Nautilus 1992 July/Nautilus-3-8/Nautilus-3-8.bin/Tools & Utilities/Techy Stuff/Doco ƒ/CSMP ƒ/CSMP-V1-050.TXT

Text File | 1992-06-03 | 43.9 KB | 1,157 lines

C.S.M.P. Digest             Thu, 16 Apr 92       Volume 1 : Issue 50

Today's Topics:

    Launching app from INIT (again)
    DRVR or background: which is better?
    Another MacsBug question...
    Is TextEdit the best solution?
    PBVolumeMount Weirdness!!!
    adding small icon to a cell of a list
    How to force Finder filesize etc. update after THINK C stdio?
    Drawing a Picture in a Modal Dialog?
    List manager problem
    Help! Filling a rect with a uniform color
    iostream package for Think C?


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

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

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

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

Send administrative mail to [e-mail].

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

From: [e-mail] (Jeffrey Verdegan)
Subject: Launching app from INIT (again)
Date: 9 Mar 92 16:43:50 GMT
Organization: Computer Aided Engineering, Univ. of Wisconsin-Madison

A while ago, I posted a question asking how to launch an application from an
INIT.  As I haven't gotten any responses yet, I'll make a pest of myself :-)
and post again.

Hence, how do I launch an application from an INIT?  Note, I have no problem
with RTFM, if somebody will give me a clue as to which one.  I've looked in
all the IM volumes' indices under "Launch"  (even VI, and although there was
plenty of info on the process manager, my program has to work under system 6)
IM-II had one or two paragraphs on _Launch, but it didn't seem particularly
useful.

Help?  Please?

- ----------

Jeff Verdegan
University of Wisconsin-Madison
Computer-Aided Engineering Center
[e-mail]
[phone]

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

From: [e-mail] (Tim Dierks, ToyMeister, Cray abuser)
Date: 13 Mar 92 19:32:52 GMT
Organization: MacDTS, Apple Computer

In article <[e-mail]>, [e-mail] (Jeffrey Verdegan) writes:
> Hence, how do I launch an application from an INIT?

What do you really want to do?  Launching an application from an INIT is a
pretty bad idea; the system software is not finished setting itself up, the
process manager hasn't started up yet, etc, etc.  You need to ask the real
question: How do I do x?  Now what do you want to do?

Do you want an application to always start up when the machine is booted?
Do you want to replace the Finder?
Do you want to do a task before the Finder starts, and you want to do it from
 an application?

All these problems would have different solutions.  You need to ask the right
question in order to get a reasonable answer.

Most of the time, when people post questions to csmp and don't get an answer,
it's because they haven't phrased their question in a way to make it anything
but an annoyance to answer.

Tim Dierks
MacDTS, but I speak for myself

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

From: [e-mail] (Jeffrey Verdegan)
Organization: College of Engineering, Univ. of Wisconsin-Madison
Date: 14 Mar 92 13:40:34 CST

In article <[e-mail]> [e-mail] (Tim Dierks, ToyMeister, Cray abuser) writes:
>In article <[e-mail]>, [e-mail] (Jeffrey Verdegan) writes:
>> Hence, how do I launch an application from an INIT?
>
>What do you really want to do?  Launching an application from an INIT is a
>pretty bad idea; the system software is not finished setting itself up, the
>process manager hasn't started up yet, etc, etc.  You need to ask the real
>question: How do I do x?  Now what do you want to do?
>
>Do you want an application to always start up when the machine is booted?
>Do you want to replace the Finder?
>Do you want to do a task before the Finder starts, and you want to do it from
> an application?
>
>All these problems would have different solutions.  You need to ask the right
>question in order to get a reasonable answer.
>
>Most of the time, when people post questions to csmp and don't get an answer,
>it's because they haven't phrased their question in a way to make it anything
>but an annoyance to answer.
>
>Tim Dierks
>MacDTS, but I speak for myself

Okay, I realize now that I phrased my question badly, but at the time I posted,
I wasn't really aware of the difference between an INIT and a stand alone code
resource, and of the fact that at INIT time, all the stuff needed to launch
an app has not been set up yet.  I guess now that it's been pointed out to me,
I realize I should have known, if I had really thought about it.  Sorry for
the lack of clarity.

Anyway, here's the scenario:  I have an INIT that patches a trap (PBControl,
to be exact).  I want to perform certain actions whenever someone logs in to
an AppleShare volume (actually, a NetWare volume, but I make the distinction
inside the patch).  One of the things I want to do, after the patch is finished
with everything else, is to launch an application.  So, as several people
surmised, launching from the INIT is not actually what I want to do.  Launching
from the trap is.  By the way, it's a head patch, not a tail patch, if that
makes a difference.

I've already gotten several responses that answer my question of basic
strategy, (Thanks everyone! =:-) ), so I should be able to make something work
with that,  but if there's any canned code out there that will do it, I don't
really feel a need to re-invent the wheel on this one.


Later,
Jeff


- ----------

Jeff Verdegan
University of Wisconsin-Madison
Computer-Aided Engineering Center
[e-mail]
[phone]



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

From: [e-mail]
Subject: DRVR or background: which is better?
Date: 10 Mar 92 08:05:04 GMT
Organization: McDnDSpaceSys, Huntington Bch, CA

Hi everyone,

  I'm building a program to open a TCP connection over a socket on startup (or
close to it) and listen for a response from a host (i.e. an event happens on
the host and a server sitting on the host will send a "bip" over the connection
to the Mac). When the Mac gets the response it will tweak the Notification
Manager. All the Macs are running 6.0.x or 7 and have MacTCP.

  Now for the $64 question....Should this be done as a DRVR (installed by an
INIT at startup) or as a background app? I don't think I need to patch anything
as long as I can be assured of getting some of the CPU once in a while (it's not
terribly important I get called the second that chars are available). A
background app would make our user support organization less nervous, just
simply because it "isn't another INIT to worry about". BTW, there will be some
prefs data input from the user (i.e. type of host event to watch for)... just
a string or two which I was thinking on gathering with a cdev (or could I get
it with a window attached to the DRVR/DA...this DRVR could be a DA, right?)

  If it's done as a background app, what's different about a background app
besides the OnlyinBackground flag in the SIZE resource (and no window calls)?
Does anyone have an example of a simple background app or point me to where I
can find source to one (C, Pascal, I'm not picky)...

Thanks in advance...!

- -Jeff Johnson
[e-mail]

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

From: [e-mail] (Aaron Wohl)
Date: 11 Mar 92 13:16:53 GMT
Organization: Special Projects, Carnegie Mellon, Pittsburgh, PA

For an example of dothing this see host akutatak anonymous ftp
/aw0g/mailcheck.sit.hqx.

This is an init/drvr/cdev collection to check unix mail and pop up a
notification manager alert (or blink the apple or play a sound) when
mail comes in.  Think C sources are included.

Aaron

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

From: [e-mail] (Joe Zobkiw)
Organization: The World Public Access UNIX, Brookline, MA
Date: Wed, 11 Mar 1992 19:47:16 GMT

Jeff - You might also want to look into writing a "Daemon" which is the
best of both worlds. It's like a background application yet it doesn't
have any interface at all (so, it's like a DRVR)

I think it is discussed in IM VI as well as one of the later Dev CDs which
has some example code.






- --
<--------------------------------------------------->
 joe zobkiw                     [e-mail]
 mac.synthesis.MIDI.development.C.asm.communications
>---------------------------------------------------<

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

From: [e-mail] (Pete Resnick)
Date: 11 Mar 92 21:27:15 GMT
Organization: University of Illinois at Urbana

jjohnson@ewsvax writes:

[...]

>All the Macs are running 6.0.x or 7 and have MacTCP.

[...]

>Should this be done as a DRVR (installed by an INIT at startup) or as a
>background app?

Things to help you decide:

Why it should be an application:
- --------------------------------
1. **Much** easier to code and debug. Source level debugging is much
   nicer with apps.
2. Doesn't take up a device driver entry in the table.
3. Communication with it is cake under System 7 using AppleEvents and
   the Process Manager.
4. Avoid "another INIT to worry about".

Why it should be a driver:
- --------------------------
1. Needs to run under Finder because Multifinder might not be there.
2. Need to be able to communicate with it easily under System 6.0.x.
   There ain't no Process Manager in System 6.
3. You need to write an INIT to launch the application under System 6
   anyway, so you might as well just write the driver.
4. I'm just like Pete and I like mucking around with other people's
   heaps instead of having one of my own.

If none of the reasons under "Why it should be a driver" are overwhelming,
write it as an application and make everybody much calmer.

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] (Donald T. Major)
Organization: SAS Institute Inc.
Date: Fri, 13 Mar 1992 18:59:40 GMT

In article <[e-mail]>, [e-mail] (Pete Resnick) writes:
[many true & useful statements deleted]
|> 3. You need to write an INIT to launch the application under System 6
|>    anyway, so you might as well just write the driver.

Not true.  System 6 supports the concept of startup applications--you
just set it up via the Startup... menu item under Special.

                                             ..
                                            dtm

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

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

From: [e-mail] (James Gustave)
Subject: Another MacsBug question...
Date: 11 Mar 92 01:46:01 GMT
Organization: University of California, Santa Cruz


When you're debugging with MacsBug, how can you find out where in the file
(ie. position in whatever CODE resource) you are looking at??

Also, since I'm already posting something anyway...
When I post news I get a message warning me of the extravagant cost of my
actions.  Is this true or is this a joke?  Just wondering...


- --
Jim Speth
[e-mail]

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

From: [e-mail] (Steve Dorner)
Date: 13 Mar 92 15:59:13 GMT
Organization: University of Illinois at Urbana-Champaign

[e-mail] (James Gustave) writes:
>When you're debugging with MacsBug, how can you find out where in the file
>(ie. position in whatever CODE resource) you are looking at??

how, sc, and sc7 all print useful information.

What they give you is routine name and an offset.  Finding that in your
source can be difficult.  You can use the "il" command to list instructions
until you can figure out just where you are (subroutine calls can be a big
help).  Personally, I print out an UltraSlimFast listing of all my source
from time to time, which makes this very easy.  (UltraSlimFast is a freeware
MPW tool that prints a combined source code & assembly listing.)

>When I post news I get a message warning me of the extravagant cost of my
>actions.  Is this true or is this a joke?

It's becoming less true as time goes on and fewer sites are paying
large marginal costs for article transmission.
- --
Steve Dorner, U of Illinois Computing Services Office
Internet: [e-mail]  UUCP: uunet!uiucuxc!uiuc.edu!s-dorner

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

From: [e-mail] (James Gustave)
Subject: Is TextEdit the best solution?
Date: 11 Mar 92 01:48:48 GMT
Organization: University of California, Santa Cruz


One feature of my latest project is a simple terminal program.
What seems like the most elegant way of displaying the test coming in?
Should I use TextEdit or just use DrawChar?
Thanks.


- --
Jim Speth
[e-mail]

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

From: [e-mail] (Juri Munkki)
Date: 11 Mar 92 14:59:57 GMT
Organization: Helsinki University of Technology, Finland

In article <[e-mail]> [e-mail] (James Gustave) writes:
>One feature of my latest project is a simple terminal program.
>What seems like the most elegant way of displaying the test coming in?
>Should I use TextEdit or just use DrawChar?

Both are wrong. Using TextEdit is totally unacceptable and DrawChar
requires more work than is necessary is also slower than optimal.

The easiest solution is to use the communications toolbox, since it
provides terminal emulation features. You only have to learn how to
use it and that's it. I think the latest Develop had an article on
how to use the terminal stuff.

If you still need to write your own terminal emulation code, use
DrawText or DrawString instead. Try to draw as many characters
as possible with one call to DrawText. Buffer all incoming characters
into an internal buffer and draw the buffered characters only when
the cursor moves or you are about to process events. This will
provide extremely smooth and fast text output.

   ____________________________________________________________________________
  / Juri Munkki        /  Helsinki University of Technology   /  Wind  / Project /
 / [e-mail]  /  Computing Center Macintosh Support  /  Surf  / Arashi  /
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

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

From: [e-mail] (Tim Endres)
Date: 14 Mar 92 02:55:54 GMT
Organization: ICE Engineering, Inc.


In article <[e-mail]> (comp.sys.mac.programmer), [e-mail] (Juri Munkki) writes:
> The easiest solution is to use the communications toolbox, since it
> provides terminal emulation features. You only have to learn how to
> use it and that's it. I think the latest Develop had an article on
> how to use the terminal stuff.

AGREED! CTB all the way.


tim endres - [e-mail]  -or-  uupsi!tbomb!time
ICE Engineering, Inc. - Phone [phone] - FAX [phone]
8840 Main Street, Whitmore Lake, MI  48189
USENET - a slow moving self parody... ph

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

From: [e-mail] (James Borquist)
Subject: PBVolumeMount Weirdness!!!
Date: 11 Mar 92 18:42:52 GMT
Organization: Microsoft Corp.

Hi there!
  I was just playing around with this new (well, not so new now, but new to
me) File Manager call PBVolumeMount.  I have found that on AppleShare servers
that allow guests to connect, but have volumes that require passwords (at
least in the Chooser!) I can mount the volume.  Of course, the volume
disappears after about 1/2 of a second, but the weird thing is that it does
mount.  Now, my questions are:

1.  Why do I not get an error from PBVolumeMount when I mount a volume like
    this?

2.  If the answer is that this is an access privilege related thing, then
    is there any way to find out when I'm trying to mount a volume like
    this?

Thanks!
James

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

From: [e-mail] (Tom Colley)
Date: 12 Mar 92 13:46:14 GMT
Organization: Pacer Software, Inc.

In article <[e-mail]>, [e-mail] (James Borquist) writes:
>
> Hi there!
>   I was just playing around with this new (well, not so new now, but new to
> me) File Manager call PBVolumeMount.  I have found that on AppleShare servers
> that allow guests to connect, but have volumes that require passwords (at
> least in the Chooser!) I can mount the volume.  Of course, the volume
>
> Thanks!
> James
>

I have noticed this also, and quite a while ago had sent a bug report to
Apple concerning PBVolumeMount and password protected volumes.  Aliases to
password protected volumes do not work either.  I never received a
satisfactory response about the situation.

What I saw:  Even though all of the information for the server, volume, user,
and volume password is correct (and put together correctly in the
AFPVolMountInfo struct, the password does not make it to the AFPOpenVol
packet that gets out on the network, hence the call will fail at the
server with an AccessDenied (that is what the server should do) (the
volume password is sent to the server in the AFPOpenVol call).

With password protected volumes on Pacer's PacerShare UNIX fileservers, the
AFPVolMountInfo does return an AccessDenied (-5000).  I'm willing to bet
some sum of money ($0.02) that the fileserver you are using is blowing the
error code on the AFPOpenVol call, and that after a subsequent
AFPGetDirParms to get access rights to the base of the volume, the Mac
realizes it has a volume that it has no privs to, and unmounts it.

Tom Colley
<[e-mail]>  <[e-mail]>
/* Watch my return address, it may be wrong */

(In case anyone is wondering, PacerShare is an AppleShare file server that
runs on UNIX and VMS.  Call 508 898 3300 for more info.)

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

From: [e-mail] (David Castell)
Date: 12 Mar 92 17:39:13 GMT
Organization: Computer Systems Group

In article <[e-mail]>, [e-mail] (Tom Colley) writes:
>
> In article <[e-mail]>, [e-mail] (James Borquist) writes:
> >
> > Hi there!
> >   I was just playing around with this new (well, not so new now, but new to
> > me) File Manager call PBVolumeMount.  I have found that on AppleShare servers
> > that allow guests to connect, but have volumes that require passwords (at
> > least in the Chooser!) I can mount the volume.  Of course, the volume
> >
> > Thanks!
> > James
> >
>
> I have noticed this also, and quite a while ago had sent a bug report to
> Apple concerning PBVolumeMount and password protected volumes.  Aliases to
> password protected volumes do not work either.  I never received a
> satisfactory response about the situation.
>
> What I saw:  Even though all of the information for the server, volume, user,
> and volume password is correct (and put together correctly in the
> AFPVolMountInfo struct, the password does not make it to the AFPOpenVol
> packet that gets out on the network, hence the call will fail at the
> server with an AccessDenied (that is what the server should do) (the
> volume password is sent to the server in the AFPOpenVol call).
>
> With password protected volumes on Pacer's PacerShare UNIX fileservers, the
> AFPVolMountInfo does return an AccessDenied (-5000).  I'm willing to bet
> some sum of money ($0.02) that the fileserver you are using is blowing the
> error code on the AFPOpenVol call, and that after a subsequent
> AFPGetDirParms to get access rights to the base of the volume, the Mac
> realizes it has a volume that it has no privs to, and unmounts it.
>
> Tom Colley
> <[e-mail]>  <[e-mail]>
> /* Watch my return address, it may be wrong */
>
> (In case anyone is wondering, PacerShare is an AppleShare file server that
> runs on UNIX and VMS.  Call 508 898 3300 for more info.)
>
>
I also found this bug in System 7 and traced through the operating system
to see what the problem was.  Apparently, the PBVolumeMount routine never
uses the Volume Password that is passed to it.  It completely ignores it!
The response I got was that it was definitely a bug and that AppleShare 3.0
drivers has its own version of PBVolumeMount which will provide a working
call to System 6 and System 7 user's alike.  They didn't say when this would
be fixed in System 7.

Hope this is of some help.

Dave Castell

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

From: [e-mail] (Charles McClary)
Organization: Indiana University
Date: Thu, 12 Mar 92 21:06:44 GMT

In article <[e-mail]>, [e-mail] (David Castell) writes:
>
> In article <[e-mail]>, [e-mail] (Tom Colley) writes:
> >
> > In article <[e-mail]>, [e-mail] (James Borquist) writes:
> > >
> > > Hi there!
> > >   I was just playing around with this new (well, not so new now, but new to
> > > me) File Manager call PBVolumeMount.  I have found that on AppleShare servers
> > > that allow guests to connect, but have volumes that require passwords (at
> > > least in the Chooser!) I can mount the volume.  Of course, the volume
> > >
> > > Thanks!
> > > James
> > >
> >
> > I have noticed this also, and quite a while ago had sent a bug report to
> > Apple concerning PBVolumeMount and password protected volumes.  Aliases to
> > password protected volumes do not work either.  I never received a
> > satisfactory response about the situation.
> >
> > What I saw:  Even though all of the information for the server, volume, user,
> > and volume password is correct (and put together correctly in the
> > AFPVolMountInfo struct, the password does not make it to the AFPOpenVol
> > packet that gets out on the network, hence the call will fail at the
> > server with an AccessDenied (that is what the server should do) (the
> > volume password is sent to the server in the AFPOpenVol call).
> >
> > With password protected volumes on Pacer's PacerShare UNIX fileservers, the
> > AFPVolMountInfo does return an AccessDenied (-5000).  I'm willing to bet
> > some sum of money ($0.02) that the fileserver you are using is blowing the
> > error code on the AFPOpenVol call, and that after a subsequent
> > AFPGetDirParms to get access rights to the base of the volume, the Mac
> > realizes it has a volume that it has no privs to, and unmounts it.
> >
> > Tom Colley
> > <[e-mail]>  <[e-mail]>
> > /* Watch my return address, it may be wrong */
> >
> > (In case anyone is wondering, PacerShare is an AppleShare file server that
> > runs on UNIX and VMS.  Call 508 898 3300 for more info.)
> >
> >
> I also found this bug in System 7 and traced through the operating system
> to see what the problem was.  Apparently, the PBVolumeMount routine never
> uses the Volume Password that is passed to it.  It completely ignores it!
> The response I got was that it was definitely a bug and that AppleShare 3.0
> drivers has its own version of PBVolumeMount which will provide a working
> call to System 6 and System 7 user's alike.  They didn't say when this would
> be fixed in System 7.
>
> Hope this is of some help.
>
> Dave Castell
>
>
I was trying to use PBVolumeMount to an AppleShare volume as a guest and it
bounced with a -58 (which states: volume in question belongs to an external
file system).  I assume, from the previous discussions, that I have fallen
into the same bug.

Is there another, reliable, way to mount an AppleShare volume?

Thanks,
Charles McClary
[e-mail]



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

From: [e-mail] (Tom Colley)
Date: 13 Mar 92 14:40:56 GMT
Organization: Pacer Software, Inc.

In article <[e-mail]>, [e-mail] (Charles McClary) writes:
> In article <[e-mail]>, [e-mail] (David Castell) writes:
> > In article <[e-mail]>, [e-mail] (Tom Colley) writes:
> > > In article <[e-mail]>, [e-mail] (James Borquist) writes:
> > > > Hi there!
> > > >   I was just playing around with this new (well, not so new now, but new to
> > > > me) File Manager call PBVolumeMount.  I have found that on AppleShare servers
> > > > that allow guests to connect, but have volumes that require passwords (at
> > > > least in the Chooser!) I can mount the volume.  Of course, the volume

  [...lots of stuff deleted...]

> I was trying to use PBVolumeMount to an AppleShare volume as a guest and it
> bounced with a -58 (which states: volume in question belongs to an external
> file system).  I assume, from the previous discussions, that I have fallen
> into the same bug.
>
> Is there another, reliable, way to mount an AppleShare volume?
>
> Thanks,
> Charles McClary
> [e-mail]
>

I had problems with logging in as guest also (but eventually had success!).
The trick was to make sure that the userNameOffset field points to a valid
place in the AFPVolMountInfo structure, and that it points to an empty string.
Use the guest UAM as if you were supplying a username and password, but
make sure the username and password are empty (for Pascal strings, make sure
the length byte is zero).  If the username field was not empty, I would
get a -5019, which is an AFP ParamErr.  I've never seen a -58.  Of course,
logging in as guest only works for non-password protected volumes, as
discussed in previous postings.

Tom Colley
<[e-mail]>  <[e-mail]>
/* Watch my return address, it may be wrong */

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

From: [e-mail] (Wilson Fu)
Subject: adding small icon to a cell of a list
Date: 12 Mar 92 07:42:45 GMT
Organization: Netcom, Bay Area, CA

Wondering is there any easy way to do the above? The list manager does not
seem to support adding a samll icon to its cell. Any help would be greatly
appreciated.

Regards,
Wilson.
- --
Wilson Fu (Nocturnal Bug)       e-mail: [e-mail]

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

From: [e-mail] (John Purlia)
Organization: Qualcomm, Inc
Date: Thu, 12 Mar 1992 23:58:52 GMT

In article <[e-mail]> [e-mail] (Wilson Fu) writes:
> Wondering is there any easy way to do the above? The list manager does
not
> seem to support adding a samll icon to its cell. Any help would be
greatly
> appreciated.

That's correct, the list manager's standard LDEF is a text only list.  In
order to display small icons in a cell, you need a custom LDEF which draws
icons in the cells.  Luckily, an LDEF is one of the simpler defProcs to
write (as it basically just needs a draw routine), and luckier still,
there is an excellent small icon LDEF example included in Apple's DTS
Sample Code folder.

- -- John

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

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

From: [e-mail] (David Swofford )
Subject: How to force Finder filesize etc. update after THINK C stdio?
Date: 12 Mar 92 18:00:32 GMT
Organization: University of Illinois at Urbana

If a Finder window is open and the size of a file is changed using THINK
C 5.0 stdio after the Finder first sees the file, it doesn't seem to
update the file size and other Finder info properly.  The simplest
example of this is:

   Open a Finder window containing file 'foo.txt' in list mode (i.e., by
       name, size, etc.).
   Do the following:

         fp = fopen("foo.text", "a");
         [write a bunch of stuff to fp]
         fclose(fp);
         [exit program]

The file size displayed in the Finder doesn't change no matter how many
bytes (or kilobytes, or megabytes) you write to it.  If you close the
window and reopen, it displays the correct size.  Obvious things like
calling SetFInfo etc. don't help.

In pre-System 7 versions of the OS, I found by trial-and-error that I
could force the Finder to update the file size and other info by
creating a new temporary file and immediately deleting it.  Somehow,
messing with the directory entries caused the Finder to recollect its
information about the files.  Even this doesn't seem to work under
System 7.  Under System 6 and earlier, this bug/feature was particular
annoying because copies mediated by the Finder (including duplicate,
etc.) used the Finder's incorrect notion of the file size correct and
did bizarre things such as spilling the progress bar outside of the
thermometer display, rapid multiple flashing back and forth between
reading and writing, etc.

I imagine that this has come up before in this group, but I don't
remember anything about it and there was no mention in the FAQ list.
Any suggestions on how I can FORCE the Finder to update its info from
within my app would be appreciated.
- --
David L. Swofford                 Phone:    [phone]
Illinois Natural History Survey   FAX:      [phone]
607 E. Peabody Drive              BITNET:   DAVESWOF@UIUCVMD
Champaign, Illinois 61820 USA     Internet: [e-mail]

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

From: [e-mail] (Jon W{tte)
Date: 13 Mar 92 17:09:13 GMT
Organization: Royal Institute of Technology, Stockholm, Sweden

.cso.uiuc.edu> [e-mail] (David Swofford ) writes:

   The file size displayed in the Finder doesn't change no matter how many
   bytes (or kilobytes, or megabytes) you write to it.  If you close the
   window and reopen, it displays the correct size.  Obvious things like
   calling SetFInfo etc. don't help.

The apple-sanctioned way is to change the modified date
of the enclosing folder.

>From the limitless depths of my hairy, ever-growing utilities.c
file:

OSErr
TouchDir ( short vol , long dir )
{
    CInfoPBRec info ;
    Name name ;
     OSErr err ;

    BlockClear ( & info , sizeof ( info ) ) ;
    info . dirInfo . ioDrDirID = dir ;
    info . dirInfo . ioVRefNum = vol ;
    info . dirInfo . ioNamePtr = name ;
    info . dirInfo . ioFDirIndex = -1 ;

    err = PBGetCatInfo ( & info , 0 ) ;

    if ( ! err ) {

        info . dirInfo . ioCompletion = 0 ;
        info . dirInfo . ioDrDirID = info . dirInfo . ioDrParID ;
        info . dirInfo . ioFDirIndex = 0 ;
        GetDateTime ( & ( info . dirInfo . ioDrMdDat ) ) ;

        err = PBSetCatInfo ( & info , 0 ) ;
    }

    return err ;
}


The implementation is left as an exercise to the reader.

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

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

From: [e-mail] (David Swofford)
Date: 13 Mar 92 00:00:32 GMT
Organization: Macnet Omaha

(From: [e-mail] (David Swofford ))
(Organization: University of Illinois at Urbana)

If a Finder window is open and the size of a file is changed using THINK
C 5.0 stdio after the Finder first sees the file, it doesn't seem to
update the file size and other Finder info properly.  The simplest
example of this is:

   Open a Finder window containing file 'foo.txt' in list mode (i.e., by
       name, size, etc.).
   Do the following:

         fp = fopen("foo.text", "a");
         [write a bunch of stuff to fp]
         fclose(fp);
         [exit program]

The file size displayed in the Finder doesn't change no matter how many
bytes (or kilobytes, or megabytes) you write to it.  If you close the
window and reopen, it displays the correct size.  Obvious things like
calling SetFInfo etc. don't help.

In pre-System 7 versions of the OS, I found by trial-and-error that I
could force the Finder to update the file size and other info by
creating a new temporary file and immediately deleting it.  Somehow,
messing with the directory entries caused the Finder to recollect its
information about the files.  Even this doesn't seem to work under
System 7.  Under System 6 and earlier, this bug/feature was particular
annoying because copies mediated by the Finder (including duplicate,
etc.) used the Finder's incorrect notion of the file size correct and
did bizarre things such as spilling the progress bar outside of the
thermometer display, rapid multiple flashing back and forth between
reading and writing, etc.

I imagine that this has come up before in this group, but I don't
remember anything about it and there was no mention in the FAQ list.
Any suggestions on how I can FORCE the Finder to update its info from
within my app would be appreciated.
- --
David L. Swofford                 Phone:    [phone]
Illinois Natural History Survey   FAX:      [phone]
607 E. Peabody Drive              BITNET:   DAVESWOF@UIUCVMD
Champaign, Illinois 61820 USA     Internet: [e-mail]

- ---
 * Origin: Interuniverse Gateway (ivgate), Omaha (11:30102/2.0)
SEEN-BY: 285/1 285/14

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

From: [e-mail] (Danny W. McCampbell)
Subject: Drawing a Picture in a Modal Dialog?
Date: 12 Mar 92 18:34:47 GMT
Organization: University of Tennessee

I have a simple program where I want to draw a picture in a modal
dialog with out having to call DrawDialog.  I notice that when I
call DrawDialog I get a flicker in the Static Text when the dialog
is redraw.  So I thought I would just draw the picture into the dialog,
but I am unsure about how to do this since I am new to programming
the mac.  Could someone please help me out?

Danny McCampbell
[e-mail]

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

From: [e-mail] (Danny W. McCampbell)
Date: 13 Mar 92 13:51:02 GMT
Organization: University of Tennessee

In article <[e-mail]>, [e-mail] (Danny W. McCampbell) writes:
>
> I have a simple program where I want to draw a picture in a modal
> dialog with out having to call DrawDialog.  I notice that when I
> call DrawDialog I get a flicker in the Static Text when the dialog
> is redraw.  So I thought I would just draw the picture into the dialog,
> but I am unsure about how to do this since I am new to programming
> the mac.  Could someone please help me out?
>
> Danny McCampbell
> [e-mail]
>
Never mind, I figured it out.  Thanks anyway.

Danny

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

From: [e-mail] (Chris Anderson)
Subject: List manager problem
Date: 11 Mar 92 22:59:07 GMT
Organization: Hewlett-Packard, Fort Collins, CO, USA

Mac programmers:

I know  questions  about the list manager  have been posted here  before, but I
didn't read them because I didn't use the list manager.  I am now trying to use
the list manager, I am having  problems, so NOW I AM INTERESTED.  Please, would
someone who knows, send me the answer to this possibly simple problem.

Set up:  Mac SE (8 MHz 68000)  running  under System  6.0.7; I am using Think C
version 4.0.

Here is the problem:

I have a Dialog  Window that has a list in it (I use  LNew(...)  to  initialize
it).  The destination rect and the scroll bar control  (vertical only) rect are
defined in the Dialog Resource as User Items.  I use  Dialog_Select to look for
mouseDown  events in those items and pass the  mouseDown  pt (in local  coords)
along to LClick.  The first  problem is that the scroll bar does not respond at
all (IM-IV implies that all of the scroll bar stuff is handled by LClick).

The second problem is really weird.  As long as the mouse button is down in the
destination rect, LClick seems to work ok, as I see it highlight the list cells
as the mouse passes over each one, and even scroll automatically when the mouse
is moved outside the  destination  rect (the scroll bar's thumb even moves with
the  scrolling).  Yet when I release the mouse  button and call  LLastClick  to
find out which list cell was last  clicked  in, it returns a cell number 2 OR 3
greater  than the top cell.  LClick also does not seem to respond to  mouseDown
events in the last 2 OR 3 cells -- it does not highlight them.

The last problem may be related to the  previous  one.  I cannot get Think C (v
4.0) to correctly  assign a structured  function  result  (like Point or Cell).
for example:
        Cell        theCell;
        .
        .
        .
        theCell = LLastClick(...);   /* compiler doesn't like this */

What I have been able to do is coerce the function result to type long and then
assign each element of the  structure  (the .h and the .v) from the  high-order
word and  low-order  word of the long  result.
for example:
        Cell        theCell;
        long        l;
        .
        .
        .
        l = (long)LLastClick(...);
        theCell.v = (int)(l >> 16);
        theCell.h = (int)(l & 0xffff); /* compiler accepts this */

This  seems like a kludge to me.  Is there a better way (say, use a UNION)?  Is
this part of the  problem of the cell  numbers  being 2 OR 3 greater  than they
should?  I don't know.

Please respond by e-mail to [e-mail] since my connection to the
Internet  is down for an  undetermined  period  of time, and I won't be able to
read this news group.

Chris Anderson ([e-mail])

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

From: [e-mail] (Jack Herrington)
Organization: University of Miami Medical School
Date: Fri, 13 Mar 1992 00:29:54 GMT

[e-mail] (Chris Anderson) writes:
>              Yet when I release the mouse  button and call  LLastClick  to
>find out which list cell was last  clicked  in, it returns a cell number 2 OR 3
>greater  than the top cell.  LClick also does not seem to respond to  mouseDown
>events in the last 2 OR 3 cells -- it does not highlight them.

1) Make sure your data rect has 0,0 as the left and top, and your data limits
   as bottom and right.
2) What you really want to know from LLastClick is what is selected, I guess?
   If it is, try this:

    Point selCell;
    selCell.h = selCell.v = 0;
    if ( LGetSelect(true,&selCell,theList) )
    {
        ...
        now you have the cell address
        ...
    }

   I know it makes sense to use LLastClick, but then you have to use LoWord
   and HiWord, and cast the result into a long, it's a drag.  This way you make
   sure what you got is selected, too.
- --
"Electric word 'life', it means forever and that's a might long time.  But I'm
 here to tell yah, there's something else... The after-life, a word of
 never-ending happiness, you can always see the sun, day or night.  So when you
 call up that shrink in Beverly hills, you know the one, Dr. everything-we-all-

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

From: [e-mail] (Chris Anderson)
Date: 13 Mar 92 16:21:15 GMT
Organization: Hewlett-Packard, Fort Collins, CO, USA

Mac programmers:

Thank you for all of your responses.  All of the problems with the List manager
I have had are now solved.

Chris Anderson ([e-mail])

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

From: [e-mail] (David Putzolu)
Subject: Help! Filling a rect with a uniform color
Date: 13 Mar 92 03:38:11 GMT
Organization: Department of Computer Science, University of California, Davis

Help! I am trying to do a very simple thing: Fill a rect that is in an
open window with the color stored in a variable of type RGBColor. But
the obvious procedure, FillCRect, requires a PixPatHandle, which,
looking at it in my IM books, I can't figure out for the life of me.
Can anyone tell me how to do this very simple thing? Thanks in
advance for any help....

Also, let me express my gratitute to Matt Hall, Warren DeLano, and
Bill Leue, whose suggestions were extremely helpful in dealing with
arrays that wouldn't fit in the 32k program segment limit. Thanks
again!

| David M. A. Putzolu                  | [e-mail]                |
| Senior Undergraduate                 | op disclaimer(opinion : ptr mine)     |
| Computer Science and Psychology      |       Aiuto! Sono caduto e            |
| University of California at Davis    |         non posso alzarmi!            |

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

From: [e-mail] (Christopher Kempke)
Date: 13 Mar 92 21:10:41 GMT
Organization: Oregon State University, Computer Science Dept.

In article <[e-mail]> [e-mail] (David Putzolu) writes:
>Help! I am trying to do a very simple thing: Fill a rect that is in an
>open window with the color stored in a variable of type RGBColor. But
>the obvious procedure, FillCRect, requires a PixPatHandle, which,

    RGBForeColor(theColor);
    PaintRect(theRect);

    "Fill" uses a pattern, "Paint" uses a color, "Frame" draws an outline.

    Hope this helps.

    -Chris

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

From: [e-mail] (Kiran Wagle)
Subject: iostream package for Think C?
Date: 13 Mar 92 08:52:01 GMT
Organization: Indiana University

I would like to use Think C 5 to test some routines that depend
heavily on simple i/o. I'd like to use the iostream routines for this,
but they're not included with Think C.

If they're written in C or C++ (which was suggested in Stroustrup's
reference) then perhaps they could be compiled to run under think c?

If anyone knows where to get source code, or has tried using them with
think c, please send any advice you may have.

Thanks in advance.
    -Kiran

- --
"The truffle is not an outright aphrodisiac, but it may in certain
 circumstances make women more affectionate and men more amiable."
                                                ...Brillat-Savarin

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

From: [e-mail] (Matthias Ulrich Neeracher)
Date: 13 Mar 92 16:22:27 GMT
Organization: Integrated Systems Laboratory, ETH, Zurich

In article <[e-mail]> [e-mail] (Kiran Wagle) writes:
>I would like to use Think C 5 to test some routines that depend
>heavily on simple i/o. I'd like to use the iostream routines for this,
>but they're not included with Think C.

For good reasons...

>If they're written in C or C++ (which was suggested in Stroustrup's
>reference) then perhaps they could be compiled to run under think c?

Uhm, I don't think so. Doesn't Think C abhor operator overloading and multiple
inheritance ? iostream is inconceivable without operator overloading and The
current implementatikon also relies on multiple inheritance.

>If anyone knows where to get source code, or has tried using them with
>think c, please send any advice you may have.

There is an iostream library coming with g++. If TC could compile it (It can't
(He He)), you could easily get it, but I'm a little unclear as to whether your
application would fall under the General Public License, which you might not
want to happen.

Matthias

- -----
Matthias Neeracher                                   [e-mail]
   "No, what he didn't like about heroes was that they were usually
    suicidally gloomy when sober and homicidally insane when drunk."
                          -- Terry Pratchett, _The Colour of Magic_

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

From: [e-mail] (Thomas Moriarty)
Date: 15 Mar 92 02:27:12 GMT
Organization: Texas A&M University


Kiram,

It is unfortunate that Think C does not have an iostream class available.

As you know, Think C is produced by Symantec. Symantec acquired Zortech
Ltd. about a year ago. Zortech does have a C++ compiler with iostream
classes for the Macintosh. It would make good sense for Symantec to merge
the Zortech C++ Macintosh version together with Think C. If Symantec could
do this, they could have a new and powerful product that solves the
deficiencies that you mentioned.

- -- Tom Moriarty

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

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