Developer CD Series 1993…ch: Other People's Memory/ADC Developer CD (1993-03) (''Other People's Memory'')_iso/Dev.CD Mar 93.iso/Development Platforms/CSMP Digests/csmp-v1-016.txt

Text File | 1992-11-18 | 51.5 KB | 1,494 lines | [TEXT/MPS ]

C.S.M.P. Digest             Sat, 14 Mar 92       Volume 1 : Issue 16

Today's Topics:

    Sockets now avaiable
    PowerOpen
    C++ for the Mac
    Sending FileSpecifications
    ResEdit 2.1 Bundle Bit and Custom Icon Woes
    getting Vrefnum without using StdGetFile
    Location 0 (Was Re: Blinking box on my screen)
    Think debugger BUG?
    WARNING: Bug in THINK Pascal 4.0.1
    FILE *'s and Mac files
    MPW: deleting resource fork
    Launching Applications
    GetMenu returning wrong ID?
    Resolving aliases for pref files etc
    finding a publisher


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 (try skinner.cs.uoregon.edu if that doesn't work).  This is also the home
of the comp.sys.mac.programmer Frequently Asked Questions list.

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] (Charlie Reiman)
Subject: Sockets now avaiable
Organization: University of Illinois at Urbana
Date: Wed, 12 Feb 1992 02:53:21 GMT

Brief announcement:

The Mac socket library is finally available from ftp.ncsa.uiuc.edu, the
official NCSA ftp server. It is in the directory /misc/unsupported.
Please take special note of the location. It is _not_ in the /Mac
subdirectory! Also, it is in _unsupported_, because it is definately
unsupported.

For the newcomers:

This is a library for MPW 3.2, witten in C, that is designed to emulate
most of the functionality of BSD sockets with regard to TCP/IP
communications. Please note it will not allow you to develop anything
if you do not have a copy of APDA's MacTCP toolkit.  Also note I do not
have a THINK C version. If someone ports it over to THINK, please drop
me a line as to exactly what changes you needed to make so I can
include a note in the README of future versions.

This is the February '92 release. It is the first one in a long time
to make an attempt at UDP support.

Have at it, folks.

Server: ftp.ncsa.uiuc.edu (141.142.20.50)
File: /misc/unsupported/Socket.sit.hqx
Author: [e-mail]
--
Charlie Reiman - [e-mail]
Submissions to RHF: 3
Times published in RHF: 4



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

From: [e-mail] (G. S. Marzot (Joe))
Subject: PowerOpen
Date: 30 Jan 92 16:49:24 GMT
Organization: The MITRE Corporation

This may be a little early to ask, but I have been reading articles that
talk about a forum for people interested in PowerOpen, the new Apple/IBM
OS for the RISC based PowerPC. Is there any info out there...possibly who
at Apple should be contacted Thanks for any Info
-GSM


Email: [e-mail]

(standard  disclaimer)



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

From: [e-mail] (Kent Sandvik)
Subject:  PowerOpen
Date: 2 Feb 92 23:39:23 GMT
Organization: MacDTS Mongols

In article <[e-mail]>, [e-mail] (G. S. Marzot (Joe)) writes:
>
> This may be a little early to ask, but I have been reading articles that
> talk about a forum for people interested in PowerOpen, the new Apple/IBM
> OS for the RISC based PowerPC. Is there any info out there...possibly who
> at Apple should be contacted Thanks for any Info
> -GSM

Latest Byte had a story about the IBM/Apple alliance, including PowerOpen.
As for more information, I'm sure the trade press, especially the UNIX one,
will provide both real information and rumor mill stuff.

Kent Sandvik
- ----------
(create-a-.signature-please)



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

From: [e-mail] (G. S. Marzot (Joe))
Subject:  PowerOpen
Date: 4 Feb 92 13:33:27 GMT
Organization: The MITRE Corporation

In article <[e-mail]> [e-mail] (Kent Sandvik) writes:
> In article <[e-mail]>, [e-mail]
(G. S. Marzot (Joe)) writes:
> >
> > This may be a little early to ask, but I have been reading articles
that
> > talk about a forum for people interested in PowerOpen, the new
Apple/IBM
> > OS for the RISC based PowerPC. Is there any info out there...possibly
who
> > at Apple should be contacted Thanks for any Info
> > -GSM
>
> Latest Byte had a story about the IBM/Apple alliance, including
PowerOpen.
> As for more information, I'm sure the trade press, especially the UNIX
one,
> will provide both real information and rumor mill stuff.

This does not sound like a very open policy for Apple to have budding
developers seek information in the form of rumors.  I thought the whole
point of this new OS was to be open, you know like Power - OPEN. Isn't
there a more direct channel for info. How does OSF/1 fit in? I thought I
remembered hearing something about a developer program for PowerOpen.
Thanks you for any more concrete information. - GSM


Email: [e-mail]

(standard  disclaimer)



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

From: [e-mail] (Matthias Ulrich Neeracher)
Subject:  PowerOpen
Organization: Integrated Systems Laboratory, ETH, Zurich
Date: 6 Feb 92 19:41:08

In article <[e-mail]> [e-mail] (G. S. Marzot (Joe)) writes:
>This does not sound like a very open policy for Apple to have budding
>developers seek information in the form of rumors.  I thought the whole
>point of this new OS was to be open, you know like Power - OPEN.

You don't want to be locked into Open operating systems anyway :-)

>How does OSF/1 fit in?

Good question. Does OSF fit in anywhere ? I remember being at a HP press
conference in Switzerland where it was announced, but that was like 5 years
ago. OSF/1 seems to be shipping, but I haven't seen any company betting its
future on it yet. But maybe I'm misunderstanding the OSF's motives, anyway.

Matthias

- ---
Matthias Neeracher                                   [e-mail]
  "... Rice walked out wondering if the world was nothing but wimps,
   pimps, psychos and sex fiends" -- James Ellroy, _Suicide Hill_



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

From: [e-mail] (Thad Humphries)
Subject:  PowerOpen
Date: 07 Feb 92  06:26:59 EST

 MU> From: [e-mail] (Matthias Ulrich Neeracher)
 MU> Newsgroups: comp.sys.mac.programmer
 MU> Organization: Integrated Systems Laboratory, ETH, Zurich
 MU> Reply-To: [e-mail]'s message of 4 Feb 92 13:33:27 GMT
...
 MU> >How does OSF/1 fit in?
 MU>
 MU> Good question. Does OSF fit in anywhere ?

Are we talking about the Open Software Foundation?  If so, their claim to fame in my eye is as the sponsors of OSF/MOTIF, the most popular window manager and widget set for X Windows.  That may change, though:  AT&T and SUN, who sponsor OPEN LOOK, have decided to bundle XView (an OPEN LOOK compliant widget set) with all UNIX V R.4.

At least I hope -- my boss is shelling out lots for SUN SPARCstations and schooling for us hacks.

Besides, MOTIF looks too much like Windows -- reason enough for me to hate it but likely to stay commercially popular.

If only I could get XView libraries on a Mac at a reasonable price...
 * Origin: Quis custodiet ipsos custodes? (1:109/70.950)



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

From: [e-mail]
Subject:  PowerOpen
Date: 11 Feb 92 08:03:36 GMT
Organization: University of Missouri Computing Services

>
> Good question. Does OSF fit in anywhere ? I remember being at a HP press
> conference in Switzerland where it was announced, but that was like 5 years
> ago. OSF/1 seems to be shipping, but I haven't seen any company betting its
> future on it yet. But maybe I'm misunderstanding the OSF's motives, anyway.

Digital at least seems to be putting all its eggs into OSF/1.  They are
now shipping OSF/1 on their unix platforms, they are supplying the same
code to be the basis of SCO's unix for ACE compliant platforms, and have
ported it to their unreleased "Alpha" platforms (100-200MHz risc chips).

As for POWER/Open, as I understand it IBM is supposed to base AIX
on OSF/1, and hence POWER/Open will be (at some level?) OSF/1
compatible.  OSF/1 requires the Motif interface, which will I believe,
be supported along with the Apple UI on POWER/Open.

Jonathan/[e-mail]



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

From: [e-mail] (Brian Russ)
Subject: C++ for the Mac
Date: 4 Feb 92 03:14:48 GMT
Organization: CSE Dept., UC San Diego



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


                    Brian
--
Brian Russ
CSE Department, UC San Diego
..!sdcsvax!bruss



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

From: [e-mail] (Thomas Alan Akins)
Subject:  C++ for the Mac
Organization: University of Oregon Computer and Information Sciences Dept.
Date: Tue, 4 Feb 1992 19:59:05 GMT

In article <[e-mail]> [e-mail] (Brian Russ) writes:
>
>    From what I've gathered, if I want to program in C++ on the Mac,
>I want to get MPW.  The question is, I can't find it anywhere.

>Where might one find it?
>
>                    Brian


I believe you can purchase it from the Apple Product Developer's Associtation.
Information and current price lists are available from ftp.apple.com.
--
*                                            Alan Akins                      *
*       "I could say 'addition'              [e-mail]            *
*        when I mean ... 'basketball.'"      University of Oregon            *
*                        - E. Tick           Department of Computer Science  *



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

From: [e-mail] (Rick Fleischman)
Subject:  C++ for the Mac
Date: 6 Feb 92 18:11:59 GMT
Organization: Apple Computer Inc., Cupertino, CA

In article <[e-mail]> [e-mail] (Brian Russ) writes:
>
>    From what I've gathered, if I want to program in C++ on the Mac,
>I want to get MPW.  The question is, I can't find it anywhere.
>I've looked over the various mail-order places in MacWorld and MacUser
>(Mac Zone, Mac Connection, others) and there is absolutely no mention
>of MPW (or any other C++) at all.  Is there a reason for this?
>Where might one find it?
>
>                    Brian
>--
>Brian Russ
>CSE Department, UC San Diego
>..!sdcsvax!bruss

MPW and C++, as well as all other Apple-published development tools are
available through APDA, Apple's mail order channel for development tools and
technical documentation.

You can reach APDA at:
[phone] (in the U.S.)
[phone] (in Canada)
[phone] (International)

via the mail at:
APDA
Apple Computer, Inc.
20525 Mariani Avenue, M/S 33-G
Cupertino, CA  95014

or via e-mail at:
[e-mail]

They will send you a free APDA Tools Catalog.

Rick Fleischman
MPW Product Manager
Apple Computer, Inc.
e-mail: [e-mail]
AppleLink: [e-mail]



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

From: [e-mail] (Lee Story)
Subject:  C++ for the Mac
Date: 13 Feb 92 23:10:03 GMT
Organization: Wang Laboratories, Inc.

In article <[e-mail]> [e-mail] (Thomas Alan Akins) writes:

   Path: wang!uunet!uunet!elroy.jpl.nasa.gov!swrinde!mips!cs.uoregon.edu!dogmatix.cs.uoregon.edu!alana
   From: [e-mail] (Thomas Alan Akins)
   Newsgroups: comp.sys.mac.programmer
   Date: 4 Feb 92 19:59:05 GMT
   References: <[e-mail]>
   Sender: [e-mail] (Netnews Owner)
   Organization: University of Oregon Computer and Information Sciences Dept.
   Lines: 17

   In article <[e-mail]> [e-mail] (Brian Russ) writes:
   >
   >    From what I've gathered, if I want to program in C++ on the Mac,
   >I want to get MPW.  The question is, I can't find it anywhere.

   >Where might one find it?
   >
   >                    Brian


 I believe you can purchase it from the Apple Product Developer's Associtation.
 Information and current price lists are available from ftp.apple.com.

Call APDA at 1-800-282-2732 for ordering info.  I believe MPW plus C++
is in the $300 to $400 range.  (don't have my catalog here)

--

- ----------------------------------------------------------------------
  Lee Story ([e-mail]) Wang Laboratories, Inc.
     (Boston and New Hampshire AMC, and Merrimack Valley Paddlers)
- ----------------------------------------------------------------------



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

From: [e-mail] (Olle Furberg)
Subject: Sending FileSpecifications
Date: 2 Feb 92 19:20:57 GMT
Organization: Satellite Image in Kiruna Corporation



I'm constructing a program which will work as a background only
subprocess and managed by another program, hereafter called "master
program", through IAC (interapplication communication). The program
will be launched by the master (through the process manager, or Finder
events if they are on different machines).

The master program will send a highlevel event to my program together
with a reference to a file which my program will have to open, do some
work, and then return the result to the master (through IAC).


My main concern is what sort of file specification I should use. If
both programs are running on the same computer, a FSSpec will do the
job. But, what if the two programs is not running on the same Mac?

If the master sends an Alias to my process on the other Mac, will it be
able to resolve this alias (created on another machine) and try to mount
the other Mac's volume through filesharing?


    /Olle

- ----------------------------------------------------------
Olle Furberg                       Internet: [e-mail]
Satellitbild in Kiruna             Phone: [phone]
Swedish Space Corporation          Fax:   [phone]
                                   Home:  [phone]



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

From: [e-mail] (Grobbins)
Subject:  Sending FileSpecifications
Date: 7 Feb 92 06:13:47 GMT
Organization: Apple DTS

In article <[e-mail]> [e-mail] (Olle Furberg) writes:
>The master program will send a highlevel event to my program together
>with a reference to a file which my program will have to open, do some
>work, and then return the result to the master (through IAC).
>...
>If the master sends an Alias to my process on the other Mac, will it be
>able to resolve this alias (created on another machine) and try to mount
>the other Mac's volume through filesharing?

Yep, that's the ticket.  Full alias records can be handed off to any other
Mac on the network.  As Inside Mac VI says on page 27-4,

  An alias record is a reliable way to identify a file system object
  when your application is communicating with a process that might be
  running on a different machine.

Because your server app is a faceless background puppy, you will have
to make certain that it does not bring up the user validation dialog
when mounting a remote volume.  Call GetAliasInfo if you want to see
where the alias points to, and use MatchAlias rather than ResolveAlias
to get full control of the alias resolution.


Grobbins        [e-mail]

Usual disclaimers apply.




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

From: [e-mail] (Nicholas Jackiw)
Subject:  Sending FileSpecifications
Date: 10 Feb 92 19:49:13 GMT
Organization: Whole Earth 'Lectronic Link, Sausalito, CA

In article <[e-mail]> [e-mail] (Olle Furberg) writes:
>The master program will send a highlevel event to my program together
>with a reference to a file which my program will have to open, do some
>work, and then return the result to the master (through IAC).
>
>My main concern is what sort of file specification I should use. If
>both programs are running on the same computer, a FSSpec will do the
>job. But, what if the two programs is not running on the same Mac?
>
>If the master sends an Alias to my process on the other Mac, will it be
>able to resolve this alias (created on another machine) and try to mount
>the other Mac's volume through filesharing?

Yes.  I use exactly this scheme.  You can even have the Apple Event
manager coerce the sent alias into an FSSpec for you.  However, recall
that logging onto a fileshared volume requires user interaction, even
if no password (Guest privileges) are enabled.  If there's no user there,
your remote mac will freeze at the password stage unless you have already
explicitly mounted the shared folder on it.  As of the present time,
there's no work around for this (allegedly, there's an init on one of
the developer cds, but no Official Workaround).



--
                              --- * ---
Nick Jackiw                  [e-mail]   | [e-mail]
Key Curriculum Press, Inc.   Applelink:KEY.EDUSOFT | [phone]
                              --- * ---



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

From: [e-mail] (Ed Lai)
Subject:  Sending FileSpecifications
Date: 11 Feb 92 00:15:02 GMT
Organization: Apple Computer Inc., Cupertino, CA

In article <[e-mail]> [e-mail] (Nicholas Jackiw) writes:
>Yes.  I use exactly this scheme.  You can even have the Apple Event
>manager coerce the sent alias into an FSSpec for you.  However, recall
>that logging onto a fileshared volume requires user interaction, even
>if no password (Guest privileges) are enabled.  If there's no user there,
>your remote mac will freeze at the password stage unless you have already
>explicitly mounted the shared folder on it.  As of the present time,
>there's no work around for this (allegedly, there's an init on one of
>the developer cds, but no Official Workaround).
>
>

This is not true. If you have the Apple Event manager coerce the alias to
fss for you using the built-in coercion handler, then if you specify no
user-interaction (the default when you send it to a remote machine) in
your Apple Event, it would not try to interact and it would just return
an error.

/* 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] (Nicholas Jackiw)
Subject:  Sending FileSpecifications
Date: 13 Feb 92 21:55:25 GMT
Organization: Whole Earth 'Lectronic Link, Sausalito, CA

In article <[e-mail]> [e-mail] (Ed Lai) writes:
>In article <[e-mail]> [e-mail] (Nicholas Jackiw) writes:
>>
[...]

>>If there's no user there,
>>your remote mac will freeze at the password stage unless you have already
>>explicitly mounted the shared folder on it.  As of the present time,
>>there's no work around for this (allegedly, there's an init on one of
>>the developer cds, but no Official Workaround).
>This is not true. If you have the Apple Event manager coerce the alias to
>fss for you using the built-in coercion handler, then if you specify no
>user-interaction (the default when you send it to a remote machine) in
>your Apple Event, it would not try to interact and it would just return
>an error.
>/* Edmund K. Lai                                               */

You're correct, of course, if userInteraction is turned off. If, on
the other hand, you assume that the reason Mac A is sending Mac B
a filespec is so that Mac B can actually *do* something with or to
the file, then what I said remains: there's no way you can get to the
file without either previously mounting the shared volume or having
a user there to click on the "guest" button.

Returned errors are nice compared to crashes, but they're still a far
cry from useful results.



--
                              --- * ---
Nick Jackiw                  [e-mail]   | [e-mail]
Key Curriculum Press, Inc.   Applelink:KEY.EDUSOFT | [phone]
                              --- * ---



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

From: [e-mail] (Bill Leue)
Subject: ResEdit 2.1 Bundle Bit and Custom Icon Woes
Date: 10 Feb 92 14:40:13 GMT
Organization: General Electric Research & Development

Has anyone else had the kind of problems I am having with ResEdit 2.1, trying
to create a custom icon for an application?  Here's a brief summary:

1. I create the 'ICN#' resource, ID 128, with all the various icons.
2. Create the 'BNDL' resource, ID 128, owner ID 'BILL', with one entry,
'APPL', which points to my 'ICN#' resource.  The icons show up in the entry,
just as they should.
3. I set the creator type of the application to 'BILL', the same as the owner
line in the 'BNDL' resource.

Checking the other resources, it appears that ResEdit has created the other
resources such as 'FREF' and the Local ID's ok.

I re-link the application and add the new resource file.  (Did I mention that
I'm using Think C 5.0?)  Then I rebuild the desktop.

Hmmm..... No custom icon, just the generic 'hand' icon.  Checking the desktop
files with ResEdit reveals that my custom icons have not been added.  When
I go back into the application's resource file with ResEdit, someone has
changed my 'APPL' entry in the 'BNDL' resource back into '????'.  Arggghhh!

I've gone through this cycle (and variations) a number of times, but to no
avail.  I found out that if I add a second (dummy) line to 'BNDL', it seems
to make it less likely that my 'APPL' entry will get changed back to '????',
but this doesn't by itself help to get the custom icon to appear.

Any guesses?  This is happening using System 7.0 (with tune-up) on a 5 meg
IIcx.

Thanks!

-Bill Leue
[e-mail]







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

Subject:  ResEdit 2.1 Bundle Bit and Custom Icon Woes
From: [e-mail] (Russell Street)
Date: Mon, 10 Feb 1992 19:23:37 GMT
Organization: University of Auckland, New Zealand.

[e-mail] (Bill Leue) writes:

>Has anyone else had the kind of problems I am having with ResEdit 2.1, trying
>to create a custom icon for an application?  Here's a brief summary:



I believe it is a bug in ResEdit -- it does NOT WRITE OUT THE FREF RESOURCES
WHEN YOU CHANGE THEM IN A BNDL.  It drove me nuts figuring this
one out.

The only way around this is to go and change the FREFs manually to
get them "changed" and written out.

If you look at the FREF resources when while the file is still open
the appear changed, but if you close and reopen the file they will be
the old values.

Curses and naughty words!
- ------------------------------------------------------------
Russell Street ([e-mail])
    It was the least I could do -- a quantity I specialize in.






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

Subject:  ResEdit 2.1 Bundle Bit and Custom Icon Woes
From: [e-mail] (Nigel Stanger)
Date: 11 Feb 92 13:57:15 +1300
Organization: University of Otago, Dunedin, New Zealand

In article <[e-mail]>,
[e-mail] (Bill Leue) writes:
> Has anyone else had the kind of problems I am having with ResEdit 2.1, trying
> to create a custom icon for an application?  Here's a brief summary:
>
> 1. I create the 'ICN#' resource, ID 128, with all the various icons.
> 2. Create the 'BNDL' resource, ID 128, owner ID 'BILL', with one entry,
> 'APPL', which points to my 'ICN#' resource.  The icons show up in the entry,
> just as they should.
> 3. I set the creator type of the application to 'BILL', the same as the owner
> line in the 'BNDL' resource.
[...]
> Hmmm..... No custom icon, just the generic 'hand' icon.  Checking the desktop
> files with ResEdit reveals that my custom icons have not been added.  When
> I go back into the application's resource file with ResEdit, someone has
> changed my 'APPL' entry in the 'BNDL' resource back into '????'.  Arggghhh!

You could try editing the BNDL resource with the BNDL template
rather than the BNDL editor (i.e. choose "Open using template..."
from the Resource menu). It's not as nice to use, but when I had
similar problems with the BNDL editor I did this and it worked,
so maybe it will for you too.

--
See ya
                                Nigel.
- --------------------------------------------------------------------
Nigel Stanger,                  Internet: [e-mail]
University of Otago,            Phone: [phone]
Dunedin, NEW ZEALAND.           Fax:   [phone]
"ACK! Pthptttt!" -- Bill the Cat, "Bloom County"



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

From: [e-mail] (Matthew Holle)
Subject:  ResEdit 2.1 Bundle Bit and Custom Icon Woes
Organization: Motorola Inc., Austin, Texas
Date: Tue, 11 Feb 1992 15:06:14 GMT

In article <[e-mail]> [e-mail] (Russell Street) writes:
>[e-mail] (Bill Leue) writes:
>
>>Has anyone else had the kind of problems I am having with ResEdit 2.1, trying
>>to create a custom icon for an application?  Here's a brief summary:
>
>
>
>I believe it is a bug in ResEdit -- it does NOT WRITE OUT THE FREF RESOURCES
>WHEN YOU CHANGE THEM IN A BNDL.  It drove me nuts figuring this
>one out.
>
>The only way around this is to go and change the FREFs manually to
>get them "changed" and written out.
>
>If you look at the FREF resources when while the file is still open
>the appear changed, but if you close and reopen the file they will be
>the old values.
>
>Curses and naughty words!
>--------------------------------------------------------------
>Russell Street ([e-mail])
>    It was the least I could do -- a quantity I specialize in.

Actually, I think another problem may be to blame.  Here's the method
I use to give my applications their own custom icons.

1. Use resedit to build an ICN# resource, then build the BNDL resource.
   At this point, when you input the Signature in the BNDL resource,
   you should see the icons appear in the window (ResEdit 2.1.1).

2. Here's where I think Bill Leue was having problems.  You need to go
   into Think C (version whatever) and do a "Set Project Type..." under
   the Project menu.  Set the project to be of type application, file type
   to be "APPL" and signature to the same signature you used in the BNDL
   resource in step one.

3. Recompile and rebuild the desktop and you should see an icon.  I think
   Think C will overwrite the BNDL resource with the preferences you set
   up in the "Set Project Type..." menu (setting the signature back to
   "????".

# include <std_disclaimer.h>

 -----------------------------------------------------------------------
|       Matt Holle                      Motorola Microprocessor         |
|       RISC Applications               & Memory Technology Group       |
|       ([e-mail])     Austin, TX                      |
 -----------------------------------------------------------------------




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

From: [e-mail] (Jerry Sy)
Subject: getting Vrefnum without using StdGetFile
Organization: Case Western Reserve University, Cleveland, OH (USA)
Date: Mon, 10 Feb 92 17:15:11 GMT


I want to create a file (or open a file) that I know the file name
and where it is located.  How can I get the vrefnum for that file
without using the StdGetFile dialog box ?

jerry




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

From: [e-mail] (Keith Rollin)
Subject:  getting Vrefnum without using StdGetFile
Date: 14 Feb 92 02:15:08 GMT
Organization: Apple Computer Inc., Cupertino, CA

In article <[e-mail]> [e-mail] (Jerry Sy) writes:
>
>I want to create a file (or open a file) that I know the file name
>and where it is located.  How can I get the vrefnum for that file
>without using the StdGetFile dialog box ?

If you know where the file is, but don't know the vRefNum of the volume
it's on, then presumably you know the name of the volume and the dirID
of the directory the file is in. What you need to do is find the
vRefNum for the volume with the name you have. Most people use
PBGetVInfo with a negative ioVolIndex for this (I think that's the
trick).

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



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

From: [e-mail] (Tim Dierks, ToyMeister, Cray abuser)
Subject: Location 0 (Was Re: Blinking box on my screen)
Date: 10 Feb 92 21:43:06 GMT
Organization: MacDTS, Apple Computer

In article <[e-mail]>, [e-mail] (David Dantowitz) writes:
> p.s. is it still valid to set memory location 0 to 'NIL!' to trap
> invalid defs?

  Yes, but there are better values.  Try $50FFC037.  This does several things:
1) It's odd, so accesses to it fail on 68000 machines
2) It will cause a bus error on all current Macs with bus error hardware
3) It's an invalid instruction, so jumping to 0 will die immediately.

  Note that under System 7, the toolbox will change location 0 to point it
into ROM.  This is done because some applications are improperly written, and
dereference location 0.  It was decided that it would be best for this fault,
taken alone, not to crash applications on a users system.  For best results,
install Even Better Bus Error on your development and testing machines;
it will catch 95% of such problems.

Tim Dierks
MacDTS, but I speak for myself



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

From: [e-mail] (Charlie Reiman)
Subject:  Location 0 (Was Re: Blinking box on my screen)
Organization: University of Illinois at Urbana
Date: Tue, 11 Feb 1992 18:22:29 GMT

[e-mail] (Tim Dierks, ToyMeister, Cray abuser) writes:
: In article <[e-mail]>, [e-mail] (David Dantowitz) writes:
 < stuff deleted>
:   Note that under System 7, the toolbox will change location 0 to point it
: into ROM.  This is done because some applications are improperly written, and

Actually, I found this to be only half true. Near as I can figure it,
System 7 reads in the value from 0 some time after boot up, remembers
it, then periodically writes it back there.

Under "clean" system 7, that value somehow, magically, gets set to
ROMbase+64k (as you probably know) during init time. This is read in
after init time then later written back on a periodic basis.

The trick comes when you run EBBE. It initializes 0 to $50FF8001 at
init time. System 7 reads this in, then writes it back periodically.
EBBE does not need to continually write that value back to 0, system 7
will gleefully do so on its own.

In case your wondering how the hell I figured all this out: I wrote my
own version of EBBE that I could switch on or off without rebooting.
Seemed like a great idea to me, except System 7 insisted on overwitting
the value I wanted in 0 with whatever happened to be there during
system startup.

A few experiments with Macsbugs' 'SS 0' and StopINIT should prove
enlightening to the curious.

--
Charlie Reiman - [e-mail]
Submissions to RHF: 3
Times published in RHF: 4



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

From: [e-mail] (Laurence Arthur Pittenger)
Subject: Think debugger BUG?
Date: 10 Feb 92 22:17:50 GMT
Organization: Yale University Computer Science Dept., New Haven, CT 06520-2158

Think C version: 5.0.2
System version: 7.0 +
Machine: Mac IIci

Usually when something screws up the debugger I assume it's the fault
of my code.  However I can't find the bug anywhere (which usually
means it's obvious, but still).  Also, this last time caused major
problems, screwing up my hard drive.  I had to re-install the system
before I could use any of my files/applications again.  So I figured
I'd best pass it along.

The problem has a consistent manifestation.  It only happens after
I've interrupted my program's execution to check a falue or something
in the debugger.  If I simply run the program via the debugger (but
don't stop it at any point) this doesn't happen.

After I've done that, the program continues to work fine.  The problem
only shows up *after* I've quit the program *and* the debugger.

Shortly thereafter all the colors for the menubar icons, icons,
folders, etc get screwed up.  Usually I notice this when the menu
bar's icons have turned into black blobs.  However it doesn't always
happen on the *first* redraw of the menu (as far as I can tell).

It used to be that I could sometimes still run my program once or
twice.  Now it consistently causes this problem and bombs the next
time I try to enter the debugger.  Sometimes it even crashes when the
think c program itself is running -- compiling or saving or suchlike.

What always happen, when it does crash, is that I get dumped into
MacsBug -- usually when I've tried to launch the debugger a second
time.  An "es" doesn't work; I just get dumped back into the debugger
again.

This last time, es sent the hard drive whirring briefly before it
dumped me back to MacsBug.  When I went to reboot, the Finder got as
far as drawing the menu bar and desktop icons.  Then those
disappeared, the arrow changed to a watch, the disk whirred, and it
drew them again. That kept happening.

I tried rebuilding the desktop, to no avail.  Again, it got as far as
the menus/icons, and started looping/whirring endlessly.

I was able to boot of a floppy, could see (and copy) all the files on
the hard drive, however when I tried to start up any of the
applications, it got as far as getting the application's name up in
the center of the menubar before it froze.  This happened for every
application I tried.  Also, it seemed to have lost track of think c,
because when I 2ble-clicked on my project it said that it couldn't
find the creator of that file.

Finally, I reinstalled the system software.  Only then could I
successfully boot from the hard drive.

Like I said, this may well be my code; heck, my project is over 1000K
so there're any *number* of places one could be hiding.  However this
recent incident was so disasterous I thought I'd best pass it along in
case (a) anyone recognizes this and can tell me what I'm screwing up,
or (b) if it actually is symptomatic of something wrong w/ the think
stuff, in which case whoever collects data on this kind of thing now
has a little more info.

Happy hacking all!
LP

--

Laurence A. Pittenger
CSNET  : [e-mail]
BITNET : pitlaua@yalevm ,  pittenger-laurence@yalecs



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

From: [e-mail] (Brady Duga)
Subject:  Think debugger BUG?
Date: 11 Feb 92 13:33:13 GMT
Organization: Center for Visual Science, U. of Rochester

In article <[e-mail]> [e-mail] (Laurence Arthur Pittenger) writes:
>
>After I've done that, the program continues to work fine.  The problem
>only shows up *after* I've quit the program *and* the debugger.
>
>Shortly thereafter all the colors for the menubar icons, icons,
>folders, etc get screwed up.  Usually I notice this when the menu
>bar's icons have turned into black blobs.  However it doesn't always
>happen on the *first* redraw of the menu (as far as I can tell).
>

I've had similar problems. The screen colors were all wrong, and a peek at
the monitors CDEV showed the first several colors to be a blend from white to
black, which I thought odd since I wasn't using any interesting color code.
After a few runs, the machine would eventually crash (I didn't have MacsBug
loaded at the time, so the machine just froze). The problem was rare enough
that I never tracked down the cause, and it has stopped altogether recently.

Anyone else with a similar experience?

--Brady ([e-mail])



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

From: [e-mail] (Scott Lindsey)
Subject:  Think debugger BUG?
Date: 13 Feb 92 23:48:06 GMT
Organization: Claris Corporation

In article <[e-mail]>, [e-mail] (Brady Duga) writes:
>
> In article <[e-mail]> [e-mail] (Laurence Arthur Pittenger) writes:
> >
> >After I've done that, the program continues to work fine.  The problem
> >only shows up *after* I've quit the program *and* the debugger.
> >
> >Shortly thereafter all the colors for the menubar icons, icons,
> >folders, etc get screwed up.  Usually I notice this when the menu
> >bar's icons have turned into black blobs.  However it doesn't always
> >happen on the *first* redraw of the menu (as far as I can tell).
> >
>
> Anyone else with a similar experience?
>

Yes, I've seen it also.  I reported it to Symantec in beta versions of Think C 5.0,
but it wasn't fixed.  It's hard to nail down to a specific user/programmer action.

--
Scott Lindsey <[e-mail]>



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

From: [e-mail] (Peter N Lewis)
Subject: WARNING: Bug in THINK Pascal 4.0.1
Date: 11 Feb 92 03:35:55 GMT
Organization: NCRPDA, Curtin University

Hi All,

When you upgrade to 4.0.1, be careful with code resources.  When it
updates (compresses) the project, it also kindly turns on the
"Custom Header" flag in the Project Type options.  This has the wholely
pleasant effect of removing the jmp to your main code, and thus producing
a resource that will simply go bang in a really helpful way when it gets
called.

You have been warned (I have been burned),
   Peter.

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



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

From: [e-mail] (Steven Langlois)
Subject:  WARNING: Bug in THINK Pascal 4.0.1
Date: 11 Feb 92 13:06:23 GMT
Organization: Bell-Northern Research Ltd.

In article <[e-mail]>, [e-mail] (Peter N Lewis) writes:
|> When you upgrade to 4.0.1, be careful with code resources.  When it
|> updates (compresses) the project, it also kindly turns on the
|> "Custom Header" flag in the Project Type options.  This has the wholely
|> pleasant effect of removing the jmp to your main code, and thus producing
|> a resource that will simply go bang in a really helpful way when it gets
|> called.
|>

I tried this with several code resource projects and couldn't reproduce
this problem. Can you give any more details? I'm sure the people at
Symantec would be interested.

Steven Langlois
ISDN Basic Rate Access
Bell-Northern Research

Internet: [e-mail]



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

From: [e-mail] (Steve Dakin)
Subject: FILE *'s and Mac files
Date: 11 Feb 92 01:34:01 GMT
Organization: Oceania Health Care Systems


Using THINK C 5.0.x:

Is there a way to get a Mac file reference number from a FILE * that is used in
standard C file I/O operations (e.g. fprintf) and vice versa?

All comments, suggestion, references, etc. appreciated.

--
    Steve Dakin      UIguy, company morale booster, Marillion worshipper
 [e-mail]        Oceania Health Care Systems, Palo Alto, CA
 [e-mail]                    [phone]
    (NeXT mail)            I'm not a doctor, I just work for them.



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

From: [e-mail] (Phil Shapiro)
Subject:  FILE *'s and Mac files
Date: 12 Feb 92 02:37:26 GMT
Organization: Symantec Corp.

In article <[e-mail]> [e-mail] (Steve Dakin) writes:

   Using THINK C 5.0.x:

   Is there a way to get a Mac file reference number from a FILE *
   that is used in standard C file I/O operations (e.g. fprintf) and
   vice versa?

In general, check out the FILE structure in the stdio.h header file.
The structure fields are pretty mnemonic -- of course, they can change
in a future release. Currently, you can use:

    FILE *fp;
    ...
    err = GetEOF(fp->refnum, &theEOF);   /* or whatever */

Also, you may want to look at the code for __open() in fopen.c.

    -phil
--
   Phil Shapiro                           Technical Support Analyst
   Language Products Group                     Symantec Corporation
        Internet: [e-mail]



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

From: [e-mail] (Thad Humphries)
Subject: MPW: deleting resource fork
Date: 11 Feb 92 04:46:57 GMT


 JD> From: [e-mail] (John Doner)
 JD> Newsgroups: comp.sys.mac.programmer
 JD> Organization: The Aerospace Corporation, El Segundo, CA
 JD>
 JD> Can anyone tell me how to use MPW commands to delete just the
 JD> resource fork of a file?

Don't know that it can be done.  But here's some THINK C code to do it...

/* delete a fork.c
 *
 * This code shows how to delete the resource and data forks of a file.
 */

#include <ResourceMgr.h>
#include <FileMgr.h>
#include <stdio.h>

void main()
{
    int refNum, err;

    /* The next three lines are used to delete the resource fork.  They require
     * you include ResourceMgr.h
     */
    refNum = OpenResFile("\pRENT-A-COMP:Misc ESSO:Junk:my_file");
    CloseResFile(refNum);
    err = OpenRF("\pRENT-A-COMP:Misc ESSO:Junk:my_file", 0, &refNum);


    /* If you want to delete the data fork, the previous three lines and use the
     * FSOpen line below.  You need not include ResourceMgr.h
     */

    /* err = FSOpen("\pRENT-A-COMP:Applications:Microsoft Word 4.0", 0, &refNum); */

    if (err) {
        printf("Cannot open file, error %d.\n", err);
        return;
        }

    err = SetEOF(refNum, 0);
    if (err) {
        printf("Cannot set EOF, error %d.\n", err);
        return;
        }

    err = FSClose(refNum);
    if (err) {
        printf("Cannot close file, error %d.\n", err);
        return;
        }
}

 * Origin: Quis custodiet ipsos custodes? (1:109/70.950)



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

From: [e-mail] (Chris D. Nebel)
Subject:  MPW: deleting resource fork
Organization: University of Maryland at College Park
Date: Tue, 11 Feb 1992 19:19:50 GMT

Try "rez -o file < Dev:Null".  Works for me.  Admittedly, what this
actually does is make the file have an empty resource fork, which is
a little different than making it have NO resource fork, but it will
still save you disk space.  This command relies on the fact that rez
initializes the resource fork of the target file before rezzing anything
into it (unless you specify -a); getting input from Dev:Null means that
nothing gets added, and voila! an empty resource fork.

Chris Nebel
[e-mail]



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

From: [e-mail] (E Gasparini)
Subject: Launching Applications
Organization: Clemson University Computer Science Dept.
Date: Tue, 11 Feb 1992 17:54:09 GMT


Hi,

I need to know how to create (text|binary) data files within a Symantec Think
Pascal application, and be able to launch this application (not Think Pascal)
by 2-clicking on the data file.

Please, be specific as my knowledge of the Mac OS is NIL (ok,
close to that).

Reply by e-mail, or post here.
    Thanks
================================
Elvis Gasparini
[e-mail]
Clemson University



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

From: [e-mail]!gluttony.reed.edu!pcalahan (Patrick John Calahan)
Subject:  Launching Applications
Date: 13 Feb 92 07:25:35 GMT
Organization: Reed College, Portland, OR

In article <[e-mail]>
[e-mail] (E Gasparini) writes:
>
> Hi,
>
> I need to know how to create (text|binary) data files within a Symantec
Think
> Pascal application, and be able to launch this application (not Think
Pascal)
> by 2-clicking on the data file.
>
> Please, be specific as my knowledge of the Mac OS is NIL (ok,
> close to that).
>
> Reply by e-mail, or post here.
>     Thanks
> ================================
> Elvis Gasparini
> [e-mail]
> Clemson University

Assuming I understand your question correctly...
    When you double click on a document file, the finder looks at its
'creator' field and searches for an APPLication of that type.  All you
need to do is pass your application's resource-signature through the
'creator' parameter of whatever file i/o routines you are using, or
perhaps this is handled automatically.
                -Pat



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

From: [e-mail]
Subject: GetMenu returning wrong ID?
Date: 11 Feb 92 18:17:23 GMT
Organization: University of Helsinki


    I would be very grateful if someone helped me a little:

    I am writing a small drawing application with
    THINK C 5.0 using TCL.  I created a new MENU
    resource with ResEdit (2.1) and gave it an ID 20.
    Then I added the commands and their numbers and
    finally added the menu ID in the MBAR resource.

    When running the application my menu commands never
    came through.  So I watched with the debugger, and
    what did I see?  When the Bartender is initializing
    itself, it makes a call of GetMenu to every menu in
    the MBAR resource.  According to IM I-419, the GetMenu
    is supposed to return the menu record of the menu, whose
    ID is given as a parameter.  All the other menus come
    out ok, but when my menu was called (with the ID 20),
    back came a menu record with an ID of 128.

    This is of course the reason why the commands were never
    sent, but how is this possible?  And what should I do now?

    Yours,
            Mikko Lindholm
            [e-mail]
            [e-mail]



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

From: [e-mail] (Keith Rollin)
Subject:  GetMenu returning wrong ID?
Date: 14 Feb 92 03:28:54 GMT
Organization: Apple Computer Inc., Cupertino, CA

In article <[e-mail]> [e-mail] writes:
>
>    I am writing a small drawing application with
>    THINK C 5.0 using TCL.  I created a new MENU
>    resource with ResEdit (2.1) and gave it an ID 20.
>    Then I added the commands and their numbers and
>    finally added the menu ID in the MBAR resource.
>
>    When running the application my menu commands never
>    came through.  So I watched with the debugger, and
>    what did I see?  When the Bartender is initializing
>    itself, it makes a call of GetMenu to every menu in
>    the MBAR resource.  According to IM I-419, the GetMenu
>    is supposed to return the menu record of the menu, whose
>    ID is given as a parameter.  All the other menus come
>    out ok, but when my menu was called (with the ID 20),
>    back came a menu record with an ID of 128.
>
>    This is of course the reason why the commands were never
>    sent, but how is this possible?  And what should I do now?

It is probable that you changed only the resource ID of the MENU, and
not the menu ID. Both must be changed in order for TCL's command
handling to work. To change the menu ID, open the MENU resource, go to
the MENU menu, and select Edit Menu and MDEF ID.

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



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

Subject: Resolving aliases for pref files etc
From: [e-mail] (Russell Street)
Date: Wed, 12 Feb 1992 02:10:01 GMT
Organization: University of Auckland, New Zealand.


Should one resolve aliases for preferences files, folders found by
FindFolder found etc?

The Finder does not, for instance do this, for instance if you
put an alias called "Finder Preferences" in the Prefs folder.
Should the rest of us bother?  It is a fairly perverse thing for
a user to do, but you never know what they will get up to.

Also does the System resolve aliases for folders found by FindFolder?
Again it is a silly thing for someone to do...

TIA
- -----------------------------------------------------------
Russell Street                     [e-mail]
"Baldrick, I believe the phrase rhymes with 'clucking bell'."
        -- Edmund Blackadder



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

From: [e-mail] (Marc T. Kaufman)
Subject:  Resolving aliases for pref files etc
Organization: CS Department, Stanford University, California, USA
Date: 12 Feb 92 04:29:19 GMT

[e-mail] (Russell Street) writes:

>Should one resolve aliases for preferences files, folders found by
>FindFolder found etc?

>The Finder does not, for instance do this, for instance if you
>put an alias called "Finder Preferences" in the Prefs folder.
>Should the rest of us bother?  It is a fairly perverse thing for
>a user to do, but you never know what they will get up to.

Under the principle of least astonishment, I believe you SHOULD resolve
aliases in the Preferences folder (and most anywhere else, too).  Your
recourse is to post an error, since you don't want TWO preference files with
the same name (one an alias and one not).  After all, for most applications
the preferences file is generated automatically on first launch, and the
only way an alias prefs file can get there is for someone to explicitly
put it there -- maybe he or she had a good reason.

Besides, it's as simple as: "ResolveAliasFile" (see IM-VI, p.9-30)

--
Marc Kaufman ([e-mail])




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

From: [e-mail] (Eric E. Braun)
Subject: finding a publisher
Date: 12 Feb 92 21:00:36 GMT
Organization: Yale University Computer Science Dept., New Haven, CT 06520-2158


Hi, I am just finishing up a new networking product and I'm looking
for a publisher.  Does any of you out there who have published
software before have any good sugestions for where to go looking (or
where to stay away from)?  Phone #s and address would be appreciated
and names of people to contact would best if possible.

If other folks are interested in this same info drop me a note and if
there are enough of you I'll post a summary of what I get (else I'll
just mail it out).

Thanks in advance (as usual)..

Eric



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

From: [e-mail] (Roger Allen)
Subject:  finding a publisher
Date: 12 Feb 92 23:59:55 GMT
Organization: Amdahl Corporation, Sunnyvale CA

In article <[e-mail]> [e-mail] (Eric E. Braun) writes:
>
>Hi, I am just finishing up a new networking product and I'm looking
>for a publisher.  Does any of you out there who have published
>software before have any good sugestions for where to go looking (or
>where to stay away from)?  Phone #s and address would be appreciated
>and names of people to contact would best if possible.
>
>If other folks are interested in this same info drop me a note and if
>there are enough of you I'll post a summary of what I get (else I'll
>just mail it out).
>
>Thanks in advance (as usual)..
>
>Eric

I would certainly like to hear other people's stories of how they got
their product to market.  It seems we have plenty of advice on how
to get our programs to work--but, after you get your program done you
are on your own!

I know "in general" what to do, so I'm not too interested in the
standard "write to a publisher and wait for a reply..."  This leaves
ALOT unanswered.  For instance--how do you negotiate your fee/royalty--
this is probably something that it would be nice to know ballpark
figures beforehand.  Another for instance--after you sold your program
to a publisher did you find out about a better deal you had overlooked?
or maybe see that going it alone would have been better? Etc, etc, etc.

C'mon Mac programming gurus! Let's hear how you got rich! (or not)

Roger.
--
> Roger Allen                   |  The preceding message was written     <
> Amdahl Computer Development   |  by 100 monkeys typing for 100 years.  <
> [e-mail]           |  I am not responsible for the monkeys. <



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

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