Info-Mac-Request@SUMEX-AIM.STANFORD.EDU (The Moderators) (11/20/89)
Info-Mac Digest Sun, 19 Nov 89 Volume 7 : Issue 205
Today's Topics:
3D graph source
[DCJCON]FLASH!Spellswell 2.2 Corrupts WriteNow & Acta Receipt
Color of a pixel
compilers: lisp,prolog,ada
dashed line object in MacDraw II
Flight simulator source
IBM PC lookalike screen font wanted
I can't get the map cdev to work
Kermit/LSC/variant records
Mac PROLOG info request
mail systems for mac/tcp
multiscreen screensavers?
PICTs and Drawing them
PopChar Problems
Procedure type in LS Pascal
Sound files format
White Knight/Red Ryder
Your Info-Mac Moderators are Bill Lipa, Lance Nakata, and Jon Pugh.
The Info-Mac archives are available (by using FTP, account anonymous,
any password) in the info-mac directory on sumex-aim.stanford.edu
[36.44.0.6]. Help files are in /info-mac/help. Indicies are in
/info-mac/help/recent-files.txt and /info-mac/help/all-files.txt.
Please send articles and binaries to info-mac@sumex-aim.stanford.edu.
Send administrative mail to info-mac-request@sumex-aim.stanford.edu.
----------------------------------------------------------------------
Date: Wed, 8 Nov 89 23:03:36 PST
From: light@ux3.lbl.gov (Francis Rubinstein)
Subject: 3D graph source
Herein find the C source for a 3D graphing program I wrote
some time ago.
The program demostrates how to do non-perspective 3D
plotting and mathematical expression parsing. All math is
floating point, though some of the code for fixed point math
remains. It is not particularly well commented, since I never
wrote it for other eyes; in most cases I was interested soley
in speed, not readability. Feel free to send me any questions,
though I don't guaruntee useful answers.
Since this program was mostly an experiment, it is guilty
of some unfortunate behavior, such as writing directly to the
screen buffer; I don't nescessarily encourage anyone to emulate
my examples in this regard. (Though if you're really determined
to draw lines very quickly...)
I've retained the copyright; feel free to steal ideas at
whim. If you extract significant sections of code unchanged,
copy the copyright message at the top of each file.
This version incorporates the ability to read the files
generated by Dave Platt's MandelZot program and plot them as
landscapes, which the executable version I posted did not. Once
I've gotten ThinkC4.0 to its knees I'll post an executable version
of this. (The project included with the source is ThinkC vers. 3.0)
If you're trying to do perspective 3D plotting, I plan to
post an example of code that shows how to do this (actually a
sort of poor man's flight simulator).
Chris Moll
chris@hobbes.lbl.gov
(415)843-2437
[Archived as /info-mac/source/c/three-d-graphics.hqx; 108K]
------------------------------
Date: Wed, 15 Nov 89 20:32:02 PST
From: "[DCJCON]FLASH" <XB.DAS@forsythe.stanford.edu>
Subject: [DCJCON]FLASH!Spellswell 2.2 Corrupts WriteNow & Acta Receipt
Date: November 14, 1989 6:00 pm PT Item: 3281791
>From: FLASH Flash (K.J.) Sheridan
To: DASNET @CONNECT-DASNET Mail Gateway
cc: WORKINGSW Working Software, Inc.
Subj: Spellswell 2.2 Corrupts WriteNow & Acta Files
Spellswell 2.2 advertises that it now checks WriteNow and Acta Application
files; it does, but it corrupts them: Acta files merely cosmetically, WriteNow
files fatally. The two bugs seem unrelated, to each other or to older, more
tolerable bugs which are still around.
Spellswell reset my second WriteNow footnote to 381, then to 81. I viewed
footnote 81, scrolled *backwards* to later feetnote, tried to view the page, and
was rewarded with "A non-existent file-path was specified." This is only my
second crash in two years with WriteNow; merely the latest of many with
Spellswell. The file seems unopenable.
In Acta font or size changes within a word cause get lost, but words nearby
acquire similar changes. I reported this bug in the last version, but nothing's
happened.
Flash@DCJCON.DAS.NET (preferred)(MacNet)
FlashsMom@cup.portal.com
------------------------------
Date: Wed, 15 Nov 89 16:16:10 -0500
From: demarsee@icarus.cns.syr.edu (Darryl E. Marsee)
Subject: Color of a pixel
How does one go about finding out the color mixture information for
a pixel on the screen? I need to get the Hue/Saturation/Brightness
(HSB) or Red/Green/Blue (RGB) values of a pixel that is currently
visible on the Mac screen. Ideally, I'd like to find something like
the FatMouse DA that displays this color information, but in lieu of
that, anything that doesn't require some big bucks package like
PixelPaint will do. Any clues out there on how to do this?
Regards,
Darryl Marsee
Syracuse University
------------------------------
Date: Thu, 16 Nov 89 10:36:26 +0100
From: Pottie Karl <GHGAQBA%BLEKUL11.BITNET@forsythe.stanford.edu>
Subject: compilers: lisp,prolog,ada
I would like to get into contact with people who use a Lisp interpreter
or Ada or Prolog compiler on their mac (not xlisp or mini-ada but the
real thing).
drop me a mail on GHGAQBA@BLEKUL11.BITNET
------------------------------
Date: Wed, 15 Nov 89 19:02:39 CST
From: kc_yeo@sngsf1.sinet.slb.com (KC Yeo*Sedco Forex S'pore*Tel-65-345-9944*Fax-65-344-2655)
Subject: dashed line object in MacDraw II
I use MacDraw II 1.0v2 - June 1988 with LaserWriter Plus and Mac II. I cannot
draw dashed object like rectangle, oval, circle though I checked Dashed line
under the Pen menu. Drawing dashed line, e.g. straight, slanted or horizontal
lines has no problem. I wonder what's wrong. Can some of the knowledgeable
MacDraw II users out there enlighten me ? I don't have the manual. Any help
will be very much appreciated. Thank you
Arthur Yeo
SINet : KC_YEO@SNGSF1.SINET.SLB.COM
------------------------------
Date: Thu, 9 Nov 89 13:02:38 PST
From: light@ux3.lbl.gov (Francis Rubinstein)
Subject: Flight simulator source
This is the source code for a program to
demonstrate how to do 3D perspective drawing and
clipping. It's written as a sort of crude flight
simulator - you "fly" through a rather simple world.
It also shows how to do simple animation.
All math is fixed-point to make the speed tolerable
on feebler Macs. On a II the speed would be roughly
the same using floating point. If some would prefer
to see a floating point version, that could be done.
The package contains the source code, a ThinkC
vers.4.0 project, and an executable version compiled
for any Mac.
Comments etc. to
Chris Moll
chris@hobbes.lbl.gov
(415)843-2437
[Archived as /info-mac/source/c/flight-simulator.hqx; 28K]
------------------------------
Date: Wed, 15 Nov 89 21:51:11 DNT
From: Jakob Nielsen <DATJN%NEUVM1.bitnet@forsythe.stanford.edu>
Subject: IBM PC lookalike screen font wanted
I want to be able to generate illustrations *on my Mac* which look as
if they are screendumps from an IBM PC. For this purpose I need a
font which looks like the monospaced characters on a standard IBM PC screen.
Could somebody please point me to such a font (if it were ftp'able,
it would best, but anything would help).
Jakob Nielsen, Technical University of Denmark
Department of Computer Science, DK-2800 Lyngby Copenhagen, DENMARK
Tel: +45-31-38 23 20 or +45-42-88 15 66. Fax: +45-42-88 22 39
Email: datJN@NEUVM1.bitnet OR jn@iddth.dk OR nielsen.chi@xerox.com
------------------------------
Date: Wed, 15 Nov 89 21:54:44 PST
From: clancy@janus.berkeley.edu (Mike Clancy)
Subject: I can't get the map cdev to work
The symptom is that all my cdevs except Map appear when I select
Control Panel. However, the BUNDLE, SYSTEM, INITED, and CACHED bit
are set in the Map file (according to DiskTop).
I'm running System 6.0.2, Finder 6.0, and a bunch of inits on a 512KE
with a 2M Monster Mac upgrade.
Renaming Map doesn't help, and copying Map from the original disk
doesn't either. Any suggestions as to what could be causing the problem,
or what I should be doing to try to track it down?
------------------------------
Date: Thu, 16 Nov 89 10:34:56 CST
From: Michael Hanrahan <C09615MH%WUVMD.BITNET@forsythe.stanford.edu>
Subject: Kermit/LSC/variant records
Hi,
I have a question regarding Kermit/LSC/variant forms of Pascal
records. Yes, all three of these in one integrated question :-)
In an attempt to imbed Kermit's file transfer protocol within
an electronic mail bridge we are writing, I am first trying to simply
port the C source to MacKermit to Lightspeed C (oh, excuse me -- Think's
C). The ckmfio.c source file (which handles all file io) uses the
CInfoPBRec data structure described in Inside Mac Volume IV on page 186.
Unfortunately, LSC's FileMgr.h header file which declares some of the
File Manager data structures DOESN'T define this data type.
The Pascal declaration for the types involved with CInfoPBRec looks
something like:
CInfoType = (hfileInfo,dirInfo);
CInfoPBPtr = ~CInfoPBRec;
CInfoPBRec = RECORD
qLink: QElemPtr;
.
.
.
filler2: SignedByte;
CASE CInfoType OF
hfileInfo:
(ioFlFndrInfo: FInfo;
.
.
.
ioFlClpSiz: LONGINT);
dirInfo:
(ioDrUsrWds: DInfo;
.
.
.
ioDrParID: LONGINT);
END;
When used, you can always use the fields such as qLink, and you can
use either the fields defined under hfileInfo or those under dirInfo.
The compiler figures out how what to pass by looking at the field names
you use.
***********************************************************************
* The question: How do you declare these variant records in C? *
***********************************************************************
I know how to use typedef struct { and typedef union but anything beyond
that is beyond me at this point. What I find particularly confusing is
that when my source code uses this record and uses one of the fields
contained in one of the two variant portions, it references it as
info.hfileinfo.ioFDirIndex = fileindex;
which (to me) seems like it's trying to tell the compiler to use the
hfileinfo variant form of the record CInfoPBRec. However, LSC seems
to be interpreting it as a Pascal compiler would: it says ioFDirIndex
is an unknown struct/union member because it's thinking ioFDirIndex
is a field within the variant part of the record. It's not. It's part
of the "main" record which is the same for both variant forms.
Any help you could provide would be greatly appreciated. Also, any other
hints for converting MPW C programs to Think C would also be quite
welcome.
Michael Hanrahan
Educational Computing Services
Washington University
St. Louis, MO 63130
------------------------------
Date: Thu, 16 Nov 89 10:06 CST
From: "tony bible, iowa state university" <GVAEB@ccvax.iastate.edu>
Subject: Mac PROLOG info request
I need to find a Macintosh PROLOG for use in a class this coming Spring
term. It would be nice if the supplier also had a version for IBM/Compatible
PC's and/or VAX's, but that is not a requirement. Any information including
recommendations/warnings would be appreciated. BTW, if, like me, you thought
Borland had a Mac version they don't.
Thanks in advance for your help.
Tony
------------------------------
Date: Thu, 16 Nov 89 09:35:55 PST
From: meldal@anna.stanford.edu
Subject: mail systems for mac/tcp
Stanford University has developed two packages addressing the need for
TCP/IP connectivity. One named Su-MacIP, which does telnet and ftp,
and the other named MacMH which does mail processing. Both use an
ethernet connection (either directly or through an appletalk-ethernet
gateway). MacMH needs a POP-server (or some such, I do not remember
the name) running on a host on the ethernet. Such server software
exists in the public domain, I believe. The price is around $100 for
academic institutions. With MacMH you keep all your mail on the mac,
and allows folder based organization of messages.
Contact Academic Information Resources
Stanford University
Stanford CA 94305
for further information.
-- Sigurd
------------------------------
Date: Wed, 15 Nov 89 12:22:37 EST
From: George McKee <mckee@corwin.ccs.northeastern.edu>
Subject: multiscreen screensavers?
Are there any screen savers available that can blank multiple screens?
Due to one of Apple's rare cases of generosity, I have both a 4-bit
grayscale monitor and an 8-bit color one on my Mac II. I've tried
six or eight different screensavers, and they generally darken only
the screen with the menubar. Since I like to have the menubar on
the grayscale screen, this leaves my expensive monitor in burn-in mode.
The only one I've found that ever darkens both screens calls itself
"blackout". It tries to cover the darkened screen with zooming circles,
but in the only arrangement of pixel depth, startup order, menubar
location, and screen arrangement that works at all (no, I haven't
tried all 16 combinations :-), these "stars" pile up on the left-hand
edge of the right hand display without making it all the way across.
Other arrangements only darken one screen.
Now that the IIci comes with built-in display hardware, I'd expect
people who want more advanced graphics to start using two displays,
rather than waste the built-in one, and my situation will become
more widespread.
- George McKee
College of Computer Science
Northeastern University, Boston 02115
Internet: mckee@Corwin.CCS.Northeastern.EDU
Phone: (617) 437-5204
uucp: ...purdue!iuvax!corwin!mckee (don't ask why)
------------------------------
Date: Wed, 15 Nov 89 22:19 CST
From: <BPB9204%TAMSTAR.BITNET@forsythe.stanford.edu>
Subject: PICTs and Drawing them
I'm having a problem that's not so obvious to me. What I'm doing seems
simple enough, but it doesn't work.
Here it is:
setPort(@myWindow);
For i := 0 To 11 Do
Begin
myPic[i] := Nil;
myPic[i] := getPicture(i * 30); { ID #s are 0,30,60...33
0}
If myPic[i] = Nil Then
Begin
sysbeep(4);
moveto(20, 20 * i + 20);
numtoString(i * 30, numberStr);
drawString(concat('PICT id# not found: '
, numberStr));
End;
myPicR[i] := myPic[i]~~.picFrame;
End;
{drawing starts here in a repeat .. until loop to 'animate'}
I know it's reading the resources OK because I have never had it beep,
but when I call
drawPicture(myPic[x],destinationRectangle);
nothing happens. I also have recently been getting an 'Address Error'
message in Think Pascal 2.0 on the drawPicture procedure.
Can someone send me an example of how to display many PICTs (for cheap
animation) ? If not, can someone tell me what I am not doing correctly?
Any and all suggestions are appreciated.
Brent Burton
BPB9204@TAMSTAR.BITNET
-=-=-=-=-=-=-=-=-=-=-=
------------------------------
Date: Thu, 16 Nov 89 09:01:59 MST
From: Bruce Long <ICBAL%ASUACAD.BITNET@forsythe.stanford.edu>
Subject: PopChar Problems
In I-M Digest 191 there was a message about a problem with PopChar INIT
when used with two monitors, where one monitor is left of the apple
(in other words, one monitor has negative horizontal coordinates).
Clicking almost anywhere in the left monitor will activate PopChar.
A partial solution may be found in the fact that PopChar INIT will not
activate if the option or command key is pressed when clicking.
This solution will be inadequate if your main monitor is left of the
apple, but if your secondary monitor is left of the apple this
workaround may be acceptable (for example if your left monitor has your
communication program which remains on-line while you work under
multifinder in the right monitor).
Bruce Long
Department of Mathematics
Arizona State University BITNET: ICBAL@ASUACAD
------------------------------
Date: Thu, 16 Nov 89 08:09:42 EST
From: chris@adms-rad.unisys.com (Chris Sterritt)
Subject: Procedure type in LS Pascal
Craig Cottingham writes:
>Date: Wed, 08 Nov 89 10:47:09 CST
>From: "Lovely Angels Fanatic Cult" <UC528665%UMCVMB.BITNET@ricevm1.rice.edu>
>Subject: Is there a procedure type in LS Pascal 1.1?
>
>I've seen the procPtr type, but so far that's only been useful for
>passing procedure addresses to system routines. What I really
>need is the ability to "late-bind" a procedure call, so that I can
>change the actual procedure called on the fly. Does LightSpeed
>Pascal 1.1 support this feature, or should I go looking elsewhere?
>(I suppose I could hack it together in assembly language, but that's
>not a desired option at this time.)
>
>Thanks in advance,
>
>craig s. cottingham
> -- "Remember to be wary of wizards,
> for to them, thou art crispy,
> and go well with strawberry jam...."
Try examining the source of the TransSkel application skeleton.
They do a 'late bind' to allow you to tell TransSkel what function to call
when a menu item is chosen, etc. I'm pretty sure TransSkel is in the archives.
--chris sterritt
chris@adms-rad.unisys.com
------------------------------
Date: Wed, 15 Nov 89 21:11:23 EST
From: Damian%UMass.BITNET@forsythe.stanford.edu
Subject: Sound files format
Wow,
I wasn't expecting such an outpouring of responses from the net regarding sound
file formats. I have received so many requests for forwarding of information
that I decided to post a summary (my first one!). Thanks to everyone who
responded...it was a great help.
Damian Roskill
damian@mars.ucc.umass.edu
=========================================================================
Eric Keller writes:
=========================================================================
Here is the info on SoundEdit files right out of the Manual
for Signalyze(TM), my speech analysis program. Signalyze does
all the conversions between ASCII, MacAdios, SoundEdit, snd
resource, plus a ton of other things.
Good luck.
Eric Keller
File type: FSSD, Creator: SFX!
SoundEdit files maintain a 64-byte resource called TINFOU (u.
c.) and numbered 1000.
Sampling rate information is given according to recording
frequencies available through Farallon A/D converters.
Although FarallonUs own documentation gives the frequen-
cies as 22, 11, 7, and 5 kHz, the actual frequencies
appear to be 1 = 22.256 kHz, 2 = 11.128 kHz, 3 = 7.418
kHz, 4 = 5.564 kHz, as in SoundWave format.
SoundEdit files have separate recording and playback
frequency parameters. Both soundEditHdr.relRecFreq and
soundEditHdr.relPlaybackFreq are set to 65536 when
recording and playback speed are normal, >65536 when the
speeds are greater and <65536 when they are less.
Signalyze calculates a multiplication factor on the basis
of the ratio between relPlaybackFreq and relRecFreq, and
adjusts the sampling frequency accordingly on input.
For file length information, Signalyze does not use the
soundEditHdr.lftChnnlLength parameter, but instead obtains
sound data length directly through the Toolbox command
GetEOF(refnum, &SoundLength) to guard against failures
>From truncated files.
The right channel stereo information and the compression
ratio parameters are ignored by Signalyze version 1.0.
Stereo will be supported at some future point in time.
typedef struct soundEditHdr {
long res0; reserved by Farallon
long res1; reserved by Farallon
Fixed relRecFreq; relative recording
frequency in fixed point format
Fixed relPlaybackFreq; relative playback
frequency in fixed point format
long compression; set to 0 by Signalyze in save
long mode; 0 for mono, 1 for stereo
long windowPos; set to standard size window by
Signalyze in save
long sRate; 1 for 22kHz, 2 for 11kHz, 3 for
7kHz, 4 for 5kHz
long lftChnnlLength; number of samples for mono
or left stereo channel
long rghtChnnlLength; number of samples for right
stereo channel
long res10; reserved by Farallon
long res11; reserved by Farallon
long res12; reserved by Farallon
long res13; reserved by Farallon
long res14; reserved by Farallon
long res15; reserved by Farallon
} soundEditHdr ; This is the file format for the MacRecorder program SoundEdit.
This is the
one that comes from Farallon, not the old do-it-yourselfer (don't know about
that one :-)
=========================================================================
Myron-Meier@Sun.COM writes:
=========================================================================
SOUNDEDIT FILE FORMAT
Tech Note - Revision 1
Copyright ) 2/24/88 Farallon Computing, Inc. All rights reserved.
The SoundEdit file format is a standard Macintosh file which contains both a
data fork and a resource fork. The data fork is used to store the actual sound
samples, while the resource fork contains a series of resources describing
various parameters of the sound and the way it is displayed on the screen. The
file type is 'FSSD' and the file creator is 'SFX!'.
DATA FORK
The data fork contains 8-bit linear sound samples, where each byte is a single
sample which can have a value between 0 and 255. The amplitude of a sample is
obtained by subtracting 128 from this value. In other words, 0 corresponds to
the minimum amplitude of -128, while 255 corresponds to the maximum amplitude of
127. Note that this conversion does not need to(and in fact should not) be done
before "playing" the sound. The Macintosh system hardware and software expects
to see the sound samples in this offset format. This conversion is only nec
essary if you are somehow manipulating the sound wave, such as amplification, or
spectrum analysis. Also note that there is no header of any kind preceding the
sound samples.
In the case of a stereo sound, the left channel sound samples appear first,
followed immediately by the right channel sound samples. There is no gap and no
odd-byte filler.
RESOURCE FORK
The resource fork contains four resource types: CLRS, INFO, LABS, and REPT.
There is a single resource of each type with a resource ID of 1000. The CLRS an
d LABS resources contain the colors and labels, if any, that have been applied
to selected portions of the sound wave. Since these are not pertinent to the
sound characteristics themselves, nothing more will be said about them.
The REPT resource contains 6 long-word (32-bit) values which correspond to the
"report" boxes in the lower right-hand corner of the SoundEdit window. These 6
values are defined as follows:
Longword 0 Mouse position. (Not really meaningful)
Longword 1 Length of sound in samples. For stereo, this value is
not the sum of the 2 channels, but rather the length
of a single channel or the longer of the 2, in case
they're uneven.
Longword 2 Sample number of the selection beginning, if there was
a selection when the file was saved; or the cursor
position otherwise.
Longword 3 Sample number of the selection end, if there was a
selection when the file was saved; or the cursor
position otherwise.
Longword 4 Sample number of the loop beginning if a loop was
defined.
Longword 5 Sample number of the loop end if a loop was defined.
The INFO resource contains 16 long-word values, of which only the first 10 are
currently used. These define various characteristics of the sound as follows:
Longword 0 (used internally)
Longword 1 (used internally)
Longword 2 Fixed point value defining the relative record
frequency selected with the "Set Pitches..." menu
item.
Longword 3 Fixed point value defining the relative playback
frequency selected with the "Set Pitches..." menu
item.
Longword 4 Compression mode. 0 for none; 1 for 8:1 compression
Longword 5 Mode. 0 for mono; 1 for stereo.
Longword 6 Window position. Vertical and horizontal coordinates
of the window when the file was saved.
Longword 7 Sample rate. 1 - 22 KHz; 2 - 11 KHz; 3 - 7 KHz; 4
-5 KHz
Longword 8 Length of left (or mono) channel in samples.
Longword 9 Length of right channel in samples if stereo.
=========================================================================
And last, but certainly not least, Margot Flowers writes:
=========================================================================
There is a two or three page document you can get them to send you
if you call them up and ask nicely. At least they did when I asked
them, but that was about 2 years ago, so things may have changed.
Margot Flowers
Flowers@CS.UCLA.EDU
=========================================================================
Disclaimer: those opinions expressed in this letter are thoses of the
individual authors...blah...blah...blah....
------------------------------
Date: Wed, 15 Nov 89 22:44:58 -0600
From: dnewman@emx.utexas.edu (David Newman)
Subject: White Knight/Red Ryder
Has anyone heard what the deal is with the new version of Red Ryder?
I am told that the program has been renamed "White Knight", and that
it will be released soon. I am also told that it has lots of new features.
Question: will upgrade notices be sent out? How do we upgrade otherwise?
Any information will be apprecciated.
>>Dave
------------------------------
End of Info-Mac Digest
******************************