Celestin Apprentice 2/Apprentice-Release2.iso/Information/Digests/CSMP Digest/volume 1/csmp-v1-157.txt

Text File | 1994-12-08 | 43.0 KB | 1,188 lines | [TEXT/R*ch]

C.S.M.P. Digest             Mon, 03 Aug 92       Volume 1 : Issue 157

Today's Topics:

    *ugh* More menu probs
    MBDF Protocol
    68030 Instruction List wanted
    How to print a postscript file?
    Think C / ANSI Gotcha
    Internationalization of a dialog box
    Magazine Subs Info - Please help
    Converting a Unix Sockets-based program to MacTCP
    How do I dequeue MacTCP calls?
    Does the PolySmooth PicComment work?
    Why is gray drawing slower than color?



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

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

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

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

The digest is also available via email.  Just send a note saying that you
want to be on the digest mailing list to [e-mail], and you will
automatically receive each new issue as it is created.  Sorry, back issues
are not available through the mailing list.

Send administrative mail to [e-mail].


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

From: [e-mail] (michael a migdol)
Subject: *ugh* More menu probs
Date: 24 Jun 92 21:24:44 GMT
Organization: The University of Texas at Austin, Austin TX

Ah, well, another day, another bug, another plea to csmp. :)

The program I'm working on uses a single menu-update routine. Here's the
meat of it:

void MakeMenu() {
    Handle       barHndl;
    MenuHandle menHndl;

    /* Reset menus */
    barHndl = GetNewMBar(MyMBarID);
    SetMenuBar(barHndl);

    /* Some stuff deleted here */

    menHndl = GetMHandle(AppleMenu);
    AddResMenu(menHndl,'DRVR');
    DrawMenuBar();
}

The stuff in the deleted section works fine, but when this routine is called
for the second time, GetMHandle(AppleMenu) returns a zero, leading to a
generally chaotic state of the universe in which the apple menu has several
copies of the DA list floating around in it. :)


Any suggestions would be greatly appreciated. I'll post any interesting replies
back to the group.
Thanks in advance!
- -Michael

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

From: [e-mail] (michael a migdol)
Date: 24 Jun 92 22:18:34 GMT
Organization: The University of Texas at Austin, Austin TX

Ok, everyone, repeat after me....
    "Think Reference is not a substitute for Inside Mac"
    "Think Reference is not a substitute for Inside Mac"
    "Think Reference is not a substitute for Inside Mac"

Thanks to Stepan Riha for pointing out that it's a no-no to call GetNewMBar
more than once. (A fact that's mentioned in IM I, but not TR) :)

Is it Friday yet?
- -Michael


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

From: Stephan Bublava <[e-mail]>
Organization: Fachschaft Informatik, TU Wien
Date: Thu, 25 Jun 1992 12:15:54 GMT

>The program I'm working on uses a single menu-update routine. Here's the
>meat of it:
>
>void MakeMenu() {
>    Handle       barHndl;
>    MenuHandle menHndl;
>
>    /* Reset menus */
>    barHndl = GetNewMBar(MyMBarID);
>    SetMenuBar(barHndl);
>
>    /* Some stuff deleted here */
>
>    menHndl = GetMHandle(AppleMenu);
>    AddResMenu(menHndl,'DRVR');
>    DrawMenuBar();
>}

I suppose by "menu-update routine" you mean a routine that disables and
enables menu items. In such a routine there is no need to call GetNewMbar.
In fact I call GetNewMBar only once in my programs (but i use only one
menu bar.)

The listing looks like a routine that initializes the menus at programm
startup. By the way you should also call DisposeHandle (barHndl) at the
end of your routine because SetMenuBar makes a copy of the handle you
pass it.

If you really want do disable and enable menu items do someting like the
following:

    menHandle := GetMHandle (menu_id);
    DisableItem (menHandle, item_number);
    ...

If this is not what you want to do, please specify the purpose of your
routine in more detail.

Hope this helps,

Stephan Bublava
krapf@email

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

From: [e-mail]
Date: 29 Jun 92 15:10:05 GMT
Organization: Univ of Miami IR

In article <[e-mail]>, [e-mail] (michael a migdol) writes:

> Thanks to Stepan Riha for pointing out that it's a no-no to call GetNewMBar
> more than once. (A fact that's mentioned in IM I, but not TR) :)

As a side note: Remember menu ID's and MENU resource ID's don't necessarily
match. If you create a MENU in ResEdit, they will match initially if there are
no resource conflicts. But if you start changing menus around, they can end up
different. This can cause all sorts of problems that would be hard to diagnose
if you didn't know about the ID problem. Just thought I'd point that out :)

Dan
/-------------------------------------------------------------------------\
|   Dan Weisman -  University of Miami - Florida   |  ||   ||   ||   ||   |
|--------------------------------------------------|  ||   ||   ||\ /||   |
|   INTERNET  -----> [e-mail]  |  ||   ||   || | ||   |
|     BITNET  -----> dweisman@umiami               |  |||||||   || | ||   |
|-------------------------------------------------------------------------|
|  "Think for yourself and feel the walls become sand beneath your feet"  |
|    -Queensryche, "Anybody Listening"                  |
\_________________________________________________________________________/

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

From: [e-mail] (Humayun Lari)
Date: 29 Jun 92 17:04:53 GMT
Organization: University of North Carolina, Chapel Hill

In article <[e-mail]> [e-mail] writes:
>As a side note: Remember menu ID's and MENU resource ID's don't necessarily
>match. If you create a MENU in ResEdit, they will match initially if there are
>no resource conflicts. But if you start changing menus around, they can end up
>different. This can cause all sorts of problems that would be hard to diagnose

A good way to prevent this from happening (at least in ResEdit 2.1.1) is to
remember to always open the menu before choosing Get Resource Info to set
the resource ID. ResEdit will warn you that the internal menu ID is different
from the new resource ID and ask if you want to change that too; just hit
Return and it'll do it for you. Nicely done, but IMHO you shouldn't have to
open the menu first...

>...Just thought I'd point that out :)

Just thought I'd point this out. :-)


Humayun Lari
([e-mail])

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

From: [e-mail] (Bruce Leban)
Subject: MBDF Protocol
Date: 25 Jun 92 16:47:27 GMT
Organization: lcs.mit.edu

I have a question about the protocol for calling an MBDF.  The GetRect
message does not appear to work as documented.

The MBDF itself is documented as returning a longint, but the GetRect
message is documented as returning a rect.  Clearly a rect cannot fit in a
longint.  (The MBDF is documented at the end of the Menu Manager chapter in
volume V; I think the respective page numbers are 251 and 254, but the book
is at home.)  I've examined the value actually returned by the MBDF and it
appears to be neither a pointer nor a handle to a rect.  I've tried passing
a pointer to a rect as an extra first or last parameter and neither works.
(Incidentally, I'm calling this from Think C.)  Does anyone know how this is
really supposed to work?

I have looked through all 6 volumes of Inside Mac and the most recent
version of the Tech Notes stack downloaded from ftp.apple.com and I cannot
find any explanation.

Thanks for any help.

    --- Bruce
    [e-mail]  @somerville.mass.usa.earth

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

From: [e-mail] (Keith Rollin)
Date: 26 Jun 92 18:37:18 GMT
Organization: Taligent

In article <[e-mail]>,
[e-mail] (Bruce Leban) writes:
>
> I have a question about the protocol for calling an MBDF.  The GetRect
> message does not appear to work as documented.
>
> The MBDF itself is documented as returning a longint, but the GetRect
> message is documented as returning a rect.  Clearly a rect cannot fit in a
> longint.  (The MBDF is documented at the end of the Menu Manager chapter in
> volume V; I think the respective page numbers are 251 and 254, but the book
> is at home.)  I've examined the value actually returned by the MBDF and it
> appears to be neither a pointer nor a handle to a rect.  I've tried passing
> a pointer to a rect as an extra first or last parameter and neither works.
> (Incidentally, I'm calling this from Think C.)  Does anyone know how this is
> really supposed to work?
>
> I have looked through all 6 volumes of Inside Mac and the most recent
> version of the Tech Notes stack downloaded from ftp.apple.com and I cannot
> find any explanation.

The MBDF is returning a pointer to a rect. Often, this is a pointer to a
low-memory global. Other times, it's to a rect in a private MBDF data structure.

BTW: I'm not sure if MBDF writing is supported. There are many messages that the
Menu Manager sends to the MBDF that aren't documented anywhere. I only know this
because I've been disassembling the 7.0 MBDF so that I, too, can write my own
:-).

Perhaps someone with access to the new Inside Mac can tell us what the
documentation for the 90's says.

- --
Keith Rollin
Phantom Programmer
Taligent, Inc.


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

From: [e-mail] (Kevin Hill)
Subject: 68030 Instruction List wanted
Date: 25 Jun 92 16:14:51 GMT
Organization: Crash TimeSharing, El Cajon, CA


  I am curious, does someone out in netland have file that has all
the opcodes for the 68000 family?  can you send it to me?  Preferably,
it will have info on each instruction (i.e. what regs are used in the
opcode)
  If no one has heard of such a thing, perhaps mention a good book that
goes over it all...
  -Kevin

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

From: [e-mail] (Nikolaus E Schuessler)
Organization: Iowa State University, Ames, IA
Date: Fri, 26 Jun 1992 02:16:46 GMT

In article <[e-mail]> [e-mail] (Kevin Hill) writes:
>  I am curious, does someone out in netland have file that has all
>the opcodes for the 68000 family?  can you send it to me?  Preferably,
>it will have info on each instruction (i.e. what regs are used in the
>opcode)
>  If no one has heard of such a thing, perhaps mention a good book that
>goes over it all...
>  -Kevin
>

 Even better than a file, you can have the 'official' Motorola documentation
by calling / writing them... It will cost you almost nothing -- If you
are involved with a educational institution of some sort and you ask nice,
they may even send it to you for free...

You want the "Enhanced 32-Bit Microprocessor User's Manual" -- what they want
to know is the following part number: MC68030UM/AD (for the 68030)

I don't know if the ISBN will help or not but anyway here it is:
ISBN 0-13-566423-3 (again for the 68030)

There is also the 68000 Programmers Reference Manual which also seems
to include FPU instructions (M68000PM/AD).

Hope this helps!
- --
Niko Schuessler                                             [e-mail]
Project Vincent Systems Manager
Iowa State University                                       voice: 294-1672
Ames IA 50010

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

From: [e-mail] (Kent Sandvik (Hacker))
Date: 1 Jul 92 01:40:07 GMT
Organization: Apple

In article <[e-mail]>, [e-mail]
(Kevin Hill) wrote:
>   I am curious, does someone out in netland have file that has all
> the opcodes for the 68000 family?  can you send it to me?  Preferably,
> it will have info on each instruction (i.e. what regs are used in the
> opcode)
>   If no one has heard of such a thing, perhaps mention a good book that
> goes over it all...
>   -Kevin

68030 Assembly Language Reference, by Williams, has MPW 68000 asm code in
the book. Or then the Motorola specs.

Kent Sandvik/DTS

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

From: [e-mail] (Dan King)
Subject: How to print a postscript file?
Date: 25 Jun 92 03:58:21 GMT
Organization: University of Southern California, Los Angeles, CA



Howdy, y'all.  A quick question: What's the easiest way to print a
postscript file from within a program?

The situation:
I've got a program that is reading a MacTCP stream (which is just a
postscript file that a Unix box is vomiting at me).  I want to redirect
this stream to a LaserWriter.  (Yes, it's a poor man's print spooler,
but there are a few other added complexities that make me unable to use
any commercial product I've seen.)

Is there a quick and easy solution?

Thanks,
dank

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

From: [e-mail] (Dana S Emery)
Date: 26 Jun 92 03:46:41 GMT
Organization: Personal

In article <[e-mail]>, [e-mail] (Dan King) writes:
>
> I've got a program that is reading a MacTCP stream (which is just a
> postscript file that a Unix box is vomiting at me).  I want to redirect
> this stream to a LaserWriter.
>...
> Is there a quick and easy solution?

Not unless Adobe has upgraded the SendPS utility to accept Apple events.

Baring that:

You will have to parse the stuff (hopefully it will follow the adobe
structured comment conventions).  You can emit PS either of 2 ways.

1 - via pic comments.
2 - via drawtext/drawchar/drawstring, with font set to 'Postscript_Escape'
(special font which your code should supply, see LW driver 'STR ' -8188 for
actual name, I think Sumex has an FTP'able example.)

The LW driver will ignore the content of the PS, and so it falls on your code
to ensure that fonts not inline with the PS stream are supplied.

Your parsing will have 2 objectives.

1 - establishing font requirements.
2 - isolateing pages for emision (remember, the LW driver reserves the
pagination and header emission for itself.)

You may have to buffer to disk in order to deal with the 'AtEnd' convention.

Good luck, Dana S Emery ([e-mail])

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

From: [e-mail] (Peter G. Cramer)
Subject: Think C / ANSI Gotcha
Date: 25 Jun 92 18:19:49 GMT
Organization: Case Western Reserve University, Cleveland, OH (USA)


I have run across a gotcha in Think C related to the use
of ANSI standard routines.  The bottom line is that if
you are using ANSI functions, make sure "native floating
point format" under "Compiler Settings" is turned OFF!
(at least in version 5.0.)

If you want to try it for yourself, here is a program
you can use:
- --------------------
#include <stdio.h>

main()
{
    char        *s = "3.14159";
    float        myVar;

    sscanf( s, "%f", &myVar );
    printf( "myVar = %f\n", myVar );
}
- -------------------

I guess it's not so surprising that this is the case.  If it isn't
mentioned in the manuals (I haven't seen it), it should be.

Best,
- --
Peter G. Cramer, Physics Department, CWRU
University Circle   Cleveland, Ohio 44106
[phone]            [e-mail]

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

From: [e-mail] (Scott Bronson)
Date: 29 Jun 92 19:55:35 GMT

In <[e-mail]> [e-mail] (Peter G. Cramer) writes:


>The bottom line is that if
>you are using ANSI functions, make sure "native floating
>point format" under "Compiler Settings" is turned OFF!
>If it isn't mentioned in the manuals (I haven't seen it), it should be.


If this isn't in the FAQ, it should be.

Though the THINK manual doesn't go out of its way to document this, it
is certainly in there.  If you browse the index for five or ten minutes
you will eventually find out where this is described.  Sorry I don't have
the manual here at school, or I could tell you right now.

If you change the "mc -68881/882" or "Use Native Floating Point" options
from their default values, you must re-compile the ANSI libraries or their
object code will not reflect the format you are using for your source
(whether or not you are using SANE's 80-bit extendeds or Motorola FPU's
96-bit ones).

    - Scott

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

From: [e-mail] (Pete Resnick)
Subject: Internationalization of a dialog box
Date: 25 Jun 92 18:57:54 GMT
Organization: University of Illinois at Urbana

I have a piece of code which I am hoping to make as
"internationalizable" as possible. Most of the stuff in the Script
Manager I am having no problems with, but I have run into two problems
concerning text in a Modal dialog box for which I cannot find the
answer. This code, by the way, must run under System 6 as well as
System 7.

1. I have a few TE boxes in this dialog. One of them can take any sort
of printable text and the others can take only numerals (i.e. things I
can convert to numbers in the program). I want to give some feedback
to the user from within my dialog filter by flagging keystrokes that
are no good and beeping. This would mean not allowing non-printables
in the first box and not allowing non-digits in the others. The
problem is that I could get double-byte characters, i.e. two separate
events. What is the best way to handle this? Can I filter the way I
want to, or do I basically have to let the user enter what they want
and go back and complain once they hit "OK"?

2. I am going to draw some text in a small box in my dialog. The box
is just a userItem in my DITL which I will do a FrameRect; it will be
wide enough for the longest possible string. What I am having a
problem with is how to actually draw the text. If I am in a
left-to-write script, I simply move to the left side of the box and
call DrawText. But if I am in a right-to-left script, where do I start
my DrawText? Should I start at the right side of the box, or
MeasureText pixels from the right side? Is there an cleaner way to do
this?

Sample code would be appreciated, but ideas are really what I am
looking for.

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]
Organization: Campus Computing Sites, University of Michigan-Ann Arbor
Date: Fri, 26 Jun 92 02:34:27 GMT

In article <[e-mail]> [e-mail] (Pete Resnick) writes:
>I have a piece of code which I am hoping to make as
>"internationalizable" as possible. Most of the stuff in the Script
>Manager I am having no problems with, but I have run into two problems
>concerning text in a Modal dialog box for which I cannot find the
>answer. This code, by the way, must run under System 6 as well as
>System 7.
>
    No problem....

>1. I have a few TE boxes in this dialog. One of them can take any sort
>of printable text and the others can take only numerals (i.e. things I
>can convert to numbers in the program). I want to give some feedback
>to the user from within my dialog filter by flagging keystrokes that
>are no good and beeping. This would mean not allowing non-printables
>in the first box and not allowing non-digits in the others. The
>problem is that I could get double-byte characters, i.e. two separate
>events. What is the best way to handle this? Can I filter the way I
>want to, or do I basically have to let the user enter what they want
>and go back and complain once they hit "OK"?
>
    Take a look at the CharByte() routine (somewhere in IM V). It will
tell you whether you are looking at the first byte of a 2-byte character,
the 2nd byte of a two byte character or a single byte character. There should
also be an example of using it in either IM V or VI.

>2. I am going to draw some text in a small box in my dialog. The box
>is just a userItem in my DITL which I will do a FrameRect; it will be
>wide enough for the longest possible string. What I am having a
>problem with is how to actually draw the text.
>
    What I usually do is to use _TextBox, which admittedly uses TextEdit
to do the drawing, but since you are talking about a small amount of text
it will work just fine.  That way TE gets to worry about all the issues and
you can get justifications (jeft/center/right) if you want.


- --
- -----------------------------------------------------------------------
Leonard Rosenthol          Internet: [e-mail]
Director of Advanced Technology   AppleLink: MACgician
Aladdin Systems, inc.          GEnie:     MACgician

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

From: [e-mail] (Jamie R. McCarthy)
Organization: Kalamazoo College
Date: Fri, 26 Jun 1992 14:51:39 GMT

[e-mail] writes:
>[e-mail] (Pete Resnick) writes:
>>
>>I have a piece of code which I am hoping to make as
>>"internationalizable" as possible.  ... but I have run into two problems
>>concerning text in a Modal dialog box ...
>>
>>2. I am going to draw some text in a small box in my dialog. The box
>>is just a userItem in my DITL which I will do a FrameRect; it will be
>>wide enough for the longest possible string. What I am having a
>>problem with is how to actually draw the text.
>>
>    What I usually do is to use _TextBox, which admittedly uses TextEdit
>to do the drawing, but since you are talking about a small amount of text
>it will work just fine.  That way TE gets to worry about all the issues and
>you can get justifications (jeft/center/right) if you want.

You might want to check out NeoTextBox, Bryan Ressler's improvement on
that trap.  I'm not that "up" on internationalization, but Bryan makes
lots of references to the Script Manager in his code, so it must be
good.  :-)

It's in the Winter 1992 develop (if you're in the Northern Hemisphere).
- --
 Jamie McCarthy      Internet: [e-mail]      AppleLink: j.mccarthy
 Never piss off a computer.

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

From: [e-mail]
Organization: Campus Computing Sites, University of Michigan-Ann Arbor
Date: Fri, 26 Jun 92 17:53:00 GMT

In article <[e-mail]> [e-mail] (Jamie R. McCarthy) writes:
>[e-mail] writes:
>>[e-mail] (Pete Resnick) writes:
>>>
>>> [drawing international text in a box]
>>>
>>    What I usually do is to use _TextBox, which admittedly uses TextEdit
>>to do the drawing, but since you are talking about a small amount of text
>>it will work just fine.  That way TE gets to worry about all the issues and
>>you can get justifications (jeft/center/right) if you want.
>
>You might want to check out NeoTextBox, Bryan Ressler's improvement on
>that trap.  I'm not that "up" on internationalization, but Bryan makes
>lots of references to the Script Manager in his code, so it must be
>good.  :-)
>
    If memory serves me correctly, the NeoTextBox stuff is NOT
Script Manager compatible - though it is fast and does do full justification.
I know someone who has done a bit of "internationalizatiion" to NTB, but it
still has a way to go.


- --
- -----------------------------------------------------------------------
Leonard Rosenthol          Internet: [e-mail]
Director of Advanced Technology   AppleLink: MACgician
Aladdin Systems, inc.          GEnie:     MACgician

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

From: [e-mail] (Pete Resnick)
Organization: University of Illinois at Urbana
Date: Sun, 28 Jun 1992 00:02:12 GMT

Thanks to Leonard, Jamie, and Dana for some answers and ideas.

1. For the numbers, CharType seems to work fine, though I have not
tested it on an interational machine yet. I think this will do it for
me.

2. As far as drawing text in my box goes: Though I used the TE
routine TextBox elsewhere in my code (thanks for pointing that one
out), it turns out that I need to do manipulation on the text in the
original item I was asking about. So, I am checking GetSysJust() and
doing some Char2Pixel's and drawing it by hand. The Script Manager
routines are working out great!

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] (ap.parkman)
Subject: Magazine Subs Info - Please help
Date: 25 Jun 92 14:40:03 GMT

I am a programmer interested in obtaining some overseas
subscriptions to U.S. Mac Magazines such as MacWorld/MacUser
which impressed me on a recent visit as being:

     * Ahead of the U.K. versions in terms of News
     * Bigger coverage then U.K. versions in terms of more
       useful articles
     * Useful for keeping an eye on U.S. Prices for comparison
       with U.K. Pricing

To speed the process of obtaining such a subscription I would
ideally like to FAX a subscription request payable against my
MasterCard. I was wondering:

     - What other magazines do you suggest I consider?
     - Could anyone be kind enough to send me the FAX
       numbers of the relevant subscription departments
       ( including MacWorld and MacUser )

You could e-mail me the responses if you prefer, although I
suspect an airing of the matter could prove interesting to
many readers of magazines who are unaware of the relative
merits of alternatives. If there is a good response I'll
summarise the opinions about each magazine.

Yours in hopeful anticipation.

Andy Parkman

Research Programmer,
Dept. of Computer Science,
Univ. Coll. of Swansea,
Singleton Park,
Swansea. SA2 8PP.
U.K.

e-mail: [e-mail]

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

From: [e-mail] (John Wallace)
Organization: TRW Data Systems Center, Redondo Beach, CA
Date: Fri, 26 Jun 92 22:18:51 GMT

In article <[e-mail]> [e-mail] (ap.parkman)
writes:
>I am a programmer interested in obtaining some overseas
>subscriptions to U.S. Mac Magazines such as MacWorld/MacUser
>which impressed me on a recent visit as being:
>
>     * Ahead of the U.K. versions in terms of News
>     * Bigger coverage then U.K. versions in terms of more
>       useful articles
>     * Useful for keeping an eye on U.S. Prices for comparison
>       with U.K. Pricing
>
>To speed the process of obtaining such a subscription I would
>ideally like to FAX a subscription request payable against my
>MasterCard. I was wondering:
>
>     - What other magazines do you suggest I consider?
>     - Could anyone be kind enough to send me the FAX
>       numbers of the relevant subscription departments
>       ( including MacWorld and MacUser )
>
>Andy Parkman
>e-mail: [e-mail]

It depends on what you want to read.

For news, I would recommend MacWeek.  As a weekly magazine, its focus is
to deliver the latest news about major dealings in the Macintosh market.
Its cost is $99 for US, $175 for Canada & Mexico, $300 for International.
Write to: Customer Service Department, MacWEEK, c/o JCI, P.O. Box
1766, Riverton, N.J. 08077-7366

For programming, I would recommend MacTutor.  It contains neat
articles that are of varying benefit depending on your programming
level on the Mac.  (I don't have a copy handy for subscription info)

For reviews and buying information, there's MacWorld and MacUser.  As
monthly publications, their emphasis is to evaluate products and make
recommendations to end users.  (Again, sorry but I don't have a copy
handy)

Cheers!
John

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

From: [e-mail] (Dana S Emery)
Date: 27 Jun 92 02:35:22 GMT
Organization: Personal

In article <[e-mail]> [e-mail] (ap.parkman)
writes:
>I am a programmer interested in obtaining some overseas
>subscriptions to U.S. Mac Magazines such as MacWorld/MacUser


Mactutor died, but is now reborn (freindly buyout)

[phone]

Xplain corp
po box 250025
Los Angeles CA, USA 90025-250025

(in US) $40/55 yr (+ $72 source code disks).

April-May 92 issue  is first under the new regime.

Dana S Emery. ([e-mail])

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

From: [e-mail] (Clay Luther)
Subject: Converting a Unix Sockets-based program to MacTCP
Organization: University of North Texas
Date: Thu, 25 Jun 1992 22:59:45 GMT

Can anyone offer advice on taking a simple unix sockets-based program, and
porting to MacTCP?

For instance, we have need to use ping.  We'd like to take the ping source and
make it a mpw tool.

Just looking for pointers and advice.

Thanks!

- --
Clay W. Luther                  [e-mail]
Macintosh/Unix Programmer for Vortech Data, Inc.
Virtual System Consultant for the UNT Center for Network Neuroscience
[phone]

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

From: [e-mail] (John Veizades)
Date: Fri, 26 Jun 1992 17:15:07 GMT
Organization: Apple Computer, Inc

In article <cluther.709513185@morticia>, [e-mail] (Clay
Luther) writes:
>
> Can anyone offer advice on taking a simple unix sockets-based program, and
> porting to MacTCP?
>
> For instance, we have need to use ping.  We'd like to take the ping source and
> make it a mpw tool.
>
> Just looking for pointers and advice.
>
On the developer CD starting at number X there are sources and objects to a ping
tool.  The tool is called MPing and it trys to emulate as much of the UNIX ping
functionality as is possible with the current release of MacTCP.

John Veizades...
Apple Computer, Inc.

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

From: [e-mail] (Jim Browne)
Organization: University of Illinois at Urbana
Date: Sat, 27 Jun 1992 00:59:39 GMT

[e-mail] (Clay Luther) writes:

>Can anyone offer advice on taking a simple unix sockets-based program, and
>porting to MacTCP?

>--
>Clay W. Luther                  [e-mail]
>Macintosh/Unix Programmer for Vortech Data, Inc.
>Virtual System Consultant for the UNT Center for Network Neuroscience
>[phone]

You can use the socket library in misc/unsupported on ftp.ncsa.uiuc.edu


- --
Jim Browne                                          | [e-mail]  |
NCSA/STG System Administrator                       | [e-mail] |
System Administration: Where the fun never ends...  | [phone]         |
"I was ionized... but now I'm better." - Bukaroo Banzai

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

From: [e-mail] (David Peterson)
Subject: How do I dequeue MacTCP calls?
Date: 25 Jun 1992 17:17:27 -0700
Organization: University of Southern California, Los Angeles, CA



Subject about sums it up...

How do I get rid of async MacTCP calls that haven't completed yet?

I can walk through my parameter blocks and find the ones that are inProgress,
but calling Dequeue() on them tells me they aren't in the queue (or bombs).
Sure enough, find the driver in Macsbug, and all the fields of dCtlQHdr are
nil.

So where are they queued, and how do I get rid of them?

Thanks,
- -dave.

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

From: [e-mail] (Pete Resnick)
Organization: University of Illinois at Urbana
Date: Fri, 26 Jun 1992 03:38:16 GMT

[e-mail] (David Peterson) writes:

>How do I get rid of async MacTCP calls that haven't completed yet?

If they are outstanding TCP calls, you can simply use *another*
parameter block, copy the tcpStream from the PB you want to canel into
the new PB, set the csCode to TCPAbort or TCPRelease (depending on
which you really want to do) and call PBControl (usually
synchronously, but you can do it asynchronously) on the new PB. The
inProgress PB will complete with connectionTerminated and you will get
a TCPNotify call with TCPTerminate and the terminReason set to
TCPULPAbort.

If they are outstanding UDP calls, you can also use another PB, copy
the udpStream from the PB you want to cancel to the new PB, set the
csCode to UDPRelease and call PBControl on the new PB. The inProgress
PB will complete with connectionTerminated as above. ***NOTE*** You
CAN'T do this if there is an outstanding UDPWrite, only on an
outstanding UDPRead!! There is no way to cancel an outstanding
UDPWrite.

>I can walk through my parameter blocks and find the ones that are inProgress,
>but calling Dequeue() on them tells me they aren't in the queue (or bombs).
>Sure enough, find the driver in Macsbug, and all the fields of dCtlQHdr are
>nil.

>So where are they queued, and how do I get rid of them?

Yeah, calling Dequeue would be a really bad idea. The Device Manager
only allows you to be working on one PB at a time, so MacTCP decided
to pull the PB's out of the queue and store them internally so that it
could work on more than one request at a time. That is why you
actually have to make another PBControl call to get things cancelled.

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] (Jason Philip Stevens)
Date: 26 Jun 1992 15:56:46 GMT
Organization: Baylor College of Medicine, Houston, Tx


In article <[e-mail]>, [e-mail] (David Peterson) writes:
|>
|>
|> Subject about sums it up...
|>
|> How do I get rid of async MacTCP calls that haven't completed yet?
|>
|> I can walk through my parameter blocks and find the ones that are inProgress,
|> but calling Dequeue() on them tells me they aren't in the queue (or bombs).
|> Sure enough, find the driver in Macsbug, and all the fields of dCtlQHdr are
|> nil.

A queued MacTCP call can be dequeued by calling TCPAbort (MacTCP Developer's
guide, p.54).  This does have unfortunate side effects, however:  your connection
is closed and all outstanding data is lost.

- -jps

- --
Jason Stevens            Internet:  [e-mail]
Network User Services        Voice:  [phone]
Baylor College of Medicine    Opinions expressed are mine alone.


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

From: [e-mail] (David Peterson)
Date: 26 Jun 1992 11:10:52 -0700
Organization: University of Southern California, Los Angeles, CA



Guess I didn't make myself clear enough the first time...

I need to dequeue calls without killing the current connection.

What I'm trying to do is hand the StreamPtr off to another program,
but I have to kill all pending IO in the current program first.

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

From: [e-mail] (Eric Bowman)
Subject: Does the PolySmooth PicComment work?
Date: 25 Jun 92 23:56:33 GMT
Organization: Reed College, Portland, OR

I've attempted make it work, but it appears to have no effect; furthermore,
it doesn't appear to make any changes to the postscript output.  Has any-
one out there gotten it to work with a LaserWriter?

Thanks,
bobo
[e-mail]

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

From: [e-mail] (David Phillip Oster)
Organization: Whole Earth 'Lectronic Link
Date: Fri, 26 Jun 1992 05:45:31 GMT

Develop Magazine in issue 9 or 110 ran an article on extracting the outlines
from true type fonts. In that article was code for producing a PICT that
used a quickdraw polygon simulation of a bezier curve when drawn on the
screen, but when drawn to a postscript printer, it used polysmooth to
draw the actual bezier curve.  It works.

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

From: [e-mail] (Eric E Johnson)
Subject: Why is gray drawing slower than color?
Organization: University of Illinois at Urbana-Champaign
Date: Fri, 26 Jun 1992 06:34:15 GMT

Well, after a lot of work, I have some code that will draw color icons
on the screen at a very rapid rate.  But, when I change the settings on
the screen from color to grays, the rate at which drawing takes place
drops considerably.  I even created icons using a gray palatte, which
helped a little bit, but the speed it still *very* slow.

Has anyone else had this experience?  Does anyone know why this might
happen?

Eric
- --
Eric Johnson               | "The American Republic will endure until the day
[e-mail] |  Congress discovers that it can bribe the public
[e-mail]  |  with the public's money" - Alexis de Toucqueville

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

From: [e-mail] (John Wallace)
Organization: TRW Data Systems Center, Redondo Beach, CA
Date: Fri, 26 Jun 92 22:39:46 GMT

In article <[e-mail]> [e-mail] (Eric E Johnson) writes:
>Well, after a lot of work, I have some code that will draw color icons
>on the screen at a very rapid rate.  But, when I change the settings on
>the screen from color to grays, the rate at which drawing takes place
>drops considerably.  I even created icons using a gray palatte, which
>helped a little bit, but the speed it still *very* slow.
>
>Has anyone else had this experience?  Does anyone know why this might
>happen?
>
>Eric
>--
>Eric Johnson               | "The American Republic will endure until the day
>[e-mail] |  Congress discovers that it can bribe the public
>[e-mail]  |  with the public's money" - Alexis de Toucqueville

It may have to do with pixel depth.  For example, drawing it in 256
colors/grays is faster than drawing in 16 colors/grays.  The reason
being that in 256 colors/grays a pixel takes up 1 byte, but in 16
colors/grays a pixel takes up 1/2 byte which must be masked in to the
surrounding data.

As far as I know, there should be no difference in drawing speed
between colors and grays at the same pixel depth (for the same graphics
card on the same Macintosh).

Cheers!
John
- --------
John Wallace

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

From: [e-mail] (Dave Leonard)
Organization: Department of Mathematics, Uni of Newcastle, Oz.
Date: Sat, 27 Jun 1992 11:58:45 GMT

[e-mail] (John Wallace) writes:

>In article <[e-mail]> [e-mail] (Eric E Johnson) writes:
>>Well, after a lot of work, I have some code that will draw color icons
>>on the screen at a very rapid rate.  But, when I change the settings on
>>the screen from color to grays, the rate at which drawing takes place
>>drops considerably.  I even created icons using a gray palatte, which
>>helped a little bit, but the speed it still *very* slow.
>>
>>Has anyone else had this experience?  Does anyone know why this might
>>happen?

>It may have to do with pixel depth.  For example, drawing it in 256
>colors/grays is faster than drawing in 16 colors/grays.  The reason
>being that in 256 colors/grays a pixel takes up 1 byte, but in 16
>colors/grays a pixel takes up 1/2 byte which must be masked in to the
>surrounding data.

>As far as I know, there should be no difference in drawing speed
>between colors and grays at the same pixel depth (for the same graphics
>card on the same Macintosh).

It may also be that quickdraw or some pallete managing code is thinking
about going from a (colourful) grey to a (grey) grey. remember how when
viewing gifs and you background the viewer, the pallete manager tries to do
its best to make the gif look okay and not a psycadelic explosion of
colour? well the same thing might be happening here. Your grey-looking
icons which are described using RGB are being translated somewhere to come
up as 8-bit grayscale images.

Just how slow is it? (More than 2 and i'd be surprised)

dave
- --
 Dave Leonard, Uni of Newcastle           [e-mail]
    "and you get a linear line" - M. Fu

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

From: [e-mail] (Eric E Johnson)
Organization: University of Illinois at Urbana-Champaign
Date: Sat, 27 Jun 1992 21:18:38 GMT

>>It may have to do with pixel depth.  For example, drawing it in 256
>>colors/grays is faster than drawing in 16 colors/grays.  The reason
>>being that in 256 colors/grays a pixel takes up 1 byte, but in 16
>>colors/grays a pixel takes up 1/2 byte which must be masked in to the
>>surrounding data.

My icons are 24x24, which means the 1/2 byte issue won't be a problem.
And, I'm making sure that the coordinates that I draw to are always byte
aligned, so that no bit shifting is necessary.

In <[e-mail]>
[e-mail] (Dave Leonard) writes:

>It may also be that quickdraw or some pallete managing code is thinking
>about going from a (colourful) grey to a (grey) grey. remember how when
>viewing gifs and you background the viewer, the pallete manager tries to do
>its best to make the gif look okay and not a psycadelic explosion of
>colour? well the same thing might be happening here. Your grey-looking
>icons which are described using RGB are being translated somewhere to come
>up as 8-bit grayscale images.

>Just how slow is it? (More than 2 and i'd be surprised)

Its about twice as slow.  That's the weird part.  Let me give some more
details on what I'm doing.

I have an offscreen gWorld in which I plot a single color icon (cicn)
into the gWorld.  Then, using some 68000 code, I mass produce it very
quickly.  Then, I use copybits to put it on the screen.  Everything
works perfectly fine under color and its lightning fast, but under gray
mode, its really slow.

When I create my gWorld, I make sure to use the current color palette of
the current screen, so copying between the gWorld and the screen would
still be fast.  I also use PlotCIcon to put the color icon into my
gWorld so PlotCIcon should fix the colors.  Keep in mind that I only use
PlotCIcon once to plot to the offscreen gWorld.  I have my own code that
takes this initial plot and mass produce it.  This works fine under
color, but under grays, the speed goes to hell.

What palette should I use when creating the icons in ResEdit?

- --
Eric Johnson               | "The American Republic will endure until the day
[e-mail] |  Congress discovers that it can bribe the public
[e-mail]  |  with the public's money" - Alexis de Toucqueville

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

From: [e-mail] (Eric E Johnson)
Organization: University of Illinois at Urbana-Champaign
Date: Sun, 28 Jun 1992 00:43:55 GMT

I discovered why my grey scale drawings were going so slow.  When I
created my offscreen gWorld, the code that I relied on assumed that the
default color table was going to be color.  So, it was grabbing a
default color table for color mode.  I changed it to check for either
gray scale or color mode, and then grab the appropriate system colot
table for the offscreen gWorld.

- --
Eric Johnson               | "The American Republic will endure until the day
[e-mail] |  Congress discovers that it can bribe the public
[e-mail]  |  with the public's money" - Alexis de Toucqueville

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

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