internet:  [e-mail]

- ---

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

From: [e-mail] (Cory Kempf)
Date: 12 Jul 92 05:55:50 GMT
Organization: EnigamI, Inc., Nashua, NH


In article <[e-mail]> (comp.sys.mac.programmer), [e-mail] (David L. Crow) writes:
>    I am writing an application to check something every so often.  If
>  something exists, I will then put up an error window.  Everything is
>  done except that I don't know the best way to sleep until the next
>  check and use as little system resources (specifically CPU) as
>  possible.  In UNIX, I would just sleep and the process would be
>  put to sleep.

MacOS Pet Peive #375: Delay doesn't give up the CPU.

The way to give up the CPU is to call WaitNextEvent, passing it a
large number of the 3rd parameter (time that WNE can give away).
I usually use 50 or so (there was this bug a while back...).  By putting
this inside a while loop that checks the tickcount, I give away as
much CPU as I can.

When an application is in the background, it doesn't get as much CPU
to start with.  With a loop as described, its impact should be minimal.

+C


- -------------------------------------------------------------
Cory Kempf                    EnigamI, Inc.
[e-mail]           ...!decvax!enigami!cory
Annon:    [e-mail]

"F' cryin' out loud, it's *your* life; shouldn't you at least examine
your lifestyle values to see if they are really what *you* want."
                                        --Keith Lim

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

From: Andreas Wuertz <[e-mail]>
Organization: Swiss Federal Institute of Technology, TIK
Date: Tue, 14 Jul 1992 11:49:05 GMT

In article <[e-mail]> David L. Crow, [e-mail]
writes:
>    Basically, I guess that I just want to know the best way to put
>  my application to sleep for a given amount of time.

Use WaitNextEvent(myEventMask, &theEvent, maxidle, mouseRgn); see inside
mac for exact definition of the parameters for this function. maxidle is
the amount of ticks you want to sleep if no event is there for you; so
you are woken up after maxidle ticks THE LATEST. If you want to make
something periodically, you'd need to write something like:
if WaitNextEvent(myEventMask, &theEvent, maxidle, mouseRgn) {
  if (theEvent.when-lastWakeup>=desiredIdle) {
    HandlePeriodicAction();
    maxIdle=desiredIdle;
  } else
    maxIdle=desiredIdle-(TickCount()-lastWakeup);
  lastWakeup=TickCount();
};

Not exactly, but I hope you get the idea. Hope this helps.

Andy

Andy from TIK

=======================================================
The earliest time for a program to be bug-free is,
when it's out of date. The latest time for a programmer
to retire is, when he writes such a program.
=======================================================

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

From: [e-mail] (Pete Resnick)
Organization: University of Illinois at Urbana
Date: Tue, 14 Jul 1992 13:54:23 GMT

[e-mail] (David Peterson) writes:

>How about during completion procs -- can't call WNE then.

>What I've been doing is parking in a SystemTask/TickCount loop. Will this
>allow things to go on in the background? Will it allow other completion
>procs to execute? (Skip reentrancy problems, I _think_ that's taken care
>of)

Eeeee! This is a design problem, Dave.

If you are getting called at interrupt and it is not time to do what
you have to do, you should do something that will cause another
interrupt later when it is time and return immediately. If I had to
wait for a certain time to roll around when my completion proc gets
called I would do the following:

My completion proc does only one thing: install a Time Manager task
that is set to go off at the appropriate time. My completion routine
would simply figure out how much time I need to wait and install the
TM task.

The completion routine for the TM task would do all of the work I need
to do. If I determined in my original completion proc that I didn't
need to wait at all, I could always call the TM task completion
routine directly from there.

Things that execute at interrupt are doing so because "they must
execute now and fast", as far as the OS is concerned. Do as little as
possible during interrupt and don't hang on to the processor.

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

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

From: [e-mail] (Lawrence D'Oliveiro, Waikato University)
Date: 15 Jul 92 18:07:17 +1200
Organization: University of Waikato, Hamilton, New Zealand

In article <[e-mail]>, Jerome Chan <[e-mail]> writes:
> In article <[e-mail]> Cory Kempf,
> [e-mail] writes:
>>The way to give up the CPU is to call WaitNextEvent, passing it a
>>large number of the 3rd parameter (time that WNE can give away).
>>I usually use 50 or so (there was this bug a while back...).  By putting
>>this inside a while loop that checks the tickcount, I give away as
>>much CPU as I can.

The bug was in MultiFinder 1.0, System 4.2. i e it's long gone.

>   Hmm. But won't we be losing events if we do this? If windows are being
> displayed and the user switches to another application under
> multi-finder, won't we lose the the Update Event? Or is that the
> Resume/Suspend Event?

It's the Resume/Suspend event, and I'm pretty sure you're right--I recall
reading this in the MultiFinder documentation a while back, though I'm having
difficulty finding the same info in IM6 chapter 5. Basically if you mask out
app4/OS events, you won't see the suspend/resume events--they simply disappear.

You could of course put in calls to your regular suspend/resume event handling
code in the delay loop. Another technique involves having only a single event
loop in your program, and being able to selectively enable/disable handling
of particular event types (e g mouse and key clicks) when your program is in a
"busy" state. And being able to specify a "background" task to perform
each time round the event loop.

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+12:00
The aliens are willing to spare the Earth, on condition we surrender to them
the country responsible for Kylie Minogue.

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

From: [e-mail]
Organization: Actrix Information Exchange
Date: Wed, 15 Jul 1992 12:33:58 GMT

In article <[e-mail]> [e-mail] (Lawrence D'Oliveiro, Waikato University) writes:
> The aliens are willing to spare the Earth, on condition we surrender to them
> the country responsible for Kylie Minogue.

... not to mention "shrimp on the barbie" and the term of endearment "bogan".
- --
[e-mail]   Twisted pair: [phone]
BIX: brucehoult                 Last Resort:  PO Box 4145 Wellington, NZ
"Cray's producing a 200 MIPS personal computer with 64MB RAM and a 1 GB
hard disk that fits in your pocket!"   "Great!  Is it PC compatable?"

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

From: [e-mail] (Thomas Chin)
Organization: Wolfram Research, Inc.
Date: Tue, 14 Jul 1992 16:15:50 GMT

>In article <[e-mail]> David L. Crow, [e-mail]
>writes:
>>    Basically, I guess that I just want to know the best way to put
>>  my application to sleep for a given amount of time.

With system 7 process manager, you can have a time manager task which do
a 'WakeUpProcess', install the timer, go into your yielding 'WaitNextEvent'
with maximum sleep ticks and empty mask, you'll get a NULL event when the
time is up!

- -Tom

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

From: [e-mail] (Joshua Rabinowitz-Summer-91)
Date: 24 Jul 92 00:09:28 GMT
Organization: NASA/ARC Information Sciences Division

In article <[e-mail]> [e-mail] (David W. Berry) writes:
>In article <[e-mail]>, you write:
>>    I am writing an application to check something every so often.  If
>>  something exists, I will then put up an error window.  Everything is
>Typed in off the type of my head, so this isn't error checked (or really
>even syntax checked) but it will give you the general idea...
>
>pascal void Delay(long ticks)
>{
>        long            wakeTime = TickCount() + ticks;
>        long            delay;
>        EventRecord     event;
>
>        while((delay = (wakeTime - TickCount()) > 0)
>                WaitNextEvent(0, &event, delay, 0);
>}

The only problem with this, as I remember it
(and please correct me if I'm wrong),
that apple events cannot be masked.  I looked in inside mac for this, but
could not find a reference to this ?fact?.  (looked at The Event Manager,
IM VI, p. 5-29)  I cannot recall where I got this idea from.

Can anyone refute or back up this belief?
- --
- ----------------------------------
#include <std/disclaimer.h>     Josh Rabinowitz, Mac TCL programmer
[e-mail]
"'I see', said the blind carpenter, as he picked up his hammer and saw".

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

From: [e-mail] (David W. Berry)
Date: 17 Jul 92 21:19:45 GMT
Organization: Greenwing Enterprises

In article <[e-mail]>, you write:
>    I am writing an application to check something every so often.  If
>  something exists, I will then put up an error window.  Everything is
Typed in off the type of my head, so this isn't error checked (or really
even syntax checked) but it will give you the general idea...

pascal void Delay(long ticks)
{
        long            wakeTime = TickCount() + ticks;
        long            delay;
        EventRecord     event;

        while((delay = (wakeTime - TickCount()) > 0)
                WaitNextEvent(0, &event, delay, 0);
}

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

From: [e-mail] (Tom Dowdy)
Date: 20 Jul 92 15:15:13 GMT
Organization: Apple Computer, Inc.

In article <[e-mail]>, [e-mail]
(Jerome Chan) wrote:
>
> In article <[e-mail]> Cory Kempf,
> [e-mail] writes:
> >The way to give up the CPU is to call WaitNextEvent, passing it a
> >large number of the 3rd parameter (time that WNE can give away).
> >I usually use 50 or so (there was this bug a while back...).  By putting
> >this inside a while loop that checks the tickcount, I give away as
> >much CPU as I can.
>
>   Hmm. But won't we be losing events if we do this? If windows are being
> displayed and the user switches to another application under
> multi-finder, won't we lose the the Update Event? Or is that the
> Resume/Suspend Event?

Actually, you won't be losing events - however, there is a potential
serious
problem with calling WNE this way all of the time.  Note that in *this*
particular case, the original poster (long since gone from this discussion,
I would fear) wasn't going to have any windows open - he was simply running
until an event happened, at which time a dialog box would be brought
up (no doubt saying something such as "Your rendering is complete").

However, if you *do* have windows open, and fail to respond to update
events, you will see very sluggish performance on your Mac as the
process manager attempts to give you updateEvents over and over again.
Eventually, when it sees some very large number of update events go
by for the same window, it will assume you are clueless and clear
the updateRgn itself - however, this takes almost a minute and a half.

The thing to learn is - always handle update events, and always call
BeginUpdate()/EndUpdate() when handling them.

 Tom Dowdy                 Internet:  [e-mail]
 Apple Computer MS:81KS    UUCP:      {sun,voder,amdahl,decwrl}!apple!dowdy
 20525 Mariani Ave         AppleLink: DOWDY1
 Cupertino, CA 95014
 "The 'Ooh-Ah' Bird is so called because it lays square eggs."

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

From: [e-mail] (Cory Kempf)
Date: Fri, 24 Jul 92 11:52:07 EST
Organization: EnigamI, Inc., Nashua, NH


In article <dowdy-200792081218@90.10.20.25> (comp.sys.mac.programmer), [e-mail] (Tom Dowdy) writes:
>In article <[e-mail]>, [e-mail]
>(Jerome Chan) wrote:
>>
>> In article <[e-mail]> Cory Kempf,
>> [e-mail] writes:
>> >The way to give up the CPU is to call WaitNextEvent, passing it a
>> >large number of the 3rd parameter (time that WNE can give away).
>> >I usually use 50 or so (there was this bug a while back...).  By putting
>> >this inside a while loop that checks the tickcount, I give away as
>> >much CPU as I can.
>>
>>   Hmm. But won't we be losing events if we do this? If windows are being
>> displayed and the user switches to another application under
>> multi-finder, won't we lose the the Update Event? Or is that the
>> Resume/Suspend Event?
>
>Actually, you won't be losing events - however, there is a potential
>serious
>problem with calling WNE this way all of the time.  Note that in *this*
>particular case, the original poster (long since gone from this discussion,
>I would fear) wasn't going to have any windows open - he was simply running
>until an event happened, at which time a dialog box would be brought
>up.

No, I'm still here, just a bit behind.

And yes, when I do actually do a delay like that, I am usually initializing
my application, which (in many cases) requires that I wait for some
hardware to respond.  The only window open at this point is the splash
screen.  Normally though there is the possibility that the user would wish
to interact with them or make a selection from a menu (like Quit!), so I
would call my Yield(long time) routine instead. Yield() repeatedly calls
my main event function (the looping is done externaly) until 'time' had
passed. This has the added benefit of making sure the mouse was correct.

Some other magic was in place to prevent re-entrancy.

Usually though, if I am doing something that is long and involved,
I would put up a dialog box with a thermometer in it (movable modal,
of course!) to let the user know how much longer until I was finished.

>However, if you *do* have windows open, and fail to respond to update
>events, you will see very sluggish performance on your Mac as the
>process manager attempts to give you updateEvents over and over again.
>Eventually, when it sees some very large number of update events go
>by for the same window, it will assume you are clueless and clear
>the updateRgn itself - however, this takes almost a minute and a half.

Would this still happen if the first parameter to WNE was 0, indicating
that I didn't want ANY events?

+C


- -------------------------------------------------------------
Cory Kempf                    EnigamI, Inc.
[e-mail]           ...!decvax!enigami!cory
Annon:    [e-mail]

"F' cryin' out loud, it's *your* life; shouldn't you at least examine
your lifestyle values to see if they are really what *you* want."
                                        --Keith Lim

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

From: [e-mail] (John T. Nelson)
Date: 4 Aug 92 03:29:43 GMT
Organization: Advanced Decision Systems, Mtn. View, CA [phone]

Why not use the time manager to schedule tasks?  Delay loops are crude.

- - Hacker

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

From: [e-mail] (Paul Robichaux)
Subject: Naming my print job: how?
Date: 4 Aug 92 05:26:31 GMT
Organization: Fairgate Technologies

After a bit of hacking to make my plotting and printing code work
properly (thanks to Apple for their source code snippets, especially
HairLines), I'd like to dress up my printing routines a bit.

Right now, my app has one main window (titled "Control"), and whenever
I print a document from it, the print job has that same title. I know
it's possible to tag the job with an arbitrary title ("foo @ HellWriter"
has sort of a nice ring to it) and would like to add this feature to
my application, so that I can identify the specific kind of plot or
graph being printed instead of just "Control."

I'll post a summary of responses.

- -Paul

- ---
                             Paul Robichaux, KD4JZG
         [e-mail] -or- ...uunet!{sci34hub,xavax}!fairgate!paul

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

From: [e-mail] (Charles B. Cranston)
Organization: UM Home for the Terminally Analytical
Date: Tue, 4 Aug 1992 21:13:46 GMT

In article <[e-mail]>, [e-mail] (Paul Robichaux)
wrote:

> Right now, my app has one main window (titled "Control"), and whenever
> I print a document from it, the print job has that same title. I know
> it's possible to tag the job with an arbitrary title ("foo @ HellWriter"
> has sort of a nice ring to it) and would like to add this feature to
> my application, so that I can identify the specific kind of plot or
> graph being printed instead of just "Control."

As I understand it the printing complex takes the frontmost window's
title as the job name.  I suppose you could always temporarily change
the window's title for just the duration of the printing.

[e-mail]     -KA3ZDF

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

From: [e-mail] (Eric C. Kofoid)
Subject: alias pathname
Date: 3 Aug 92 22:18:32 GMT
Organization: Dept. Biology, University of Utah

I want a getFile dialog to return the pathname to an alias itself, *not*
the target of an alias. Since StandardGetFile always resolves the alias,
I decided to use a call to CustomGetFile. My dialog hook function traps
the pseudo-items "sfHookOpenAlias" and "sfHookGoToaliasTarget", but I
discover that I haven't the slightest idea what to do at this point.
Simply returning a null event at this point selects nothing in the dialog
list. Any ideas?

I want my users to be able to create a default list of of input files
which
will be maintained during future activations of the program. Say that one
of the files lives on a file server. When users wish to add this to the
list, they use a getFile dialog. This in turn resolves the alias and adds
its target to the list of files. This is fine, as long as there is a
connection to the file server. However, if naive users activate the
program
when the connection is not there, then they will be informed that the
volume is not mounted or that the file does not exist. If the alias
itself
were named in the list, a login dialog would appear allowing the user to
create the connection immediately (no connection), or the alias target
would be opened in the normal fashion (connection already there).

 __________________________________________________________________
|  Eric Kofoid                         Dept.Biology, U. of Utah    |
|  "Being certain of the conclusion assists in finding the proof"  |
|                                      Galileo, 17th century       |
|__________________________________________________________________|

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

From: [e-mail] (Ross Brown)
Date: 4 Aug 92 14:13:44 GMT
Organization: Bell-Northern Research

In article <[e-mail]> Eric C. Kofoid
<[e-mail]> writes:
>I want a getFile dialog to return the pathname to an alias itself, *not*
>the target of an alias. Since StandardGetFile always resolves the alias,
>I decided to use a call to CustomGetFile. My dialog hook function traps
>the pseudo-items "sfHookOpenAlias" and "sfHookGoToaliasTarget", but I
>discover that I haven't the slightest idea what to do at this point.
>Simply returning a null event at this point selects nothing in the dialog
>list. Any ideas?

You need to "map" sfHookOpenAlias to sfItemOpenButton.
This works for me in MPW C:

pascal short DefeatAliasDialogHook( short item, DialogPtr theDialog,
                                    Ptr myDataPtr )
{
#pragma    unused( theDialog, myDataPtr )
    if( item == sfHookOpenAlias )
        return sfItemOpenButton;
    else
        return item;
}

...
CustomGetFile( (FileFilterYDProcPtr) AliasOnlyFileFilter, -1, nil, &reply,
               0, centerScreen, DefeatAliasDialogHook, nil, nil, nil, nil );
...

Your file filter (first argument) would probably be "nil".

W. Ross Brown                 |        from Brown's Bestiary of the Macintosh:
Advisor, Telemanagement Svcs. |  PBCatMoof:  A purring system trap shaped like
Bell-Northern Research Ltd.   | a dog with cow spots and a 3-hr. battery life.
Ottawa, Ontario, Canada       | ----------------------------------------------
[e-mail]                   |      Any opinion expressed is mine, not BNR's.

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

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