From icon-group-sender  Tue Jan  5 08:45:29 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id IAA24057
	for icon-group-addresses; Tue, 5 Jan 1999 08:41:39 -0700 (MST)
Message-Id: <199901051541.IAA24057@baskerville.CS.Arizona.EDU>
Date: Mon, 04 Jan 1999 17:34:32 -0600
From: "Charles Hethcoat" <chethcoa@oss.oceaneering.com>
To: icon-group@optima.CS.Arizona.EDU
Subject:  EditPAD
Content-Disposition: inline
X-MIME-Autoconverted: from quoted-printable to 8bit by baskerville.CS.Arizona.EDU id QAA13387
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

MJE <evans@gte.net> wrote:
>Windows users might be interested in EditPAD,

I never cared for Notepad, either.  For several years I've been pretty happy using PFE (http://www.lancs.ac.uk/people/cpaap/pfe/)  as a general purpose editor.  A quick reading of the EditPAD description makes the two sound very similar.  Has anyone got any comparative experience with both editors?  I'd be interested in your comments.  I've got so much stuff installed on my computer now that I'm reluctant to install both.  I might, however, swap out PFE for EditPad if it seems clearly superior.

Charles Hethcoat



From icon-group-sender  Tue Jan  5 16:31:24 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id QAA10984
	for icon-group-addresses; Tue, 5 Jan 1999 16:31:07 -0700 (MST)
Message-Id: <199901052331.QAA10984@baskerville.CS.Arizona.EDU>
To: icon-group@optima.CS.Arizona.EDU
Date: 05 Jan 1999 12:23:51 -1000
From: Robert Valliant <valliant@hawaii.edu>
Subject: Icon graphics
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO


I was on vacation when I found some statistics that would allow me to
make a population pyramid for Russia that I could put on my website.

I thought I would do it in postscript; then I remembered reading in the
Icon graphics book that I bought that Icon can output .gif files.

Fine, sez I, I'll just do it. And I did. I had to figure out how to do
the stuff I needed and play with it and other things, but I got it to
work.

Just save the file and it is a .gif, and I can use the full X font
specification to get any font I need!

Wonderful. I am now a happy camper. I should even be able to
superimpose one pyramid on another. I love it!

Thanks Icon.

Robert Valliant
Center for Russia in Asia, University of Hawaii at Manoa
www.hawaii.edu/shaps/russia
valliant@hawaii.edu Fax: 808.956.2682 Tel: 808.956.7814



From icon-group-sender  Mon Jan 11 08:17:18 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id IAA05343
	for icon-group-addresses; Mon, 11 Jan 1999 08:15:16 -0700 (MST)
Message-Id: <199901111515.IAA05343@baskerville.CS.Arizona.EDU>
Date: Fri, 8 Jan 1999 19:32:15 -0700
From: Ralph Griswold <ralph>
To: icon-group
Subject: A Small Programming Problem
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO


We've processed the solutions to the problem of sorting
the digits of an integer.

Preliminary result are on our Web site:

	http://www.cs.arizona.edu/icon/spp.htm

If you have comments, please send them to me or to icon-group.

	Ralph Griswold



From icon-group-sender  Thu Jan 14 08:50:52 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id IAA05681
	for icon-group-addresses; Thu, 14 Jan 1999 08:47:35 -0700 (MST)
Message-Id: <199901141547.IAA05681@baskerville.CS.Arizona.EDU>
Date: Wed, 13 Jan 1999 16:45:37 -0800
From: Jim Mehl <mehl@ihot.com>
X-Accept-Language: en
To: icon-group@optima.CS.Arizona.EDU
Subject: Getting started with QEI
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

  I'm having trouble getting QEI to work as advertised. I am using
Icon 9.3.1 on Windows 95. I have plenty of memory, disk space, etc.
I compiled QEI and linked into an EXE file. When I run it I get a
QEI console window saying Icon Expression Evaluator, Version 1.2
and a > prompt. I enter something like 3 + 4; but get nothing but
another > prompt. If I then say :list it shows me #r1_ := (3 + 4)
indicating that it accepted my input, but I get no direct output 
no matter what expression I enter. Any suggestions?
 
Jim Mehl
Los Gatos, CA



From icon-group-sender  Fri Jan 15 16:44:02 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id QAA04732
	for icon-group-addresses; Fri, 15 Jan 1999 16:42:03 -0700 (MST)
Message-Id: <199901152342.QAA04732@baskerville.CS.Arizona.EDU>
To: icon-group@optima.CS.Arizona.EDU
Date: Fri, 15 Jan 1999 21:35:12 GMT
From: Mario.Dupuis@mfg.canadair.ca
Subject: Icon and GUI development
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Hi,

I'm presently just looking at Icon and I would like to know if there
are people doing GUI with it.

Is there a widget set available (scrollbars,menus ...) ?

How would it compare to , TK/TCl as far as GUI dev. ?
(please no language wars !!)

Where could I see example of GUI developped with Icon ?

Please reply directly to my E-mail

Thanks

-----------== Posted via Deja News, The Discussion Network ==----------
http://www.dejanews.com/       Search, Read, Discuss, or Start Your Own    


From icon-group-sender  Mon Jan 18 13:22:53 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id NAA03013
	for icon-group-addresses; Mon, 18 Jan 1999 13:22:46 -0700 (MST)
Message-Id: <199901182022.NAA03013@baskerville.CS.Arizona.EDU>
Date: Sat, 16 Jan 1999 10:37:40 -0700
From: "Sandra L. Miller" <slm@optima.CS.Arizona.EDU>
X-Accept-Language: en
To: icon-group@optima.CS.Arizona.EDU
Subject: [Fwd: www site, pattern-matching research]
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO



-------- Original Message --------
Subject: www site, pattern-matching research
Date: Mon, 14 Dec 1998 10:36:32 -0800
From: Mike Shafto <mshafto@mail.arc.nasa.gov>
Reply-To: mshafto@mail.arc.nasa.gov
Organization: NASA Ames Research Center
To: icon-project@optima.CS.Arizona.EDU

http://www.cs.purdue.edu/homes/stelo/pattern.html


This www site seems potentially to have high relevance
to the Icon Group. It is an index-site to people
and information source on pattern-matching, including
theoretical and applied work in biology, text processing,
etc.

Mike

-- 

----- ----- ----- ----- ----- ----- ----- ----- ----- ----- -------
Mike Shafto (mshafto@mail.arc.nasa.gov)
Chief, Human-Automation Integration Research Branch
http://olias.arc.nasa.gov/
Mail Stop 262-4
NASA-Ames Research Center
Moffett Field,  CA       94035-1000
--------------------------------------------------------
The best-laid plans of mice and men oft-times go astray.
Especially mice.
--------------------------------------------------------


From icon-group-sender  Thu Jan 21 16:49:57 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id QAA26896
	for icon-group-addresses; Thu, 21 Jan 1999 16:47:33 -0700 (MST)
Message-Id: <199901212347.QAA26896@baskerville.CS.Arizona.EDU>
To: icon-group@optima.CS.Arizona.EDU
Date: Thu, 21 Jan 1999 12:54:55 -0600
From: Scott Mcarthur <kaltorak@satexas.com>
Subject: Re: Icon and GUI development
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

I am personally doing some GUI development with icon.  My development is
for Windows, and not UNIX however.  There is a widget set available
(called vidgits) and a visual interface available (called the VIB Visual
Interface Builder)  Personally I have never used TK/TCL so I can not
compare the ease of use.  I have however used xforms and I find that the
VIB is very similar to it.  Personally I don't use the VIB much because I
am developing for Windows and VIB creates GUI's that have a strong UNIX
feel for them.  Instead, I use the graphics facilities to create my own
look and feel.  (I do however use some of the Notice() vidgits at
times).
I have a tetris program that I am working on, but the source code is not
available as I am planning on making this into a commercial product
(sorry!).  There are some uses of the notice boxes and a couple of other
vidgits in the program that you can see visually though if you would like
(plus it's just a fun game!)

you can download it from http://www.cs.utsa.edu/~cmcarthu/sdfftetris.htm
remember that this is intended for Windows!  (Color map limitations have
made me decide not to work on a unix version)
I hope that this helps!

Scott



Mario.Dupuis@mfg.canadair.ca wrote:

> Hi,
>
> I'm presently just looking at Icon and I would like to know if there
> are people doing GUI with it.
>
> Is there a widget set available (scrollbars,menus ...) ?
>
> How would it compare to , TK/TCl as far as GUI dev. ?
> (please no language wars !!)
>
> Where could I see example of GUI developped with Icon ?
>
> Please reply directly to my E-mail
>
> Thanks
>
> -----------== Posted via Deja News, The Discussion Network ==----------
> http://www.dejanews.com/       Search, Read, Discuss, or Start Your Own

--
Scott McArthur
cmcarthu@cs.utsa.edu
-------------------------------
If we aren't supposed to eat animals, why are they made of meat?





From icon-group-sender  Mon Jan 25 09:34:46 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id JAA12959
	for icon-group-addresses; Mon, 25 Jan 1999 09:32:47 -0700 (MST)
Message-Id: <199901251632.JAA12959@baskerville.CS.Arizona.EDU>
From: "Yang, De (Dianna)" <deyang@lucent.com>
To: icon-group@optima.CS.Arizona.EDU
Subject: qei questions
Date: Sat, 23 Jan 1999 17:18:53 -0600
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Dear Sirs, 

I download qei.icn and compiled it to run on my pc.  But I did not get any
echo of result 
from the input.  For example,  I input
> 1+3; 

It does not return r1_:=4 but only another ">" for me.  What did I do wrong?
This is not 
described at the help page either. 

Thank you. 

De Yang


From icon-group-sender  Tue Jan 26 08:08:57 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id IAA19953
	for icon-group-addresses; Tue, 26 Jan 1999 08:06:21 -0700 (MST)
Message-Id: <199901261506.IAA19953@baskerville.CS.Arizona.EDU>
Date: Mon, 25 Jan 1999 21:23:22 -0800
From: Jim Mehl <mehl@ihot.com>
X-Accept-Language: en
To: icon-group@optima.CS.Arizona.EDU
Subject: Re: qei questions
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

  It sounds like Dianna is having the same problem I was having a 
couple of weeks ago. Thanks to help from Steve Wampler, I was able 
to sort it out.
  Dianna didn't say, but I assume she is using the Windows version 
of Icon.
  The problem is that qei (like a lot of other programs in the Icon
library) makes a call to system("icont ...) to invoke the Icon 
compiler to evaluate expressions. Unfortunately with the Windows 
version, the Icon compiler is named wicont (for the windowing 
version) or nticont (for the command line version).
  The way I solved this was to make a copy of the nticont.exe 
compiler named icont.exe. From a Windows DOS command line issue
     copy nticont.exe icont.exe
     patchstr icont.exe icont.exe
Patchstr comes with Windows Icon and is described under 
Windows Icon Help -> Known Bugs and Limitations.
It is also necessary to compile qei.icn with either nticont or the 
icont you just created. What you get is a command line only version 
of qei, but it's better than nothing.
  It might be helpful if the documentation for Windows Icon mentioned 
the fact that the library includes all sorts of calls to icont, but 
that no icont.exe is included.
 
Jim Mehl
Los Gatos, CA
mehl@ihot.com



From icon-group-sender  Tue Jan 26 13:09:42 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id NAA10508
	for icon-group-addresses; Tue, 26 Jan 1999 13:09:02 -0700 (MST)
Message-Id: <199901262009.NAA10508@baskerville.CS.Arizona.EDU>
Date: Tue, 26 Jan 1999 09:23:55 -0600
From: MJE <evans@gte.net>
To: icon-group@optima.CS.Arizona.EDU
Subject: Re: qei questions
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

What the program needs is conditional compilation around the name of the
Icon compiler...such is typical of cross-platform code...don't know
whether Icon supports it.

Mark


Jim Mehl wrote:
> 
>   It sounds like Dianna is having the same problem I was having a
> couple of weeks ago. Thanks to help from Steve Wampler, I was able
> to sort it out.
>   Dianna didn't say, but I assume she is using the Windows version
> of Icon.
>   The problem is that qei (like a lot of other programs in the Icon
> library) makes a call to system("icont ...) to invoke the Icon
> compiler to evaluate expressions. Unfortunately with the Windows
> version, the Icon compiler is named wicont (for the windowing
> version) or nticont (for the command line version).
>   The way I solved this was to make a copy of the nticont.exe
> compiler named icont.exe. From a Windows DOS command line issue
>      copy nticont.exe icont.exe
>      patchstr icont.exe icont.exe
> Patchstr comes with Windows Icon and is described under
> Windows Icon Help -> Known Bugs and Limitations.
> It is also necessary to compile qei.icn with either nticont or the
> icont you just created. What you get is a command line only version
> of qei, but it's better than nothing.
>   It might be helpful if the documentation for Windows Icon mentioned
> the fact that the library includes all sorts of calls to icont, but
> that no icont.exe is included.
> 
> Jim Mehl
> Los Gatos, CA
> mehl@ihot.com




From icon-group-sender  Tue Jan 26 13:11:02 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id NAA10636
	for icon-group-addresses; Tue, 26 Jan 1999 13:10:54 -0700 (MST)
Message-Id: <199901262010.NAA10636@baskerville.CS.Arizona.EDU>
To: icon-group@optima.CS.Arizona.EDU
Date: 21 Jan 1999 18:55:16 -0500
From: Scott Mcarthur <kaltorak@satexas.com>
Subject: Re: Icon and GUI development
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

I am personally doing some GUI development with icon.  My development is
for Windows, and not UNIX however.  There is a widget set available
(called vidgits) and a visual interface available (called the VIB Visual
Interface Builder)  Personally I have never used TK/TCL so I can not
compare the ease of use.  I have however used xforms and I find that the
VIB is very similar to it.  Personally I don't use the VIB much because I
am developing for Windows and VIB creates GUI's that have a strong UNIX
feel for them.  Instead, I use the graphics facilities to create my own
look and feel.  (I do however use some of the Notice() vidgits at
times).
I have a tetris program that I am working on, but the source code is not
available as I am planning on making this into a commercial product
(sorry!).  There are some uses of the notice boxes and a couple of other
vidgits in the program that you can see visually though if you would like
(plus it's just a fun game!)

you can download it from http://www.cs.utsa.edu/~cmcarthu/sdfftetris.htm
remember that this is intended for Windows!  (Color map limitations have
made me decide not to work on a unix version)
I hope that this helps!

Scott



Mario.Dupuis@mfg.canadair.ca wrote:

> Hi,
>
> I'm presently just looking at Icon and I would like to know if there
> are people doing GUI with it.
>
> Is there a widget set available (scrollbars,menus ...) ?
>
> How would it compare to , TK/TCl as far as GUI dev. ?
> (please no language wars !!)
>
> Where could I see example of GUI developped with Icon ?
>
> Please reply directly to my E-mail
>
> Thanks
>
> -----------== Posted via Deja News, The Discussion Network ==----------
> http://www.dejanews.com/       Search, Read, Discuss, or Start Your Own

--
Scott McArthur
cmcarthu@cs.utsa.edu
-------------------------------
If we aren't supposed to eat animals, why are they made of meat?






From icon-group-sender  Tue Jan 26 16:59:40 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id QAA20820
	for icon-group-addresses; Tue, 26 Jan 1999 16:59:07 -0700 (MST)
Message-Id: <199901262359.QAA20820@baskerville.CS.Arizona.EDU>
To: icon-group@optima.CS.Arizona.EDU
Date: Tue, 26 Jan 1999 23:05:02 +0100
From: "name" <jas@lander.es>
Subject: se busca profesor particular de programación
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Preferentemente con experiencia en C, Pascal y bases de datos. Las clases
serán sobre TCL y Icon, de modo que el conocimiento previo de estos
lenguajes, o uno de ellos, será valorado.

Residencia en Madrid.

Respuestas a jas@lander.es

Saludos,

JaS






From icon-group-sender  Tue Jan 26 17:01:30 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id RAA21041
	for icon-group-addresses; Tue, 26 Jan 1999 17:01:23 -0700 (MST)
Message-Id: <199901270001.RAA21041@baskerville.CS.Arizona.EDU>
To: icon-group@optima.CS.Arizona.EDU
Date: Tue, 26 Jan 1999 22:27:45 GMT
From: "Marek Budyn" <rayn@infostrade.com.pl>
Subject: the beginning
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Hi!
I am new in programming in ICON. So, I am looking for manuals, examples,
specyfications, libraries for programming in Windows enviroment (I have
downloaded almost everything from www.cs.arizona.edu). If somebody knows any
web page that shares eg. books (id pdf or ps format) how to programm in
Icon.
Could somebody help me, please ?

Marek Budyn
rayn@infostrade.com.pl





From icon-group-sender  Wed Jan 27 09:09:00 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id JAA18913
	for icon-group-addresses; Wed, 27 Jan 1999 09:07:04 -0700 (MST)
Message-Id: <199901271607.JAA18913@baskerville.CS.Arizona.EDU>
Date: Wed, 27 Jan 1999 00:14:27 -0800
From: Jim Mehl <mehl@ihot.com>
X-Accept-Language: en
To: icon-group@optima.CS.Arizona.EDU
Subject: Re: qei questions
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

  Mark is certainly right about that. And Icon does have preprocessing
with conditional compilation. In fact there is a preprocessor symbol 
defined called _MS_WINDOWS that would allow something like
     $ifdef _MS_WINDOWS
       system("nticont ...)
     $else
       system("icont ...)
     $endif
It just hasn't been used in the Icon library.
  Jim Mehl

MJE wrote:
> 
> What the program needs is conditional compilation around the name of the
> Icon compiler...such is typical of cross-platform code...don't know
> whether Icon supports it.
> 
> Mark
>



From icon-group-sender  Fri Jan 29 09:00:21 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id IAA02505
	for icon-group-addresses; Fri, 29 Jan 1999 08:58:19 -0700 (MST)
Message-Id: <199901291558.IAA02505@baskerville.CS.Arizona.EDU>
Date: Thu, 28 Jan 1999 17:14:37 -0600
From: "George K. Thiruvathukal" <gkt@math.luc.edu>
X-Accept-Language: en
To: icon-group@optima.CS.Arizona.EDU
Subject: file and directory manipulation
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Help!
Is it possible, other than by using system(...), to perform various file
and directory manipulation functions on the Unix platform? I'm
desperately trying to use Icon as a perl replacement, but I cannot seem
to create directories easily. It seems like functions such as chdir are
available, though the rest appear to be absent.

Thanks,
George


From icon-group-sender  Fri Jan 29 17:20:31 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id RAA20402
	for icon-group-addresses; Fri, 29 Jan 1999 17:20:12 -0700 (MST)
Message-Id: <199901300020.RAA20402@baskerville.CS.Arizona.EDU>
Date: Fri, 29 Jan 1999 14:53:17 -0600
From: Clinton Jeffery <jeffery@segfault.cs.utsa.edu>
To: gkt@math.luc.edu
CC: icon-group@optima.CS.Arizona.EDU
Subject: Re: file and directory manipulation
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO


>   Is it possible, other than by using system(...), to perform various file
>   and directory manipulation functions on the Unix platform? I'm
>   desperately trying to use Icon as a perl replacement.

Hi George,

You want Unicon, http://www.drones.com/unicon, an Icon with POSIX extensions
developed by Shamim Mohamed.  Unicon forms the basis for some multiplatform
file, directory, and internet facilities that we are working on, which will
be described in a forthcoming book.

>  I cannot seem to create directories easily.

I really think that system() ought to work great for this, but in any case,
Unicon has these types of functions as built-ins for Icon.

Clint Jeffery, jeffery@cs.utsa.edu
Research http://www.cs.utsa.edu/research/plss.html


From icon-group-sender  Fri Jan 29 17:23:01 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id RAA20480
	for icon-group-addresses; Fri, 29 Jan 1999 17:22:55 -0700 (MST)
Message-Id: <199901300022.RAA20480@baskerville.CS.Arizona.EDU>
Date: Fri, 29 Jan 1999 15:59:01 -0700
From: Gregg Townsend <gmt@optima.CS.Arizona.EDU>
To: gkt@math.luc.edu, icon-group@optima.CS.Arizona.EDU
Subject: Re:  file and directory manipulation
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

    From: "George K. Thiruvathukal" <gkt@math.luc.edu>

    Is it possible, other than by using system(...), to perform various file
    and directory manipulation functions on the Unix platform?...

The usual solution is indeed a call to system().  Chdir is implemented
in Icon as a special case because it can't be done by calling system().

There is an extended version of Icon ("Unicon") with additional calls; see
	http://www.drones.com/unicon/
for details.

   ---------------------------------------------------------------------------
   Gregg Townsend              Gould-Simpson Building   gmt@cs.arizona.edu
   Staff Scientist             1040 E. 4th St.          32 13 45N  110 57 16W
   Dept. of Computer Science   PO Box 210077            tel: +1 520 621 4325
   The University of Arizona   Tucson, AZ  85721-0077   fax: +1 520 621 4246


From icon-group-sender  Fri Jan 29 17:23:34 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id RAA20515
	for icon-group-addresses; Fri, 29 Jan 1999 17:23:28 -0700 (MST)
Message-Id: <199901300023.RAA20515@baskerville.CS.Arizona.EDU>
Date: Fri, 29 Jan 1999 18:10:00 -0600
From: "George K. Thiruvathukal" <thiruvathukal@acm.org>
X-Accept-Language: en
To: Gregg Townsend <gmt@optima.CS.Arizona.EDU>
CC: gkt@math.luc.edu, icon-group@optima.CS.Arizona.EDU
Subject: Re: file and directory manipulation
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

This is a multi-part message in MIME format.
--------------63F982C9BEE3D06D4563F24A
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Hi Gregg,

Thanks for your reply.

I know how to use the system() call; however, I really do not want the overhead of
the shell for every operation I need to do. I can live with the interpreted nature
of Icon. I'm still awaiting news on Unicon. My thought was to use loadable
functions (or external calls)  to support many of the features I need, since I
have done this before to include DBM support in Icon.

Best,
George


Gregg Townsend wrote:

>     From: "George K. Thiruvathukal" <gkt@math.luc.edu>
>
>     Is it possible, other than by using system(...), to perform various file
>     and directory manipulation functions on the Unix platform?...
>
> The usual solution is indeed a call to system().  Chdir is implemented
> in Icon as a special case because it can't be done by calling system().
>
> There is an extended version of Icon ("Unicon") with additional calls; see
>         http://www.drones.com/unicon/
> for details.
>
>    ---------------------------------------------------------------------------
>    Gregg Townsend              Gould-Simpson Building   gmt@cs.arizona.edu
>    Staff Scientist             1040 E. 4th St.          32 13 45N  110 57 16W
>    Dept. of Computer Science   PO Box 210077            tel: +1 520 621 4325
>    The University of Arizona   Tucson, AZ  85721-0077   fax: +1 520 621 4246

--------------63F982C9BEE3D06D4563F24A
Content-Type: text/x-vcard; charset=us-ascii;
 name="thiruvathukal.vcf"
Content-Description: Card for George K. Thiruvathukal
Content-Disposition: attachment;
 filename="thiruvathukal.vcf"
Content-Transfer-Encoding: 7bit

begin:vcard 
n:Thiruvathukal;George K.
tel;fax:773.508.2123
tel;work:773.508.3447
x-mozilla-html:TRUE
url:http://www.jhpc.org/george
org:Loyola University;Mathematical and Computer Sciences
adr:;;6525 Sheridan Road;Chicago;IL;60613;USA
version:2.1
email;internet:thiruvathukal@acm.org
note:Go Cubs!
fn:George K. Thiruvathukal
end:vcard

--------------63F982C9BEE3D06D4563F24A--



From icon-group-sender  Fri Jan 29 17:23:56 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id RAA20550
	for icon-group-addresses; Fri, 29 Jan 1999 17:23:49 -0700 (MST)
Message-Id: <199901300023.RAA20550@baskerville.CS.Arizona.EDU>
Date: Fri, 29 Jan 1999 17:21:13 -0700
From: Gregg Townsend <gmt@optima.CS.Arizona.EDU>
To: thiruvathukal@acm.org
Subject: Re: file and directory manipulation
Cc: gkt@math.luc.edu, icon-group@optima.CS.Arizona.EDU
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

    From: "George K. Thiruvathukal" <thiruvathukal@acm.org>

    ...My thought was to use loadable functions (or external calls)
    to support many of the features I need, since I
    have done this before to include DBM support in Icon.

There are a few such loadable functions, mostly written as examples,
in the cfuncs directory of the Icon program library.



From icon-group-sender  Mon Feb  1 08:28:46 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id IAA28893
	for icon-group-addresses; Mon, 1 Feb 1999 08:25:45 -0700 (MST)
Message-Id: <199902011525.IAA28893@baskerville.CS.Arizona.EDU>
To: icon-group@optima.CS.Arizona.EDU
Date: 29 Jan 1999 19:22:38 -0500
From: Clinton Jeffery <jeffery@segfault.cs.utsa.edu>
Subject: Re: file and directory manipulation
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO


>   Is it possible, other than by using system(...), to perform various file
>   and directory manipulation functions on the Unix platform? I'm
>   desperately trying to use Icon as a perl replacement.

Hi George,

You want Unicon, http://www.drones.com/unicon, an Icon with POSIX extensions
developed by Shamim Mohamed.  Unicon forms the basis for some multiplatform
file, directory, and internet facilities that we are working on, which will
be described in a forthcoming book.

>  I cannot seem to create directories easily.

I really think that system() ought to work great for this, but in any case,
Unicon has these types of functions as built-ins for Icon.

Clint Jeffery, jeffery@cs.utsa.edu
Research http://www.cs.utsa.edu/research/plss.html



From icon-group-sender  Mon Feb  1 08:30:11 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id IAA29133
	for icon-group-addresses; Mon, 1 Feb 1999 08:30:02 -0700 (MST)
Message-Id: <199902011530.IAA29133@baskerville.CS.Arizona.EDU>
To: icon-group@optima.CS.Arizona.EDU
Date: Sun, 31 Jan 1999 19:05:37 +0800
From: "©ú¥Õ" <mingbai@ms3.hinet.net>
Subject: [Q]:Latest version of ICON & tutorial
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Greetings:

I currently have version 9.3 of ICON. Checking the ftp site, it seems
that that is the lastest version available. But, I'd like to confirm.

Also, in the documentation there is a statement concerning a toturial
"to be added".....is that tutorial now available anywhere?

Thanks!

--
"Mathematics is the foundation of all science and those who are
profficent in it become the builders of society."

                                         The Honorable Ellijah Mohammed

"A mathematician is a machine for turning coffee into theorems."

                                          Paul Erdos
                                          Mathematician





From icon-group-sender  Mon Feb  1 08:30:30 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id IAA29174
	for icon-group-addresses; Mon, 1 Feb 1999 08:30:24 -0700 (MST)
Message-Id: <199902011530.IAA29174@baskerville.CS.Arizona.EDU>
To: icon-group@optima.CS.Arizona.EDU
Date: Mon, 01 Feb 1999 13:21:42 -0800
From: ©ú¥Õ <mingbai@ms3.hinet.net>
Subject: Re: [Q]:Latest version of ICON & tutorial
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

©ú¥Õ wrote:
> 
> Greetings:
> 
> I currently have version 9.3 of ICON. Checking the ftp site, it seems
> that that is the lastest version available. But, I'd like to confirm.
> 
> Also, in the documentation there is a statement concerning a toturial
> "to be added".....is that tutorial now available anywhere?
> 
> Thanks!

Addendum to my own post:

I now realize I'm not sure which version I have.

The documentation that came with it says Verion 9.3 on the Introduction
page. But this may be common to all packages. There is mention on the
ICON website of a version 9.3.1 but it seems that is only for unix
platforms. Is this true?

I notice at the ICON site (as U. Arizona), there are two MS-DOS versions
in the ftp archive. But I still am not clear as to what represents the
most current version for a Win NT machine.

Also, I am not sure (searching the archive) which (if any) document is
the "tutorial" menioned in the introduction as "to be written".

Answers and assistance still appreciated.



> 
> --
> "Mathematics is the foundation of all science and those who are
> profficent in it become the builders of society."
> 
>                                          The Honorable Ellijah Mohammed
> 
> "A mathematician is a machine for turning coffee into theorems."
> 
>                                           Paul Erdos
>                                           Mathematician


From icon-group-sender  Mon Feb  1 08:30:52 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id IAA29203
	for icon-group-addresses; Mon, 1 Feb 1999 08:30:45 -0700 (MST)
Message-Id: <199902011530.IAA29203@baskerville.CS.Arizona.EDU>
X-Sender: acsander@csat.gatech.edu
Date: Mon, 01 Feb 1999 08:23:09 -0800
To: Clinton Jeffery <jeffery@segfault.cs.utsa.edu>, gkt@math.luc.edu
From: Alan Sandercock <acsander@mir.gis.gatech.edu>
Subject: Re: file and directory manipulation
Cc: icon-group@optima.CS.Arizona.EDU
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

I've been successfully using Icon on a Unix platform (DGUX) for small
programming jobs related to Geographical Information Processessing (GIS)
and some text processing. We are moving to a Windows NT platform. Will I be
able to use Unicon in this environment? I'm especially concerned about
gathering a good set of software tools to ensure that our move from Unix to
NT is as painless as possible, and I hope that I can continue to use Icon.
I guess I can always resort to Perl which seems to enjoy good support on NT
and fot that matter straight Icon will work, but I'm not sure about those
POSIX extensions.

At 02:53 PM 1/29/99 -0600, Clinton Jeffery wrote:
>
>>   Is it possible, other than by using system(...), to perform various file
>>   and directory manipulation functions on the Unix platform? I'm
>>   desperately trying to use Icon as a perl replacement.
>
>Hi George,
>
>You want Unicon, http://www.drones.com/unicon, an Icon with POSIX extensions
>developed by Shamim Mohamed.  Unicon forms the basis for some multiplatform
>file, directory, and internet facilities that we are working on, which will
>be described in a forthcoming book.
>
>>  I cannot seem to create directories easily.
>
>I really think that system() ought to work great for this, but in any case,
>Unicon has these types of functions as built-ins for Icon.
>
>Clint Jeffery, jeffery@cs.utsa.edu
>Research http://www.cs.utsa.edu/research/plss.html


--
Alan Sandercock  -  acsander@csat.gatech.edu
Phone (404) 894-0129



From icon-group-sender  Mon Feb  1 12:46:28 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id MAA09894
	for icon-group-addresses; Mon, 1 Feb 1999 12:45:51 -0700 (MST)
Message-Id: <199902011945.MAA09894@baskerville.CS.Arizona.EDU>
To: icon-group@optima.CS.Arizona.EDU
Date: 1 Feb 1999 15:30:59 GMT
From: corre@alpha1.csd.uwm.edu (Alan D Corre)
Subject: Dead link
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

The very first link in
http://www.drones.com/unicon/
is dead. There is no email address for the author.

Clarification, please.

-- 
			Alan D. Corre
			Emeritus Professor of Hebrew Studies
			University of Wisconsin-Milwaukee
			http://www.uwm.edu/~corre/


From icon-group-sender  Mon Feb  1 13:05:22 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id NAA10611
	for icon-group-addresses; Mon, 1 Feb 1999 13:05:04 -0700 (MST)
Message-Id: <199902012005.NAA10611@baskerville.CS.Arizona.EDU>
Date: Mon,  1 Feb 1999 08:48:25 -0800 (PST)
From: Shamim Mohamed <spm@drones.com>
To: Alan Sandercock <acsander@mir.gis.gatech.edu>
Cc: icon-group@optima.CS.Arizona.EDU
Subject: Re: file and directory manipulation
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

 > I've been successfully using Icon on a Unix platform (DGUX) for small
 > programming jobs related to Geographical Information Processessing (GIS)
 > and some text processing. We are moving to a Windows NT platform. Will I
 > be able to use Unicon in this environment?

Yes, the goal for the project is to be able to use the Unicon extensions
both on Unix and on Windows. One of Clint's students has already ported most 
of the features to Windows. Some of the features Unix programmers expect are 
absent in NT, so those are deprecated. One example is fork/dup/exec that is
a common idiom for spawning programs on Unix; it has been subsumed by
extending system(). Please read http://www.drones.com/unicon/ for more
details. (fork, exec and dup are still available in Unix versions but are
deprecated.)

-s


From icon-group-sender  Wed Feb  3 17:23:01 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id RAA25123
	for icon-group-addresses; Wed, 3 Feb 1999 17:21:05 -0700 (MST)
Message-Id: <199902040021.RAA25123@baskerville.CS.Arizona.EDU>
To: icon-group@optima.CS.Arizona.EDU
Date: Wed, 3 Feb 1999 16:49:59 -0600
From: <thchr@ibm.net>
Subject: Re: the beginning
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

The Icon Programming Language Handbook is available on
    http://www.toolsofcomputing.com/freesoftware.htm
Marek Budyn wrote in message ...

If somebody knows any
>web page that shares eg. books (id pdf or ps format) how to programm in
>Icon.





From icon-group-sender  Tue Feb  9 08:53:24 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id IAA27464
	for icon-group-addresses; Tue, 9 Feb 1999 08:50:52 -0700 (MST)
Message-Id: <199902091550.IAA27464@baskerville.CS.Arizona.EDU>
To: icon-group@optima.CS.Arizona.EDU
Date: 08 Feb 1999 19:11:44 -0500
From: clayton@jazz.cc.gatech.edu (R. Clayton)
Subject: Calling C functions under Solaris.
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

I'd like to call some C functions from with an Icon program under Solaris, and
I'm having a little trouble.  First, here's what I'm doing and what I'd like to
have happen (I don't care that the error number is the wrong one):

  ja uname -a
  SunOS jazz 4.1.3 1 sun4m

  ja cat ps.c
  #include "rt.h"

  int printstring(argc, argv) 

    int argc;
    struct descrip * argv;

   {char str[MaxCvtLen];
    if (qtos(argv[1], str) == Error) return 101;
    printf("printstring parameter is \"%s\".\n", str);
    return 0;
    }

  ja cat ps.icn
  procedure main()

    local printstring

    printstring := loadfunc("./ps.so", "printstring") |
      stop("Couldn't load printstring().")

    printstring("Hello world!")

    end # main

  ja make
  cc -I/usr/local/icon-9.3/src/h -I/usr/openwin/include   -c ps.c -o ps.o
  ld -o ps.so ps.o
  icont -u ps.icn
  Translating:
  ps.icn:
    main
  No errors
  Linking:

  ja ./ps
  printstring parameter is "Hello world!".

  ja 

So everything works under SunOS.  Here's Solaris:

  ma uname -a
  SunOS mayzie 5.5.1 Generic_103640-24 sun4u sparc SUNW,Ultra-2

  ma rm *o

  ma make
  cc -I/usr/local/icon-9.3/src/h -I/usr/openwin/include   -c ps.c -o ps.o
  "/usr/local/icon-9.3/src/h/../h/config.h", line 424: warning: macro
  replacement within a string literal 
  cc -G -K pic -o ps.so ps.o
  icont -u ps.icn
  Translating:
  ps.icn:
    main
  No errors
  Linking:

  ma ./ps

  loadfunc("./ps.so","printstring"): ld.so.1: /usr/local/public/bin/iconx:
  fatal: relocation error: file ./ps.so: symbol qtos: referenced symbol not
  found 

  Run-time error 216
  File ps.icn; Line 5
  external function not found
  Traceback:
     main()
     loadfunc("./ps.so\x00","printstring\x00") from line 5 in ps.icn

  ma

I'm thinking I'm missing an Icon library under Solaris, but I can't find one in
the source tree or the executable directories.

And while I have your attention: does anybody know if there are C-callable
routines for creating and manipulating Icon lists?  Eventually I'm going to
want to write some C routines that return a list of strings.


From icon-group-sender  Wed Feb 10 08:30:59 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id IAA17702
	for icon-group-addresses; Wed, 10 Feb 1999 08:30:16 -0700 (MST)
Message-Id: <199902101530.IAA17702@baskerville.CS.Arizona.EDU>
Date: Tue, 9 Feb 1999 21:47:01 -0700
From: Gregg Townsend <gmt@optima.CS.Arizona.EDU>
To: clayton@jazz.cc.gatech.edu, icon-group@optima.CS.Arizona.EDU
Subject: Re:  Calling C functions under Solaris.
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

    From: clayton@jazz.cc.gatech.edu (R. Clayton)
    
    I'd like to call some C functions from with an Icon program under Solaris,
    and I'm having a little trouble....
    
      loadfunc("./ps.so","printstring"): ld.so.1: /usr/local/public/bin/iconx:
      fatal: relocation error: file ./ps.so: symbol qtos: referenced symbol not
      found 
    
    I'm thinking I'm missing an Icon library under Solaris, but I can't
    find one in the source tree or the executable directories.

No, there is no separate Icon library, so that's not it.  I'm not sure
what the problem is; it worked for me on a Solaris 2.6 system using
the Sun C compiler.  I did have to edit your code to fix a wrong
parameter to qtos, though; if you didn't see that error, maybe you
have a different configuration, or a version of Icon older than 9.3.1?

   ---------------------------------------------------------------------------
   Gregg Townsend              Gould-Simpson Building   gmt@cs.arizona.edu
   Staff Scientist             1040 E. 4th St.          32 13 45N  110 57 16W
   Dept. of Computer Science   PO Box 210077            tel: +1 520 621 4325
   The University of Arizona   Tucson, AZ  85721-0077   fax: +1 520 621 4246
    


From icon-group-sender  Wed Feb 10 08:31:51 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id IAA17765
	for icon-group-addresses; Wed, 10 Feb 1999 08:31:43 -0700 (MST)
Message-Id: <199902101531.IAA17765@baskerville.CS.Arizona.EDU>
Date: Wed, 10 Feb 1999 00:19:44 -0800 (PST)
From: Shamim Mohamed <spm@drones.com>
To: icon-group@optima.CS.Arizona.EDU
Subject: BART schedule, Tufte and Icon
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

I wrote a little Icon program to draw Bart (San Francisco area metro
railway) schedules as first drawn by E. J. Marey and popularised by 
E. R. Tufte. I think it's not bad for a couple of hours of work, and thought 
others might be interested. It's about 500 lines of poorly documented
(but easy to understand, I think!) code, and I'd be glad to send it to
anyone that wants to play with it and improve it.

For a picture, http://www.drones.com/bart.html

-s


From icon-group-sender  Thu Feb 11 09:03:37 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id JAA06690
	for icon-group-addresses; Thu, 11 Feb 1999 09:00:30 -0700 (MST)
Message-Id: <199902111600.JAA06690@baskerville.CS.Arizona.EDU>
To: icon-group@optima.CS.Arizona.EDU
Date: Thu, 11 Feb 1999 04:54:49 GMT
From: eodell@pobox.com (Eric O'Dell)
Subject: Re: BART schedule, Tufte and Icon
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

On 10 Feb 1999 10:33:02 -0500, Shamim Mohamed <spm@drones.com> wrote:

>I wrote a little Icon program to draw Bart (San Francisco area metro
>railway) schedules as first drawn by E. J. Marey and popularised by 
>E. R. Tufte.

I feel much less geeky now knowing that I'm not the only person in the
world who thinks E. R. Tufte is really cool. ;-)

-E.



+-------------------------------------------------------------------+
| "I have come a very long way from myself only to realize that     |
| identity is a skill and self-betrayal is a habit. Once lost, the  |
| former is very hard to regain; once gained, the latter is very    |
| hard to lose."  ---I. Corvus, _The Europe of Our Dreams_          |
+-------------------------------------------------------------------+


From icon-group-sender  Thu Feb 11 09:10:17 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id JAA07128
	for icon-group-addresses; Thu, 11 Feb 1999 09:08:35 -0700 (MST)
Message-Id: <199902111608.JAA07128@baskerville.CS.Arizona.EDU>
To: icon-group@optima.CS.Arizona.EDU
Date: 11 Feb 1999 06:02:35 GMT
From: jeffery@cs.utsa.edu (Clinton Jeffery)
Subject: Re: BART schedule, Tufte and Icon
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Eric O'Dell (eodell@pobox.com) wrote:

: SM> I wrote a little Icon program to draw Bart (San Francisco area metro
: SM> railway) schedules as first drawn by E. J. Marey and popularised by 
: SM> E. R. Tufte.

: I feel much less geeky now knowing that I'm not the only person in the
: world who thinks E. R. Tufte is really cool. ;-)

Tufte is the antithesis of geeky. Geeks do not go for $995-a-day seminars
of the kind Tufte has been doing.  His first book is especially cool, so I
still recommend him to anyone interested in information visualization. Has
anyone checked out his third book, to see if it is better than his second?

--
Clint Jeffery, jeffery@cs.utsa.edu
Research http://www.cs.utsa.edu/research/plss.html
Will Hack Icon For Food!


From icon-group-sender  Tue Feb 16 11:37:18 1999
Return-Path: <icon-group-sender>
Received: from ursus.CS.Arizona.EDU (ursus.CS.Arizona.EDU [192.12.69.63])
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) with SMTP id LAA20509
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Tue, 16 Feb 1999 11:37:18 -0700 (MST)
Received: by ursus.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA15852; Tue, 16 Feb 1999 11:37:17 -0700
From: WHITSTON@ltu.edu
Date: Mon, 15 Feb 1999 13:26:24 -0500 (EST)
Subject: Weaving Patterns and Analyst Issue #52 - A Question
To: ICON-GROUP@optima.CS.Arizona.EDU
Message-Id: <01J7RVJX4NOW90NYIT@LTU.EDU>
X-Vms-To: ICON-GROUP@OPTIMA.CS.ARIZONA.EDU
X-Vms-Cc: WHITSTON
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; CHARSET=US-ASCII
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

To the group:

In the latest Analyst, there is an article about "weaving language" with
examples.

The first example is:  1346[]9231
which expands to:  111111111334441

Shouldn't it read:  111111111334446
                                  -

The next example is:  1345[]12{10}3
which expands to:  1334444444444666

Shouldn't it read:  1334444444444555
                                 ---

Did the typo imp strike twice or I'm missing the obvious?

Thank you for your assistance in this matter.

Howard Whitston
Lawrence Tech. Univ.
whitston@ltu.edu


From icon-group-sender  Tue Feb 16 11:48:43 1999
Return-Path: <icon-group-sender>
Received: from ursus.CS.Arizona.EDU (ursus.CS.Arizona.EDU [192.12.69.63])
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) with SMTP id LAA20936
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Tue, 16 Feb 1999 11:48:43 -0700 (MST)
Received: by ursus.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA15885; Tue, 16 Feb 1999 11:48:42 -0700
Date: Tue, 16 Feb 1999 11:48:01 -0700
From: Ralph Griswold <ralph>
Message-Id: <9902161848.AA15033@jupiter.CS.Arizona.EDU>
To: icon-group
Subject: Re:  Weaving Patterns and Analyst Issue #52 - A Question
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO


You're right in both cases.  The typos are my fault; sorry for
the confusion.


	Ralph



From icon-group-sender  Thu Feb 18 11:47:26 1999
Return-Path: <icon-group-sender>
Received: from ursus.CS.Arizona.EDU (ursus.CS.Arizona.EDU [192.12.69.63])
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) with SMTP id LAA17905
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Thu, 18 Feb 1999 11:47:26 -0700 (MST)
Received: by ursus.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA19830; Thu, 18 Feb 1999 11:47:25 -0700
Content-Identifier: EditPAD
Hop-Count: 1
Date: Thu, 18 Feb 1999 09:21 -0600
From: Chris D Tenaglia <Chris.D.Tenaglia@jci.com>
Subject: EditPAD
To: icon-group@optima.CS.Arizona.EDU
Message-Id: <154482@jsux04.corp.jci.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

I tried them both. The EditPad has some convenience features like sitting
in the "tray" preloaded.

But I switched to PFedit because it has "auto-indent" which is something
I value highly in a programmers' editor.

Chris.

---------------------- Forwarded by Chris D Tenaglia/CORP/Johnson_Controls on 
02/18/99 09:16 AM ---------------------------


"Charles Hethcoat" <chethcoa@oss.oceaneering.com> on 01/04/99 05:34:32 PM
To: icon-group@optima.CS.Arizona.EDU
cc:  (bcc: Chris D Tenaglia/CORP/Johnson_Controls)
bcc: Chris D Tenaglia/CORP/Johnson_Controls

Subject: EditPAD



MJE <evans@gte.net> wrote:
>Windows users might be interested in EditPAD,

I never cared for Notepad, either.  For several years I've been pretty happy 
using PFE (http://www.lancs.ac.uk/people/cpaap/pfe/)  as a general purpose 
editor.  A quick reading of the EditPAD description makes the two sound very 
similar.  Has anyone got any comparative experience with both editors?  I'd be 
interested in your comments.  I've got so much stuff installed on my computer 
now that I'm reluctant to install both.  I might, however, swap out PFE for 
EditPad if it seems clearly superior.

Charles Hethcoat






From icon-group-sender  Sun Feb 21 18:06:14 1999
Return-Path: <icon-group-sender>
Received: from ursus.CS.Arizona.EDU (ursus.CS.Arizona.EDU [192.12.69.63])
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) with SMTP id SAA18108
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Sun, 21 Feb 1999 18:06:13 -0700 (MST)
Received: by ursus.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA26019; Sun, 21 Feb 1999 18:06:13 -0700
From: "Frank Lhota" <lhotaf@lexma.meitech.com>
To: <icon-group@optima.CS.Arizona.EDU>
Subject: Icon and the Codewright Editor
Date: Fri, 19 Feb 1999 12:17:14 -0500
Message-Id: <01be5c2b$b1b12bc0$51cba326@freetown>
Mime-Version: 1.0
Content-Type: multipart/alternative;
	boundary="----=_NextPart_000_0036_01BE5C01.C8DB23C0"
X-Priority: 3
X-Msmail-Priority: Normal
X-Mailer: Microsoft Outlook Express 4.71.1712.3
X-Mimeole: Produced By Microsoft MimeOLE V4.71.1712.3
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

This is a multi-part message in MIME format.

------=_NextPart_000_0036_01BE5C01.C8DB23C0
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

We've had some recent posts regarding code editors: the god-awful  =
Notepad versus Editpad versus PFE. I am currently using Codewright for =
Windows NT / 95 for most of my program editing for C, Ada, and Icon. I =
have written a Codewright extension DLL for Icon, including =
chroma-coding and templates. I would be happy to share this DLL and =
sources with any other Icon programmer using Codewright.

------=_NextPart_000_0036_01BE5C01.C8DB23C0
Content-Type: text/html;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD W3 HTML//EN">
<HTML>
<HEAD>

<META content=3Dtext/html;charset=3Diso-8859-1 =
http-equiv=3DContent-Type>
<META content=3D'"MSHTML 4.71.1712.3"' name=3DGENERATOR>
</HEAD>
<BODY bgColor=3D#ffffff>
<DIV><FONT size=3D2>We've had some recent posts regarding code editors: =
the=20
god-awful&nbsp; Notepad versus Editpad versus PFE. I am currently using=20
Codewright for Windows NT / 95 for most of my program editing for C, =
Ada, and=20
Icon. I have written a Codewright extension DLL for Icon, including=20
chroma-coding and templates. I would be happy to share this DLL and =
sources with=20
any other Icon programmer using Codewright.</FONT></DIV></BODY></HTML>

------=_NextPart_000_0036_01BE5C01.C8DB23C0--



From icon-group-sender  Mon Feb 22 07:32:04 1999
Return-Path: <icon-group-sender>
Received: from ursus.CS.Arizona.EDU (ursus.CS.Arizona.EDU [192.12.69.63])
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) with SMTP id HAA07372
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Mon, 22 Feb 1999 07:32:04 -0700 (MST)
Received: by ursus.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA27100; Mon, 22 Feb 1999 07:32:03 -0700
Message-Id: <4.1.19990221195319.00a27c90@pop5.ibm.net>
X-Sender: laturk@pop5.ibm.net
X-Mailer: QUALCOMM Windows Eudora Pro Version 4.1 
Date: Sun, 21 Feb 1999 20:08:02 -0600
To: icon-group@optima.CS.Arizona.EDU
From: "Dr. Louis A. Turk" <laturk@ibm.net>
Subject: find()
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Dear friends on the Icon group,

My last post was many months ago, right before a major computer crash shut
me down for awhile.  As a result of that crash, I also lost a lot of email
that I needed to answer.  I owed some of you thank you notes, but I can't
remember who, after so long a time.  Please accept my apology and thanks now.

I am now back with a new computer, having fun programming in Icon.  

I have a question: is there a simple way to find only whole words in a
document?  I need "on" to be found, but not the "on" in "Icon."  I also
need whatever method I use to recognize that puncuation marks are not part
of words.

Many thanks in advance,
Louis Turk




From icon-group-sender  Mon Feb 22 07:32:21 1999
Return-Path: <icon-group-sender>
Received: from ursus.CS.Arizona.EDU (ursus.CS.Arizona.EDU [192.12.69.63])
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) with SMTP id HAA07375
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Mon, 22 Feb 1999 07:32:21 -0700 (MST)
Received: by ursus.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA27104; Mon, 22 Feb 1999 07:32:21 -0700
Date: Mon, 22 Feb 1999 09:04:37 -0500
Message-Id: <199902221404.JAA16135@mir.gis.gatech.edu>
From: Alan Sandercock <acsander@mir.gis.gatech.edu>
To: icon-group@optima.CS.Arizona.EDU
In-Reply-To: <01be5c2b$b1b12bc0$51cba326@freetown> (lhotaf@lexma.meitech.com)
Subject: Re: Icon and the Codewright Editor
References:  <01be5c2b$b1b12bc0$51cba326@freetown>
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

My favorite editor for Icon is Gnu Emacs for Windows Version 20.3.1
since it's free and it comes with color syntax highlighting already
defined for Icon.  It is also extremely easy to write and debug Icon
programs from within the editor with output going to an emacs window.
In fact this is by far the fastest and most interactive Icon
programming environment that I've encountered so far.  It is trivial
to define function keys to compile and execute the contents of the
buffer - the nticont/execute cycle.  

>  We've had some recent posts regarding code editors: the god-awful  =
>  Notepad versus Editpad versus PFE. I am currently using Codewright for =
>  Windows NT / 95 for most of my program editing for C, Ada, and Icon. I =
>  have written a Codewright extension DLL for Icon, including =
>  chroma-coding and templates. I would be happy to share this DLL and =
>  sources with any other Icon programmer using Codewright.
>  

-- 
Alan Sandercock
(404) 894-0129
acsander@csat.gatech.edu



From icon-group-sender  Mon Feb 22 09:21:32 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id JAA11214
	for icon-group-addresses; Mon, 22 Feb 1999 09:21:25 -0700 (MST)
Message-Id: <199902221621.JAA11214@baskerville.CS.Arizona.EDU>
Date: Mon, 22 Feb 1999 08:42:31 -0600
From: "Charles Hethcoat" <chethcoa@oss.oceaneering.com>
To: laturk@ibm.net, icon-group@optima.CS.Arizona.EDU
Subject: Re: find()
Content-Disposition: inline
X-MIME-Autoconverted: from quoted-printable to 8bit by baskerville.CS.Arizona.EDU id HAA07568
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

See "concord.icn" in the Icon Programming Library.  This is an excellent tool in its own right and serves as a perfect starting point for custom work.

Charles Hethcoat

>>> "Dr. Louis A. Turk" <laturk@ibm.net> 99-02-21 8:08:02 PM >>>
...is there a simple way to find only whole words in a
document?  I need "on" to be found, but not the "on" in "Icon."  I also
need whatever method I use to recognize that puncuation marks are not part
of words.




From icon-group-sender  Mon Feb 22 16:26:19 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id QAA26732
	for icon-group-addresses; Mon, 22 Feb 1999 16:25:10 -0700 (MST)
Message-Id: <199902222325.QAA26732@baskerville.CS.Arizona.EDU>
Date: Mon, 22 Feb 1999 14:25:59 -0800 (PST)
From: Shamim Mohamed <spm1@drones.com>
To: icon-group@optima.CS.Arizona.EDU
Subject: Icon on Linux/PPC?
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

A friend of mine wants to run Icon (with graphics) on the Power PC running
Linux. I told him it should be straightforward to port, except for the
rswitch stuff. Does anyone know if there are any Power PC ports of Icon?
(For now I've told him to disable coexpressions.)

Thanks,

-s


From icon-group-sender  Tue Feb 23 08:44:08 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id IAA21982
	for icon-group-addresses; Tue, 23 Feb 1999 08:42:29 -0700 (MST)
Message-Id: <199902231542.IAA21982@baskerville.CS.Arizona.EDU>
Date: Mon, 22 Feb 1999 17:08:36 -0700
From: Gregg Townsend <gmt@optima.CS.Arizona.EDU>
To: icon-group@optima.CS.Arizona.EDU, spm1@drones.com
Subject: Re:  Icon on Linux/PPC?
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

    From: Shamim Mohamed <spm1@drones.com>
    
    ...Does anyone know if there are any Power PC ports of Icon?
    
Try starting from the rswitch.s file in the rs6000_aix configuration.

   ---------------------------------------------------------------------------
   Gregg Townsend              Gould-Simpson Building   gmt@cs.arizona.edu
   Staff Scientist             1040 E. 4th St.          32 13 45N  110 57 16W
   Dept. of Computer Science   PO Box 210077            tel: +1 520 621 4325
   The University of Arizona   Tucson, AZ  85721-0077   fax: +1 520 621 4246


From icon-group-sender  Tue Feb 23 14:51:48 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id OAA04732
	for icon-group-addresses; Tue, 23 Feb 1999 14:50:53 -0700 (MST)
Message-Id: <199902232150.OAA04732@baskerville.CS.Arizona.EDU>
From: "Frank Lhota" <lhotaf@lexma.meitech.com>
To: <icon-group@optima.CS.Arizona.EDU>
Subject: Bridging Icon and C Calls
Date: Tue, 23 Feb 1999 12:03:24 -0500
X-Priority: 3
X-MSMail-Priority: Normal
X-MimeOLE: Produced By Microsoft MimeOLE V4.71.1712.3
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Recently, there have been some discussion of the Icon facility for calling
external C functions. This facility varies widely from platform to platform,
but it basically consists of calling loadfunc to load a C function from a
dynamic library, then calling the function via the Icon callout function (or
directly in the case of Unix).

The problem with this facility is that it can only be used for C functions
specifically written for Icon. The C function must accept Icon descriptors
as parameters. You generally cannot call a typical external C function from
Icon without first writing a "skin" function that does Icon-to-C argument
translation.

This would be particularly useful under Win 95 / 98 / NT, where the system
services all take the form of functions in dynamic link libraries. The
Windows API has hundreds of functions, many of which would be handy to use
under Icon. These functions cannot be used, however, without some additional
C work to turn Icon arguments into C arguments.

What would be nicer if we could create a facility by which callout would
automatically convert Icon parameters to C, e.g. if there is an external
function foo declared as

    int foo ( int n, double x );

then to call this function from Icon, we first load it with loadfunc, then
do this call:

    callout ( "foo", 3, 4.5 )

There are numerous complications with doing this. C includes integers and
floating point numbers of various sizes, whereas Icon programmers are
shielded from such concerns. C also uses pointers for many parameters, so we
need to come up with Icon support for this. Pointer parameters can be
implemented using an Icon record, e.g.

    record Pointer ( value )

    # ...
    # function bar declared as
    #     void bar ( int *ip );
    callout ( "bar", ip := Pointer ( 0 ) )
    # ip.value updated by callout

I have some ideas about how this facility could be designed, and
implemented. Before proceeding this much further, however, I would like some
feedback. Would other Icon programmers find this useful? If I complete
something like this for NT, could others help me port this to other
platforms?







From icon-group-sender  Wed Feb 24 08:40:49 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id IAA18284
	for icon-group-addresses; Wed, 24 Feb 1999 08:39:57 -0700 (MST)
Message-Id: <199902241539.IAA18284@baskerville.CS.Arizona.EDU>
X-Authentication-Warning: pandora.cs.utsa.edu: rpereda owned process doing -bs
Date: Tue, 23 Feb 1999 22:02:14 -0600 (CST)
From: Ray Pereda <rpereda@cs.utsa.edu>
X-Sender: rpereda@pandora
To: Frank Lhota <lhotaf@lexma.meitech.com>
cc: icon-group@optima.CS.Arizona.EDU
Subject: Re: Bridging Icon and C Calls
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

On Tue, 23 Feb 1999, Frank Lhota wrote:

> I have some ideas about how this facility could be designed, and
> implemented. Before proceeding this much further, however, I would like some
> feedback. Would other Icon programmers find this useful? If I complete
> something like this for NT, could others help me port this to other
> platforms?

I and many others would find this very useful.  Particularly if Icon to C
compiler technology continues to improve. 

Look at www.swig.org 
Here is a bit on their title page:
Welcome to SWIG 

SWIG is a software development tool that connects programs
written in C, C++, and Objective-C with a variety of high-level
programming languages. SWIG is primarily used with common
scripting languages such as Perl, Python, and Tcl/Tk, but has also
been extended to include languages such as Java, Eiffel, and Guile.
SWIG is most commonly used to create high-level interpreted
programming environments, systems integration, and as a tool for
building user interfaces. SWIG is distributed as OpenSource and
may be freely used, distributed, and modified for commercial or
noncommercial use. 

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

I looked at it a while back.  It looks like a well-thought out design for
interfacing to C.  I don't remember much right now.  Let me know what you
think. 

-ray 



From icon-group-sender  Wed Feb 24 13:51:43 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id NAA29285
	for icon-group-addresses; Wed, 24 Feb 1999 13:50:51 -0700 (MST)
Message-Id: <199902242050.NAA29285@baskerville.CS.Arizona.EDU>
Date: Wed, 24 Feb 1999 10:24:56 -0600
From: MJE <evans@gte.net>
To: icon-group@optima.CS.Arizona.EDU
Subject: Re: Icon and the Codewright Editor
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Why not post your CodeWrite DLL to the Icon FTP site?  Then it will be
available anytime anywhere to anybody.

Mark

P.S.  Can you fix your email program to send ASCII instead of HTML.


Frank Lhota wrote:
> 
> This is a multi-part message in MIME format.
> 
> ------=_NextPart_000_0036_01BE5C01.C8DB23C0
> Content-Type: text/plain;
>         charset="iso-8859-1"
> Content-Transfer-Encoding: quoted-printable
> 
> We've had some recent posts regarding code editors: the god-awful  =
> Notepad versus Editpad versus PFE. I am currently using Codewright for =
> Windows NT / 95 for most of my program editing for C, Ada, and Icon. I =
> have written a Codewright extension DLL for Icon, including =
> chroma-coding and templates. I would be happy to share this DLL and =
> sources with any other Icon programmer using Codewright.
> 
> ------=_NextPart_000_0036_01BE5C01.C8DB23C0
> Content-Type: text/html;
>         charset="iso-8859-1"
> Content-Transfer-Encoding: quoted-printable
>



From icon-group-sender  Thu Feb 25 12:32:44 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id MAA12131
	for icon-group-addresses; Thu, 25 Feb 1999 12:30:44 -0700 (MST)
Message-Id: <199902251930.MAA12131@baskerville.CS.Arizona.EDU>
From: "Frank Lhota" <lhotaf@lexma.meitech.com>
To: <icon-group@optima.CS.Arizona.EDU>
Subject: Designing the Bridge between Icon and C
Date: Thu, 25 Feb 1999 13:04:29 -0500
X-Priority: 3
X-MSMail-Priority: Normal
X-MimeOLE: Produced By Microsoft MimeOLE V4.71.1712.3
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

As I mentioned in a previous posting, I would like to develop an Icon
facility for calling a native C function in a dynamic library, that is, a C
function that is not an Icon "wrapper" function. I believe this can be done
in reasonably flexible and portable way. In my proposed design, a C function
is loaded into Icon using the predefined function loadfunc, which takes the
form

    loadfunc ( modname, funcname, conv )

where

- modname is the name of the dynamic library file,
- funcname is the name of the function as it appears in the library; and
- conv is a string indicating the calling convention ("C" or "Pascal"); the
conv parameter defaults to "C".

The loaded external C function is called using the callout function, invoked
as follows:

    callout ( funcname, retval, argv[] )

where

- funcname is the name of the external function;
- retval is a prototype for the return value (defaults to &null, indicating
that the C function return type is void); and
- argv[] is the list of function arguments, which are to be converted to C
format before the procedure is called.

To free the resources associated with loaded C function, call the Icon
function unloadfunc, called with the name of the loaded function:

    unloadfunc(funcname)

The real sticking point is how to design the Icon specification of C data.
Icon is relatively free of the data representation concerns that pervade C.
Part of what we need to do when we to convert Icon data to C is to add a
specification on how this data is to be represented.

All C types fall into two categories: basic types and derived types. The
basic data types of C can be divided into three categories:

1.    The void type;
2.    The integer types, including char, short, int, and long (as well as
long long on some systems), along with their unsigned versions; and
3.    The floating point types float, double, and on some systems, long
double.

Since enumerations always have integral values, we can include enumeration
types with the integer types.

Representing the C void is easy. The Icon null type is the perfect
counterpart to the C void type.

Integers and floats are more problematic. Icon is not concerned with number
sizes, and Icon integers always have a sign. An Icon real can be directly be
translated into a C double. An Icon integer can also be used to represent a
signed C int. To translate Icon to other C numeric types, we need some way
to specify the number format. To do this, I propose a standard Icon include
file that defines the following Icon record types:

    record int ( value, size, unsigned )
    record float ( value, size )

When records of these types are used as callout parameters, they have the
following meanings:

- A record of the form int ( value, size, unsigned ) represents a C integer
with a value given by the value component and with a length of size bytes.
This C integer is unsigned if the unsigned component of the record is
non-null.

- A record of the form float ( value, size ) represents a C real number with
a value given by the value component of length of size bytes.

These numeric formatting records should suffice for specifying any of the
basic C numeric types.

Generating the C derived types from Icon is a bit more challenging. Starting
with the basic C types, all C types can be derived using the following
constructs:

1.    An array of objects of a given type;
2.    A structure containing a sequence of objects of various types;
3.    A union that is capable of containing any one of several objects of
various types;
4.    A pointer to an object of a given type; or
5.    A pointer to a function.

The Icon representation of a C array or structure is relatively
straightforward. From a certain point of view, an array is quite similar to
a structure. Both an array and a structure are a sequence of values; the
difference is that in an array, the values are all of the same type. A C
array or structure can be modeled in Icon using an Icon list. When callout
translates an Icon list to C, it will create a C region consisting of a
sequence of the C translations of the elements in the Icon list. The int and
float record types can be used to specify structure member sizes.

One of the most popular C arrays are, of course, arrays of characters. We
should be able to use Icon strings to represent C strings.

Representing a C union in Icon is trickier, since there is no obvious Icon
equivalent. Some important API functions use unions, so we should come up
with some way of representing them in Icon. For now, I'm stumpted on this
one. I would greatly appreciate anyone's input on this matter.

Pointers are very important. This is how C implements "call by reference".
For many C functions, the most important outputs are returned by pointers.
Icon technically uses only "call by value", but the effect of "call by
reference" can be obtained by using Icon structures. We can therefore get
the desired effect using the following record types:

    record Pointer ( value )
    record CPointer ( value )

In the context of callout, a Pointer record represents a pointer to the C
representation of the value component. When callout encounters a record of
this type, it will allocate a C region to store the C translation of the
Icon value. The pointer to this C region can then be used in the call to the
C function. After the C function is called, the region value can then be
copied back to the Pointer record, and the pointer is deallocated. For
example,

    # function bar declared as
    #     void bar ( int *ip );
    callout ( "bar", &null, ip := Pointer ( 0 ) )
    # ip.value updated by callout

The CPointer record is similar to pointer, except that it is used to
indicate a pointer to a region that will not be modified by the C function.
The callout function processes a CPointer record in the same way, except
that the region value is not copied back to pointer value.

For now, I'm punting on coming up with an Icon representation of C function
pointers. Ideally, it would be nice to be able to automatically generate C
wrapper functions that call an Icon procedure, but that is a project for
another day.

I've started work on this interface design, but before I take it further, I
would greatly some feedback from the Icon programmer community. In
particular,

- Does this design meet your interface needs?
- Is there a simpler and/or more intuitive design for this interface?
- How would you implement C unions in Icon?



From icon-group-sender  Mon Mar  1 09:00:53 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id IAA28134
	for icon-group-addresses; Mon, 1 Mar 1999 08:57:45 -0700 (MST)
Message-Id: <199903011557.IAA28134@baskerville.CS.Arizona.EDU>
To: icon-group@optima.CS.Arizona.EDU
Date: Sat, 27 Feb 1999 04:23:00 GMT
From: eodell@pobox.com (Eric O'Dell)
Subject: Re: Bridging Icon and C Calls
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

On 23 Feb 1999 16:55:50 -0500, "Frank Lhota"
<lhotaf@lexma.meitech.com> wrote:

>I have some ideas about how this facility could be designed, and
>implemented. Before proceeding this much further, however, I would like some
>feedback. Would other Icon programmers find this useful?

I think this would be enormously useful even if it wasn't portable
from platform to platform. Being able to code speed-critical sections
in C, or just to use pre-existing C functions would be a godsend. If
it _was_ portable from one platform to another, it would be even more
useful. I'm all for it.

-E.




+-------------------------------------------------------------------+
| "I have come a very long way from myself only to realize that     |
| identity is a skill and self-betrayal is a habit. Once lost, the  |
| former is very hard to regain; once gained, the latter is very    |
| hard to lose."  ---I. Corvus, _The Europe of Our Dreams_          |
+-------------------------------------------------------------------+


From icon-group-sender  Wed Mar 10 17:37:48 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id RAA00150
	for icon-group-addresses; Wed, 10 Mar 1999 17:35:13 -0700 (MST)
Message-Id: <199903110035.RAA00150@baskerville.CS.Arizona.EDU>
To: icon-group@optima.CS.Arizona.EDU
Date: Wed, 10 Mar 1999 16:57:04 -0500
From: Charles Evans <cko@chyden.net>
Subject: Mac version of 9.3?
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Any guesstimates on when the Mac version of 9.3 will be released?

Also, is anyone working on a BeOS port?

Sincerely,

Charles Evans
President, Chyden.Net
Home of the World's First BeOS ISP
http://www.chyden.net


From icon-group-sender  Thu Mar 11 09:05:29 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id JAA25487
	for icon-group-addresses; Thu, 11 Mar 1999 09:04:36 -0700 (MST)
Message-Id: <199903111604.JAA25487@baskerville.CS.Arizona.EDU>
To: icon-group@optima.CS.Arizona.EDU
Date: Thu, 11 Mar 1999 04:25:33 GMT
From: hueras@world.std.com (Jon Hueras)
Subject: Re: Mac version of 9.3?
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

In article <199903110035.RAA00150@baskerville.CS.Arizona.EDU>, Charles
Evans <cko@chyden.net> wrote:

> Any guesstimates on when the Mac version of 9.3 will be released?

Although I have made a couple of attempts to kick-start a port of 9.3 to
the Mac, various personal factors (not the least of which is my real job
:-) have conspired to rob me of the time necessary to move it forward. At
this point I cannot predict when this logjam will clear for a sufficient
interval to permit me to wrap up the project.

I apologize profusely to those who have been anxiously awaiting this version. 

   Jon


From icon-group-sender  Mon Mar 15 16:41:55 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id QAA08488
	for icon-group-addresses; Mon, 15 Mar 1999 16:37:00 -0700 (MST)
Message-Id: <199903152337.QAA08488@baskerville.CS.Arizona.EDU>
Date: Mon, 15 Mar 1999 13:03:15 -0700
From: "Sandra L. Miller" <slm@optima.CS.Arizona.EDU>
X-Accept-Language: en
To: Mikko Saarela <druga@clinet.fi>
CC: ralph@optima.CS.Arizona.EDU, icon-group@optima.CS.Arizona.EDU
Subject: Re: Icon for my company?
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

I have forwarded your email to the Icon mailing list.  If you would like
to subscribe to the list, send mail to icon-group-request@cs.arizona.edu
requesting to be added to the list.  The members of the list can be very
helpful with questions such as yours.

Mikko Saarela wrote:
> 
> Hello!
> 
> I wonder if I sent this e-mail to right address, but I would I have a
> question about Icon.
> I develop professionally shopping and ordering applications for web, and
> it's critical that these applications function accurately. (As always whe
> it'ss a question about money :) )
> 
> So the question is... is Icon suitable for my purposes? In other words,
> does Icon as a language offer such features that allow the development of
> highly reliable software? And what are these characteristics?
> 
> Icon seems to me very promising and intelligent language.
> 
> I liked the Icon - website too, it's clear and loads fast beside the
> excellent content.
> 
> I would appreciate a reply.
> 
> Yours,
> 
> Mikko Saarela, Druga Company, Finland
> (e-mail: mikko.saarela@druga.com, www: www.druga.com)

-- 
Sandra L. Miller
Department of Computer Science
University of Arizona


From icon-group-sender  Mon Mar 15 16:47:53 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id QAA08793
	for icon-group-addresses; Mon, 15 Mar 1999 16:46:34 -0700 (MST)
Message-Id: <199903152346.QAA08793@baskerville.CS.Arizona.EDU>
Date: Mon, 15 Mar 1999 21:53:27 +0000
From: Ian Gilbert <igil@globalnet.co.uk>
To: icon-group@optima.CS.Arizona.EDU
Subject: compiling icon
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Hello,
            I have recently downloaded and installed icon version 9 and
was wondering if you
could answer a question about it. How do I compile and run an icon
program to work outside of the wicon enviroment ??? and not in the
console window that appears when you run an .exe ??

Hope you can see what I,m getting at , and can help.

All the best , Ian Gilbert.














From icon-group-sender  Tue Mar 16 16:55:19 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id QAA23680
	for icon-group-addresses; Tue, 16 Mar 1999 16:53:12 -0700 (MST)
Message-Id: <199903162353.QAA23680@baskerville.CS.Arizona.EDU>
To: icon-group@optima.CS.Arizona.EDU
Date: Tue, 16 Mar 1999 17:58:16 -0400
From: "Gerardo H Calderon" <snooper@coqui.net>
Subject: Questions about Icon
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Hello. I'm a computer engineering student at the University of Puerto Rico,
Mayaguez campus. I'm currently in my Programming Languages course and I have
to do a project regarding the Icon Programming language. The professor
assigned some questions to be answered, regarding it's creation,
implementation, and parts. I would really appreciate if anyone could help me
in acquiring this information. I'll try to post some of the questions later
on.

Attentively,

Gerardo H Calderon
Email: snooper@coqui.net




From icon-group-sender  Wed Mar 17 08:25:43 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id IAA17327
	for icon-group-addresses; Wed, 17 Mar 1999 08:23:49 -0700 (MST)
Message-Id: <199903171523.IAA17327@baskerville.CS.Arizona.EDU>
Date: Tue, 16 Mar 1999 17:06:58 -0700 (MST)
From: Gregg Townsend <gmt@baskerville.CS.Arizona.EDU>
To: icon-group@optima.CS.Arizona.EDU, snooper@coqui.net
Subject: Re: Questions about Icon
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Icon's history is covered in ACM SIGPLAN Notices for March 1993 (Vol 28, No 3):
Preprints from the second HOPL (History of Programming Languages) conference.
Other information about Icon can be found at the Icon web site,
	http://www.cs.arizona.edu/icon/

   ---------------------------------------------------------------------------
   Gregg Townsend              Gould-Simpson Building   gmt@cs.arizona.edu
   Staff Scientist             1040 E. 4th St.          32 13 45N  110 57 16W
   Dept. of Computer Science   PO Box 210077            tel: +1 520 621 4325
   The University of Arizona   Tucson, AZ  85721-0077   fax: +1 520 621 4246



From icon-group-sender  Wed Mar 17 13:05:20 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id NAA02061
	for icon-group-addresses; Wed, 17 Mar 1999 13:04:12 -0700 (MST)
Message-Id: <199903172004.NAA02061@baskerville.CS.Arizona.EDU>
To: icon-group@optima.CS.Arizona.EDU
Date: Wed, 17 Mar 1999 08:08:30 -0800
From: "Alan B. Saichek" <abs@sgi.com>
Subject: Re: Icon Programming Language FAQ: Y2K ?
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

icon-project@cs.arizona.edu wrote:

> Archive-name: comp-lang-icon-faq
> Posting-Frequency: monthly
>
> Frequently Asked Questions About The Icon Programming Language
>
> Last updated October 12, 1998.
>

Sorry if I've missed any discussion on Y2K and Icon previously.

Is there a general observation about this topic? Might go on the FAQ?

--
Alan B. Saichek                                         abs@engr.sgi.com
Member Technical Staff - QA Team Lead                   (650) 933-2507
Advanced Entertainment Systems, Video Compute R&D





From icon-group-sender  Thu Mar 18 08:39:53 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id IAA00942
	for icon-group-addresses; Thu, 18 Mar 1999 08:34:59 -0700 (MST)
Message-Id: <199903181534.IAA00942@baskerville.CS.Arizona.EDU>
X-Authentication-Warning: pandora.cs.utsa.edu: rpereda owned process doing -bs
Date: Wed, 17 Mar 1999 19:25:50 -0600 (CST)
From: Ray Pereda <rpereda@cs.utsa.edu>
X-Sender: rpereda@pandora
To: "Alan B. Saichek" <abs@sgi.com>
cc: icon-group@optima.CS.Arizona.EDU
Subject: Re: Icon Programming Language FAQ: Y2K ?
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

On Wed, 17 Mar 1999, Alan B. Saichek wrote:

> Sorry if I've missed any discussion on Y2K and Icon previously.
> 
> Is there a general observation about this topic? Might go on the FAQ?

I wrote a program to automatically fix this bug and have already 
been testing it.  Below is the code: 

# Filename: y2k.icn

procedure main()
   while write(map(read(), 'y', 'k'))
end

Here is the output when I run it on a calendar program:

$ cal 1999 | y2k
                               1999

      Januark                Februark                March
Su Mo Tu We Th Fr Sa   Su Mo Tu We Th Fr Sa   Su Mo Tu We Th Fr Sa
                1  2       1  2  3  4  5  6       1  2  3  4  5  6
 3  4  5  6  7  8  9    7  8  9 10 11 12 13    7  8  9 10 11 12 13
10 11 12 13 14 15 16   14 15 16 17 18 19 20   14 15 16 17 18 19 20
17 18 19 20 21 22 23   21 22 23 24 25 26 27   21 22 23 24 25 26 27
24 25 26 27 28 29 30   28                     28 29 30 31
31

       April                   Mak                    June
Su Mo Tu We Th Fr Sa   Su Mo Tu We Th Fr Sa   Su Mo Tu We Th Fr Sa
             1  2  3                      1          1  2  3  4  5
 4  5  6  7  8  9 10    2  3  4  5  6  7  8    6  7  8  9 10 11 12
11 12 13 14 15 16 17    9 10 11 12 13 14 15   13 14 15 16 17 18 19
18 19 20 21 22 23 24   16 17 18 19 20 21 22   20 21 22 23 24 25 26
25 26 27 28 29 30      23 24 25 26 27 28 29   27 28 29 30
                       30 31

        Julk                  August               September
Su Mo Tu We Th Fr Sa   Su Mo Tu We Th Fr Sa   Su Mo Tu We Th Fr Sa
             1  2  3    1  2  3  4  5  6  7             1  2  3  4
 4  5  6  7  8  9 10    8  9 10 11 12 13 14    5  6  7  8  9 10 11
11 12 13 14 15 16 17   15 16 17 18 19 20 21   12 13 14 15 16 17 18
18 19 20 21 22 23 24   22 23 24 25 26 27 28   19 20 21 22 23 24 25
25 26 27 28 29 30 31   29 30 31               26 27 28 29 30


      October                November               December
Su Mo Tu We Th Fr Sa   Su Mo Tu We Th Fr Sa   Su Mo Tu We Th Fr Sa
                1  2       1  2  3  4  5  6             1  2  3  4
 3  4  5  6  7  8  9    7  8  9 10 11 12 13    5  6  7  8  9 10 11
10 11 12 13 14 15 16   14 15 16 17 18 19 20   12 13 14 15 16 17 18
17 18 19 20 21 22 23   21 22 23 24 25 26 27   19 20 21 22 23 24 25
24 25 26 27 28 29 30   28 29 30               26 27 28 29 30 31
31

-ray



From icon-group-sender  Fri Mar 19 09:48:37 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id JAA06899
	for icon-group-addresses; Fri, 19 Mar 1999 09:46:15 -0700 (MST)
Message-Id: <199903191646.JAA06899@baskerville.CS.Arizona.EDU>
Content-Identifier: Building Icon
Hop-Count: 1
Date: Thu, 18 Mar 1999 16:48 -0600
From: Chris D Tenaglia <Chris.D.Tenaglia@jci.com>
Subject: Building Icon
To: icon-group@optima.CS.Arizona.EDU
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Simple question : HP ux on risc
I built icon, but it seems to be missing
X-windows stuff. How do I configure it
to build with the X-windows?

I remember years ago there was a
#define XWINDOWS
in some file???

Perhaps point me at a good webpage 
about building icon with various features.
Thanks in advance.

Chris.



From icon-group-sender  Mon Mar 22 16:57:58 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id QAA07364
	for icon-group-addresses; Mon, 22 Mar 1999 16:54:36 -0700 (MST)
Message-Id: <199903222354.QAA07364@baskerville.CS.Arizona.EDU>
Date: Fri, 19 Mar 1999 17:44:45 -0700 (MST)
From: Gregg Townsend <gmt@baskerville.CS.Arizona.EDU>
To: Chris.D.Tenaglia@jci.com, icon-group@optima.CS.Arizona.EDU
Subject: Re: Building Icon
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

    From: Chris D Tenaglia <Chris.D.Tenaglia@jci.com>
    
    Simple question : HP ux on risc
    I built icon, but it seems to be missing
    X-windows stuff. How do I configure it
    to build with the X-windows?

To configure for graphics, type
	make X-Configure name=hp_risc
instead of
	make Configure name=hp_risc

The installation guide is
	http://www.cs.arizona.edu/icon/docs/ipd243.htm

   ---------------------------------------------------------------------------
   Gregg Townsend              Gould-Simpson Building   gmt@cs.arizona.edu
   Staff Scientist             1040 E. 4th St.          32 13 45N  110 57 16W
   Dept. of Computer Science   PO Box 210077            tel: +1 520 621 4325
   The University of Arizona   Tucson, AZ  85721-0077   fax: +1 520 621 4246


From icon-group-sender  Tue Mar 23 09:20:19 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id JAA27920
	for icon-group-addresses; Tue, 23 Mar 1999 09:19:10 -0700 (MST)
Message-Id: <199903231619.JAA27920@baskerville.CS.Arizona.EDU>
X-Authentication-Warning: helios.dmu.ac.uk: hgs owned process doing -bs
Date: Tue, 23 Mar 1999 09:58:35 +0000 (GMT)
From: Hugh Sasse <hgs@dmu.ac.uk>
To: Icon Group <icon-group@optima.CS.Arizona.EDU>
Subject: opendir/readdir in Icon?
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

I think every machine I have used since about 1978 which had a disk
supported heirarchical directory structures.  However, I cannot find
anything in the Icon book or the online docs about how to test a "file" to
see if it is a file or a directory, or how to list a directory without
making a call to some program outside Icon.  Is there really no 
equivalent to opendir and readdir in Icon?  Or where should I be looking?
	Hugh
	hgs@dmu.ac.uk



From icon-group-sender  Tue Mar 23 09:20:56 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id JAA28013
	for icon-group-addresses; Tue, 23 Mar 1999 09:20:49 -0700 (MST)
Message-Id: <199903231620.JAA28013@baskerville.CS.Arizona.EDU>
To: icon-group@optima.CS.Arizona.EDU
Date: Tue, 23 Mar 1999 08:54:14 -0500
From: Robert Wyesham <rwyesham@earthlink.net>
Subject: Help for the Newbie!
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

I'd like to define an "assert" macro for use in my Icon programs, using ipp, of
course.
I need criticism and advice regarding the definition that I'm proposing.

I decided to steal from the best, so I looked up the definition that P.J.
Plauger used in his book, "The Standard C Library." Here it is, with
line-numbering added, in order to make it easy to refer to (any mistakes are my
own):

1. #undef assert
2. #ifdef NDEBUG
3. #define assert(test) ((void)0)
4. #else
5. #define _STR(x) _VAL(x)
6. #define _VAL(x) #x
7. #define assert(test) (test) ? (void)0 : \
8.                                _Assert( __FILE__ ":" _STR(__LINE__) #test )
9. #endif

The _STR and _VAL macros turn the __LINE__ macro from an integer into a string.
Dr. Plauger defined assert as a ternary expression, rather than as an 'if'
statement, so that crazy people could use the comma operator to do deranged
things like:

                                ( assert(0 < x), x < y )

Without an assert macro, I've been writing:

$ifndef NDEBUG
if test then
    (&null)
else
    stop( "--- assertion failed: file ", &file, ", line ", &line )
$endif

I'm proposing this:

10. $undef assert
11. $ifdef NDEBUG
12. $define assert(test) (&null)
13. $else
14. $define assert(test) if test then (&null) else \
15.                                 stop( "--- assertion failed: file ", &file,
", line ", &line )
16. $endif

Does anyone see anything basically wrong with this definition? Are there
mistakes in it?
Are there better ways to do this? A big problem with it is that the test never
gets printed.
I have no idea how to turn the test into a string that I can print. ipp and Icon
must have a way to do this, but I haven't thought of it. Of course, in a macro,
everything's a string. At this point, I get confused. Any ideas?

Thanks and best regards,
- rob

Rob Wyesham,
Wilmington, Delaware





From icon-group-sender  Tue Mar 23 09:21:28 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id JAA28101
	for icon-group-addresses; Tue, 23 Mar 1999 09:21:19 -0700 (MST)
Message-Id: <199903231621.JAA28101@baskerville.CS.Arizona.EDU>
Date: Tue, 23 Mar 1999 10:07:54 -0600
From: "Charles Hethcoat" <chethcoa@oss.oceaneering.com>
To: icon-group@optima.CS.Arizona.EDU
Subject: Windows Unicon?
Content-Disposition: inline
X-MIME-Autoconverted: from quoted-printable to 8bit by baskerville.CS.Arizona.EDU id JAA27306
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Some earlier posts to this group indicated (it seemed) that an evolution of Unicon to a common implementation for UNIX and Windows might be under way.  Perhaps "convergence" would be the better word.  I was wondering if or when such a thing might occur, and what the status might be, and is there anything I might do to help.  (I run Windows when I can't run Linux, which is most of the time, and it would sure be a nice tool for me.)  Thanks.

Charles Hethcoat
Oceaneering Space Systems



From icon-group-sender  Wed Mar 24 08:18:37 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id IAA02598
	for icon-group-addresses; Wed, 24 Mar 1999 08:15:44 -0700 (MST)
Message-Id: <199903241515.IAA02598@baskerville.CS.Arizona.EDU>
Content-Identifier: Building on HP
Hop-Count: 1
Date: Tue, 23 Mar 1999 11:20 -0600
From: Chris D Tenaglia <Chris.D.Tenaglia@jci.com>
Subject: Building on HP
To: icon-group@optima.CS.Arizona.EDU
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Finally got it.
The target systems have a good C compiler, and good XWindows support.
They were just configured a little oddly and this complicated the build.

My steps were  (and work here, maybe nowhere else, but this is a methodolgy?):

make Clean
make X-Configure name=hp_risc
make Icon
*** oops, missing Xlib.h and other Xwindows files
downloaded them
*** oops, not on the path, can't find anything under X11/
hardcoded X11 occurences to fullpath (i.e. #include 
/user/is/cdt/i9/src/h/X11/Xlib.h)
in all C and .h files.
*** oops, linking problems
HPUX uses the icon var LPATH, setenv LPATH /usr/lib
*** oops, one more link problem
setenv LPATH /usr/lib:/usr/lib/X11R5
Done.

Note : When I built it without the X-Configure, there were no problems at all.

Chris.



From icon-group-sender  Wed Mar 24 08:22:00 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id IAA02739
	for icon-group-addresses; Wed, 24 Mar 1999 08:21:47 -0700 (MST)
Message-Id: <199903241521.IAA02739@baskerville.CS.Arizona.EDU>
Date: Tue, 23 Mar 1999 12:18:00 -0600
From: Clinton Jeffery <jeffery@segfault.cs.utsa.edu>
To: chethcoa@oss.oceaneering.com
CC: icon-group@optima.CS.Arizona.EDU
Subject: Re: Windows Unicon?
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO


[Charles Hethcoat wrote:]
>   Some earlier posts to this group indicated (it seemed) that an evolution
>   of Unicon to a common implementation for UNIX and Windows might be under
>   way.  Perhaps "convergence" would be the better word.  I was wondering if
>   or when such a thing might occur, and what the status might be, and is
>   there anything I might do to help.

A student of mine, Li Lin, has been working on porting Unicon to Windows for
the past 7-8 months. I believe much of the core file system access is
working at this point.  He is presently debugging the networking code.  When
his port works on reasonable tests, we will advertise for beta testers.

Shamim Mohamed and I have revised and extended the Unicon facilities to be
more platform-neutral.  The new multiplatform system interface is covered
in a forthcoming book, "Programming with Icon".  The timetable for releasing
everything was to be this spring, but the Windows Unicon port and some new
Icon versions of the byacc and flex compiler-writer's tools are both behind
schedule at this point.  We won't do a beta release until the software is
ready, hopefully later this spring.  The new distributions also include an
extensive set of database connectivity facilities, and a major rewrite of the
object-oriented facilities.

Clint Jeffery, jeffery@cs.utsa.edu
Research http://www.cs.utsa.edu/research/plss.html
Will Hack For Food!


From icon-group-sender  Wed Mar 24 08:23:24 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id IAA02782
	for icon-group-addresses; Wed, 24 Mar 1999 08:23:13 -0700 (MST)
Message-Id: <199903241523.IAA02782@baskerville.CS.Arizona.EDU>
Date: Wed, 24 Mar 1999 11:29:38 +1200 (NZST)
From: "Richard A. O'Keefe" <ok@atlas.otago.ac.nz>
To: hgs@dmu.ac.uk, icon-group@optima.CS.Arizona.EDU
Subject: Re:  opendir/readdir in Icon?
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Hugh Sasse wrote:
	I think every machine I have used since about 1978 which had a disk
	supported heirarchical directory structures.

Then you haven't used VM/CMS or RT-11 or the old Mac filesystem.
(The old Mac file system had nested *folders* but no actual
directories, file names had to be unique on a disc.)

	However, I cannot find
	anything in the Icon book or the online docs about how to test a "file" to
	see if it is a file or a directory, or how to list a directory without
	making a call to some program outside Icon.

Note that there are no such functions in standard C either.

	Is there really no equivalent to opendir and readdir in Icon?
"Portable" Icon, right.l
	Or where should I be looking?
Unicon.

	


From icon-group-sender  Wed Mar 24 08:23:48 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id IAA02816
	for icon-group-addresses; Wed, 24 Mar 1999 08:23:42 -0700 (MST)
Message-Id: <199903241523.IAA02816@baskerville.CS.Arizona.EDU>
X-Authentication-Warning: helios.dmu.ac.uk: hgs owned process doing -bs
Date: Wed, 24 Mar 1999 09:34:10 +0000 (GMT)
From: Hugh Sasse <hgs@dmu.ac.uk>
To: "Richard A. O'Keefe" <ok@atlas.otago.ac.nz>
cc: icon-group@optima.CS.Arizona.EDU
Subject: Re:  opendir/readdir in Icon?
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

On Wed, 24 Mar 1999, Richard A. O'Keefe wrote:

> Hugh Sasse wrote:
> > I think every machine I have used since about 1978 which had a disk
> > supported heirarchical directory structures.
> 
> Then you haven't used VM/CMS or RT-11 or the old Mac filesystem.
> (The old Mac file system had nested *folders* but no actual
> directories, file names had to be unique on a disc.)

No, I never used a Mac, and it explains why I found RT-11 
baffling when I encountered it :-) And I vaguely remember RSX-11M
being a bit odd in this department too, but am not sure.  But we
digress (a little)... 

> 
> > However, I cannot find
> > anything in the Icon book or the online docs about how to test a
> > "file" to
> > see if it is a file or a directory, or how to list a directory without
> > making a call to some program outside Icon.
> 
> Note that there are no such functions in standard C either.
> 
OK, though Unix provides stat, and Perl, Tcl, Python do also in a
fairly cross platform way, and standard C doesn't provide anything
like Icon's string handling so I would not have chosen that as a
baseline :-)

> > Is there really no equivalent to opendir and readdir in Icon?
> "Portable" Icon, right.

I want to keep things portable as far as possible.

> > Or where should I be looking?
> Unicon.

Discussions I have had confirm that this is looking a good way to 
go.  However, I am surprised at the number of answers on this list
that say "Use Unicon" rather than "It is available in Unicon but in 
Icon 9.4 we plan to have ...".  So I have to ask: what is the
future of Icon, as opposed to its descendents Idol, Jcon, Unicon?
The language has some really cunning features, (readable string
scanning, result or failure, generators...) so it would be a shame
if development just stopped.  Or is it just that the Icon Project
are trying to avoid "feeping creaturism", as Larry Wall called it?

	Thank you,
	Hugh
	hgs@dmu.ac.uk



From icon-group-sender  Thu Mar 25 08:37:54 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id IAA18367
	for icon-group-addresses; Thu, 25 Mar 1999 08:35:23 -0700 (MST)
Message-Id: <199903251535.IAA18367@baskerville.CS.Arizona.EDU>
To: icon-group@optima.CS.Arizona.EDU
Date: Wed, 24 Mar 1999 21:03:51 -0400
From: "Gerardo H Calderon" <snooper@coqui.net>
Subject: Icon Inquiries
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Hello again. Here are some of the questions I have to answer for my
Programming Languages project about Icon. If any of you can help me out,
I'll be very grateful.

Name of the language.
When was it created?
Who created it?
Where was it created? A university or private company?
Paradigm to which it belongs.
>From which language did it inherit characteristics? (Mention all languages
of which it inherited any characteristics from.)
Is it compatible with another language?
Does a formal definition of the language exist? (i.e. IEEE, ANSI, ISO, etc.)
What was the purpose for its creation?
Did it accomplish those purposes?
Is the language still in use?
Is it used for the purpose it was created for?
How many official revisions have there been?

Any help is fully appreciated. Thank you all.

Gerardo H. Calderon
Email: snooper@coqui.net




From icon-group-sender  Thu Mar 25 12:25:56 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id MAA13234
	for icon-group-addresses; Thu, 25 Mar 1999 12:24:58 -0700 (MST)
Message-Id: <199903251924.MAA13234@baskerville.CS.Arizona.EDU>
To: icon-group@optima.CS.Arizona.EDU
Date: Thu, 25 Mar 1999 16:58:33 +0100
From: "Rommert J. Casimir" <casimir@kub.nl>
Subject: Re: Icon Inquiries
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Gerardo H Calderon wrote:
> 
> Hello again. Here are some of the questions I have to answer for my
> Programming Languages project about Icon. If any of you can help me out,
> I'll be very grateful.

Many answers to your questions can be found in

 
                The Icon programming language / Ralph E. Griswold and
Madge T.
                Griswold
                Englewoods Cliffs, N.J. : Prentice Hall
 
                2nd ed
 
                Prentice-Hall software series


-- 
--------------------------------------------
Rommert J. Casimir
Tilburg University, Room B435
P.O. Box 90153
5000LE Tilburg, The Netherlands
tel 31 13 4662016, fax 31 13 4663377
http://cwis.kub.nl/~few/few/BIKA/rc_home.htm
---------------------------------------------


From icon-group-sender  Thu Mar 25 12:26:33 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id MAA13392
	for icon-group-addresses; Thu, 25 Mar 1999 12:26:20 -0700 (MST)
Message-Id: <199903251926.MAA13392@baskerville.CS.Arizona.EDU>
From: Bob Ardler <ardler@argonet.co.uk>
To: icon-group@optima.CS.Arizona.EDU
Date: Thu, 25 Mar 1999 18:30:11 +0000 (GMT)
Subject: Re: Icon Inquiries
User-Agent: Pluto/1.11f (RISC-OS/3.6)
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Gerardo H Calderon:
>questions I have to answer for my Programming Languages project
>about Icon

Quote from a Griswold posting re Icon:-
>Frequently Asked Questions About The Icon Programming Language
>Last updated October 12, 1998.

>The FAQ answers questions about the Icon programming language,
>ranging from what it is to how you can get it. The master copy of
>this FAQ is the Web page http://www.cs.arizona.edu/icon/faq.htm.
>Other on-line documentation is available via the main Icon page at
>http://www.cs.arizona.edu/icon/

-- 

Bob Ardler,  ardler@argonet.co.uk



From icon-group-sender  Fri Mar 26 07:27:36 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id HAA22468
	for icon-group-addresses; Fri, 26 Mar 1999 07:26:44 -0700 (MST)
Message-Id: <199903261426.HAA22468@baskerville.CS.Arizona.EDU>
X-Authentication-Warning: helios.dmu.ac.uk: hgs owned process doing -bs
Date: Fri, 26 Mar 1999 10:09:03 +0000 (GMT)
From: Hugh Sasse <hgs@dmu.ac.uk>
To: Icon Group <icon-group@optima.CS.Arizona.EDU>
Subject: Installation step 8...
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

With reference to :
http://www.cs.arizona.edu/icon/docs/ipd243.htm
for the last step (8) in the Install, (which I have not needed till now
because I have been using Icon where I built it), I would like
to put the files in these positions:

icont	/usr/local/bin/icont
iconx	/usr/local/bin/iconx
Icon.1  /usr/local/man/man1
ilib/   /usr/local/lib/icon/
[plus others??]

but as far as I can see Setup will not do this -- thee is no flexibility
in where things go, patchstr doesn't seem to tell Icont where the
libraries (ilib) are -- at least I didn't see it in the above doc.

Could someone tell me how to do this correctly?  I cannot continue
to use it from ~hgs/icon because I will be NFS mounting ~hgs for
Sparc and for Alpha, and they both need different binaries.  They
will both have different /usr/local partitions.  Maybe someone has
written install_thyself.icn to accomplish this? :-)

Thank you,
Hugh
hgs@dmu.ac.uk



From icon-group-sender  Fri Mar 26 12:38:25 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id MAA04189
	for icon-group-addresses; Fri, 26 Mar 1999 12:35:38 -0700 (MST)
Message-Id: <199903261935.MAA04189@baskerville.CS.Arizona.EDU>
Date: Fri, 26 Mar 1999 08:39:53 -0700 (MST)
From: Gregg Townsend <gmt@baskerville.CS.Arizona.EDU>
To: hgs@dmu.ac.uk, icon-group@optima.CS.Arizona.EDU
Subject: Re: Installation step 8...
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

    Date: Fri, 26 Mar 1999 10:09:03 +0000 (GMT)
    From: Hugh Sasse <hgs@dmu.ac.uk>
    
    I would like to put the files in these positions:
    
    icont	/usr/local/bin/icont
    iconx	/usr/local/bin/iconx
    Icon.1  /usr/local/man/man1
    ilib/   /usr/local/lib/icon/
    [plus others??]

You'd run patchstr like this:
	patchstr /usr/local/bin/icont /usr/local/bin/iconx

Icont has no embedded knowledged about library locations; it gets this
from environment variables.  You would typically put something like
	setenv IPATH /usr/local/lib/icon
	setenv LPATH /usr/local/lib/icon
in your shell startup file (.cshrc or .profile or whatever).

   ---------------------------------------------------------------------------
   Gregg Townsend              Gould-Simpson Building   gmt@cs.arizona.edu
   Staff Scientist             1040 E. 4th St.          32 13 45N  110 57 16W
   Dept. of Computer Science   PO Box 210077            tel: +1 520 621 4325
   The University of Arizona   Tucson, AZ  85721-0077   fax: +1 520 621 4246


From icon-group-sender  Fri Mar 26 12:39:44 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id MAA04286
	for icon-group-addresses; Fri, 26 Mar 1999 12:39:33 -0700 (MST)
Message-Id: <199903261939.MAA04286@baskerville.CS.Arizona.EDU>
X-Authentication-Warning: helios.dmu.ac.uk: hgs owned process doing -bs
Date: Fri, 26 Mar 1999 17:12:55 +0000 (GMT)
From: Hugh Sasse <hgs@dmu.ac.uk>
To: Gregg Townsend <gmt@baskerville.CS.Arizona.EDU>
cc: icon-group@optima.CS.Arizona.EDU
Subject: Re: Installation step 8...
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

On Fri, 26 Mar 1999, Gregg Townsend wrote:

>     Date: Fri, 26 Mar 1999 10:09:03 +0000 (GMT)
>     From: Hugh Sasse <hgs@dmu.ac.uk>
>     
>     I would like to put the files in these positions:
>     
	[...]
> 
> You'd run patchstr like this:
> 	patchstr /usr/local/bin/icont /usr/local/bin/iconx
> 
> Icont has no embedded knowledged about library locations; it gets this
> from environment variables.  You would typically put something like
> 	setenv IPATH /usr/local/lib/icon
> 	setenv LPATH /usr/local/lib/icon
> in your shell startup file (.cshrc or .profile or whatever).
> 
>    ---------------------------------------------------------------------------
>    Gregg Townsend              Gould-Simpson Building   gmt@cs.arizona.edu

	Thank you.
	Hugh
	hgs@dmu.ac.uk



From icon-group-sender  Tue Mar 30 15:30:02 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id PAA13864
	for icon-group-addresses; Tue, 30 Mar 1999 15:25:42 -0700 (MST)
Message-Id: <199903302225.PAA13864@baskerville.CS.Arizona.EDU>
Date: Tue, 30 Mar 1999 11:02:54 -0600
From: Steve Graham <graham@cowboy.biomed.com>
X-Accept-Language: en
To: icon-group@optima.CS.Arizona.EDU
Subject: Problem reading binary file
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Hello.  

I am using MS-DOS Icon 9.1 to read a binary file.  After reading 6
characters, the program apparently quits reading.  I have attached the
relevant portion of the code:

-------------------------- Code ------------------------------
procedure main(args)
#
  cnt := 0
  in  := open(args[1],"r") | stop("Unable to open ",args[1])
  out := open(args[2],"w") | stop("Unable to open ",args[2])
  while character := reads(in,1) do {
     write(character," -->", ord(character))
     cnt := cnt+1
    }
  close(in,out)
  write(cnt," characters transferred")
end
-----------------------End of Code-------------------------

I invoke the program with:

word r99-0059.doc r99-0059.txt


I get output of (some unprintable characters are represent by _):

- -->208
- -->207
 -->17
_ -->224
í -->161
_ -->177
6 characters transferred


Any ideas?

Thanks in advance.


-- Steve


______________________________________________________

Steve Graham
Laboratory Corporation of America (LabCorp)                
12160 Abrams Road, Suite 601
Dallas, Texas 75243

E-mail: graham@cowboy.biomed.com
Phone:  972.643.6124  (direct to desk)
        972.437.5255 ext 5224
FAX:    972.454.1040
_______________________________________________________


From icon-group-sender  Tue Mar 30 15:32:46 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id PAA14056
	for icon-group-addresses; Tue, 30 Mar 1999 15:32:39 -0700 (MST)
Message-Id: <199903302232.PAA14056@baskerville.CS.Arizona.EDU>
Date: Tue, 30 Mar 1999 16:04:23 -0600
From: Steve Graham <graham@cowboy.biomed.com>
X-Accept-Language: en
To: Icon Mailing List <icon-group@optima.CS.Arizona.EDU>
Subject: [Fwd: Problem reading binary file]
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Hello again.

Turns out that the problem with this was that the 7th character was an
ASCII 26 (^Z) and Icon would not read any further than that.  I created
a 2nd file with a ^Z in a different position and got a similar result. 
This is despite the fact that there were over 25,000 characters
following the Control Z.

So, my question is:  Why does Icon stop reading when it hits a ^Z?  I
can kind of understand this if I were reading a text file; but I was
specifically using reads() because this is a binary file?  Should not
reads() support all 256 ASCII codes and only stop reading when there are
REALLY no more characters to read?  I ended up using another language,
which was significantly harder for me, to read the binary characters.

In Icon's defense, I notice that when I TYPE the above-mentioned 2nd
file, I only see the characters occurring prior to the ^Z.  And I do not
want it to appear that I am questioning Icon's usefulness.  It is
normally one of the 1st languages I choose for utilities of this type.

Comments?


Steve Graham

===

Steve Graham wrote:
> 
> Hello.
> 
> I am using MS-DOS Icon 9.1 to read a binary file.  After reading 6
> characters, the program apparently quits reading.  I have attached the
> relevant portion of the code:
> 
> -------------------------- Code ------------------------------
> procedure main(args)
> #
>   cnt := 0
>   in  := open(args[1],"r") | stop("Unable to open ",args[1])
>   out := open(args[2],"w") | stop("Unable to open ",args[2])
>   while character := reads(in,1) do {
>      write(character," -->", ord(character))
>      cnt := cnt+1
>     }
>   close(in,out)
>   write(cnt," characters transferred")
> end
> -----------------------End of Code-------------------------
> 
> I invoke the program with:
> 
> word r99-0059.doc r99-0059.txt
> 
> I get output of (some unprintable characters are represent by _):
> 
> - -->208
> - -->207
>  -->17
> _ -->224
> í -->161
> _ -->177
> 6 characters transferred
> 
> Any ideas?
> 
> Thanks in advance.
> 
> -- Steve
> 
> ______________________________________________________
> 
> Steve Graham
> Laboratory Corporation of America (LabCorp)
> 12160 Abrams Road, Suite 601
> Dallas, Texas 75243
> 
> E-mail: graham@cowboy.biomed.com
> Phone:  972.643.6124  (direct to desk)
>         972.437.5255 ext 5224
> FAX:    972.454.1040
> _______________________________________________________


From icon-group-sender  Wed Mar 31 09:23:13 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id JAA09145
	for icon-group-addresses; Wed, 31 Mar 1999 09:21:52 -0700 (MST)
Message-Id: <199903311621.JAA09145@baskerville.CS.Arizona.EDU>
Date: Tue, 30 Mar 1999 17:27:24 -0600
From: Steve Graham <graham@cowboy.biomed.com>
X-Accept-Language: en
To: Ralph Griswold <ralph@optima.CS.Arizona.EDU>
Subject: Re: [Fwd: Problem reading binary file]
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Ralph,

   This makes sense now.  I modified my program based on this and
earlier responses and I was able to read all the characters.

   Thanks a lot.


Steve Graham


P.S.  My sister is involved in computers and research libraries and she
was saying that computers are now being applied in new areas in
education, such as the humanities.  It seems to me that I read that Icon
had been put to use in this area, but I don't remember any details.  Can
you elaborate on this or point me in the right direction?  Thanks again.

===

Ralph Griswold wrote:
> 
> Icon doesn't stop reading on a ^Z.  The DOS/WIndows I/O system
> does.

-- 


______________________________________________________

Steve Graham
Laboratory Corporation of America (LabCorp)                
12160 Abrams Road, Suite 601
Dallas, Texas 75243

E-mail: graham@cowboy.biomed.com
Phone:  972.643.6124  (direct to desk)
        972.437.5255 ext 5224
FAX:    972.454.1040
_______________________________________________________


From icon-group-sender  Wed Mar 31 09:26:05 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id JAA09302
	for icon-group-addresses; Wed, 31 Mar 1999 09:25:57 -0700 (MST)
Message-Id: <199903311625.JAA09302@baskerville.CS.Arizona.EDU>
Date: Tue, 30 Mar 1999 17:28:55 -0600
From: Steve Graham <graham@cowboy.biomed.com>
X-Accept-Language: en
To: Alan Saichek <abs@reba.engr.sgi.com>,
        Icon Mailing List <icon-group@optima.CS.Arizona.EDU>
Subject: Re: [Fwd: Problem reading binary file]
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Alan,

   I tried it and that did the trick.  Thanks a lot.


Steve Graham

===

Alan Saichek wrote:
> 
> 
>         Hi Steve,
> 
>                 I haven't tried this, but on the surface...
> 
>                 Ctrl-Z is end-of-file in DOS.
> 
>         Check out Chap 11 Input and Output (2nd edition)
>         Text Files and Binary Files
> 
>         which mentions translated mode. Use "u" qualifier for
>         untranslated mode.
> 
> > >   in  := open(args[1],"ru") | stop("Unable to open ",args[1])
> > >   out := open(args[2],"wu") | stop("Unable to open ",args[2])
> 
>         Regards, Alan
> 
> >
> > Hello again.
> >
> > Turns out that the problem with this was that the 7th character was an
> > ASCII 26 (^Z) and Icon would not read any further than that.  I created
> > a 2nd file with a ^Z in a different position and got a similar result.
> > This is despite the fact that there were over 25,000 characters
> > following the Control Z.
> >
> > So, my question is:  Why does Icon stop reading when it hits a ^Z?  I
> > can kind of understand this if I were reading a text file; but I was
> > specifically using reads() because this is a binary file?  Should not
> > reads() support all 256 ASCII codes and only stop reading when there are
> > REALLY no more characters to read?  I ended up using another language,
> > which was significantly harder for me, to read the binary characters.
> >
> > In Icon's defense, I notice that when I TYPE the above-mentioned 2nd
> > file, I only see the characters occurring prior to the ^Z.  And I do not
> > want it to appear that I am questioning Icon's usefulness.  It is
> > normally one of the 1st languages I choose for utilities of this type.
> >
> > Comments?
> >
> >
> > Steve Graham
> >
> > ===
> >
> > Steve Graham wrote:
> > >
> > > Hello.
> > >
> > > I am using MS-DOS Icon 9.1 to read a binary file.  After reading 6
> > > characters, the program apparently quits reading.  I have attached the
> > > relevant portion of the code:
> > >
> > > -------------------------- Code ------------------------------
> > > procedure main(args)
> > > #
> > >   cnt := 0
> > >   in  := open(args[1],"r") | stop("Unable to open ",args[1])
> > >   out := open(args[2],"w") | stop("Unable to open ",args[2])
> > >   while character := reads(in,1) do {
> > >      write(character," -->", ord(character))
> > >      cnt := cnt+1
> > >     }
> > >   close(in,out)
> > >   write(cnt," characters transferred")
> > > end
> > > -----------------------End of Code-------------------------
> > >
> > > I invoke the program with:
> > >
> > > word r99-0059.doc r99-0059.txt
> > >
> > > I get output of (some unprintable characters are represent by _):
> > >
> > > - -->208
> > > - -->207
> > >  -->17
> > > _ -->224
> > > í -->161
> > > _ -->177
> > > 6 characters transferred
> > >
> > > Any ideas?
> > >
> > > Thanks in advance.
> > >
> > > -- Steve
> > >
> > > ______________________________________________________
> > >
> > > Steve Graham
> > > Laboratory Corporation of America (LabCorp)
> > > 12160 Abrams Road, Suite 601
> > > Dallas, Texas 75243
> > >
> > > E-mail: graham@cowboy.biomed.com
> > > Phone:  972.643.6124  (direct to desk)
> > >         972.437.5255 ext 5224
> > > FAX:    972.454.1040
> > > _______________________________________________________
> >
> 
> --
> Alan B. Saichek                                         abs@engr.sgi.com
> Member Technical Staff - QA Team Lead                   (650) 933-2507
> Advanced Entertainment Systems, Video Compute R&D       07L-808

-- 


______________________________________________________

Steve Graham
Laboratory Corporation of America (LabCorp)                
12160 Abrams Road, Suite 601
Dallas, Texas 75243

E-mail: graham@cowboy.biomed.com
Phone:  972.643.6124  (direct to desk)
        972.437.5255 ext 5224
FAX:    972.454.1040
_______________________________________________________


From icon-group-sender  Wed Mar 31 09:26:35 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id JAA09337
	for icon-group-addresses; Wed, 31 Mar 1999 09:26:27 -0700 (MST)
Message-Id: <199903311626.JAA09337@baskerville.CS.Arizona.EDU>
Date: Tue, 30 Mar 1999 17:29:50 -0600
From: Steve Graham <graham@cowboy.biomed.com>
X-Accept-Language: en
To: Ralph Griswold <ralph@optima.CS.Arizona.EDU>
Subject: Re: Problem reading binary file
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Ralph,

   I tried that and it worked great.  Thanks a lot.


Steve Graham

===

Ralph Griswold wrote:
> 
> If you're working under DOS or windows, open the file in untranslate
> mode.  Otherwise the I/O system interprets ^Z and an end of file.
> 
> For example,
> 
>         open(..., "ur")

-- 


______________________________________________________

Steve Graham
Laboratory Corporation of America (LabCorp)                
12160 Abrams Road, Suite 601
Dallas, Texas 75243

E-mail: graham@cowboy.biomed.com
Phone:  972.643.6124  (direct to desk)
        972.437.5255 ext 5224
FAX:    972.454.1040
_______________________________________________________


From icon-group-sender  Wed Mar 31 09:26:55 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id JAA09376
	for icon-group-addresses; Wed, 31 Mar 1999 09:26:48 -0700 (MST)
Message-Id: <199903311626.JAA09376@baskerville.CS.Arizona.EDU>
Date: Tue, 30 Mar 1999 22:47:27 -0500
From: Phillip Lee Thomas <teruthom@ix.netcom.com>
X-Accept-Language: en
To: Steve Graham <graham@cowboy.biomed.com>, icon-group@optima.CS.Arizona.EDU
Subject: Re: Problem reading binary file
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Steve,

Under DOS, the files need to be opened in "untranslated" mode:


  in  := open(args[1],"ru") | stop("Unable to open ",args[1])
  out := open(args[2],"wu") | stop("Unable to open ",args[2])

This stops the translation of <CR><LF> and other characters with special
meaning in DOS.

Phillip

Steve Graham wrote:

> Hello.
>
> I am using MS-DOS Icon 9.1 to read a binary file.  After reading 6
> characters, the program apparently quits reading.  I have attached the
> relevant portion of the code:
>
> -------------------------- Code ------------------------------
> procedure main(args)
> #
>   cnt := 0
>   in  := open(args[1],"r") | stop("Unable to open ",args[1])
>   out := open(args[2],"w") | stop("Unable to open ",args[2])
>   while character := reads(in,1) do {
>      write(character," -->", ord(character))
>      cnt := cnt+1
>     }
>   close(in,out)
>   write(cnt," characters transferred")
> end
> -----------------------End of Code-------------------------
>
> I invoke the program with:
>
> word r99-0059.doc r99-0059.txt
>
> I get output of (some unprintable characters are represent by _):
>
> - -->208
> - -->207
>  -->17
> _ -->224
> í -->161
> _ -->177
> 6 characters transferred
>
> Any ideas?
>
> Thanks in advance.
>
> -- Steve
>
> ______________________________________________________
>
> Steve Graham
> Laboratory Corporation of America (LabCorp)
> 12160 Abrams Road, Suite 601
> Dallas, Texas 75243
>
> E-mail: graham@cowboy.biomed.com
> Phone:  972.643.6124  (direct to desk)
>         972.437.5255 ext 5224
> FAX:    972.454.1040
> _______________________________________________________



From icon-group-sender  Wed Mar 31 09:27:26 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id JAA09430
	for icon-group-addresses; Wed, 31 Mar 1999 09:27:18 -0700 (MST)
Message-Id: <199903311627.JAA09430@baskerville.CS.Arizona.EDU>
Date: Wed, 31 Mar 1999 00:13:29 -0400 (AST)
From: Larry Bezeau <bezeau@unb.ca>
X-Sender: bezeau@sol.sun.csd.unb.ca
To: Icon list <icon-group@optima.CS.Arizona.EDU>
Subject: Re: Problem reading binary file
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Steve,

>I am using MS-DOS Icon 9.1 to read a binary file.  After reading 6
>characters, the program apparently quits reading.  I have attached the
>relevant portion of the code:
>
        I encountered this problem myself sometime ago.  To overcome
it you must open the file in unstranslated mode as shown below.
Note the "u" in the open statement.  A "t" in the same position means
translated but this is the default if neither is specified.

  in  := open(args[1],"ru") | stop("Unable to open ",args[1])

This will allow your program to read the file character by character
without having to worry about whether the character is an end of file
character or a line-ending character.  In general, if you are reading
a file with reads, it should be opened in untranslated mode and if
you are reading it with read it should be opened in translated mode.
To put it another way, if the file is being used as a text file it
should be opened in translated mode but if it is being used as a
binary file it should be opened in untranslated mode.

                                    Larry     Bezeau@UNB.Ca

>-------------------------- Code ------------------------------
>procedure main(args)
>#
>  cnt := 0
>  out := open(args[2],"w") | stop("Unable to open ",args[2])
>  while character := reads(in,1) do {
>     write(character," -->", ord(character))
>     cnt := cnt+1
>    }
>  close(in,out)
>  write(cnt," characters transferred")
>end
>-----------------------End of Code-------------------------
>
>I invoke the program with:
>
>word r99-0059.doc r99-0059.txt
>
>
>I get output of (some unprintable characters are represent by _):
>
>- -->208
>- -->207
> -->17
>_ -->224
>í -->161
>_ -->177
>6 characters transferred
>
>
>Any ideas?
>
>Thanks in advance.
>
>
>-- Steve
>
>
>______________________________________________________
>
>Steve Graham
>Laboratory Corporation of America (LabCorp)
>12160 Abrams Road, Suite 601
>Dallas, Texas 75243
>
>E-mail: graham@cowboy.biomed.com
>Phone:  972.643.6124  (direct to desk)
>        972.437.5255 ext 5224
>FAX:    972.454.1040
>_______________________________________________________
>
>From icon-group-sender@baskerville.CS.Arizona.EDU  Tue Mar 30 18:42:50 1999
>Date: Tue, 30 Mar 1999 16:04:23 -0600
>From: Steve Graham <graham@cowboy.biomed.com>
>Subject: [Fwd: Problem reading binary file]
>Status: O
>X-Status:
>X-Keywords:
>X-UID: 2
>
>Hello again.
>
>Turns out that the problem with this was that the 7th character was an
>ASCII 26 (^Z) and Icon would not read any further than that.  I created
>a 2nd file with a ^Z in a different position and got a similar result.
>This is despite the fact that there were over 25,000 characters
>following the Control Z.
>
>So, my question is:  Why does Icon stop reading when it hits a ^Z?  I
>can kind of understand this if I were reading a text file; but I was
>specifically using reads() because this is a binary file?  Should not
>reads() support all 256 ASCII codes and only stop reading when there are
>REALLY no more characters to read?  I ended up using another language,
>which was significantly harder for me, to read the binary characters.
>
>In Icon's defense, I notice that when I TYPE the above-mentioned 2nd
>file, I only see the characters occurring prior to the ^Z.  And I do not
>want it to appear that I am questioning Icon's usefulness.  It is
>normally one of the 1st languages I choose for utilities of this type.
>
>Comments?
>
>
>Steve Graham
>
>===
>
>Steve Graham wrote:
>>
>> Hello.
>>
>> I am using MS-DOS Icon 9.1 to read a binary file.  After reading 6
>> characters, the program apparently quits reading.  I have attached the
>> relevant portion of the code:
>>
>> -------------------------- Code ------------------------------
>> procedure main(args)
>> #
>>   cnt := 0
>>   in  := open(args[1],"r") | stop("Unable to open ",args[1])
>>   out := open(args[2],"w") | stop("Unable to open ",args[2])
>>   while character := reads(in,1) do {
>>      write(character," -->", ord(character))
>>      cnt := cnt+1
>>     }
>>   close(in,out)
>>   write(cnt," characters transferred")
>> end
>> -----------------------End of Code-------------------------
>>
>> I invoke the program with:
>>
>> word r99-0059.doc r99-0059.txt
>>
>> I get output of (some unprintable characters are represent by _):
>>
>> - -->208
>> - -->207
>>  -->17
>> _ -->224
>> í -->161
>> _ -->177
>> 6 characters transferred
>>
>> Any ideas?
>>
>> Thanks in advance.
>>
>> -- Steve
>>
>> ______________________________________________________
>>
>> Steve Graham
>> Laboratory Corporation of America (LabCorp)
>> 12160 Abrams Road, Suite 601
>> Dallas, Texas 75243
>>
>> E-mail: graham@cowboy.biomed.com
>> Phone:  972.643.6124  (direct to desk)
>>         972.437.5255 ext 5224
>> FAX:    972.454.1040
>> _______________________________________________________
>



From icon-group-sender  Wed Mar 31 09:27:48 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id JAA09479
	for icon-group-addresses; Wed, 31 Mar 1999 09:27:39 -0700 (MST)
Message-Id: <199903311627.JAA09479@baskerville.CS.Arizona.EDU>
Date: Wed, 31 Mar 1999 16:21:53 +0200 (CEST)
From: Bohdan Rau <ethanak@bigfoot.com>
X-Sender: ethanak@guru
To: icon-group@optima.CS.Arizona.EDU
Subject: Re: Problem reading binary file
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

On Tue, 30 Mar 1999, Steve Graham wrote:

> Hello.  

Hello!

>   cnt := 0
>   in  := open(args[1],"r") | stop("Unable to open ",args[1])
                         ^^
In MSDOS use "rb" for binary files!

ethanak



From icon-group-sender  Wed Mar 31 09:28:10 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id JAA09520
	for icon-group-addresses; Wed, 31 Mar 1999 09:28:00 -0700 (MST)
Message-Id: <199903311628.JAA09520@baskerville.CS.Arizona.EDU>
Date: Wed, 31 Mar 1999 09:07:15 -0700 (MST)
From: Gregg Townsend <gmt@baskerville.CS.Arizona.EDU>
To: graham@cowboy.biomed.com, icon-group@optima.CS.Arizona.EDU
Subject: Re: Problem reading binary file
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

    From: Steve Graham <graham@cowboy.biomed.com>
    Date: Tue, 30 Mar 1999 11:02:54 -0600
    
    I am using MS-DOS Icon 9.1 to read a binary file.  After reading 6
    characters, the program apparently quits reading...
      
You might try opening the file in untranslated mode:  open(filename, "ru")

   ---------------------------------------------------------------------------
   Gregg Townsend              Gould-Simpson Building   gmt@cs.arizona.edu
   Staff Scientist             1040 E. 4th St.          32 13 45N  110 57 16W
   Dept. of Computer Science   PO Box 210077            tel: +1 520 621 4325
   The University of Arizona   Tucson, AZ  85721-0077   fax: +1 520 621 4246


From icon-group-sender  Wed Mar 31 09:28:28 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id JAA09577
	for icon-group-addresses; Wed, 31 Mar 1999 09:28:19 -0700 (MST)
Message-Id: <199903311628.JAA09577@baskerville.CS.Arizona.EDU>
From: Lloyd Uhler <luhler@excel.com>
To: "'icon-group@cs.arizona.edu'" <icon-group@optima.CS.Arizona.EDU>
Subject: Selecting and moving Graphics Objects
Date: Wed, 31 Mar 1999 10:17:01 -0600
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

What modules (subroutines) in the Graphics Library (or in ICON) support
the selection of an object with the mouse and then moving that object?

In VIB, you  select somthing like a "slider" and move it to the Design
Window.
I was curious on how this is done in general.

Thanks.


From icon-group-sender  Wed Mar 31 09:28:46 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id JAA09634
	for icon-group-addresses; Wed, 31 Mar 1999 09:28:38 -0700 (MST)
Message-Id: <199903311628.JAA09634@baskerville.CS.Arizona.EDU>
Date: Wed, 31 Mar 1999 10:18:31 -0600
From: Steve Graham <graham@cowboy.biomed.com>
X-Accept-Language: en
To: Icon Mailing List <icon-group@optima.CS.Arizona.EDU>
Subject: Re: Problem Reading Binary File
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

To all Iconists:

   I would like to take a moment to publicly thank the group for the
rapid and voluminous response to my above question.  I received e-mail
from 7 individuals, which proved to be just what I needed.


-- Steve Graham


______________________________________________________

Steve Graham
Laboratory Corporation of America (LabCorp)                
12160 Abrams Road, Suite 601
Dallas, Texas 75243

E-mail: graham@cowboy.biomed.com
Phone:  972.643.6124  (direct to desk)
        972.437.5255 ext 5224
FAX:    972.454.1040
_______________________________________________________


From icon-group-sender  Wed Mar 31 13:26:46 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id NAA20561
	for icon-group-addresses; Wed, 31 Mar 1999 13:25:36 -0700 (MST)
Message-Id: <199903312025.NAA20561@baskerville.CS.Arizona.EDU>
Date: Wed, 31 Mar 1999 09:40:06 -0700 (MST)
From: Gregg Townsend <gmt@baskerville.CS.Arizona.EDU>
To: icon-group@optima.CS.Arizona.EDU, luhler@excel.com
Subject: Re: Selecting and moving Graphics Objects
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

    From: Lloyd Uhler <luhler@excel.com>
    Subject: Selecting and moving Graphics Objects
    
    What modules (subroutines) in the Graphics Library (or in ICON) support
    the selection of an object with the mouse and then moving that object?

For dragging, see Drag() from the Icon program library (ipl/gprocs/drag.icn).
This is what VIB uses.  For selection, VIB just uses a simple loop that
compares the mouse position with the bounds of the objects it knows.

   ---------------------------------------------------------------------------
   Gregg Townsend              Gould-Simpson Building   gmt@cs.arizona.edu
   Staff Scientist             1040 E. 4th St.          32 13 45N  110 57 16W
   Dept. of Computer Science   PO Box 210077            tel: +1 520 621 4325
   The University of Arizona   Tucson, AZ  85721-0077   fax: +1 520 621 4246


From icon-group-sender  Wed Mar 31 13:27:37 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id NAA20624
	for icon-group-addresses; Wed, 31 Mar 1999 13:27:29 -0700 (MST)
Message-Id: <199903312027.NAA20624@baskerville.CS.Arizona.EDU>
Date: Wed, 31 Mar 1999 10:47:00 -0600
From: Steve Graham <graham@cowboy.biomed.com>
X-Accept-Language: en
To: Icon Mailing List <icon-group@optima.CS.Arizona.EDU>
Subject: Icon in the Humanities
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

My sister is involved with research libraries and recently mentioned the
use of computers in fields such as the Humanities.  If there is any
related work in Icon, I would like to pass that on to her.  Does anyone
know of any such endeavors using Icon?

Thanks in advance.


-- Steve Graham


______________________________________________________

Steve Graham
Laboratory Corporation of America (LabCorp)                
12160 Abrams Road, Suite 601
Dallas, Texas 75243

E-mail: graham@cowboy.biomed.com
Phone:  972.643.6124  (direct to desk)
        972.437.5255 ext 5224
FAX:    972.454.1040
_______________________________________________________


From icon-group-sender  Thu Apr  1 12:59:41 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id MAA00234
	for icon-group-addresses; Thu, 1 Apr 1999 12:57:27 -0700 (MST)
Message-Id: <199904011957.MAA00234@baskerville.CS.Arizona.EDU>
From: Art Eschenlauer <art.eschenlauer@sufsys.com>
To: "'icon-group@cs.arizona.edu'" <icon-group@optima.CS.Arizona.EDU>
Subject: has anyone written 'make' utility in Icon
Date: Thu, 1 Apr 1999 11:30:50 -0600 
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

I am wondering if anyone has written 'make' using Icon. 
Not wishing to reinvent the wheel, I checked the Icon Programming Library,
but saw nothing there that seemed to be like 'make'.
Yes, I understand the operating-system-specific implications, but I have
always been able to get a lot of mileage out of the system(s) call.

> Art Eschenlauer, Systems Analyst
> Sufficient Systems, Inc.
> 2860 Patton Road
> Roseville, MN  55113
> Phone: (651) 638-9190, ext. 29. Fax: (651) 638-9290
> eMail: Art.Eschenlauer@sufsys.com
> 


From icon-group-sender  Mon Apr  5 08:59:07 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id IAA20054
	for icon-group-addresses; Mon, 5 Apr 1999 08:47:02 -0700 (MST)
Message-Id: <199904051547.IAA20054@baskerville.CS.Arizona.EDU>
From: "Mark Evans" <evans@gte.net>
To: "Icon List" <icon-group@optima.CS.Arizona.EDU>
Subject: Ccon and Icon-based Compilers
Date: Sat, 3 Apr 1999 10:34:49 -0600
X-Priority: 3 (Normal)
X-MSMail-Priority: Normal
Importance: Normal
X-MimeOLE: Produced By Microsoft MimeOLE V4.72.3155.0
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO


The Jcon effort involved a Java translator plus a runtime system in Java.  My
question is whether anyone has ideas about the same concept for the C/C++
language.  Let's call it "Ccon."  That is, a runtime system (library/classes),
plus a translator to turn Icon into C/C++ targeted for that runtime system.

I remember something called "iconc" but it seems to have fallen into disfavor,
is not being upgraded, all attention is now on Unicon, etc., etc.

Another idea for which I would like to solicit opinions is writing a
machine-code compiler in the Icon language directly.  Good?  Bad?  Ugly?  What
about using a GNU compiler and then Icon code as the front end and/or back end.
Any advantages?

The kind of comments I don't need are people telling me that I don't need C
code, I can use Icon as it is, why don't I get with it, etc.  If I didn't need C
code I wouldn't bother with this note.

Regards,

Mark Evans




Believe Evolution?  Save this block as FAITH.HTM and open with your browser.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
<HTML><HEAD></HEAD><script language="JavaScript">function doFaithJump(){ site =
"http://www." + document.FF01.FL01.value; W = open(site)}</script><FORM NAME =
"FF01"><P><SELECT NAME="FL01"><OPTION SELECTED
VALUE="netgoal.com/lacf.nsf/general/debate+highlights">Los Alamos Origins
Debate<OPTION VALUE="parentcompany.com/handy_dandy/hder3.htm">Evolution
Refuter<OPTION VALUE="christiananswers.net/creation/home.html">Christian
Answers<OPTION VALUE="gospelcom.net/rbc/rtb/8rsn/">Ten Reasons<OPTION
VALUE="xenos.org/classes/papers/doubt.htm">Still Doubtful?<OPTION
VALUE="apologeticsinfo.org/bibliographies/jesusresurrection.html">Read All About
It<OPTION VALUE="persecutedchurch.org/home.htm">The Price is Still
Blood</SELECT><INPUT TYPE="button" NAME="Go" VALUE="Go"
onClick="doFaithJump()"></P></FORM></BODY></HTML>



From icon-group-sender  Mon Apr  5 09:23:07 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id JAA21689
	for icon-group-addresses; Mon, 5 Apr 1999 09:22:52 -0700 (MST)
Message-Id: <199904051622.JAA21689@baskerville.CS.Arizona.EDU>
X-Authentication-Warning: pluto.mscc.huji.ac.il: mslamm owned process doing -bs
Date: Mon, 5 Apr 1999 15:13:05 +0300 (WET)
From: Ehud Lamm <mslamm@mscc.huji.ac.il>
To: icon-group@optima.CS.Arizona.EDU
Subject: How programs fail
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Had a funny problem today, which may give a lesson to us all.

I had some files that needed to be hacked a bit (a chunk was to be removed
from each of them), and I decided to use Icon. 

I am using Win95, but my Icon version is quite old - 8.10 I think.

It took me a few minutes to find filter.icn in the IPL, change it to my
needs and test to see it works. All was fine.

Then I tried to run it on my files, but it didn't work. No error messages,
just that the output files didn't get created, and the input files, which
weren't supposed to change were suddenly empty.

It occured to me that Win95 allows long file names, so I changed my
directory names to one word shorter than 8 letters. However things still
didn't work.

I tried moving the files arround, to my Icon direcotry, so no path will be
used. No improvement. 

It tried &trace - didn't give a clue.

Only thing I found out was that the read() from the input file fails.
However, I didn't see any reason for this.

I almost decided to think of another solution.

And than it hit me...

The input files were named *.htm, and I was so happy I found a way to get
meaningful output names with little effort - just call them *.html. This
means my code simply appended an "l".

However this made the extension longer than 3 letters... Took me an hour
to notice. A minute to fix. Than it worked.

Aside from feeling stupid, I think it would have been nice if Icon would
have yeeld saying that the output file name I gave it was bad. I am not
sure if this can be done in any nice way, but it does show that software
must be maintained, and that the dictum "if it ain't broke don't fix it"
is wrong when applied to software. Other software will be upgraded and
will cause yours to fail.

It is also quite sad that instead of producing an error, when the open was
done - the code continued to run and corrupted the input files. Even
sadder is the fact that I had no tools to help analyze the problem. My
Icon version is very lacking in this regard.

I seem to recall seeing that Ralph Griswold is teaching Software Eng. Well
I may use this story on my SE students...

   

Ehud Lamm     mslamm@pluto.mscc.huji.ac.il



From icon-group-sender  Mon Apr  5 12:24:02 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id MAA29776
	for icon-group-addresses; Mon, 5 Apr 1999 12:22:27 -0700 (MST)
Message-Id: <199904051922.MAA29776@baskerville.CS.Arizona.EDU>
From: Todd Proebsting <toddpro@microsoft.com>
To: "'Ehud Lamm'" <mslamm@mscc.huji.ac.il>, icon-group@optima.CS.Arizona.EDU
Subject: RE: How programs fail
Date: Mon, 5 Apr 1999 09:42:11 -0700 
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

I like Icon's model of what is failure and what is an error, although I,
too, have been nailed by a failing expression that in a particular context
is fatal to my application.  To find such, I use the following pattern:

	f := open(filename) | runerr(500, filename)

runerr(500) is for "Programmer-Specified Errors" and reports "program
malfunction", which is precisely what you want in this circumstance.
Getting the stack trace is useful.

Todd

-----Original Message-----
From: Ehud Lamm [mailto:mslamm@mscc.huji.ac.il]
Sent: Monday, April 05, 1999 5:13 AM
To: icon-group@optima.CS.Arizona.EDU
Subject: How programs fail


Had a funny problem today, which may give a lesson to us all.

I had some files that needed to be hacked a bit (a chunk was to be removed
from each of them), and I decided to use Icon. 

I am using Win95, but my Icon version is quite old - 8.10 I think.

It took me a few minutes to find filter.icn in the IPL, change it to my
needs and test to see it works. All was fine.

Then I tried to run it on my files, but it didn't work. No error messages,
just that the output files didn't get created, and the input files, which
weren't supposed to change were suddenly empty.

It occured to me that Win95 allows long file names, so I changed my
directory names to one word shorter than 8 letters. However things still
didn't work.

I tried moving the files arround, to my Icon direcotry, so no path will be
used. No improvement. 

It tried &trace - didn't give a clue.

Only thing I found out was that the read() from the input file fails.
However, I didn't see any reason for this.

I almost decided to think of another solution.

And than it hit me...

The input files were named *.htm, and I was so happy I found a way to get
meaningful output names with little effort - just call them *.html. This
means my code simply appended an "l".

However this made the extension longer than 3 letters... Took me an hour
to notice. A minute to fix. Than it worked.

Aside from feeling stupid, I think it would have been nice if Icon would
have yeeld saying that the output file name I gave it was bad. I am not
sure if this can be done in any nice way, but it does show that software
must be maintained, and that the dictum "if it ain't broke don't fix it"
is wrong when applied to software. Other software will be upgraded and
will cause yours to fail.

It is also quite sad that instead of producing an error, when the open was
done - the code continued to run and corrupted the input files. Even
sadder is the fact that I had no tools to help analyze the problem. My
Icon version is very lacking in this regard.

I seem to recall seeing that Ralph Griswold is teaching Software Eng. Well
I may use this story on my SE students...

   

Ehud Lamm     mslamm@pluto.mscc.huji.ac.il



From icon-group-sender  Mon Apr  5 13:20:34 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id NAA02010
	for icon-group-addresses; Mon, 5 Apr 1999 13:20:09 -0700 (MST)
Message-Id: <199904052020.NAA02010@baskerville.CS.Arizona.EDU>
From: "Mark Evans" <evans@gte.net>
To: "Icon List" <icon-group@optima.CS.Arizona.EDU>
Subject: RE: Ccon and Icon-based Compilers
Date: Mon, 5 Apr 1999 11:52:15 -0500
X-Priority: 3 (Normal)
X-MSMail-Priority: Normal
Importance: Normal
X-MimeOLE: Produced By Microsoft MimeOLE V4.72.3155.0
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Thanks, Todd!  Helpful comments.

Garbage collection is not a big deal to me, e.g.
http://reality.sgi.com/boehm_mti/gc.html

As for runtime support, it already exists in C.  The Icon interpreter is
implemented in C.  All I have to do is tweak the code that is there already.
(With Jcon, you did not have Java code for the runtime system; your case is not
the same as mine.)

I like co-expressions but detest their current implementation, which involves
custom assembly language.  Yuk!  I want to rewrite that stuff, whether with
threads or something simpler.

Type inferencing is fine if it helps performance, and if there is code already
out there.  Presumably I can get my hands on the source to "iconc" and determine
how it was done.

We can talk ourselves blue in the face about whether C will perform better than
native Icon.  To me that is not the point.  The point is, I want to extend the C
language.  I want Icon code that integrates cleanly into C/C++ projects under
Windows.  The reason I said that I wanted to avoid discussion of "why C?" is
that I already know my reasons, and they are valid.

Best regards,

Mark



-----Original Message-----
From: Todd Proebsting [mailto:toddpro@microsoft.com]
Sent: Monday, April 05, 1999 11:24 AM
To: 'Mark Evans'; Icon List
Subject: RE: Ccon and Icon-based Compilers


Jcon's translator is written in Icon.  It would be very easy to adapt it to
emit C++ (or assembly for your favorite machine).  In fact, the translator
was structured to be easy to retarget.  (I'm toying with the idea of
retargetting Jcon to emit Postscript---with only a limited run-time
system---just for kicks....)

The problem is not translating Icon, however, but providing the run-time
support for all of Icon's built-in types and functions.  This can be done
easily in C++, too, with enough time.  With the exception of the graphics
stuff, most of Jcon should translate pretty cleanly into C++.

Two issues that any Icon implementation must face are how to handle
co-expressions, and how to handle garbage collection.  Co-expressions are
easy if you have a threads package.  Garbage collection is a royal pain.
Jcon relied on Java's garbage collection, so GC in Jcon was free.  Unless
you use a conservative collector in a C++ implementation, I think GC would
be complete mess: a source of lots of complications, lots of performance
problems, and lots of hard-to-track bugs.

I should comment, however, on emitting machine code.  It's not going to help
performance much at all.  The vast majority of the time in an Icon program
is spent in the run-time system, not in compiled code.  In Jcon, its around
10% in compiled code, if I recall correctly.  (The old Icon compiler, iconc,
had significant performance gains not because it compiled to C rather than
being interpreted, but because it did type inferencing to avoid having to do
lots of type conversions and type tests.  This is an orthogonal issue to
what target or implementation languages are used.)

I'd be surprised if any implementation of Icon is going to be significantly
faster than the current interpreter unless the new implementation does
type-specific optimizations.

Todd

-----Original Message-----
From: Mark Evans [mailto:evans@gte.net]
Sent: Saturday, April 03, 1999 8:35 AM
To: Icon List
Subject: Ccon and Icon-based Compilers



The Jcon effort involved a Java translator plus a runtime system in Java.
My
question is whether anyone has ideas about the same concept for the C/C++
language.  Let's call it "Ccon."  That is, a runtime system
(library/classes),
plus a translator to turn Icon into C/C++ targeted for that runtime system.

I remember something called "iconc" but it seems to have fallen into
disfavor,
is not being upgraded, all attention is now on Unicon, etc., etc.

Another idea for which I would like to solicit opinions is writing a
machine-code compiler in the Icon language directly.  Good?  Bad?  Ugly?
What
about using a GNU compiler and then Icon code as the front end and/or back
end.
Any advantages?

The kind of comments I don't need are people telling me that I don't need C
code, I can use Icon as it is, why don't I get with it, etc.  If I didn't
need C
code I wouldn't bother with this note.

Regards,

Mark Evans




Believe Evolution?  Save this block as FAITH.HTM and open with your browser.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
<HTML><HEAD></HEAD><script language="JavaScript">function doFaithJump(){
site =
"http://www." + document.FF01.FL01.value; W = open(site)}</script><FORM NAME
=
"FF01"><P><SELECT NAME="FL01"><OPTION SELECTED
VALUE="netgoal.com/lacf.nsf/general/debate+highlights">Los Alamos Origins
Debate<OPTION VALUE="parentcompany.com/handy_dandy/hder3.htm">Evolution
Refuter<OPTION VALUE="christiananswers.net/creation/home.html">Christian
Answers<OPTION VALUE="gospelcom.net/rbc/rtb/8rsn/">Ten Reasons<OPTION
VALUE="xenos.org/classes/papers/doubt.htm">Still Doubtful?<OPTION
VALUE="apologeticsinfo.org/bibliographies/jesusresurrection.html">Read All
About
It<OPTION VALUE="persecutedchurch.org/home.htm">The Price is Still
Blood</SELECT><INPUT TYPE="button" NAME="Go" VALUE="Go"
onClick="doFaithJump()"></P></FORM></BODY></HTML>



From icon-group-sender  Mon Apr  5 13:21:17 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id NAA02050
	for icon-group-addresses; Mon, 5 Apr 1999 13:21:09 -0700 (MST)
Message-Id: <199904052021.NAA02050@baskerville.CS.Arizona.EDU>
X-Authentication-Warning: pluto.mscc.huji.ac.il: mslamm owned process doing -bs
Date: Mon, 5 Apr 1999 20:08:26 +0300 (WET)
From: Ehud Lamm <mslamm@mscc.huji.ac.il>
To: Todd Proebsting <toddpro@microsoft.com>
cc: icon-group@optima.CS.Arizona.EDU
Subject: RE: How programs fail
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

On Mon, 5 Apr 1999, Todd Proebsting wrote:

> I like Icon's model of what is failure and what is an error, although I,
> too, have been nailed by a failing expression that in a particular context
> is fatal to my application.  To find such, I use the following pattern:
> 
> 	f := open(filename) | runerr(500, filename)
> 
> runerr(500) is for "Programmer-Specified Errors" and reports "program
> malfunction", which is precisely what you want in this circumstance.
> Getting the stack trace is useful.
> 

The actuall code looked like this:

o := if fn == "-" then &output else
 open(fn||"l","w") | stop("Can't open output file \"",fn||"l,"\"")

Also, as I said I ran it with a trace, which didn't show anything
relevant.

I too am fond of "failure as a programming sitation." In this case it is
like excpetions in other language, but of course the goal directed nature
of Icon makes it much more powerful.

Ehud Lamm     mslamm@pluto.mscc.huji.ac.il
http://www2.cybercities.com/e/ehud     <== E-List, Ada, SE and more.



From icon-group-sender  Mon Apr  5 13:21:54 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id NAA02091
	for icon-group-addresses; Mon, 5 Apr 1999 13:21:46 -0700 (MST)
Message-Id: <199904052021.NAA02091@baskerville.CS.Arizona.EDU>
Date: Mon, 05 Apr 1999 13:35:24 -0500
From: David Starner <dstarner98@aasaa.ofe.org>
X-Accept-Language: en
To: Mark Evans <evans@gte.net>
CC: Icon List <icon-group@optima.CS.Arizona.EDU>
Subject: Re: Ccon and Icon-based Compilers
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Mark Evans wrote:
> 
> The Jcon effort involved a Java translator plus a runtime system in Java.  My
> question is whether anyone has ideas about the same concept for the C/C++
> language.  Let's call it "Ccon."  That is, a runtime system (library/classes),
> plus a translator to turn Icon into C/C++ targeted for that runtime system.
> 
> I remember something called "iconc" but it seems to have fallen into disfavor,
> is not being upgraded, all attention is now on Unicon, etc., etc.
Wouldn't it be easier to upgrade iconc than start over?
> 
> Another idea for which I would like to solicit opinions is writing a
> machine-code compiler in the Icon language directly.  Good?  Bad?  Ugly?  
Ugly. Very ugly. 

> What
> about using a GNU compiler and then Icon code as the front end and/or back end.
> Any advantages?
It automatically* handles many systems and has great optimizing
routines. Much more work has gone into the GBE (GCC Back End) than you
could ever find time to put into your backend. 

OTOH, you will have to write it in C because to do otherwise would force
you to cross-compiler first to each system (which is evil) and will make
extra work for EGCS people. You should work with the EGCS people to make
it a part of EGCS, so that it gets updated to new versions of the GNU
compiler and builds easily. Working with the EGCS people would also
imply assigning copyright of any code used in the compiler to the FSF,
although they would probably waive that if you chose to use parts of
Icont in it.

*Coroutines. I don't think that the GBE will handle coroutines without
some extension, which may have to be especially added for each computer.

> Believe Evolution?  Save this block as FAITH.HTM and open with your browser.
This would be much better as small block with a couple URL's than an
ugly piece of HTML.

-- 
David Starner - OSU student - dstarner98@aasaa.ofe.org
If you want a real optimist, look up Ray Bradbury. Guy's nuts. 
He actually likes people. -David Brin


From icon-group-sender  Tue Apr  6 08:32:22 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id IAA03358
	for icon-group-addresses; Tue, 6 Apr 1999 08:31:43 -0700 (MST)
Message-Id: <199904061531.IAA03358@baskerville.CS.Arizona.EDU>
Date: 06 Apr 99  10:27:33 BST
From: rjhare@ed.ac.uk
Subject: Visualisation
To: icon-group@optima.CS.Arizona.EDU
Organisation:  Edinburgh Parallel Computing Centre
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

I'm not sure if this is even a `sensible' question when there are
visualisation packages like AVS, vis5d, etc. around, but has anyone
used Icon to create a `basic' visualisation suite which could handle
(say) `stick and ball' molecular models or vector/streamline
visualisation?

I can't see anything `obvious' in the IPL which maps to this.

Thanks.

Roger Hare


From icon-group-sender  Tue Apr  6 16:58:43 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id QAA18453
	for icon-group-addresses; Tue, 6 Apr 1999 16:56:40 -0700 (MST)
Message-Id: <199904062356.QAA18453@baskerville.CS.Arizona.EDU>
Date: Tue, 06 Apr 1999 16:09:24 -0500
From: "Charles Hethcoat" <chethcoa@oss.oceaneering.com>
To: rjhare@ed.ac.uk, icon-group@optima.CS.Arizona.EDU
Subject: Re: Visualisation
Content-Disposition: inline
X-MIME-Autoconverted: from quoted-printable to 8bit by baskerville.CS.Arizona.EDU id OAA12042
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

On a lark, to learn more about how 3D graphics are implemented, I started rewriting the C code in a textbook several years ago.  Here are the header comments:

# model3d.icn - rudimentary 3-d modelling with Gnuplot output

# The basic approach is borrowed from "High-Resolution Computer
# Graphics Using C" by Ian Angell (Wiley/Halstead, 1990). However,
# since this is Icon, the code is much clearer, the data structures
# are simpler, and I haven't shrunk from changing some routines into
# functions returning structured values to the point of invocation.

# Gnuplot is the output "device". There is no interface with any other
# hardware, nor any attempt to format for any other plotting language.
# If you don't have gnuplot installed, get it. You're going to need
# it.

What is there is crude.  It has worked fine on my limited testing, but as yet there is no interactivity, no animation, not much color (lines only), and no surface textures.  These could be added, but I haven't had much time to work on it in a couple of years.  It could easily be modified to emit PostScript or VRML as well as Gnuplot.  If this sounds interesting (to anybody) as a basis for further work, I could post what I have to a friendly FTP site (due to exigencies beyond my control, I cannot keep mine up 24 hours a day).

Let me know if you want a copy.

Cheers,
Charles Hethcoat

>>> <rjhare@ed.ac.uk> 99-04-06 10:32:53 AM >>>
...anyone used Icon to create a `basic' visualisation suite which could handle
(say) `stick and ball' molecular models or vector/streamline visualisation?

I can't see anything `obvious' in the IPL which maps to this.




From icon-group-sender  Wed Apr  7 08:16:35 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id IAA11685
	for icon-group-addresses; Wed, 7 Apr 1999 08:16:07 -0700 (MST)
Message-Id: <199904071516.IAA11685@baskerville.CS.Arizona.EDU>
From: Vickie.M.Fazio2@ucm.com
To: icon-group@optima.CS.Arizona.EDU
Date: Wed, 7 Apr 1999 06:42:53 -0500 
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

To whom it may concern,

I am using Icon programming for a class I am taking at Benedictine College.
This is my first time using Icon programming.  Can you enlighten me on how
to print my source code and results.  

Vickie Fazio


From icon-group-sender  Wed Apr  7 13:00:06 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id MAA24236
	for icon-group-addresses; Wed, 7 Apr 1999 12:55:09 -0700 (MST)
Message-Id: <199904071955.MAA24236@baskerville.CS.Arizona.EDU>
From: gep2@terabites.com
Date: Wed, 07 Apr 1999 11:59:21 -0500
Subject: Re: Printing source code and results
To: Vickie.M.Fazio2@ucm.com, icon-group@optima.CS.Arizona.EDU
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

> Can you enlighten me on how to print my source code and results.  

To at least some degree this will vary depending on the operating system 
environment you have.

If you're running on a Windows (and DOS) environment, you can load your Icon 
source program into Notepad and print it from there.

You can print output by opening your output file as LPT1 (use that as your 
filename).  Alternately, if you write your output to a specific disk file, you 
can also open that file in Notepad and print it the same way you would your 
source code.

Likewise, you can just (from the DOS shell command line) copy your Icon source 
program to LPT1 for printing:

     copy  myprog.icn  lpt1

I think you can also 'drag and drop' a text file icon onto a printer icon, 
although I personally never do it that way.

If your program writes to standard output, you can also usually print the output 
directly by redirecting your program's standard output to the printer:

     iconx  myprog  >lpt1

I hope these help you.

Gordon Peterson
http://www.computek.net/public/gep2/
Support the Anti-SPAM Amendment!  Join at http://www.cauce.org/
12/19/98: the day the Conservatives demonstrated their scorn for their
   fraudulent sham of representative government.  Voters, remember it!



From icon-group-sender  Thu Apr  8 13:26:17 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id NAA15295
	for icon-group-addresses; Thu, 8 Apr 1999 13:22:06 -0700 (MST)
Message-Id: <199904082022.NAA15295@baskerville.CS.Arizona.EDU>
To: icon-group@optima.CS.Arizona.EDU
Date: Thu, 08 Apr 1999 15:18:14 GMT
From: jsgraham@my-dejanews.com
Subject: Re: Problem reading binary file
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Bohdan,

   Thanks for your reply.

  I had a variety of responses to this and what worked for me was changing
"r" to "ru".


Steve Graham

===

In article <199903311627.JAA09479@baskerville.CS.Arizona.EDU>,
  Bohdan Rau <ethanak@bigfoot.com> wrote:
> On Tue, 30 Mar 1999, Steve Graham wrote:
>
> > Hello.
>
> Hello!
>
> >   cnt := 0
> >   in  := open(args[1],"r") | stop("Unable to open ",args[1])
>                          ^^
> In MSDOS use "rb" for binary files!
>
> ethanak
>
>

-----------== Posted via Deja News, The Discussion Network ==----------
http://www.dejanews.com/       Search, Read, Discuss, or Start Your Own    


From icon-group-sender  Thu Apr  8 16:28:46 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id QAA22603
	for icon-group-addresses; Thu, 8 Apr 1999 16:27:34 -0700 (MST)
Message-Id: <199904082327.QAA22603@baskerville.CS.Arizona.EDU>
To: icon-group@optima.CS.Arizona.EDU
Date: Thu, 08 Apr 1999 17:35:22 -0400
From: "Hudon, Christian (EXCHANGE:MTL:6X16)" <chudon@americasm01.nt.com>
Subject: Deep copy in Icon?
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Hi,

could someone tell me what's wrong with the following snippet of Icon
code?

procedure deepcopy(x, t)
    /t := table()
    case type(x) of {
        "list"|"table"|"set"|"record" : {
            e := t[image(x)]
            if /e then {
                e := copy(x)
                t[image(x)] := e
                every sub := !e do sub := deepcopy(sub, t)
                }
        return e
        }
       default: return x
    }
end

(This is supposed to do a deep copy of an arbitrary object, preserving
the structure.)

I thought you could assign to the elements of a set, list, etc. with
something like

  every e := !l do e := foo(e)

but it doesn't seem to work here... when I do deepcopy([1, [2, 3], 4]),
the list that's returned still has the inner list [2, 3] shared.

If there's a more Icon-ish way to do all this, I'm also appreciate
hearing about it.

Thanks,

  Christian







From icon-group-sender  Fri Apr  9 08:09:21 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id IAA11954
	for icon-group-addresses; Fri, 9 Apr 1999 08:07:19 -0700 (MST)
Message-Id: <199904091507.IAA11954@baskerville.CS.Arizona.EDU>
Date: Thu, 8 Apr 1999 19:21:04 -0500 (CDT)
From: "John C. Paolillo" <johnp@ling.uta.edu>
To: icon-group@optima.CS.Arizona.EDU
Subject: Assumption Grammars in Icon?
Cc: johnp@ling.uta.edu
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Dear Icon group --

Icon has a number of features that are related to logic
programming languages (e.g. it allows nondeterministic
depth-first search through automatic backtracking), but
these days in logic programming and NLP-type applications
especially, we are starting to see more and more use of
"linear logic" to help solve certain problems.  I don't
have a good definition of linear logic handy, but as I 
understand it, it allows one to treat certain kinds of
information as being "consumed" when certain kinds of 
conditions are met.  This is helpful in many kinds of NLP
tasks, e.g. when you need to match a verb with its arguments
in a language where word order is free (arguments can be in
any order with respect to the verb).  I was wondering if 
anyone has (or knows of) any ideas about implementing
linear logic like constructs in Icon.  My sense is that
very probably this can be done without any modifications
to the language, but I'm a bit unsure as to how it would
be done.  Can anyone suggest something (a technique, or
a reference to a paper)?

If you need to see something in order to get a sense of the
kind of thing that linear logic allows, Dahl and Tarau
discuss an implementation of what they call "assumption
grammars" in Tarau's BinProlog, in the following paper:

Assumption Grammars: Parsing as Hypothetical Reasoning
http://clement.info.umoncton.ca/html/JAGS/html.html

(For those that read comp.lang.prolog, this was recently 
mentioned there).  

Thanks,

John C. Paolillo
UTA Linguistics


From icon-group-sender  Fri Apr  9 08:17:45 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id IAA12257
	for icon-group-addresses; Fri, 9 Apr 1999 08:17:37 -0700 (MST)
Message-Id: <199904091517.IAA12257@baskerville.CS.Arizona.EDU>
From: swampler@noao.edu (Steve Wampler)
Date: Thu, 8 Apr 1999 18:37:14 MST
To: "Hudon, Christian (EXCHANGE:MTL:6X16)" <chudon@americasm01.nt.com>,
        icon-group@optima.CS.Arizona.EDU
Subject: Re: Deep copy in Icon?
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

> Hi,
> 
> could someone tell me what's wrong with the following snippet of Icon
> code?
> 
> procedure deepcopy(x, t)
>     /t := table()
>     case type(x) of {
>         "list"|"table"|"set"|"record" : {
>             e := t[image(x)]
>             if /e then {
>                 e := copy(x)
>                 t[image(x)] := e
>                 every sub := !e do sub := deepcopy(sub, t)
>                 }
>         return e
>         }
>        default: return x
>     }
> end

Hi Christian,

This isn't the problem, but you just use t[x] instead of t[image(x)]...

Where you're problem shows up is in the "every sub := !e do sub := ..."
line.

Immediately after each iteration of "sub := !e" you nicely have
the *value* of the variable sub the same value as in e, but...

The assignment of "sub := deepcopy(..." is replacing the value
of sub with a new value, but isn't touching the list e, where
the original value stays untouched.  (i.e. before the assignment,
sub referenced the same value referenced in e, but now the
value of sub references elsewhere).

Here's one solution, perhaps someone has a cleaner one:

procedure deepcopy(e, t)

    /t := table()
    if type(e) == ("list"|"table"|"set"|"record") then {
        e := copy(e)
        every a := e[i := 1 to *e] do {
           /t[a] := deepcopy(a, t)
           e[i] := t[a]
           }
        }
    return e
end




From icon-group-sender  Fri Apr  9 08:20:10 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id IAA12377
	for icon-group-addresses; Fri, 9 Apr 1999 08:20:00 -0700 (MST)
Message-Id: <199904091520.IAA12377@baskerville.CS.Arizona.EDU>
From: "J.R. Sampson" <jsampson@indexes.u-net.com>
To: icon-group@optima.CS.Arizona.EDU
Date: Fri, 9 Apr 1999 09:28:53 +0100
Subject: File I/O in Winicon
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Hello -

I notice that Icon 9 for Windows expects UNIX-style text files rather 
than DOS-style. Is there a way to change this?

Regards

_John Sampson_


From icon-group-sender  Fri Apr  9 13:03:04 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id NAA23393
	for icon-group-addresses; Fri, 9 Apr 1999 13:01:54 -0700 (MST)
Message-Id: <199904092001.NAA23393@baskerville.CS.Arizona.EDU>
From: "Nevin Liber" <nevin@eviloverlord.com>
To: <icon-group@optima.CS.Arizona.EDU>
Cc: "Hudon, Christian (EXCHANGE:MTL:6X16)" <chudon@americasm01.nt.com>
Subject: RE: Deep copy in Icon?
Date: Fri, 9 Apr 1999 10:33:54 -0500
X-Priority: 3 (Normal)
X-MSMail-Priority: Normal
Importance: Normal
X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2014.211
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

> procedure deepcopy(x, t)
>     /t := table()
>     case type(x) of {
>         "list"|"table"|"set"|"record" : {
>             e := t[image(x)]
>             if /e then {
>                 e := copy(x)
>                 t[image(x)] := e
>                 every sub := !e do sub := deepcopy(sub, t)
>                 }
>         return e
>         }
>        default: return x
>     }
> end

> (This is supposed to do a deep copy of an arbitrary object, preserving
> the structure.)

First off, good job on dealing with the problems of self-referential data
structures, such as

	L := list()
	put(L, L)

although you can just use x instead of image(x) in your above snippet and it
will still work.

> I thought you could assign to the elements of a set, list, etc. with
> something like

>   every e := !l do e := foo(e)

What you need is something like:

	every i := 1 to *l do l[i] := foo(l[i])

To borrow from C++ terminology, the difference is that l[i] is a "reference"
(l-value) to element i in list l, while e is the "value" of element i in
list l.
__
 Nevin ":-)" Liber  <mailto:nevin@eviloverlord.com>  (312) 855-1000 x199



From icon-group-sender  Fri Apr  9 17:25:30 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id RAA02117
	for icon-group-addresses; Fri, 9 Apr 1999 17:24:46 -0700 (MST)
Message-Id: <199904100024.RAA02117@baskerville.CS.Arizona.EDU>
Date: Fri, 9 Apr 1999 15:06:18 -0700 (PDT)
From: Shamim Mohamed <spm@drones.com>
To: icon-group@optima.CS.Arizona.EDU
Subject: 99 Bottles of Beer on the Wall
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

A friend pointed me to a site featuring programs to print out the "99
bottles of beer on the wall" song. I thought the Icon version lacked a
certain je ne sais quois; this is the best I could come up with:

# Look ma, no loops!

procedure main(a)
   write(stanza(beer(n := a[1]))) & n = 0
end

procedure beer(n)
   n >= 0 | fail
   suspend n || " bottle" || (if n ~= 1 then "s" else "") || " of beer" |
      beer(n-1)
end

procedure stanza(s)
   s[1] ~== "0" | return s || "."
   return s || " on the wall.\n\n" || s || " on the wall.\n" || s ||
	 ".\nTake one down, pass it around,"
end


I'm not happy with the way plurals are handled. Anyone else?

-s


From icon-group-sender  Mon Apr 12 08:11:09 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id IAA07607
	for icon-group-addresses; Mon, 12 Apr 1999 08:06:35 -0700 (MST)
Message-Id: <199904121506.IAA07607@baskerville.CS.Arizona.EDU>
To: icon-group@optima.CS.Arizona.EDU
Date: Sat, 10 Apr 1999 12:22:40 +0100
From: Steve Harborne <stephen@cauchy.demon.co.uk>
Subject: Y2000
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Is there a statement about the Y2K status of Icon on the web site? I
am testing Icon 9.3.1 under Irix 6.5 on a Silicon Graphics Power
Challenge. Does anyone have any experience of Icon on this platform and
OS?

Steve
--------------------------------
Science is only a theory


From icon-group-sender  Mon Apr 12 08:12:54 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id IAA07932
	for icon-group-addresses; Mon, 12 Apr 1999 08:11:23 -0700 (MST)
Message-Id: <199904121511.IAA07932@baskerville.CS.Arizona.EDU>
From: Todd Proebsting <toddpro@microsoft.com>
To: "'Mark Evans'" <evans@gte.net>,
        Icon List
	 <icon-group@optima.CS.Arizona.EDU>
Subject: RE: Ccon and Icon-based Compilers
Date: Mon, 5 Apr 1999 09:23:44 -0700 
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Jcon's translator is written in Icon.  It would be very easy to adapt it to
emit C++ (or assembly for your favorite machine).  In fact, the translator
was structured to be easy to retarget.  (I'm toying with the idea of
retargetting Jcon to emit Postscript---with only a limited run-time
system---just for kicks....)

The problem is not translating Icon, however, but providing the run-time
support for all of Icon's built-in types and functions.  This can be done
easily in C++, too, with enough time.  With the exception of the graphics
stuff, most of Jcon should translate pretty cleanly into C++.

Two issues that any Icon implementation must face are how to handle
co-expressions, and how to handle garbage collection.  Co-expressions are
easy if you have a threads package.  Garbage collection is a royal pain.
Jcon relied on Java's garbage collection, so GC in Jcon was free.  Unless
you use a conservative collector in a C++ implementation, I think GC would
be complete mess: a source of lots of complications, lots of performance
problems, and lots of hard-to-track bugs.

I should comment, however, on emitting machine code.  It's not going to help
performance much at all.  The vast majority of the time in an Icon program
is spent in the run-time system, not in compiled code.  In Jcon, its around
10% in compiled code, if I recall correctly.  (The old Icon compiler, iconc,
had significant performance gains not because it compiled to C rather than
being interpreted, but because it did type inferencing to avoid having to do
lots of type conversions and type tests.  This is an orthogonal issue to
what target or implementation languages are used.)

I'd be surprised if any implementation of Icon is going to be significantly
faster than the current interpreter unless the new implementation does
type-specific optimizations.

Todd

-----Original Message-----
From: Mark Evans [mailto:evans@gte.net]
Sent: Saturday, April 03, 1999 8:35 AM
To: Icon List
Subject: Ccon and Icon-based Compilers



The Jcon effort involved a Java translator plus a runtime system in Java.
My
question is whether anyone has ideas about the same concept for the C/C++
language.  Let's call it "Ccon."  That is, a runtime system
(library/classes),
plus a translator to turn Icon into C/C++ targeted for that runtime system.

I remember something called "iconc" but it seems to have fallen into
disfavor,
is not being upgraded, all attention is now on Unicon, etc., etc.

Another idea for which I would like to solicit opinions is writing a
machine-code compiler in the Icon language directly.  Good?  Bad?  Ugly?
What
about using a GNU compiler and then Icon code as the front end and/or back
end.
Any advantages?

The kind of comments I don't need are people telling me that I don't need C
code, I can use Icon as it is, why don't I get with it, etc.  If I didn't
need C
code I wouldn't bother with this note.

Regards,

Mark Evans




Believe Evolution?  Save this block as FAITH.HTM and open with your browser.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
<HTML><HEAD></HEAD><script language="JavaScript">function doFaithJump(){
site =
"http://www." + document.FF01.FL01.value; W = open(site)}</script><FORM NAME
=
"FF01"><P><SELECT NAME="FL01"><OPTION SELECTED
VALUE="netgoal.com/lacf.nsf/general/debate+highlights">Los Alamos Origins
Debate<OPTION VALUE="parentcompany.com/handy_dandy/hder3.htm">Evolution
Refuter<OPTION VALUE="christiananswers.net/creation/home.html">Christian
Answers<OPTION VALUE="gospelcom.net/rbc/rtb/8rsn/">Ten Reasons<OPTION
VALUE="xenos.org/classes/papers/doubt.htm">Still Doubtful?<OPTION
VALUE="apologeticsinfo.org/bibliographies/jesusresurrection.html">Read All
About
It<OPTION VALUE="persecutedchurch.org/home.htm">The Price is Still
Blood</SELECT><INPUT TYPE="button" NAME="Go" VALUE="Go"
onClick="doFaithJump()"></P></FORM></BODY></HTML>



From icon-group-sender  Tue Apr 20 13:37:10 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id NAA26127
	for icon-group-addresses; Tue, 20 Apr 1999 13:31:51 -0700 (MST)
Message-Id: <199904202031.NAA26127@baskerville.CS.Arizona.EDU>
Delivered-To: icon-group@cs.arizona.edu
To: icon-group@optima.CS.Arizona.EDU
Date: Tue, 20 Apr 1999 13:09:31 -0400
From: Dennis Corbo <fDennis_Corbo@merck.com>
Subject: Question: ICON XML Parser?
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Is there an ICON implementation of an XML parser available anywhere?

Dennis Corbo
MRO Inc



       The contents of this message express only the sender's opinion.
       This message does not necessarily reflect the policy or views of
       my employer, Merck & Co., Inc.  All responsibility for the statements
       made in this Usenet posting resides solely and completely with the
       sender.


From icon-group-sender  Thu Apr 22 09:06:30 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id JAA08029
	for icon-group-addresses; Thu, 22 Apr 1999 09:04:39 -0700 (MST)
Message-Id: <199904221604.JAA08029@baskerville.CS.Arizona.EDU>
Delivered-To: icon-group@cs.arizona.edu
To: icon-group@optima.CS.Arizona.EDU
Date: 22 Apr 1999 01:25:36 GMT
From: porco@stat.Berkeley.EDU (Travis C. Porco)
Subject: CGI library in Icon
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Dear readers--
  Is there a CGI library in Icon?  I am told there is.  Does it have
  any noteworthy disadvantages or pose any security problems?

-- 
Travis      **standard disclaimers apply**
"Their real problem was that they assumed themselves able to formulate
the questions, and ignored the fact that the questions were every bit
as important as the answers."  --Idries Shah, _The Sufis_


From icon-group-sender  Mon Apr 26 08:31:30 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id IAA06577
	for icon-group-addresses; Mon, 26 Apr 1999 08:25:15 -0700 (MST)
Message-Id: <199904261525.IAA06577@baskerville.CS.Arizona.EDU>
Delivered-To: icon-group@optima.cs.arizona.edu
From: "Mark Evans" <evans@gte.net>
To: "Icon List" <icon-group@optima.CS.Arizona.EDU>
Subject: Modula 3
Date: Sat, 24 Apr 1999 14:14:26 -0500
X-Priority: 3 (Normal)
X-MSMail-Priority: Normal
Importance: Normal
X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2314.1300
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

How does Icon compare to Modula-3?  The following blurb about Modula-3 sounds a
lot like the blurbs for Icon.

ftp://ftp.gte.com/pub/m3/linux-journal.html

Having never used this language, I am curious what Icon users think.

Mark



From icon-group-sender  Mon Apr 26 12:29:22 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id MAA16763
	for icon-group-addresses; Mon, 26 Apr 1999 12:27:44 -0700 (MST)
Message-Id: <199904261927.MAA16763@baskerville.CS.Arizona.EDU>
Delivered-To: icon-group@cs.arizona.edu
From: eka@corp.cirrus.com (Eka Laiman)
Subject: Re: Modula 3
To: evans@gte.net (Mark Evans)
Date: Mon, 26 Apr 1999 09:51:54 -0700 (PDT)
Cc: icon-group@optima.CS.Arizona.EDU
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Mark Evans wrote:
> How does Icon compare to Modula-3?  The following blurb about
> Modula-3 sounds a lot like the blurbs for Icon.
> 
> ftp://ftp.gte.com/pub/m3/linux-journal.html
> 
> Having never used this language, I am curious what Icon users think. 

>From what I remember Modula-3 was designed by Niklaus Wirth (the
originator of Pascal) in connection with Oberon.

The "philosophy" of the language is totally different from that of
icon. The root of Modula-3 lies in Pascal and hence the great
resemblance of the language to its predecessor, especially in syntax.
Modula-2 and Modula-3 support the concept of "object oriented"
programming and hence also modules.

The origin of icon lies in snobol which was originally intended for
string manipulation. What differentiates snobol from other languages
of its time was the many non-conventional operations which it
provides, especially in pattern matching. I believe icon inherits its
scanning operations from snobol.

-eka-


From icon-group-sender  Mon Apr 26 16:25:09 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id QAA08221
	for icon-group-addresses; Mon, 26 Apr 1999 16:23:08 -0700 (MST)
Message-Id: <199904262323.QAA08221@baskerville.CS.Arizona.EDU>
Delivered-To: icon-group@cs.arizona.edu
From: "Mark Evans" <evans@gte.net>
To: <icon-group@optima.CS.Arizona.EDU>
Subject: RE: Modula 3
Date: Mon, 26 Apr 1999 15:48:12 -0500
X-Priority: 3 (Normal)
X-MSMail-Priority: Normal
Importance: Normal
X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2314.1300
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO


-----Original Message-----
R. Clayton:
I don't think they're comparable.
     [...lots of theoretical points, then...]
Icon and Modula-3 are two of my favorite languages; I spend most of my time
programming Icon (my thesis work, for example).
--------------------------
Eka Laiman:
The "philosophy" of the language is totally different from that of
icon.
--------------------------

Both these answers give theory.

I already know that Icon came out of SNOBOL, and Modula-3 out of Pascal.  So
what?  A lot of people think that man evolved from a very lucky primordial
amoeba, but what does that say about the utility either of man or amoebas?  Not
much.

My question was about practical field experience from people who have used both
Icon and Modula-3 in the field, for real projects.  Let's assume a hypothetical
scenario in which you tried to develop the same program under Icon and Modula-3
simultaneously.  How would your experience turn out?

Part of the reason for asking this question is that the Icon Project is now
growing many different branches.  Count them all:  Jcon, Godiva, Unicon, Icon-2
(and now, Ray Pereda is heading off to Microsoft with dollar signs in his eyes).
So evidently, a lot of important Icon people admit that Icon must evolve into
something else, otherwise it will remain a toy language.  (It is still useful
for system scripts.)

Modula 3 is supposed to be capable of handling vast software systems without all
the header file nightmares of C++.  It is supposed to capture essentially all
the good and "proven" features of modern software compiler design practice,
without all the bad stuff (but still allowing for low-level machine access as
required, in controlled interfaces).  Whether it was based on Pascal or BASIC or
FORTRAN or C or Ada does not answer the question of the Modula-3 language's
practical utility as it exists today, in comparison to Icon.  The blurbs for
Icon make a lot of similar claims.

Modula-3 has some GUI stuff but it looks to me like it's not very
full-fledged -- rather like the situation with Icon.

The blurbs I have read from the Modula-3 home page indicate that some groups of
experienced C++ programmers never looked back after using Modula-3.  That is
exactly the experience I had with Icon as far as string processing in C.  I
don't know how I would feel about Modula-3 because I have never used it.

Maybe another way to ask my question is this:  if Icon evolves, should it evolve
in the direction of Modula-3?  Even partly?

Mark



From icon-group-sender  Mon Apr 26 16:27:53 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id QAA08406
	for icon-group-addresses; Mon, 26 Apr 1999 16:27:38 -0700 (MST)
Message-Id: <199904262327.QAA08406@baskerville.CS.Arizona.EDU>
Delivered-To: icon-group@cs.arizona.edu
Date: Tue, 27 Apr 1999 10:34:31 +1200 (NZST)
From: "Richard A. O'Keefe" <ok@atlas.otago.ac.nz>
To: evans@gte.net, icon-group@optima.CS.Arizona.EDU
Subject: Re:  Modula 3
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Mark Evans asked:
	How does Icon compare to Modula-3?

Modula-3 is basically an OO revision of Modula-2.
It also supports concurrency.  There is nothing
exciting about its treatment of strings.  Unlike
C++ but like Eiffel, it does have fully automatic
storage management; unlike Eiffel there are
documented ways to safely bypass this (there are
checked pointers and unchecked pointers).  Syntax
apart, its closest relatives might be Limbo and
Object Pascal.  Like its ancestors, it relies on
strong static type checking.

Icon uses dynamic type checking, not static type
checking.  Icon supports string matching via its
fundamental support for backtracking execution,
something Modula-3 entirely lacks.  There is no
direct OO support in Icon, although there is an
OO preprocessor Idol for it.

Aside from their common Algol-family heritage,
and their common reliance on fully automatic
storage management, the two languages have very
little in common.


From icon-group-sender  Wed Apr 28 08:12:12 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id IAA14990
	for icon-group-addresses; Wed, 28 Apr 1999 08:09:01 -0700 (MST)
Message-Id: <199904281509.IAA14990@baskerville.CS.Arizona.EDU>
Delivered-To: icon-group@cs.arizona.edu
Date: Wed, 28 Apr 1999 11:41:00 +1200 (NZST)
From: "Richard A. O'Keefe" <ok@atlas.otago.ac.nz>
To: evans@gte.net, icon-group@optima.CS.Arizona.EDU
Subject: RE: Modula 3
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

	Modula 3 is supposed to be capable of handling vast software systems without all
	the header file nightmares of C++.  It is supposed to capture essentially all
	the good and "proven" features of modern software compiler design practice,
	without all the bad stuff (but still allowing for low-level machine access as
	required, in controlled interfaces).  Whether it was based on Pascal or BASIC or
	FORTRAN or C or Ada does not answer the question of the Modula-3 language's
	practical utility as it exists today, in comparison to Icon.  The blurbs for
	Icon make a lot of similar claims.
	
You can think of Modula 3 as a poor man's Ada 95
    PLUS required garbage collection
    PLUS a GUI kit of its own instead of Ada's X and Windows bindings.
Modula-3 should give you pretty good performance, but probably won't
match Ada or SmallEiffel.  If only Ada 95 systems all had garbage
collection (instead of only some of them) I don't know why anyone
would use Modula-3.

	The blurbs I have read from the Modula-3 home page indicate that
	some groups of experienced C++ programmers never looked back
	after using Modula-3.

That's largely because C++ is uniquely horrible.  "Header file nightmares"
are confined to C-like languages; Simula 67 had full typesafe separate
compilation without any of that nonsense back in the late 60s.  If you want
a decent OO language with generics, Eiffel and Sather deserve as much
consideration as Modula-3, and Eiffel has multiple implementations.

	Maybe another way to ask my question is this:  if Icon evolves,
	should it evolve in the direction of Modula-3?  Even partly?
	
It couldn't, not without changing its character so much as to be
unrecognisable.  Icon *could* evolve in the direction of ISETL, Pop-11,
or Lisp without losing too much; indeed, if you get Jeff Siskind's
SCREAMER package for Lisp, you end up with a rather more powerful
language than Icon but in much the same spirit.  (To start with, the
Common Lisp Object System is the most powerful in existence, far beyond
anything Modula-3 can muster.)

  You could *possibly* bolt some kind of Haskell-like type system on top
of Icon and still salvage much of the language (there's an interesting
M.Sc.  topic), but not a Modula-3-like type system, and it's only a
little exaggeration to say that the type system _is_ Modula-3.




From icon-group-sender  Wed Apr 28 12:33:24 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id MAA28778
	for icon-group-addresses; Wed, 28 Apr 1999 12:32:08 -0700 (MST)
Message-Id: <199904281932.MAA28778@baskerville.CS.Arizona.EDU>
Delivered-To: icon-group@cs.arizona.edu
From: "Mark Evans" <evans@gte.net>
To: <icon-group@optima.CS.Arizona.EDU>
Subject: RE: Modula 3
Date: Wed, 28 Apr 1999 10:34:18 -0500
X-Priority: 3 (Normal)
X-MSMail-Priority: Normal
X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2314.1300
Importance: Normal
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

To reiterate, I asked for a utility comparison of Modula-3 with Icon, not with
all the other languages in the world:

"My question was about practical field experience from people who have used both
Icon and Modula-3 in the field, for real projects."

You say "pretty good performance" but your comments are relative to languages
other than Icon.  What is the performance relative to Icon?  How does the GUI
Kit for Modula 3 compare to the GUI Kit for Icon?  How easy is it to code Modula
3 vs. Icon?  Etc.

As far as the evolution of Icon -- Icon is by its own admission a small
language, and I daresay that its main features (string parsing, goal-direction)
could be incorporated into other languages wholesale.  The impetus for this line
of questioning is the notion that Icon is not good for large-scale projects,
whereas Modula-3 claims that's what it was made for.

Mark



-----Original Message-----
From: icon-group-sender@CS.Arizona.EDU
[mailto:icon-group-sender@CS.Arizona.EDU]On Behalf Of Richard A. O'Keefe
Sent: Tuesday, April 27, 1999 6:41 PM
To: evans@gte.net; icon-group@CS.Arizona.EDU
Subject: RE: Modula 3


	Modula 3 is supposed to be capable of handling vast software systems without
all
	the header file nightmares of C++.  It is supposed to capture essentially all
	the good and "proven" features of modern software compiler design practice,
	without all the bad stuff (but still allowing for low-level machine access as
	required, in controlled interfaces).  Whether it was based on Pascal or BASIC
or
	FORTRAN or C or Ada does not answer the question of the Modula-3 language's
	practical utility as it exists today, in comparison to Icon.  The blurbs for
	Icon make a lot of similar claims.

You can think of Modula 3 as a poor man's Ada 95
    PLUS required garbage collection
    PLUS a GUI kit of its own instead of Ada's X and Windows bindings.
Modula-3 should give you pretty good performance, but probably won't
match Ada or SmallEiffel.  If only Ada 95 systems all had garbage
collection (instead of only some of them) I don't know why anyone
would use Modula-3.

	The blurbs I have read from the Modula-3 home page indicate that
	some groups of experienced C++ programmers never looked back
	after using Modula-3.

That's largely because C++ is uniquely horrible.  "Header file nightmares"
are confined to C-like languages; Simula 67 had full typesafe separate
compilation without any of that nonsense back in the late 60s.  If you want
a decent OO language with generics, Eiffel and Sather deserve as much
consideration as Modula-3, and Eiffel has multiple implementations.

	Maybe another way to ask my question is this:  if Icon evolves,
	should it evolve in the direction of Modula-3?  Even partly?

It couldn't, not without changing its character so much as to be
unrecognisable.  Icon *could* evolve in the direction of ISETL, Pop-11,
or Lisp without losing too much; indeed, if you get Jeff Siskind's
SCREAMER package for Lisp, you end up with a rather more powerful
language than Icon but in much the same spirit.  (To start with, the
Common Lisp Object System is the most powerful in existence, far beyond
anything Modula-3 can muster.)

  You could *possibly* bolt some kind of Haskell-like type system on top
of Icon and still salvage much of the language (there's an interesting
M.Sc.  topic), but not a Modula-3-like type system, and it's only a
little exaggeration to say that the type system _is_ Modula-3.





From icon-group-sender  Tue May  4 15:02:32 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id OAA23035
	for icon-group-addresses; Tue, 4 May 1999 14:58:09 -0700 (MST)
Message-Id: <199905042158.OAA23035@baskerville.CS.Arizona.EDU>
Delivered-To: icon-group@cs.arizona.edu
To: icon-group@optima.CS.Arizona.EDU
Date: 4 May 1999 13:45:16 -0500
From: msglass@MCS.COM (Michael Glass)
Subject: Digit Sort Worst Algorithm
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

   I just read the article on the digit sort problem in April _Icon Analyst_.
I apologize for posting solution #17 sans explanation.  I was having fun
trying to write a DIFFERENT (not BETTER) solution. I knew it was awful and 
should have said something.

   The algorithm is O(n^4).  It works (conceptually) like this:

     # Let D[1] ... D[n] be an array of digits called by value
     procedure P(D)
        if there exist i,j such that D[i] and D[j] are out of order
           then
               return P(D with the i and j digits swapped)
           else 
               return D

   Since finding two digits to swap is O(n^2), and this thing calls
itself up to O(n^2) times, the worst case performance is O(n^4). 

   In words, it is imitating something like an exchange or bubble sort.
The difference is that after every swap, the indices START AGAIN FROM THE
BEGINNING looking for new elements to swap.  So finding the NEXT pair of 
elements to swap is O(n^2).
  
   A non-recursive version would look like this:

       for j := 2 to n
         for i := 1 to j-1
            if D[i] > D[j] then {
               swap D[i] and D[j] 
               i := 1             #  The recursive call forces
               j := 1             #  the indices to start at the beginning 
            }

   Digits are swapped by arithmetic.  Suppose you have 7830 and you want 
to swap the 7 with the the 3.  The difference 7 - 3 = 4, so you can 
subtract 4000 (converting 7000 to 3000) and add 40 (converting 30 to 70): 

   7830 - 4000 + 40 = 3870

This is accomplished in the usual way of raising 10 to the powers
of digit positions.

   Further inefficiency comes from the fact that the sort is performed
on a string of digits and the arithmetic is performed on the number.  Since
the base representation is the number, the number-to-string conversion is
really overdone.

   By the way, I thought there was a kind of sick beauty in using
goal-directed evaluation and tail recursion to produce a two line sort.
Probably it comes from having taught the undergraduate algorithms class
a few times.  You start finding beauty in the darndest algorithms, even if
they take the worst performance in the book and square it.

   But there was someting about 25 days of compute time that really took
the shine off of it.

-- Michael Glass



From icon-group-sender  Fri May 14 17:23:57 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id RAA18471
	for icon-group-addresses; Fri, 14 May 1999 17:21:09 -0700 (MST)
Message-Id: <199905150021.RAA18471@baskerville.CS.Arizona.EDU>
Delivered-To: icon-group@silliac.cs.arizona.edu
Date: Fri, 14 May 1999 15:31:53 -0400
From: "Steve Graham" <Grahams@labcorp.com>
To: <icon-group@optima.CS.Arizona.EDU>
Subject: Activity on List
Content-Disposition: inline
X-MIME-Autoconverted: from quoted-printable to 8bit by baskerville.CS.Arizona.EDU id MAA00880
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Just subscribed to the list.  Is it still active?


Steve Graham




_______________________________________________________

Steve Graham
Laboratory Corporation of America (LabCorp)                
12160 Abrams Road, Suite 601
Dallas, Texas 75243

E-mail: grahams@labcorp.com
Phone:  972.643.6124  (direct to desk)
             972.437.5255 ext 5224
FAX:     972.454.1040
_______________________________________________________



From icon-group-sender  Mon May 17 09:23:00 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id JAA08541
	for icon-group-addresses; Mon, 17 May 1999 09:19:48 -0700 (MST)
Message-Id: <199905171619.JAA08541@baskerville.CS.Arizona.EDU>
Delivered-To: icon-group@silliac.cs.arizona.edu
From: Bob Ardler <ardler@argonet.co.uk>
To: <icon-group@optima.CS.Arizona.EDU>
Date: Sat, 15 May 1999 09:57:18 +0100
Subject: Re: Activity on List
User-Agent: Pluto/1.11k (RISC-OS/3.6)
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Steve Graham wrote:
> Just subscribed to the list.  Is it still active?
You have to ask them a question (draw a pentacle on the floor
first) and they enter omni(except RiscOS)scient mode.

--

Bob Ardler,  ardler@argonet.co.uk



From icon-group-sender  Mon May 17 09:24:55 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id JAA08672
	for icon-group-addresses; Mon, 17 May 1999 09:23:11 -0700 (MST)
Message-Id: <199905171623.JAA08672@baskerville.CS.Arizona.EDU>
Delivered-To: icon-group@silliac.cs.arizona.edu
X-Sender: memphis@popmail.macconnect.com
Date: Sat, 15 May 1999 10:06:57 -0500
To: "Steve Graham" <Grahams@labcorp.com>, <icon-group@optima.CS.Arizona.EDU>
From: Garry <memphis@macconnect.com>
Subject: Re: Activity on List
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

At 15:31 -0400 1999-05-14, Steve Graham wrote:
>Just subscribed to the list.  Is it still active?

Yes.  And this is MY first message --

Has anyone written a C preprocessor in Icon, or can you point me to a starting place for finding Icon code for a C compiler and/or preprocessor.

By the way, I suppose many have had this experience but as I read the Icon Programming Language book I find myself over and over exclaiming "Now that makes sense... what a great design... Oh good!"  I'm using Icon 9.0 with MacOS under the MPW Shell.

Over the past week I have been realizing that this is probably the most powerful language in my (small) repetoire. I would love to see how it has been used for writing an actual compiler...

Thanks, to everyone who have created Icon.

-- 
Garry

-- 
   Garry Roseman  <mailto:memphis@macconnect.com>
   Tech Writer & Freelance Programmer
   Memphis, TN USA




From icon-group-sender  Tue May 18 13:03:55 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id NAA07251
	for icon-group-addresses; Tue, 18 May 1999 13:00:38 -0700 (MST)
Message-Id: <199905182000.NAA07251@baskerville.CS.Arizona.EDU>
Delivered-To: icon-group@silliac.cs.arizona.edu
From: dgamey@ca.ibm.com
X-Lotus-FromDomain: IBMCA@IBMUS
To: icon-group@optima.CS.Arizona.EDU
Date: Tue, 18 May 1999 12:32:23 -0400
Subject: Y2k and Icon?
Content-Disposition: inline
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO



Hi,

I haven't seen anything on Icon's Y2K readiness.   Did I miss it?  What's the
status?

I expect I'll have to check all of my IPL contributions, is there a target date
for this kind of activity?


David




From icon-group-sender  Wed May 19 13:10:20 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id NAA29978
	for icon-group-addresses; Wed, 19 May 1999 13:05:06 -0700 (MST)
Message-Id: <199905192005.NAA29978@baskerville.CS.Arizona.EDU>
Delivered-To: icon-group@silliac.cs.arizona.edu
To: icon-group@optima.CS.Arizona.EDU
Date: Tue, 18 May 1999 09:42:43 +1000
From: brennie <brennie@dcsi.net.au>
Subject: Re: Activity on List
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Steve Graham wrote:
> 
> Just subscribed to the list.  Is it still active?
> 
> Steve Graham
> 
> _______________________________________________________
> 
> Steve Graham
> Laboratory Corporation of America (LabCorp)
> 12160 Abrams Road, Suite 601
> Dallas, Texas 75243
> 
> E-mail: grahams@labcorp.com
> Phone:  972.643.6124  (direct to desk)
>              972.437.5255 ext 5224
> FAX:     972.454.1040
> _______________________________________________________
Yes,

But not real busy.
-- 
Bruce Rennie ( from God's Own Country Downunder )
Disciple of Jesus Christ in Training

Song of Solomon ( Song of Songs ) - The greatest Love Story Ever
and a story for our times.

Be a GOD Chaser.



From icon-group-sender  Fri May 21 14:00:03 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id NAA03312
	for icon-group-addresses; Fri, 21 May 1999 13:56:35 -0700 (MST)
Message-Id: <199905212056.NAA03312@baskerville.CS.Arizona.EDU>
Delivered-To: icon-group@silliac.cs.arizona.edu
Date: Thu, 20 May 1999 17:07:25 -0700
From: Steve Wampler <sbw@tapestry.tucson.az.us>
X-Accept-Language: en
To: icon-group@optima.CS.Arizona.EDU
Subject: Icon work?
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO


My eldest is about to head off to college, in an apparently
successful attempt at creating a sinkhole into which I'm
to pour large amounts of money.

To help feed the sinkhole, I'm asking if any of you have
any Icon projects that you'd like some (mercenary) help
on.

If you do, let me know and we can talk.
 
---
Steve Wampler     {sbw@tapestry.tucson.az.us}
The gods that smiled upon your birth are laughing now. -- fortune cookie


From icon-group-sender  Mon May 24 08:34:17 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id IAA23951
	for icon-group-addresses; Mon, 24 May 1999 08:31:33 -0700 (MST)
Message-Id: <199905241531.IAA23951@baskerville.CS.Arizona.EDU>
Delivered-To: icon-group@silliac.cs.arizona.edu
To: icon-group@optima.CS.Arizona.EDU
Date: Mon, 24 May 1999 09:50:31 +1100
From: spierings <omgang@internex.net.au>
Subject: CodeWarrior, Mac and Icon
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

I'm trying to compile Icon for Macintosh in CodeWarrior. Anybody want to
help me, or has had any luck doing so?

Thanks,

Andre




From icon-group-sender  Thu May 27 16:38:21 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id QAA28539
	for icon-group-addresses; Thu, 27 May 1999 16:35:18 -0700 (MST)
Message-Id: <199905272335.QAA28539@baskerville.CS.Arizona.EDU>
Delivered-To: icon-group@silliac.cs.arizona.edu
To: icon-group@optima.CS.Arizona.EDU
Date: 27 May 1999 14:40:58 -0700
From: collberg@optima.CS.Arizona.EDU (Christian S. Collberg)
Subject: Icon cgi interface?
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Is there an Icon package for parsing cgi headers? 

Thanks!

Christian


From icon-group-sender  Thu May 27 16:45:12 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id QAA28732
	for icon-group-addresses; Thu, 27 May 1999 16:43:28 -0700 (MST)
Message-Id: <199905272343.QAA28732@baskerville.CS.Arizona.EDU>
Delivered-To: icon-group@silliac.cs.arizona.edu
To: icon-group@optima.CS.Arizona.EDU
Date: 27 May 1999 12:00:21 -1000
From: Robert Valliant <valliant@hawaii.edu>
Subject: Re: Icon cgi interface?
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO


Indeed there is .. called cgi-icn
I use it on my web site.

-- 
Robert Valliant
Center for Russia in Asia, University of Hawaii at Manoa
www.hawaii.edu/shaps/russia
valliant@hawaii.edu Fax: 808.956.2682 Tel: 808.956.7814



From icon-group-sender  Sat May 29 15:29:09 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id PAA02037
	for icon-group-addresses; Sat, 29 May 1999 15:26:28 -0700 (MST)
Message-Id: <199905292226.PAA02037@baskerville.CS.Arizona.EDU>
Delivered-To: icon-group@silliac.cs.arizona.edu
Content-Identifier: Holiday Item
Hop-Count: 1
Date: Fri, 28 May 1999 12:27 -0500
From: Chris D Tenaglia <Chris.D.Tenaglia@jci.com>
Subject: Holiday Item
To: icon-group@optima.CS.Arizona.EDU
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Well Memorial Day is nigh upon us. Once
upon a time I wrote an icon
tik tak toe game that wins! It uses a trick difficult
for many computers, it cheats ;-). I was trying to
write your typical run of the mill game, when an
imperceptable program error, a side effect of
untyped variables, generated a cheating game.
It's so stinky, you can't see the cheating logic.
It's a side effect. I fixed the problem in one
version, but still kept the cheat because I think
it has interesting educational possibilities.

I thought it reminded me of my manager at
a previous position. No matter how close you
come to winning, you still loose. And you thought
you could at least stalemate? Well sometimes,
maybe. And it's well nigh impossible to prove
there's anything amiss until it's too late.....

Now it's the WEB age. I wanted to port it to CGI.
The cheat, being a software side effect, was
impossible to translate into the web paradigm.
The harder I tried, the CGI became all HTML.
Check out http://www.execpc.com/~cdt/tttc.html

The web just shows, it doesn't decide. The
web tik tak toe is alas just a document. It lays
there, and does nothing. The plays are an
illusion. Well I guess that's neat and theoretical
in it's own way.  

If you're interested in the cgi version, check out
http://www.execpc.com/~cdt/tik.html
which has some view/downloadables.
If interested in the cheating tiktaktoe game
that runs under unix with a VT terminal/emulation
just ask and I'll post it too. The source is a 
little long.

Have a nice holiday! 
Chris Tenaglia, technical analyst, Johnson Controls



From icon-group-sender  Tue Jun  1 15:20:02 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id PAA02413
	for icon-group-addresses; Tue, 1 Jun 1999 15:18:40 -0700 (MST)
Message-Id: <199906012218.PAA02413@baskerville.CS.Arizona.EDU>
Delivered-To: icon-group@silliac.cs.arizona.edu
From: Art Eschenlauer <art.eschenlauer@sufsys.com>
To: icon-group@optima.CS.Arizona.EDU
Subject: Jcon on windows NT? anybody running it on that platform?
Date: Tue, 1 Jun 1999 13:23:27 -0500 
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

I see from the webpage 
  http://www.cs.arizona.edu/icon/jcon/
that Jcon requires Unix and the Korn shell, Icon 9.3 and Java 1.1.

I am wondering if anyone has done the work to port Jcon to a Windows NT
platform so that one can run an executable on Windows NT to translate Icon
programs into Java classes, and, if so, what they did to achieve that feat.
Thank you.


From icon-group-sender  Thu Jun  3 16:45:03 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id QAA28313
	for icon-group-addresses; Thu, 3 Jun 1999 16:43:49 -0700 (MST)
Message-Id: <199906032343.QAA28313@baskerville.CS.Arizona.EDU>
Delivered-To: icon-group@silliac.cs.arizona.edu
Date: Thu, 03 Jun 1999 16:59:34 -0500
From: de yang <deyang@lucent.com>
To: icon-group@optima.CS.Arizona.EDU
Subject: Find the longest matching substrings
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Dear Sirs,

I am working on a project that will mark the inserted or deleted phases
between two strings.  I am wondering whether you know any library
procedure that will anchor ( or report ) the longest phases that exists
in both  strings.  

For example, assume that we have two strings:
    old:="other strings ab cd e and more substrings"
    new:="some more substrings ab cd ab cd e"

This function should return phases and position in both strings, i.e.
[15,  28, "ab cd e"]  for "ab cd e" is the longest phase that matched (
it has three continuous matching elements ).

Thanks,

De Yang
Lucent Technologies Inc.
deyang@lucent.com


From icon-group-sender  Fri Jun  4 08:01:40 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id IAA19355
	for icon-group-addresses; Fri, 4 Jun 1999 08:01:01 -0700 (MST)
Message-Id: <199906041501.IAA19355@baskerville.CS.Arizona.EDU>
Delivered-To: icon-group@silliac.cs.arizona.edu
From: swampler@noao.edu (Steve Wampler)
Date: Thu, 3 Jun 1999 21:17:38 MST
To: icon-group@optima.CS.Arizona.EDU
Subject: Re: Find the longest matching substrings
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

On Jun 3 at  4:59pm, de yang writes:
} Dear Sirs,
} 
} I am working on a project that will mark the inserted or deleted phases
} between two strings.  I am wondering whether you know any library
} procedure that will anchor ( or report ) the longest phases that exists
} in both  strings.  
} 
} For example, assume that we have two strings:
}     old:="other strings ab cd e and more substrings"
}     new:="some more substrings ab cd ab cd e"
} 
} This function should return phases and position in both strings, i.e.
} [15,  28, "ab cd e"]  for "ab cd e" is the longest phase that matched (
} it has three continuous matching elements ).

I've send a solution to De Yang, but it occurred to me that others
might like to see it (and improve on it - it's a bit, uh, ugly).
This solutions makes two assumptions:

   (1) There is a simple definition for what a phrase element is
       (in fact, it assumes an element is a sequence of letters!)

   (2) Punctuation should be considered when matching phrases, so
       "that that is is that that is not is not" and
       "that that is, is. that that is not, is not." do not
       match each other (instead, the longest common phrase in
       this case is "that that is not")

I have another solution that drops assumption 2, if any one is
interested.

Anyway, here's a solution, plus a simple test driver...

-----
procedure main(args)

    if result := longPhrase(read(),read()) then
        write("[",result[1],",",result[2],",\"",result[3],"\"]")

end

# Find longest phrase common to s1 and s2.  Fails if no common phrase.
#
#  Assumes simple definition for a phrase (sequence of letters)
#  Assumes punctuation is to be considered when matching phrases
#
procedure longPhrase(s1,s2)
    static pChars
    initial pChars := &letters		# chars legal in phrase element

    maxlen := 0

    every j := findPhrase(s1[ (i := pStart(s1,pChars)) :
                              (k := rpStop(s1,pChars))  ], s2, pChars) do {

        if maxlen <:= numPhrases(s1[i:k], pChars) then {
            write(s1[i:k])
            saveStart1 := i
            saveStop1  := k
            saveStart2 := j
            }

       }

    return [\saveStart1, saveStart2, s1[saveStart1:saveStop1]]
end

procedure pStart(s, pc)			# generate start of all phrases in s
    s ? {
        while tab(i := upto(pc)) do {
            tab(many(pc))
            suspend i
            }
        }
end

procedure rpStop(s, pc)			# generates end of all phrases in s,
                                        #   starting from right of string
    suspend *s+2 - pStart(reverse(s), pc)
end

procedure numPhrases(s, pc)		# Count number of phrases in s
    local count

    count := 0
    every pStart(s,pc) do count +:= 1

    return count
end

procedure findPhrase(p, s, pc)		# find phrase p in s
    s ? {
        while tab(i := upto(pc)) do {
            if match(p) then suspend i
            tab(many(pc))
            }
        }
end

-- 
--
Steve Wampler-  SOLIS Project, National Solar Observatory
swampler@noao.edu


From icon-group-sender  Fri Jun  4 08:03:30 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id IAA19581
	for icon-group-addresses; Fri, 4 Jun 1999 08:03:22 -0700 (MST)
Message-Id: <199906041503.IAA19581@baskerville.CS.Arizona.EDU>
Delivered-To: icon-group@silliac.cs.arizona.edu
Date: Fri, 4 Jun 99 00:51:50 -0500 (CDT)
From: gep2@terabites.com
Subject: Find the longest matching substrings
To: icon-group@optima.CS.Arizona.EDU
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

> I am working on a project that will mark the inserted or deleted phases
between two strings.  I am wondering whether you know any library
procedure that will anchor ( or report ) the longest phases that exists
in both  strings.  

I'm not aware of any such ready-made routine, although certainly "longest common 
substring" logic is frequently (and heavily!) used in file compression software 
(PKZIP for example).

> For example, assume that we have two strings:
    old:="other strings ab cd e and more substrings"
    new:="some more substrings ab cd ab cd e"

> This function should return phases and position in both strings, i.e.
[15,  28, "ab cd e"]  for "ab cd e" is the longest phase that matched (
it has three continuous matching elements ).

Intriguing that you consider the "three matching elements" phrase (only seven 
characters) 'longer' than the sixteen-character common phrase " more 
substrings".  I guess the criteria here could be most anything you want to make 
it, but still...  :-)

Generally the logic to do stuff like this (in good programs, anyhow) is coded at 
very low levels using very heavily optimized code... since it tends to be very, 
very compute-intensive (this is especially true as the two strings get longer).

Of course, the problem's solution might be VERY different depending on whether 
you're looking for something that's "elegant" in terms of the source code, 
versus something that's FAST, versus something that is storage-efficient, or 
something else.  (The rule about what substring is 'longer' also is going to 
have an effect too!).

If I were going to try to make something like this FAST (and especially if one 
of the strings were relatively constant) I'd probably want to use a hash (or 
even direct index?) table pre-computed with the offsets of character substrings 
(I'd probably use two-character adjacent pairs) in one of those strings;  this 
would enable stepping rapidly through the second string and quickly finding at 
least (only just the) plausible candidates for launching more detailed substring 
compare (and length computation) operations.  This kind of thing would likely be 
storage-hungry, but that's less of a concern today than it once was.

Gordon Peterson
http://web2.airmail.net/gep2/
Support the Anti-SPAM Amendment!  Join at http://www.cauce.org/
12/19/98: the day the Conservatives demonstrated their scorn for their
   fraudulent sham of representative government.  Voters, remember it!



From icon-group-sender  Fri Jun  4 14:45:35 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id OAA03881
	for icon-group-addresses; Fri, 4 Jun 1999 14:41:55 -0700 (MST)
Message-Id: <199906042141.OAA03881@baskerville.CS.Arizona.EDU>
Delivered-To: icon-group@silliac.cs.arizona.edu
Date: Fri, 04 Jun 1999 09:09:29 -0700
From: Steve Wampler <swampler@noao.edu>
X-Accept-Language: en
To: icon-group@optima.CS.Arizona.EDU
Subject: Re: Find the longest matching substrings
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

This is a multi-part message in MIME format.
--------------09EB4280837648B91A00FBDF
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

gep2@terabites.com wrote:
> 
> > I am working on a project that will mark the inserted or deleted phases
> between two strings.  I am wondering whether you know any library
> procedure that will anchor ( or report ) the longest phases that exists
> in both  strings.
> 
> > For example, assume that we have two strings:
>     old:="other strings ab cd e and more substrings"
>     new:="some more substrings ab cd ab cd e"
> 
> > This function should return phases and position in both strings, i.e.
> [15,  28, "ab cd e"]  for "ab cd e" is the longest phase that matched (
> it has three continuous matching elements ).
> 
> If I were going to try to make something like this FAST (and especially if one
> of the strings were relatively constant) I'd probably want to use a hash (or
> even direct index?) table pre-computed with the offsets of character substrings
> (I'd probably use two-character adjacent pairs) in one of those strings;  this
> would enable stepping rapidly through the second string and quickly finding at
> least (only just the) plausible candidates for launching more detailed substring
> compare (and length computation) operations.  This kind of thing would likely be
> storage-hungry, but that's less of a concern today than it once was.

Taking Gordon's comment on speed to heart, here's an adaptation of the
previous solution that is faster (uses simple heuristics to avoid
looking at phrases that can't possibly be longer than ones that are
found already).  Note that it can probably be sped up even more by
improving
these heuristics a bit.

This version is about 2.5x faster on my test case (the "that that is..."
nonsense).  Applying this adaptation to my solution that ignores
punctuation
when comparing phrases results in a speedup of about 25x!  (Because the
test case strings match each other when ignoring punctuation.)

Oh, I also modified the test driver to do time comparisions.

Incidently, it would be 'better' if the phrase element procedures were
converted into string scanning procedures.  Scanning strings for phrase
elements sounds useful to me and can considered as a generalization
of scanning strings for characters - in fact, the code here is just
an adaptation of code to find the longest common substring.

It would also be better if the code were rewritten to accept different
definitions of a phrase element more easily!

--
Steve Wampler-  SOLIS Project, National Solar Observatory
swampler@noao.edu
--------------09EB4280837648B91A00FBDF
Content-Type: text/plain; charset=us-ascii;
 name="phrase3.icn"
Content-Disposition: inline;
 filename="phrase3.icn"
Content-Transfer-Encoding: 7bit

procedure main(args)

    n := args[1]|10
    start := &time
    every 1 to n do {
        result := longPhrase(read(),read())
        }
    len := &time - start
    write("[",result[1],",",result[2],",\"",result[3],"\"]")
    write(real(len)/n,"ms per loop for ",n," loops")

end

# Find longest phrase common to s1 and s2.  Fails if no common phrase.
#
#  Assumes simple definition for a phrase (sequence of letters)
#  Assumes punctuation is to be considered when matching phrases
#
procedure longPhrase(s1,s2)
    static pChars
    initial pChars := &letters		# chars legal in phrase element

    maxlen := 0

    every i := pStart(s1, pChars) do {
        every k := rpStop(s1, i, pChars) do {
            if j := findPhrase(s1[i:k], s2, pChars) then {
                if maxlen <:= numPhrases(s1[i:k], pChars) then {
                    saveStart1 := i
                    saveStop1  := k
                    saveStart2 := j
                    }
                break			# no need to find shorter phrases
                }			#   that start here
            }

        }

    return [\saveStart1, saveStart2, s1[saveStart1:saveStop1]]
end

procedure pStart(s, pc)			# generate start of all phrases in s
    local i

    s ? {
        while tab(i := upto(pc)) do {
            tab(many(pc))
            suspend i
            }
        }
end

procedure rpStop(s, i, pc)		# generates end of all phrases in s,
                                        #   starting from right of string
    suspend *s+2 - pStart(reverse(s[i:0]), pc)
end

procedure numPhrases(s, pc)		# Count number of phrases in s
    local count

    count := 0
    every pStart(s,pc) do count +:= 1

    return count
end

procedure findPhrase(p, s, pc)		# find phrase p in s
    local i

    s ? {
        while tab(i := upto(pc)) do {
            if match(p) then suspend i
            tab(many(pc))
            }
        }
end

--------------09EB4280837648B91A00FBDF--



From icon-group-sender  Fri Jun  4 16:27:05 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id QAA11368
	for icon-group-addresses; Fri, 4 Jun 1999 16:26:34 -0700 (MST)
Message-Id: <199906042326.QAA11368@baskerville.CS.Arizona.EDU>
Delivered-To: icon-group@silliac.cs.arizona.edu
To: icon-group@optima.CS.Arizona.EDU
Date: 4 Jun 1999 22:45:41 GMT
From: guy@orion.math.tau.ac.il (Guy Laden)
Subject: parsing java .class files
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Hi all, does anybody have an icon library for reading java 
.class files that they are willing to share? (Jcon seems to 
come with one for writing .class files, not reading)
thanks,
guy


From icon-group-sender  Mon Jun  7 09:33:19 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id JAA23335
	for icon-group-addresses; Mon, 7 Jun 1999 09:31:20 -0700 (MST)
Message-Id: <199906071631.JAA23335@baskerville.CS.Arizona.EDU>
Delivered-To: icon-group@silliac.cs.arizona.edu
Date: Sat, 05 Jun 1999 14:42:52 -0500
From: "Charles Hethcoat" <CHETHCOA@oss.oceaneering.com>
To: icon-group@optima.CS.Arizona.EDU
Subject: File Translators in Icon
Content-Disposition: inline
X-MIME-Autoconverted: from quoted-printable to 8bit by baskerville.CS.Arizona.EDU id MAA04406
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

There always seems to be a nagging need to be able to translate between different application file formats.

The problem is that many of these file formats are incredibly complex.  I recently had the humbling experience of reviewing the definitions of MS Word 97 and RTF formats.  My idea was that it couldn't be so hard to write code to read or write these formats.  My hopes were dashed pretty quickly!

Here is an assortment of converters I have wished for recently

LaTeX to RTF:  A C version exists, but it hasn't worked for me yet.  (Gives me nonsense RTF.)  I think it requires considerable amounts of customization.

LaTeX to Word 95/97:  Haven't found one that seems worth the effort.

VRML to Encapsulated PostScript (EPS):  We use Pro/Engineer, which generates pretty good VRML.  The idea is to select your view, then generate a "still" for technical publication (preferably in B&W vector line art format).  The viewers I have seen will dump various raster formats, but there is no logical reason why a vector projection to the screen can't also be mapped to PostScript.

IGES to EPS.  (BTW, an IGES reader/checker/viewer would be very handy for troubleshooting this error-prone format.  A couple of commercial viewers exist, but they are expensive and don't make EPS.)

Mathcad to LaTeX.  The problem here is updating engineering reports with new data and re-evaluating all the formulas.  Mathcad is good at dynamically updating the math, but crappy at typesetting and page layout; LaTeX is the opposite.  How to get the benefits of both?

If anyone has code (or pointers thereto), I would like to know about it.  I also would be interested in knowing if any of the above-mentioned conversion tools would be of interest to any of you.  If there were some demand, I might be persuaded to go to work on one of them.

Charles Hethcoat



From icon-group-sender  Mon Jun  7 09:37:14 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id JAA23547
	for icon-group-addresses; Mon, 7 Jun 1999 09:37:04 -0700 (MST)
Message-Id: <199906071637.JAA23547@baskerville.CS.Arizona.EDU>
Delivered-To: icon-group@silliac.cs.arizona.edu
From: Art Eschenlauer <art.eschenlauer@sufsys.com>
To: "'guy@orion.math.tau.ac.il'" <guy@orion.math.tau.ac.il>
Cc: "'icon-group@optima.CS.Arizona.EDU'" <icon-group@optima.CS.Arizona.EDU>
Subject: RE: parsing java .class files
Date: Mon, 7 Jun 1999 10:27:01 -0500 
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Why not use a java decompiler?
I use JAD.EXE for Win NT/9x.
It is a fast, slick, one file, no-messing-with-the-registry C++ program
that converts .class files into .java files.
It can even convert .jars into the a java source tree.
You can find out more at
http://meurrens.ml.org/ip-Links/Java/CodeEngineering/jad15.html

-----Original Message-----
From: guy@orion.math.tau.ac.il [mailto:guy@orion.math.tau.ac.il]
Sent: Friday, June 04, 1999 5:46 PM
To: icon-group@optima.CS.Arizona.EDU
Subject: parsing java .class files


Hi all, does anybody have an icon library for reading java 
.class files that they are willing to share? (Jcon seems to 
come with one for writing .class files, not reading)
thanks,
guy


From icon-group-sender  Tue Jun  8 08:26:40 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id IAA08100
	for icon-group-addresses; Tue, 8 Jun 1999 08:25:37 -0700 (MST)
Message-Id: <199906081525.IAA08100@baskerville.CS.Arizona.EDU>
Delivered-To: icon-group@silliac.cs.arizona.edu
To: icon-group@optima.CS.Arizona.EDU
Date: 7 Jun 1999 20:03:40 GMT
From: guy@orion.math.tau.ac.il (Guy Laden)
Subject: RE: parsing Java .class files
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

> Why not use a java decompiler?

I plan to write some programs to analyze the bytecode: a decompiler 
won't help here.

Guy


From icon-group-sender  Tue Jun  8 08:27:08 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id IAA08145
	for icon-group-addresses; Tue, 8 Jun 1999 08:26:59 -0700 (MST)
Message-Id: <199906081526.IAA08145@baskerville.CS.Arizona.EDU>
Delivered-To: icon-group@silliac.cs.arizona.edu
To: icon-group@optima.CS.Arizona.EDU
Date: Tue, 08 Jun 1999 00:23:56 GMT
From: cbbrowne@news.isp.giganews.com (Christopher Browne)
Subject: Re: Find the longest matching substrings
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

On 4 Jun 1999 11:05:40 -0400, gep2@terabites.com <gep2@terabites.com> wrote:
>If I were going to try to make something like this FAST (and especially if one 
>of the strings were relatively constant) I'd probably want to use a hash (or 
>even direct index?) table pre-computed with the offsets of character substrings 
>(I'd probably use two-character adjacent pairs) in one of those strings;  this 
>would enable stepping rapidly through the second string and quickly finding at 
>least (only just the) plausible candidates for launching more detailed substring 
>compare (and length computation) operations.  This kind of thing would likely be 
>storage-hungry, but that's less of a concern today than it once was.

Alternatively, treat the main string as a semi-infinite string
(si-string), and insert pointers to each successive character of the
si-string into a Patricia tree.  Then do a tree walk to find the longest
matches that were found. 
-- 
"Unlike computers, guns don't have Y2K problems..."
cbbrowne@hex.net- <http://www.hex.net/~cbbrowne/computing.html>


From icon-group-sender  Tue Jun  8 13:30:37 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id NAA20978
	for icon-group-addresses; Tue, 8 Jun 1999 13:29:25 -0700 (MST)
Message-Id: <199906082029.NAA20978@baskerville.CS.Arizona.EDU>
Delivered-To: icon-group@silliac.cs.arizona.edu
Date: Tue, 08 Jun 1999 13:38:25 -0500
From: "Charles Hethcoat" <CHETHCOA@oss.oceaneering.com>
To: icon-group@optima.CS.Arizona.EDU
Subject: Fwd: Re: File Translators in Icon
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

This is a MIME message. If you are reading this text, you may want to 
consider changing to a mail reader or gateway that understands how to 
properly handle MIME multipart messages.

--=_4B1D2C2A.95F49ED1
Content-Type: text/plain; charset=US-ASCII
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

Anonymous <nobody@replay.com> wrote:

>Try Mathematica.

I'd love to give it a whirl.  If I could just get my PHB to spring the =
long green for it.  It's a little dear, and in that it conflicts with my =
innate cheapskate tendencies.  What the world needs here is a good 5c =
cigar with a public license!

>RTF is a monstrosity.

Beyond doubt.

>Compare the file sizes of RTF versus equivalent HTML or even Word.

I found that out the hard way.  One 30KB Word file grew to 650-odd KB when =
exported to RTF.  But you know, fter I learned a little bit about =
PostScript, I understood why so many automatically-generated PS files are =
also huge.  They basically carry a lot of baggage in the form of stock =
libraries of PS code that in many cases is not even used in a given =
graphic!  If you write your own and are careful, you can usually make it =
_much_ smaller.  A good example of a pretty efficient PS-generator is =
gnuplot.

I suspect the same thing happens with RTF.  The files I have inspected =
seem to contain, e.g. font data and other junk that has a lot of overhead. =
 Trouble is, I don't know enough about it to test that theory, and I'm not =
sure it's worth doing in the end.

Charles Hethcoat

--=_4B1D2C2A.95F49ED1
Content-Type: message/rfc822

Received: from mail.replay.com
	([194.109.9.44])
	by hougwia1.oceaneering.com; Mon, 07 Jun 1999 13:11:40 -0500
Received: (from remailer@localhost)
	by mail.replay.com (8.9.2/8.9.2/Replay Associates LLP) id UAA18278;
	Mon, 7 Jun 1999 20:11:34 +0200 (CEST)
Date: Mon, 7 Jun 1999 20:11:34 +0200 (CEST)
Message-Id: <199906071811.UAA18278@mail.replay.com>
From: Anonymous <nobody@replay.com>
Comments: This message did not originate from the Sender address above.
	It was remailed automatically by anonymizing remailer software.
	Please report problems or inappropriate use to the
	remailer administrator at <abuse@replay.com>.
Subject: Re: File Translators in Icon
To: CHETHCOA@oss.oceaneering.com
MIME-Version: 1.0

Try Mathematica.  A new version just came out (v.4).  It can handle all the math
typesetting you want.  Converts to EPS, HTML, TeX, etc.  Imports/exports a
variety of standard formats.  There is an addon that lets you run Mathematica
from within MS Word directly.
http://www.wolfram.com/

RTF is a monstrosity.  Compare the file sizes of RTF versus equivalent HTML or
even Word.


-----Original Message-----
From: Charles Hethcoat [mailto:CHETHCOA@oss.oceaneering.com]
Sent: Saturday, June 05, 1999 2:43 PM
To: icon-group@optima.CS.Arizona.EDU
Subject: File Translators in Icon


There always seems to be a nagging need to be able to translate between
different application file formats.


--=_4B1D2C2A.95F49ED1--


From icon-group-sender  Tue Jun  8 17:06:25 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id RAA00220
	for icon-group-addresses; Tue, 8 Jun 1999 17:05:52 -0700 (MST)
Message-Id: <199906090005.RAA00220@baskerville.CS.Arizona.EDU>
Delivered-To: icon-group@silliac.cs.arizona.edu
Date: Tue, 8 Jun 1999 15:41:22 -0500 (CDT)
From: "John C. Paolillo" <johnp@ling.uta.edu>
To: CHETHCOA@oss.oceaneering.com, icon-group@optima.CS.Arizona.EDU
Subject: Re: Fwd: Re: File Translators in Icon
Cc: john@ling.uta.edu
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

I don't know if this will help, but I once had reason to
write the following Icon program to assist in typesetting
Feature-Attribute matrices in MS Word.  In MS Word for mac
v 4.0 and 5.0, there is a family of mathematical typesetting
formulas (listed unde "Formula" in the user manuals) which
employ a special typesetting character (that turns out to be 
ASCII 006) as an escape character.  The formulas are acceptably
typeset (or they were for my purposes) but very cumbersome to
type.  I wrote this program to translate from a more transparent
syntax into the form actually used by MS Word.  As for newer
versions of MS Word, I don't know if you can type the formula
typesetting sequences directly (I couldn't figure out how to
do it), but my files with the complex formulas in them that
were created in MS Word v 4.0 still display correctly when
converted.  

I haven't tried this in a long time, so I am not 100% certain
if this is the best version.  There are probably improvements
that can be made to it at any rate.  Some of the typesetting
that is possible (e.g. integrals) are not implemented, but that
could easily be added by referring to the appropriate pages in 
the Word manuals.  The program was written to be used as a 
clipboard filter with the macintosh interpreter IClip.

John C. Paolillo
Program in Linguistics
University of Texas at Arlington

##########################################################################
#
#	This program parses a description of a attribute-value matrix using
#	ordinary ascii characters (using upper and lowercase letters for 
#	feature values and labels, [] to delimit matrices, {} to delimit sets
#	and <> to delimit lists), and outputs a string of the appropriate
#	description, using MS Word's formula typesetting expressions.  The
#	resulting string, when pasted into a Word document, will appear in
#	the correctly typeset form.  For example, the following description
#
#	[	[	this	|	is	]
#		[	an		|	example 
#			of		|	typesetting]	]
#
#	Will be typeset as below (ascii impression).
#
#		+-							 -+
#		|	[ this | is ]			  |
#		|	+-					   -+ |
#		|	|	an | example		| |
#		|	|	of | typesetting	| |
#		|	+-					   -+ |
#		+-							 -+
#
#	The program will parse as many such expressions as are in the input,
#	until it encounters a non well-formed description.  All remaining
#	input (well formed or not) is discarded.  
#
########################################################################

global white_sp			#	white space character set

procedure main()
	white_sp := ' \t'		#	white space is tab and space
	line := ""
	while line ||:= read()	#	concatenate input into one long string
							#	(removes return/new line characters)
	line ?					#	and scan the line for parsing
		{
			while a_parse := parse() do
				{			#	write each parse as a separate line
					write(lst_2_str(a_parse))
				}
		}
end

#	This procedure converts an input list (a parse) into a formatted
#	string.  Many variations are possible.  This one replaces the 
#	string, list and set delimiter characters with the appropriate
#	typesetting commands.  The "" character below is ascii 006, Word's
#	formula typesetting command character.  Using backslash to represent
#	this character, the strings replacing the delimiters are:
#
#		[	\b\bc\[(\a\al(		# begin a bracket delimited by "[]"
#								# with a left-alligned array as argument
#
#		<	\b\bc\<(\a\ac(		# begin a bracket delimited by "<>"
#								# with a center-alligned array as argument
#
#		{	\b\bc\[(\a\ac(		# begin a bracket delimited by "{}"
#								# with a center-alligned array as argument
#
#		]	))					# close two argument lists
#		>	))
#		}	))
#	
procedure lst_2_str(alist)
	if x := string(alist) then	#	don't convert it if its already a string
		return case x of
			{					#	do the above replacements
				"["		:	"bbc[(aal("
				"<"		:	"bbc<(aac("
				"{"		:	"bbc{(aac("
				"]"		:	"))"
				">"		:	"))"
				"}"		:	"))"
				default	:	x	#	default to the incoming value
			}
	else						#	if it's a list then convert it
		{
			mid_str := ""		#	mnemonically "middle string"
			every x := !alist do
				{				#	iterate through the arguments
					mid_str ||:= lst_2_str(x)	#	concatenate
					mid_str ||:= " "			#	pad with space
				}
			return mid_str[1:-1]	#	don't include the last space
		}
end

#	the parser looks for each of the following possibilities
#	label	-- a string of letters
#	expr	-- a label followed by a value
#	box		-- labels with values inside []
#	l_list	-- boxes grouped inside <>
#	set		-- boxes grouped inside {}
procedure parse()
	suspend label() | expr() | box() | set() | l_list()
end

#	a label is a bunch of letters together with no whitespace or punctuation
#	no digits are allowed
procedure label()
	suspend tab(many(&letters))
end

#	an expression is a label with a value.  The value may be
#	a label (separated by "|"), a box, a set a list or another expression
procedure expr()
	suspend [ one_of(label) ] |||	(	[ ="|", one_of(label) ] |
										[ one_of( box | set | l_list | expr ) ] )
end

#	a box is some arbitrary number of expressions enclosed in []
#	the parsed expressions will have "," items between them in the parse
#	so that Word will know to typeset them vertically in the array.  
#	DO NOT use the commas in your description, it will not parse if you do.
#	box()es must have at least one exprtession in them
procedure box()
	suspend [ ="[" ] ||| com_arbno(expr) ||| [ ="]" ]
end

#	a set is an arbitrary number of boxes enclosed in {} -- the same
#	caveats apply as for box()
procedure set()
	suspend [ ="{" ] ||| com_arbno(box) ||| [ ="}" ]
end

#	a l_list is an arbitrary number of boxes enclosed by <>, with no commas
#	between them so Word will know to typeset them horizontally.  There
#	may be zero items in a l_list.  
procedure l_list()
	suspend [ ="<" ] ||| arbno(box) ||| [ =">" ]
end

#	com_abrno() produces a list containing an arbitrary number of 
#	some matching expression p, separated by one fewer "," items
#	on the list than there are p's.  This is for things that need to
#	be typeset vertically by Word.  The base case is a single p, with
#	no commas in the putput list.
procedure com_arbno(p)
	suspend [one_of(p)] ||| ( [] | ( [ "," ] ||| arbno(p) ) )
end

#	arbno() produces a list containing an arbitrary number of some
#	matching procedure p.  This is for things that need to be typeset 
#	horizontally.  The base case is an empty list
procedure arbno(p)
	suspend [] | [one_of(p)] ||| arbno(p)
end

#	one_of() produces a single match of the expression p, discarding
#	any surrounding whitespace it encounters.  This is a useful technique
#	to use in a parser, with a little modification, it can also be
#	used to get optional constituents.
procedure one_of(p)
	tab(many(white_sp)) 
	x <- p()
	tab(many(white_sp))
	suspend \x
end



From icon-group-sender  Thu Jun 10 16:55:22 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id QAA03665
	for icon-group-addresses; Thu, 10 Jun 1999 16:54:23 -0700 (MST)
Message-Id: <199906102354.QAA03665@baskerville.CS.Arizona.EDU>
Delivered-To: icon-group@silliac.cs.arizona.edu
From: Art Eschenlauer <art.eschenlauer@sufsys.com>
To: icon-group@optima.CS.Arizona.EDU
Subject: Syntax coloring .vim file for editing .icn programs with vim
Date: Thu, 10 Jun 1999 14:22:05 -0500
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

I like vim for editing code.

It allows you to syntax color, 
using .vim files stored in the syntax directory.

Has anyone written one for .icn files.
I'm sure that I can do it, but I have a lot of
other wheels to reinvent already....

Thanks


From icon-group-sender  Fri Jun 11 08:37:47 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id IAA15033
	for icon-group-addresses; Fri, 11 Jun 1999 08:37:00 -0700 (MST)
Message-Id: <199906111537.IAA15033@baskerville.CS.Arizona.EDU>
Delivered-To: icon-group@silliac.cs.arizona.edu
To: icon-group@optima.CS.Arizona.EDU
Date: 11 Jun 1999 12:48:53 GMT
From: espie@liafa.jussieu.fr (Marc Espie)
Subject: Re: Syntax coloring .vim file for editing .icn programs with vim
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

In article <199906102354.QAA03665@baskerville.CS.Arizona.EDU>,
Art Eschenlauer  <art.eschenlauer@sufsys.com> wrote:
>I like vim for editing code.
>
>It allows you to syntax color, 
>using .vim files stored in the syntax directory.
>
>Has anyone written one for .icn files.
>I'm sure that I can do it, but I have a lot of
>other wheels to reinvent already....
This is definitely *not* perfect, as this was just a quick hack, copied over
from other syntax files.

If you make improvements to that small file, I'd like a copy :)


" Vim syntax file
" Language:	Icon

" Remove any old syntax stuff hanging around
syn clear

syn case ignore

syn keyword iconBlock procedure end link
syn keyword iconLabel		case of goto default
syn keyword iconConditional	if else then
syn keyword iconRepeat		do every while repeat
syn keyword iconStatement	return next suspend

syn keyword iconTodo contained	TODO

" String
syn region  iconString	start=+"+  end=+"+

syn match  iconMatrixDelimiter	"[][]"

syn match  iconNumber		"-\=\<\d\+\>"

syn match	iconComment		"#.*$" contains=iconTodo

"syn match iconKeyword		"&(null|[lrm]press|resize|fail)"
syn match iconKeyword		"&null"
syn match iconKeyword		"&fail" 
syn match iconKeyword		"&shift"
syn match iconKeyword		"&[lmr]press"
syn match iconKeyword		"&resize"
syn match iconKeyword		"&window"
syn match iconKeyword		"&time"
syn keyword iconFunction	read write reads writes open close
syn keyword iconFunction	put push pop pull
syn keyword iconFunction	insert member key tab set
syn keyword iconFunction	many any upto move table list trim find
syn keyword iconFunction	match bal
syn keyword iconFunction	sin cos dtor rtod
syn keyword iconFunction	integer string left right map image
syn keyword iconOperator	to by not 

syn keyword iconType	static initial global record

syn sync lines=250

if !exists("did_icon_syntax_inits")
  let did_icon_syntax_inits = 1
  " The default methods for highlighting.  Can be overridden later
  hi link iconBlock		PreProc
  hi link iconKeyword		PreProc
  hi link iconStatement		Statement
  hi link iconLabel			Label
  hi link iconConditional		Conditional
  hi link iconRepeat			Repeat
  hi link iconTodo			Todo
  hi link iconString			String
  hi link iconMatrixDelimiter		Identifier
  hi link iconNumber			Number
  hi link iconByte			Number
  hi link iconOperator			Operator
  hi link iconFunction			Function
  hi link iconType			Type
  hi link iconComment			Comment
  hi link iconStatement		Statement

"optional highlighting
  "hi link iconDelimiter		Identifier

  "hi link iconShowTab			Error
  "hi link iconShowTabc		Error

  "hi link iconIdentifier		Identifier
endif

let b:current_syntax = "icon"

" vim: ts=8


-- 
	Marc Espie		
|anime, sf, juggling, unicycle, acrobatics, comics...
|AmigaOS, OpenBSD, C++, perl, Icon, PostScript...
| `real programmers don't die, they just get out of beta'


From icon-group-sender  Mon Jun 14 12:47:18 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id MAA04250
	for icon-group-addresses; Mon, 14 Jun 1999 12:45:25 -0700 (MST)
Message-Id: <199906141945.MAA04250@baskerville.CS.Arizona.EDU>
Delivered-To: icon-group@silliac.cs.arizona.edu
Date: Mon, 14 Jun 1999 16:51:45 +0100
From: "C.R. Smith" <m7crs@csc.liv.ac.uk>
To: icon-group@optima.CS.Arizona.EDU
Subject: help with vib
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Hi

I wonder if you could possible help me.

I am a student studying computer science at Liverpool University in the
UK. I have been given a project to do concerning creating a students
admission procedure amd linking to a database. Our supervisor informs us
that we should use the VIB tool to build the interface. 



My problem is that although I have bought the Icon book "The Icon
Programming Language" and have ordered the "Graphics programming in
Icon" i am struggling for information especially small simple programs
using VIB. I like to use VIB i find it very easy to create nice looking
interface.s
My problem is linking code to the VIB generated source code. For example
a have creaated a simple interface that opens and displays a file.
However although I have used to code supplied in the book I cannot get
it to save the displayed file as another name. Also I am unsure of how I
should print a file - (program must work on unix and win95.

To get to the crux of the matter what I am asking is if you could send
me any simple programs which uses the VIB interface as this would help
me enormously. I have searched th online documentation but cannot find
any such programs.

Thanks

christine smith


From icon-group-sender  Tue Jun 15 16:21:59 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id QAA12496
	for icon-group-addresses; Tue, 15 Jun 1999 16:20:36 -0700 (MST)
Message-Id: <199906152320.QAA12496@baskerville.CS.Arizona.EDU>
Delivered-To: icon-group@silliac.cs.arizona.edu
Date: Tue, 15 Jun 1999 16:41:16 -0300 (ADT)
From: Larry Bezeau <bezeau@unb.ca>
X-Sender: bezeau@sol.sun.csd.unb.ca
To: Icon list <icon-group@optima.CS.Arizona.EDU>
Subject: Problems with noncase.icn from program library
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

      I have recently been using the Icon program library for
dealing with strings using regular expressions and have run into
a few problems, not with the regular expression procedures
themselves, but with a set of procedures called "noncase.icn",
used for case insensitive searching with regular expressions.
An example follows.

      The following procedure from noncase.icn is intended to
duplicate find in a case insensitive way.

procedure c_find(s1,s2,i1,i2)
  local scanPos,endPos
  scanPos := match("",s2,i1,i2)
  endPos := many(&cset,s2,i1,i2)
  suspend scanPos - 1 + find(map(s1),
      map((if \s2 then s2 else &subject)[scanPos:endPos]))
end

A problem occurs if the subject string is empty.  In this case,
the statement to set the value of endPos fails leaving endPos
undefined.  The last statement produces a program error when
an attempt is made to extract a substring of &subject with
one index having the null value.  The analogous built-in function
(find) simply fails under the same circumstances.

        My solution was to replace the fourth line in the procedure
with
          endPos := (many(&cset,s2,i1,i2) | scanPos)

which at least prevents the program error.  I have not tested this
rigorously.  Can anyone see problems with this or make other
suggestions?
                                      Larry   Bezeau@UNB.Ca



From icon-group-sender  Wed Jun 16 08:47:44 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id IAA11063
	for icon-group-addresses; Wed, 16 Jun 1999 08:47:10 -0700 (MST)
Message-Id: <199906161547.IAA11063@baskerville.CS.Arizona.EDU>
Delivered-To: icon-group@silliac.cs.arizona.edu
Date: Wed, 16 Jun 1999 05:32:12 -0700 (PDT)
From: Arthur Eschenlauer <eschen42@yahoo.com>
Subject: algebraic to spreadsheet translator (was File Translators in Icon)
To: icon-group@optima.CS.Arizona.EDU
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

--0-1714636915-929536332=:16201
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline

> -----Original Message-----
> From: John C. Paolillo
> Sent: Tuesday, June 08, 1999 3:41 PM
> To: CHETHCOA@oss.oceaneering.com; icon-group@optima.CS.Arizona.EDU
> Cc: john@ling.uta.edu
> Subject: Re: Fwd: Re: File Translators in Icon
>
> I don't know if this will help, but I once had reason to
> write the following Icon program to assist in typesetting
> Feature-Attribute matrices in MS Word.  . . .

Along the same lines, I once wrote a program (see attachment)
to be able to write reasonably algebraic expressions that could
then be translated into spreadsheet formulas, which I fid a
lot more difficult to read and understand.


===
Arthur Eschenlauer (eschen@alumni.princeton.edu)
When possible, please direct business email to: 
Art.Eschenlauer@sufsys.com
_________________________________________________________
Do You Yahoo!?
Get your free @yahoo.com address at http://mail.yahoo.com

--0-1714636915-929536332=:16201
Content-Type: application/x-unknown-content-type-icn_auto_file; name="spread.icn"
Content-Description: spread.icn
Content-Disposition: attachment; filename="spread.icn"
Content-Transfer-Encoding: base64
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--0-1714636915-929536332=:16201--


From icon-group-sender  Wed Jun 16 12:30:54 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id MAA19875
	for icon-group-addresses; Wed, 16 Jun 1999 12:30:23 -0700 (MST)
Message-Id: <199906161930.MAA19875@baskerville.CS.Arizona.EDU>
Delivered-To: icon-group@silliac.cs.arizona.edu
From: "Steven Gross" <ssgross@netvision.net.il>
To: "Larry Bezeau" <bezeau@unb.ca>,
        "Icon list" <icon-group@optima.CS.Arizona.EDU>
Subject: Re: Problems with noncase.icn from program library
Date: Wed, 16 Jun 1999 19:50:35 +0200
X-Priority: 3
X-MSMail-Priority: Normal
X-MimeOLE: Produced By Microsoft MimeOLE V4.72.3110.37
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

The function c_match in the same library seems to have a typo in 
scanPos := match("",s,i1,i2)
since s is not defined in the procedure.



From icon-group-sender  Wed Jun 16 16:23:11 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id QAA28885
	for icon-group-addresses; Wed, 16 Jun 1999 16:21:57 -0700 (MST)
Message-Id: <199906162321.QAA28885@baskerville.CS.Arizona.EDU>
Delivered-To: icon-group@silliac.cs.arizona.edu
Date: Wed, 16 Jun 1999 17:49:55 -0300 (ADT)
From: Larry Bezeau <bezeau@unb.ca>
X-Sender: bezeau@sol.sun.csd.unb.ca
To: Icon list <icon-group@optima.CS.Arizona.EDU>
Subject: Re: Problems with noncase.icn from program library
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

On Wed, 16 Jun 1999, Steven Gross wrote:

>The function c_match in the same library seems to have a typo in 
>scanPos := match("",s,i1,i2)
>since s is not defined in the procedure.
>
        This problem will be flagged if noncase.icn is compiled
with the -u option.  The identifier "s" is neither defined in
the procedure nor global.  I ended up replacing it with s2, but
I don't believe my program actually uses that particular 
procedure so I haven't verified my change.
 
                                  Larry  Bezeau@UNB.Ca



From icon-group-sender  Fri Jun 18 08:19:42 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id IAA13448
	for icon-group-addresses; Fri, 18 Jun 1999 08:17:35 -0700 (MST)
Message-Id: <199906181517.IAA13448@baskerville.CS.Arizona.EDU>
Delivered-To: icon-group@silliac.cs.arizona.edu
To: icon-group@optima.CS.Arizona.EDU
Date: Fri, 18 Jun 1999 12:12:34 GMT
From: "Ian Trudel" <ian@monk.cgocable.ca>
Subject: Icon implementation detail: run time type convertion?
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Hi,

I would like to know how the type convertion is applied on compiled programs
(binary file, such .exe on WinNT) from Icon. Actually, I know that the type
convertion is made in run time. But I cannot figure out how they generate
code for it. Is the run time system included in every executables?

Ian




From icon-group-sender  Tue Jun 22 09:53:00 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id JAA25306
	for icon-group-addresses; Tue, 22 Jun 1999 09:48:05 -0700 (MST)
Message-Id: <199906221648.JAA25306@baskerville.CS.Arizona.EDU>
From: Kostas Oikonomou <oikonomou@att.com>
Date: Mon, 21 Jun 1999 17:37:20 -0400 (EDT)
To: icon-group@optima.CS.Arizona.EDU
Subject: Assertions in Icon
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO


Hello,

Does anyone have a suggestion about how to implement a procedure "assert",
which, when called with an expression "e" as argument, implements

		e | stop("Assertion on line ", &line, " failed!")

Here &line should be the line on which assert(e) appears.
Thanks.

				Kostas Oikonomou
 


From icon-group-sender  Tue Jun 22 12:32:44 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id MAA00996
	for icon-group-addresses; Tue, 22 Jun 1999 12:32:19 -0700 (MST)
Message-Id: <199906221932.MAA00996@baskerville.CS.Arizona.EDU>
Date: Tue, 22 Jun 1999 10:22:34 -0700
From: Steve Wampler <swampler@noao.edu>
X-Accept-Language: en
To: Kostas Oikonomou <oikonomou@att.com>, icon-group@optima.CS.Arizona.EDU
Subject: Re: Assertions in Icon
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Kostas Oikonomou wrote:
> 
> Hello,
> 
> Does anyone have a suggestion about how to implement a procedure "assert",
> which, when called with an expression "e" as argument, implements
> 
>                 e | stop("Assertion on line ", &line, " failed!")
> 
> Here &line should be the line on which assert(e) appears.
> Thanks.

Hard!

As far as I know, there is no easy way to get the line number from
which a procedure was called (too bad, the interpreter must have this
information around or it couldn't generate the traceback messages).
Maybe someone knows of a way to get to this information (I assume it
can be an expensive solution, since the program is going to stop
immediately afterwards)?

About the best I can is to have the call pass down the current line
number, as shown in the following example:

----
procedure main()
    assert{3 > 5, &line}
end

procedure assert(e)
    if not @e[1] then {
        stop(&errout, "Assertion failed on line ", @e[2] | "unknown")
        }
end
----

Note that the call to assert is really a PDCO (Programmer Defined
Control Operation)
to delay evaluation of the expression until inside the assert procedure.




--
Steve Wampler-  SOLIS Project, National Solar Observatory
swampler@noao.edu


From icon-group-sender  Tue Jun 22 12:33:06 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id MAA01037
	for icon-group-addresses; Tue, 22 Jun 1999 12:32:55 -0700 (MST)
Message-Id: <199906221932.MAA01037@baskerville.CS.Arizona.EDU>
From: dgamey@ca.ibm.com
X-Lotus-FromDomain: IBMCA@IBMUS
To: Kostas Oikonomou <oikonomou@att.com>
cc: icon-group@optima.CS.Arizona.EDU
Date: Tue, 22 Jun 1999 13:41:34 -0400
Subject: Re: Assertions in Icon
Content-Disposition: inline
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO



>Does anyone have a suggestion about how to implement a procedure "assert",
>which, when called with an expression "e" as argument, implements
>
>         e | stop("Assertion on line ", &line, " failed!")
>
>Here &line should be the line on which assert(e) appears.


I don't think there is a way to get your callers &line, unless you provide it as
a parameter to each call.  There may be some other possibilities but they seem
messy to me.

Another route is to use runerr() instead of stop which will cause a program
termination with a traceback.   Not as pretty but serviceable.

Hope that helps.

David




From icon-group-sender  Tue Jun 22 12:34:32 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id MAA01135
	for icon-group-addresses; Tue, 22 Jun 1999 12:34:24 -0700 (MST)
Message-Id: <199906221934.MAA01135@baskerville.CS.Arizona.EDU>
To: icon-group@optima.CS.Arizona.EDU
Date: Tue, 22 Jun 1999 19:05:11 GMT
From: "Ian Trudel" <ian@monk.cgocable.ca>
Subject: Re: Assertions in Icon
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Well, I guess I cannot answer properly to this question since I am not an
Icon programmer. Though, I am studying Icon's implementation and I can tell
you a bit about. First of all, I know that you can use &trace diagnostic
facility for tracing. You could certainly assign a non-zero value to &trace
before evaluate "e" and put it back to zero after (yeah, if it fails, you'd
still have to halt the program). Anyway, Icon implementation is very
straightforward and at worst case, you could implement an assertion
function.

Ian

Kostas Oikonomou wrote in message
<199906221648.JAA25306@baskerville.CS.Arizona.EDU>...
:
:Hello,
:
:Does anyone have a suggestion about how to implement a procedure "assert",
:which, when called with an expression "e" as argument, implements
:
: e | stop("Assertion on line ", &line, " failed!")
:
:Here &line should be the line on which assert(e) appears.
:Thanks.
:
: Kostas Oikonomou
:
:




From icon-group-sender  Wed Jun 23 09:06:16 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id JAA10769
	for icon-group-addresses; Wed, 23 Jun 1999 09:05:33 -0700 (MST)
Message-Id: <199906231605.JAA10769@baskerville.CS.Arizona.EDU>
Date: Wed, 23 Jun 1999 12:31:37 +1200 (NZST)
From: "Richard A. O'Keefe" <ok@atlas.otago.ac.nz>
To: dgamey@ca.ibm.com, oikonomou@att.com
Subject: Re: Assertions in Icon
Cc: icon-group@optima.CS.Arizona.EDU
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

	>Does anyone have a suggestion about how to implement a procedure "assert",
	>which, when called with an expression "e" as argument, implements
	>
	>         e | stop("Assertion on line ", &line, " failed!")
	>
	>Here &line should be the line on which assert(e) appears.
	
	
The obvious answer is "do it the way C does; use a preprocessor".
In what follows, I'm going to use M4, because
 - it's the traditional UNIX macro processor
 - there are two free versions: GNU M4 and pdm4
 - I know it fairly well.
 
Method 1.
 
    define(assert, `real_assert((`$1'), &line)')

where the real_assert procedure is plain Icon code.    
 
Method 2.

    define(assert, `((`$1') | stop("Assertion on line ", &line, " failed!"))')

where now everything is in the macro.

That's the quick and easy way.  A better way, long term, would be for
someone to patch the Icon translator to support `assert'.  It's *FAR*
too powerful a debugging tool to leave out of the language much longer.




From icon-group-sender  Wed Jun 23 09:06:42 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id JAA10826
	for icon-group-addresses; Wed, 23 Jun 1999 09:06:30 -0700 (MST)
Message-Id: <199906231606.JAA10826@baskerville.CS.Arizona.EDU>
To: icon-group@optima.CS.Arizona.EDU
Date: 23 Jun 1999 10:17:28 GMT
From: eddie@holyrood.ed.ac.uk (Eddie Corns)
Subject: Re: Assertions in Icon
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Kostas Oikonomou <oikonomou@att.com> writes:


>Hello,

>Does anyone have a suggestion about how to implement a procedure "assert",
>which, when called with an expression "e" as argument, implements

>		e | stop("Assertion on line ", &line, " failed!")

>Here &line should be the line on which assert(e) appears.
>Thanks.

>				Kostas Oikonomou
> 

Easiest soln which occurs to me is to use a preprocessor to convert

	assert(e)

to the form you want.  Either inline it or replace with do_assert(e,nnn) where
nnn is the line no.

I usually use symobolic tags like

	assert(e,"RDF4")

then use a file that maps tags to meaningful messages like "Read failed in
Open Directory".  (Well to be truthful, most of the time I just print the tag
and check it in the source).

Eddie


From icon-group-sender  Thu Jun 24 17:11:15 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id RAA07451
	for icon-group-addresses; Thu, 24 Jun 1999 17:10:04 -0700 (MST)
Message-Id: <199906250010.RAA07451@baskerville.CS.Arizona.EDU>
From: Art Eschenlauer <art.eschenlauer@sufsys.com>
To: icon-group@optima.CS.Arizona.EDU
Subject: Using Icon to execute SQL queries?
Date: Thu, 24 Jun 1999 16:48:28 -0500
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Has anyone had experience using Icon to execute SQL queries?

I am imaging an external C function that would be invoked
   sql(query)
where query would be a string with the SQL query.

I would imagine that it would produce a list of N lists 
each of which has M members, where N would be the 
number of rows selected and M would be the number 
of columns specified in the SQL query.

Has anyone out there tried something like this?


From icon-group-sender  Wed Jul  7 08:11:25 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id IAA00687
	for icon-group-addresses; Wed, 7 Jul 1999 08:09:25 -0700 (MST)
Message-Id: <199907071509.IAA00687@baskerville.CS.Arizona.EDU>
From: "Bob Alexander" <bobalex@ix.netcom.com>
To: "Icon Project" <icon-project@optima.CS.Arizona.EDU>,
        "Icon Group" <icon-group@optima.CS.Arizona.EDU>
Subject: noncase.icn
Date: Sat, 3 Jul 1999 19:28:57 -0700
X-Priority: 3
X-MSMail-Priority: Normal
X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2615.200
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

This is a multi-part message in MIME format.

------=_NextPart_000_001B_01BEC58A.4BC65380
Content-Type: multipart/alternative;
	boundary="----=_NextPart_001_001C_01BEC58A.4BCDF4A0"


------=_NextPart_001_001C_01BEC58A.4BCDF4A0
Content-Type: text/plain;
	charset="Windows-1252"
Content-Transfer-Encoding: quoted-printable

There were a few messages recently mentioning bugs in the IPL module =
procs/noncase.icn. Sorry I've taken so long to respond to this (I'm the =
author).

I had stumbled across and corrected those same bugs a while back (Jan =
'98, if my source code comments are to be believed), but it looks like I =
never submitted the corrected version to the Icon Project. The changes I =
made were virtually identical to the suggestions made in this forum. The =
updated file is attached -- it's small (Icon Project, can you make this =
update to the library?).

-- Bob

------=_NextPart_001_001C_01BEC58A.4BCDF4A0
Content-Type: text/html;
	charset="Windows-1252"
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META content=3D"text/html; charset=3Dwindows-1252" =
http-equiv=3DContent-Type>
<META content=3D"MSHTML 5.00.2614.3500" name=3DGENERATOR>
<STYLE></STYLE>
</HEAD>
<BODY bgColor=3D#ffffff>
<DIV><FONT size=3D2>There were a few messages recently mentioning bugs =
in the IPL=20
module procs/noncase.icn. Sorry I've taken so long to respond to this =
(I'm the=20
author).</FONT></DIV>
<DIV>&nbsp;</DIV>
<DIV><FONT size=3D2>I had stumbled across and corrected those same bugs =
a while=20
back (Jan '98, if my source code comments are to be believed), but it =
looks like=20
I never submitted the corrected version to the Icon Project. The changes =
I made=20
were virtually identical to the suggestions made in this forum. The =
updated file=20
is attached -- it's small (Icon Project, can you make this update to the =

library?).</FONT></DIV>
<DIV>&nbsp;</DIV>
<DIV><FONT size=3D2>-- Bob</FONT></DIV></BODY></HTML>

------=_NextPart_001_001C_01BEC58A.4BCDF4A0--

------=_NextPart_000_001B_01BEC58A.4BC65380
Content-Type: text/icon;
	name="noncase.icn"
Content-Transfer-Encoding: quoted-printable
Content-Disposition: attachment;
	filename="noncase.icn"
Content-Transfer-Encoding: quoted-printable

#########################################################################=
###
#
#	File:     noncase.icn
#
#	Subject:  Procedures for case-independent matching
#
#	Author:   Robert J. Alexander
#
#	Date:     January 9, 1998
#
#########################################################################=
###
#
#   This file is in the public domain.
#
#########################################################################=
###
#
# Kit of case-independent versions of Icon's built-in string-analysis
# procedures.
#
#########################################################################=
###

procedure c_find(s1,s2,i1,i2)
  local scanPos,endPos
  scanPos :=3D match("",s2,i1,i2)
  endPos :=3D many(&cset,s2,i1,i2) | scanPos
  suspend scanPos - 1 + find(map(s1),
      map((if \s2 then s2 else &subject)[scanPos:endPos]))
end

procedure c_many(c,s,i1,i2)
  return many(c_cset(c),s,i1,i2)
end

procedure c_match(s1,s2,i1,i2)
  local scanPos,endPos
  scanPos :=3D match("",s2,i1,i2)
  endPos :=3D scanPos + *s1
  return (map(s1) =3D=3D map((if \s2 then s2 else =
&subject)[scanPos:endPos]),endPos)
end

procedure c_upto(c,s,i1,i2)
  suspend upto(c_cset(c),s,i1,i2)
end

procedure c_cset(c)
  static lstring,ustring
  initial {
    lstring :=3D string(&lcase)
    ustring :=3D string(&ucase)
  }
  return cset(map(c) || map(c,lstring,ustring))
end

------=_NextPart_000_001B_01BEC58A.4BC65380--



From icon-group-sender  Wed Jul  7 08:11:58 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id IAA00817
	for icon-group-addresses; Wed, 7 Jul 1999 08:11:48 -0700 (MST)
Message-Id: <199907071511.IAA00817@baskerville.CS.Arizona.EDU>
From: Tim Patti <timp@clarityconnect.com>
To: icon-group@optima.CS.Arizona.EDU
Subject: Dynamic function loading on Linux
Date: Mon, 5 Jul 1999 12:21:33 -0500
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

	I'm looking for a simple solution to using dynamic loading on Linux.
In order to get the test program in icon/tests/calling/loadfunc to work, I've
got to remove or rename icon/src/h/features.h.

What seems to be happening is -

1.  bitcount.c includes rt.h
2.  icon/src/h/rt.h gets included, which includes
3.  icon/src/h/sys.h which includes
4.  /usr/local/ctype.h which wants to include /usr/local/features.h
5.  because ISRCH in the Makefile points to icon/src/h/ the
     icon/src/h/features.h file gets included instead, breaking the
     compile.

	I created a icon/src/h/dynamic_linux  directory that  dosen't have 
features.h in it that I can point ISRCH to in the Makefile but, I have to
edit a lot of the header files.  

The includes in h/rt.h are written as  #include "../h/define.h",
which I changed to  #include "define.h" in order to make it  work
in a different directory.  Is there an easier way of dealing with this ?

	Tim 



From icon-group-sender  Wed Jul  7 13:21:57 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id NAA21228
	for icon-group-addresses; Wed, 7 Jul 1999 13:21:34 -0700 (MST)
Message-Id: <199907072021.NAA21228@baskerville.CS.Arizona.EDU>
Date: Wed, 7 Jul 1999 13:05:21 -0300 (ADT)
From: Larry Bezeau <bezeau@unb.ca>
X-Sender: bezeau@sol.sun.csd.unb.ca
To: Icon list <icon-group@optima.CS.Arizona.EDU>
Subject: Re: Problems with noncase.icn from program library
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

        The postings from Nevin Liber and later Bob Alexander 
concerning fixes for the problems in the Icon library procedure
set "noncase.icn" were certainly welcome.  Nevin Liber's fix
worked well in my application.  I have just tested Bob Alexander's
fix and it also works except for the omission of the procedure
entitled "c_any" from the new noncase set.  This procedure did
not require a fix and so I pasted the original version into
the new noncase.icn and that seems to work fine.

                                   Larry     Bezeau@UNB.Ca



From icon-group-sender  Thu Jul  8 08:19:06 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id IAA23530
	for icon-group-addresses; Thu, 8 Jul 1999 08:17:30 -0700 (MST)
Message-Id: <199907081517.IAA23530@baskerville.CS.Arizona.EDU>
Date: Wed, 7 Jul 1999 17:55:25 -0700 (MST)
From: Gregg Townsend <gmt@baskerville.CS.Arizona.EDU>
To: icon-group@optima.CS.Arizona.EDU, timp@clarityconnect.com
Subject: Re: Dynamic function loading on Linux
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

    From: Tim Patti <timp@clarityconnect.com>
    
    	I'm looking for a simple solution to using dynamic loading on Linux.
    In order to get the test program in icon/tests/calling/loadfunc to work,
    I've got to remove or rename icon/src/h/features.h....

It appears that nobody's looked at tests/calling/loadfunc in a long time,
and I'd forgotten all about it.  There is a newer example of dynamic
loading in the Icon Program Library, in the packs/loadfunc directory
of the IPL source package.  *That* one works.  For the IPL see
	http://www.cs.arizona.edu/icon/ipl93.htm

There's also a larger set of functions, but no driver programs, in the
IPL's cfuncs directory.

   ---------------------------------------------------------------------------
   Gregg Townsend         Staff Scientist      The University of Arizona
   gmt@cs.arizona.edu     Computer Science     Tucson, Arizona, USA



From icon-group-sender  Mon Jul 12 11:39:15 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id LAA27422
	for icon-group-addresses; Mon, 12 Jul 1999 11:37:18 -0700 (MST)
Message-Id: <199907121837.LAA27422@baskerville.CS.Arizona.EDU>
To: icon-group@optima.CS.Arizona.EDU
Date: 10 Jul 1999 19:09:52 -0700
From: gmt@optima.CS.Arizona.EDU (Gregg Townsend)
Subject: Versions 9.3.2 of Icon and IPL available for Unix
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Version 9.3.2 of Icon for Unix, and version 9.3.2 of the Icon
program library, are now available from the Icon web site:
    http://www.cs.arizona.edu/icon/v93u.htm   (Icon for Unix)
    http://www.cs.arizona.edu/icon/ipl93.htm  (Icon program library)

Source code is available for Icon and for the library.  Binaries of
Icon are available for several common Unix platforms.  Precompiled
ucode files of the library procedures are also available.

For the Icon source code, this is primarily a maintenance release.
The code has been reworked and simplified to increase portability.
An ANSI C compiler is now required, and UNIX systems are expected
to supply a POSIX (1003.1-1988) library.  This new source release
is sufficiently portable that it also builds on the Be operating
system (see www.be.com), albeit without graphics.

Version 9.3.2 adds one new feature to Icon:  The files in a directory
can be listed by opening the directory as a file.  Subsequent reads
return the names of the files contained in the directory.  The names
are returned in no particular order, and for Unix, the directories
"." and ".." are included. 

The program library update reflects all the changes that have
accumulated since the release of version 9.3.1 in December of 1997,
and includes several new programs, procedures, and packages.

For more information, visit the Icon web site at
    http://www.cs.arizona.edu/icon/

   ---------------------------------------------------------------------------
   Gregg Townsend         Staff Scientist      The University of Arizona
   gmt@cs.arizona.edu     Computer Science     Tucson, Arizona, USA


From icon-group-sender  Tue Jul 13 07:57:47 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id HAA01858
	for icon-group-addresses; Tue, 13 Jul 1999 07:56:47 -0700 (MST)
Message-Id: <199907131456.HAA01858@baskerville.CS.Arizona.EDU>
Date: Tue, 13 Jul 1999 12:18:39 +0100 (BST)
From: Hugh Sasse <hgs@dmu.ac.uk>
To: Gregg Townsend <gmt@optima.CS.Arizona.EDU>
cc: icon-group@optima.CS.Arizona.EDU
Subject: Re: Versions 9.3.2 of Icon and IPL available for Unix
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

On 10 Jul 1999, Gregg Townsend wrote:

> Version 9.3.2 adds one new feature to Icon:  The files in a directory
> can be listed by opening the directory as a file.  Subsequent reads
> return the names of the files contained in the directory.  The names
> are returned in no particular order, and for Unix, the directories
> "." and ".." are included. 
> 
I cannot see a function which will allow one to determine if the list of
names one is reading is due to reading a genuine directory or a file
produced by 'ls -a > files'.  There seems to be no stat or lstat function.  
Have I missed something?

	Thank you.
	Hugh
	hgs@dmu.ac.uk



From icon-group-sender  Tue Jul 13 12:31:56 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id MAA10696
	for icon-group-addresses; Tue, 13 Jul 1999 12:30:15 -0700 (MST)
Message-Id: <199907131930.MAA10696@baskerville.CS.Arizona.EDU>
Date: Tue, 13 Jul 1999 09:36:28 -0700 (MST)
From: Gregg Townsend <gmt@baskerville.CS.Arizona.EDU>
To: hgs@dmu.ac.uk
Cc: icon-group@optima.CS.Arizona.EDU
Subject: Re: checking for directory
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

    Date: Tue, 13 Jul 1999 12:18:39 +0100 (BST)
    From: Hugh Sasse <hgs@dmu.ac.uk>
    
    I cannot see a function which will allow one to determine if the list of
    names one is reading is due to reading a genuine directory or a file
    produced by 'ls -a > files'.  There seems to be no stat or lstat function.  

You can't tell by the file contents, but you can determine whether a
filename s is a directory by checking to see whether open(s || "/.")
succeeds.  This is encapsulated in the directory(s) procedure in the
Icon program library ("link io").  There's no general stat()/lstat()
function, though.

   ---------------------------------------------------------------------------
   Gregg Townsend         Staff Scientist      The University of Arizona
   gmt@cs.arizona.edu     Computer Science     Tucson, Arizona, USA


From icon-group-sender  Tue Jul 13 16:36:15 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id QAA21471
	for icon-group-addresses; Tue, 13 Jul 1999 16:35:48 -0700 (MST)
Message-Id: <199907132335.QAA21471@baskerville.CS.Arizona.EDU>
From: "Iain Alexander" <ia@stryx.demon.co.uk>
To: icon-group@optima.CS.Arizona.EDU
Date: Tue, 13 Jul 1999 23:16:49 +0100
Subject: Re: Versions 9.3.2 of Icon and IPL available for Unix
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

On 13 Jul 99, at 12:18, Hugh Sasse wrote:

> On 10 Jul 1999, Gregg Townsend wrote:
> 
> > Version 9.3.2 adds one new feature to Icon:  The files in a directory
> > can be listed by opening the directory as a file.  Subsequent reads
> > return the names of the files contained in the directory.  The names
> > are returned in no particular order, and for Unix, the directories
> > "." and ".." are included. 
> > 
> I cannot see a function which will allow one to determine if the list of
> names one is reading is due to reading a genuine directory or a file
> produced by 'ls -a > files'.  There seems to be no stat or lstat function.  
> Have I missed something?

Perhaps.  You appear to want to open("directory/files"), whereas you actually 
open the directory itself - open("directory").

Or perhaps *I've* misunderstood. :-)

-- 
Iain Alexander               PGP 1024-bit key id B501A0AD
ia@stryx.demon.co.uk   I.Alexander@bra0105.wins.icl.co.uk


From icon-group-sender  Wed Jul 14 07:53:31 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id HAA19313
	for icon-group-addresses; Wed, 14 Jul 1999 07:53:23 -0700 (MST)
Message-Id: <199907141453.HAA19313@baskerville.CS.Arizona.EDU>
Date: Wed, 14 Jul 1999 09:32:53 +0100 (BST)
From: Hugh Sasse <hgs@dmu.ac.uk>
To: Gregg Townsend <gmt@baskerville.CS.Arizona.EDU>
cc: icon-group@optima.CS.Arizona.EDU
Subject: Re: checking for directory
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

On Tue, 13 Jul 1999, Gregg Townsend wrote:

> You can't tell by the file contents, but you can determine whether a
> filename s is a directory by checking to see whether open(s || "/.")

	Yes, that would do the trick.  Thank you.

> succeeds.  This is encapsulated in the directory(s) procedure in the
> Icon program library ("link io").  There's no general stat()/lstat()
> function, though.

	I think it is worth considering.  It is not so unix specific:
	DOS C compilers had it in the 1980's.  I get the impression
	that the Unicon people are keen to contribute their extensions
	to the Icon core, but you would have to ask them of course :-)
	If Icon is regarded as being complete now, then I suppose this
	won't get included.
> 
>    ---------------------------------------------------------------------------
>    Gregg Townsend         Staff Scientist      The University of Arizona
>    gmt@cs.arizona.edu     Computer Science     Tucson, Arizona, USA
> 
	Thank you,
	Hugh



From icon-group-sender  Sun Jul 18 15:22:32 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id PAA07448
	for icon-group-addresses; Sun, 18 Jul 1999 15:22:24 -0700 (MST)
Message-Id: <199907182222.PAA07448@baskerville.CS.Arizona.EDU>
X-Sender: memphis@popmail.macconnect.com
Date: Wed, 14 Jul 1999 15:05:05 -0500
To: icon-group@optima.CS.Arizona.EDU
From: Garry <memphis@macconnect.com>
Subject: Basename.icn in Library 9.3.2
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Am I not correct that it's not correct?

The implementation of Basename in release 9.3.1 was

   procedure basename(name, suffix)  #: base name of file
      local i

      name ? {
         every i := upto('/\\:')
         tab(integer(i) + 1)       	  # get rid of path, if any
         return tab(find(\suffix | ".") | 0) # return part before suffix, if any
         }
   end

In release 9.3.2 Basename has become a little "safer" ("Basename xyzzy.c  .cpp" returns  xyzzy.c  not  xyzzy) 

   procedure basename(name, suffix)  #: base name of file
      local i, base

      name ? {
         every i := upto('/\\:')
         tab(integer(i) + 1)       	  # get rid of path, if any
         if base := 1(tab(find(\suffix)),  pos(-*suffix)) then return base
         else return tab(0)
         }
   end

I like the new version, it's the Basename that I know and love, but the behavior has actually changed and the documentation doesn't reflect that (so I don't know if the behavior is intended or not).  The file documentation says that
   #  If no suffix is provided, the portion of the name up to the first
   #  "." is returned.

I think that that is no longer true and that the new Basename returns the simple filename, with suffix, if no suffix is specified.

Am I correct? (I'm an Icon newbie and I'm confused as to whether I'm confused...)
Is the behavior intended?
Should the documentation be corrected?

(* BTW 
I'm using Icon 9.0 on MacOS 8 under the MPW Shell.
I'm looking forward to using Icon graphics procedures when 9.3 is done for Macintosh.
I'll implement it myself if necessary but right now I'm working on compiling release 9.0 with MrC.
*)

--
  Garry Roseman  <mailto:memphis@macconnect.com>
  Tech Writer & Freelance Programmer




From icon-group-sender  Sun Jul 18 15:23:12 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id PAA07496
	for icon-group-addresses; Sun, 18 Jul 1999 15:22:58 -0700 (MST)
Message-Id: <199907182222.PAA07496@baskerville.CS.Arizona.EDU>
To: icon-group@optima.CS.Arizona.EDU
Date: Thu, 15 Jul 1999 21:08:44 -0500
From: memphis@macconnect.com (Garry)
Subject: [Q][newbie] Why use a useless loop?
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

There is a wonderful little procedure, lastname(s), in the Icon Library
for extracting the last name from a string s representing a person's
full name. (It seems wonderful to me considering how compact yet
tolerant it is, and how I would write it in other languages <ugh>). 

e.g. lastname("Dr. I. M. Zany, Jr.") produces a "Zany" result.

The implementation is
#       File:     lastname.icn
#       Subject:  Procedure to produce last name
#       Author:   Ralph E. Griswold
#       Date:     April 30, 1993
#
#  Produces the last name of a name in conventional form.  Obviously, it
#  doesn't work for every possibility.

procedure lastname(s)
   local line, i

   while line := trim(s) do {
      line ?:= tab(upto(',')) # Get rid of things like " ... , Jr."
      line ? {
          every i := upto(' ')
          tab(\i + 1)
          return tab(0)
          }
      }
end


My problem is that I can't think of any reason for writing
  while line := trim(s) do { ... }
instead of just
   line := trim(s)

Can anyone see a need for the "while"?  If so, please teach me. I did of
course already spend more than one full minute thinking about this and
looked up trim() in the Icon book. Maybe there really is no need for the
"while" but then I've observed that I'm blind to my blind spots.

-- 
Garry
Tech Writer & Freelance Programmer


From icon-group-sender  Sun Jul 18 15:23:50 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id PAA07540
	for icon-group-addresses; Sun, 18 Jul 1999 15:23:25 -0700 (MST)
Message-Id: <199907182223.PAA07540@baskerville.CS.Arizona.EDU>
Date: Fri, 16 Jul 1999 17:59:13 +0200 (CEST)
From: Anonymous <nobody@replay.com>
Comments: This message did not originate from the Sender address above.
	It was remailed automatically by anonymizing remailer software.
	Please report problems or inappropriate use to the
	remailer administrator at <abuse@replay.com>.
Subject: 9.3.2 for Windows
To: icon-group@optima.CS.Arizona.EDU
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

is there 9.3.2 for MS Windows - would be nice - also how soon for the posix port
to MS Windows...?



From icon-group-sender  Mon Jul 19 08:29:01 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id IAA05230
	for icon-group-addresses; Mon, 19 Jul 1999 08:28:17 -0700 (MST)
Message-Id: <199907191528.IAA05230@baskerville.CS.Arizona.EDU>
From: "Frank Lhota" <lhotaf@lexma.meitech.com>
To: <icon-group@optima.CS.Arizona.EDU>, "Garry" <memphis@macconnect.com>
Subject: Re: Basename.icn in Library 9.3.2
Date: Mon, 19 Jul 1999 10:39:06 -0400
X-Priority: 3
X-MSMail-Priority: Normal
X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2314.1300
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

> I like the new version, it's the Basename that I know and love, but the
behavior has actually changed and the documentation doesn't reflect that (so
I don't know if the behavior is intended or not).  The file documentation
says that
>    #  If no suffix is provided, the portion of the name up to the first
>    #  "." is returned.
>
> I think that that is no longer true and that the new Basename returns the
simple filename, with suffix, if no suffix is specified.
>
> Am I correct? (I'm an Icon newbie and I'm confused as to whether I'm
confused...)
> Is the behavior intended?

Yes, that is true. Obviously, the new version has no search for "." if
suffix is undefined or not present. The best way to verify this is by unit
testing, e.g.,

    procedure main ( arg )
        write ( basename ( arg [ 1 ], ( arg [ 2 ] | &null ) )
    end

> Should the documentation be corrected?
>

That all depends on how we want basename to behave. If we actually want the
old behavior, we should correct the procedure instead. I think that the
original idea was to make basename mimic the Unix utility of the same name.
The latest version of basename does seem to match the Unix basename, so if
this is what is desired, we should update the documentation.

Here is how I would implement basename:

procedure basename(name, suffix)  #: base name of file
   local base

   name ? {
      while tab ( upto('/\\:') + 1 )  # get rid of path, if any
      if ( base := tab ( -*\suffix ) ) & =suffix then return base
$ifdef OLD_STYLE
      else return tab ( find ( "." ) | 0 )
$else    # OLD_STYLE
      else return tab ( 0 )
$endif   # OLD_STYLE
      }
end

To get the old behavior (cut off everything after the first "." if suffix is
null or not at the end of the name), define the symbol OLD_STYLE. Otherwise,
you'll get the Unix style, i.e. if the suffix is null or not at the end,
return everything past the path part.



From icon-group-sender  Mon Jul 19 08:29:31 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id IAA05286
	for icon-group-addresses; Mon, 19 Jul 1999 08:29:23 -0700 (MST)
Message-Id: <199907191529.IAA05286@baskerville.CS.Arizona.EDU>
From: "Frank Lhota" <lhotaf@lexma.meitech.com>
To: <icon-group@optima.CS.Arizona.EDU>, "Garry" <memphis@macconnect.com>
Subject: Re: [Q][newbie] Why use a useless loop?
Date: Mon, 19 Jul 1999 11:22:52 -0400
X-Priority: 3
X-MSMail-Priority: Normal
X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2314.1300
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

You're right. the loop appears to be unnecessary. My guess is that it is a
fossil remain of a previous implementation.

One thing that strikes me about the implementation of both this procedure
and of basename is the use of a variable to store a position in the subject
string. Frankly, I do not know why this is done when a simpler and more
straightforward solution can be obtained by using the tab function. for
example, we can implement lastname as follows:

   procedure lastname(s)
      local line

      line := trim(s)
      line ?:= tab(find(",")) # Get rid of things like " ... , Jr."
      line ? {
          while tab(find(" ")+1) # Move past all blanks.
          return tab(0)
          }
   end

Now isn't that clearer than the version that uses the local variable i to
store positions?

----- Original Message -----
From: Garry <memphis@macconnect.com>
To: <icon-group@optima.CS.Arizona.EDU>
Sent: Thursday, July 15, 1999 10:08 PM
Subject: [Q][newbie] Why use a useless loop?


> There is a wonderful little procedure, lastname(s), in the Icon Library
> for extracting the last name from a string s representing a person's
> full name. (It seems wonderful to me considering how compact yet
> tolerant it is, and how I would write it in other languages <ugh>).
>
> e.g. lastname("Dr. I. M. Zany, Jr.") produces a "Zany" result.
>
> The implementation is
> #       File:     lastname.icn
> #       Subject:  Procedure to produce last name
> #       Author:   Ralph E. Griswold
> #       Date:     April 30, 1993
> #
> #  Produces the last name of a name in conventional form.  Obviously, it
> #  doesn't work for every possibility.
>
> procedure lastname(s)
>    local line, i
>
>    while line := trim(s) do {
>       line ?:= tab(upto(',')) # Get rid of things like " ... , Jr."
>       line ? {
>           every i := upto(' ')
>           tab(\i + 1)
>           return tab(0)
>           }
>       }
> end
>
>
> My problem is that I can't think of any reason for writing
>   while line := trim(s) do { ... }
> instead of just
>    line := trim(s)
>
> Can anyone see a need for the "while"?  If so, please teach me. I did of
> course already spend more than one full minute thinking about this and
> looked up trim() in the Icon book. Maybe there really is no need for the
> "while" but then I've observed that I'm blind to my blind spots.
>
> --
> Garry
> Tech Writer & Freelance Programmer
>



From icon-group-sender  Mon Jul 19 12:46:50 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id MAA14861
	for icon-group-addresses; Mon, 19 Jul 1999 12:46:09 -0700 (MST)
Message-Id: <199907191946.MAA14861@baskerville.CS.Arizona.EDU>
Date: Mon, 19 Jul 1999 09:03:13 -0700
From: Blake Chapman <blakec@pacbell.net>
To: icon-group@optima.CS.Arizona.EDU
Subject: Where does EXE file go ?
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

I am attempting to use WinIcon.  In Wi I create a program, save it as an
.icn file, and then in the Compile menu select "Make (compile + link)
executable".  Then I select Run|Run and it works.  However, there is no
.EXE file SAVED anywhere I can find.  It isn't in the directory the
*.icn file was saved into, and using the Microsoft Find utility I can't
find it anywhere on the disk.  What is wrong ?

Thanks,
Blake Chapman




From icon-group-sender  Mon Jul 19 12:47:36 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id MAA14947
	for icon-group-addresses; Mon, 19 Jul 1999 12:47:29 -0700 (MST)
Message-Id: <199907191947.MAA14947@baskerville.CS.Arizona.EDU>
From: "Nevin Liber" <nevin@eviloverlord.com>
To: <icon-group@optima.CS.Arizona.EDU>
Cc: "Frank Lhota" <lhotaf@lexma.meitech.com>, "Garry" <memphis@macconnect.com>
Subject: RE: [Q][newbie] Why use a useless loop?
Date: Mon, 19 Jul 1999 10:57:42 -0500
X-Priority: 3 (Normal)
X-MSMail-Priority: Normal
Importance: Normal
X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2014.211
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Frank Lhota [mailto:lhotaf@lexma.meitech.com] wrote:

> One thing that strikes me about the implementation of both this procedure
> and of basename is the use of a variable to store a position in
> the subject
> string. Frankly, I do not know why this is done when a simpler and more
> straightforward solution can be obtained by using the tab function.

I vaguely remember that in older implementations of Icon there were some
pecularities when having nested string scanning environments.  There is some
reference to this at <http://www.cs.arizona.edu/icon/docs/ipd112.htm>,
although I might be thinking of something else.

Most of these procedures are probably old enough that they were coded to
explicitly keep track of the position within a string as to avoid problems
when being called within a string scanning environment.  If this is indeed
the reason, I'm sure that most of the authors today would use string
scanning to implement their solutions.
__
 Nevin ":-)" Liber  <mailto:nevin@eviloverlord.com>  (312) 855-1000 x199



From icon-group-sender  Tue Jul 20 08:50:22 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id IAA20055
	for icon-group-addresses; Tue, 20 Jul 1999 08:49:42 -0700 (MST)
Message-Id: <199907201549.IAA20055@baskerville.CS.Arizona.EDU>
To: icon-group@optima.CS.Arizona.EDU
Date: Tue, 20 Jul 1999 10:46:02 +1000
From: ir <ionr@library.lib.rmit.edu.au>
Subject: Re: Where does EXE file go ?
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

If the installation was done correctly the Exe file is saved in Winicon,
next to the source file.
Anyway, check the Readme.txt file from the Winicon directory.
Start with line 9 - Setup your Path and Environment variables

Regards
Ion


Blake Chapman wrote:

> I am attempting to use WinIcon.  In Wi I create a program, save it as an
> .icn file, and then in the Compile menu select "Make (compile + link)
> executable".  Then I select Run|Run and it works.  However, there is no
> .EXE file SAVED anywhere I can find.  It isn't in the directory the
> *.icn file was saved into, and using the Microsoft Find utility I can't
> find it anywhere on the disk.  What is wrong ?
>
> Thanks,
> Blake Chapman



From icon-group-sender  Thu Jul 22 16:43:50 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id QAA21384
	for icon-group-addresses; Thu, 22 Jul 1999 16:42:02 -0700 (MST)
Message-Id: <199907222342.QAA21384@baskerville.CS.Arizona.EDU>
From: "Iain Alexander" <ia@stryx.demon.co.uk>
To: icon-group@optima.CS.Arizona.EDU
Date: Thu, 22 Jul 1999 23:12:07 +0100
Subject: Re: Where does EXE file go ?
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Correct, there is no .EXE, there is instead a .BAT :-)
(At least, with the version I'm using.  I believe the situation may have 
changed with more recent versions of WinIcon, but if you've got an old 
version, it may be worth checking.)

On 19 Jul 99, at 9:03, Blake Chapman wrote:

> I am attempting to use WinIcon.  In Wi I create a program, save it as an
> .icn file, and then in the Compile menu select "Make (compile + link)
> executable".  Then I select Run|Run and it works.  However, there is no
> .EXE file SAVED anywhere I can find.  It isn't in the directory the
> *.icn file was saved into, and using the Microsoft Find utility I can't
> find it anywhere on the disk.  What is wrong ?
> 
> Thanks,
> Blake Chapman
> 
> 
> 


-- 
Iain Alexander               PGP 1024-bit key id B501A0AD
ia@stryx.demon.co.uk   I.Alexander@bra0105.wins.icl.co.uk


From icon-group-sender  Tue Jul 27 07:37:56 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id HAA08828
	for icon-group-addresses; Tue, 27 Jul 1999 07:34:35 -0700 (MST)
Message-Id: <199907271434.HAA08828@baskerville.CS.Arizona.EDU>
To: icon-group@optima.CS.Arizona.EDU
Date: 27 Jul 1999 01:39:06 GMT
From: nr@viper.cs.Virginia.EDU (Norman Ramsey)
Subject: Re: Basename.icn in Library 9.3.2
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

In article <199907191948.MAA14993@baskerville.CS.Arizona.EDU>,
Frank Lhota <lhotaf@lexma.meitech.com> wrote:
>> Should the documentation be corrected?
>>
>
>That all depends on how we want basename to behave. If we actually want the
>old behavior, we should correct the procedure instead. I think that the
>original idea was to make basename mimic the Unix utility of the same name.
>The latest version of basename does seem to match the Unix basename, so if
>this is what is desired, we should update the documentation.

A wise man once said, ``If you change the specification of a
procedure, also change its name.''



From icon-group-sender  Tue Jul 27 12:22:12 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id MAA18656
	for icon-group-addresses; Tue, 27 Jul 1999 12:21:44 -0700 (MST)
Message-Id: <199907271921.MAA18656@baskerville.CS.Arizona.EDU>
From: "Frank Lhota" <lhotaf@lexma.meitech.com>
To: <icon-group@optima.CS.Arizona.EDU>
Subject: The Correct Basename Documentation
Date: Tue, 27 Jul 1999 12:31:17 -0400
X-Priority: 3
X-MSMail-Priority: Normal
X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2314.1300
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

This is a multi-part message in MIME format.

------=_NextPart_000_0022_01BED82B.EC3433B0
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

Recently, we have been discussing the basename procedure in the 9.3.2 =
library. It has been noted that the behavior of the procedure has =
changed without any modification in the documentation. There is general =
agreement that either the procedure or the documentation should be =
changed in order to bring them into agreement. Norman Ramsey made a very =
good point regarding this discussion:

  A wise man once said, ``If you change the specification of a =
procedure, also change its name.''

In general, I strongly agree with this principle. Identifier names are =
one of the most valuable forms of internal documentation. It is a shame =
that so many programmers treat identifier naming as a second class =
issue. The name of a procedure should express what a procedure does, and =
if the actions of that procedure changes, its name should change to =
express its current actions.

In this case, however, we have a case where the procedure has not =
behaved as its name suggests until recently. The name suggests an =
internal version of the Unix basename utility. The procedure header =
makes this link. Until the 9.3.2 library, however, the basename =
procedure has not behaved the same as its namesake Unix utility. If the =
Unix basename utility is not provided with a suffix, or if the specified =
suffix is not at the end of the name, then basename prints everything =
past the path info. Try this at a Unix command line:

  $ basename /usr/icon/ipl/procs/basename.icn .icn
  basename
  $ basename /usr/icon/ipl/procs/basename.icn
  basename.icn
  $ basename /usr/icon/ipl/procs/basename.icn .c
  basename.icn

In the previous versions of the Icon basename procedure, if the suffix =
was absent or not found at the end, the procedure would lop off both the =
path and everything past the first period. Before the 9.3.2 library, the =
name of this procedure has therefore been misleading.

As a result, the right way to keep basename true to its name is too keep =
its current behavior, that is, matching its Unix namesake. The =
documentation should be updated to reflect this. Here is how =
basename.icn should look:

#########################################################################=
###
#
# File:     basename.icn
#
# Subject:  Procedures to produce base name of a file
#
# Author:   Ralph E. Griswold
#
# Date:     September 22, 1998
#
#########################################################################=
###
#
#   This file is in the public domain.
#
#########################################################################=
###
#
#  Contributor:  Charles Shartsis
#
#  Updated By: Frank J. Lhota
#
#########################################################################=
###
#
#  This procedure is based on the UNIX basename(1) utility.  It strips =
off
#  any path information. If a suffix is specified, and the stripped name
#  ends with the suffix, the suffix will be stripped off too.
#
#  It should work under UNIX, MS-DOS, and the Macintosh.
#
#########################################################################=
###

procedure basename(name, suffix)  #: base name of file
   local base

   name ? {
      while tab ( upto('/\\:') + 1 )  # get rid of path, if any
      if ( base :=3D tab ( -*\suffix ) ) & =3Dsuffix
         then return base
         else return tab ( 0 )
      }
end



------=_NextPart_000_0022_01BED82B.EC3433B0
Content-Type: text/html;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META content=3D"text/html; charset=3Diso-8859-1" =
http-equiv=3DContent-Type>
<META content=3D"MSHTML 5.00.2614.3401" name=3DGENERATOR>
<STYLE></STYLE>
</HEAD>
<BODY bgColor=3D#ffffff>
<DIV><FONT size=3D2>Recently, we have been discussing the basename =
procedure in=20
the 9.3.2 library. It has been noted that the behavior of the procedure =
has=20
changed without any modification in the documentation. There is general=20
agreement that either the procedure or the documentation should be =
changed in=20
order to bring them into agreement. Norman Ramsey made a very good point =

regarding this discussion:</FONT></DIV>
<DIV>&nbsp;</DIV>
<BLOCKQUOTE style=3D"MARGIN-RIGHT: 0px">
  <DIV><FONT size=3D2>A wise man once said, ``If you change the =
specification of a=20
  procedure, also change its name.''</FONT></DIV></BLOCKQUOTE>
<DIV style=3D"MARGIN-RIGHT: 0px"><FONT size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT size=3D2>In general, I strongly agree with this principle. =
Identifier=20
names are one of the most valuable forms of internal documentation. It =
is a=20
shame that so many programmers treat identifier naming as a second class =
issue.=20
The name of a procedure should express what a procedure does, and if the =
actions=20
of that procedure changes, its name should change to express its current =

actions.</FONT></DIV>
<DIV>&nbsp;</DIV>
<DIV><FONT size=3D2>In this case, however, we have a case where the =
procedure has=20
not behaved as its name suggests until recently. The name suggests an =
internal=20
version of the Unix basename utility. The procedure header makes this =
link.=20
Until the 9.3.2 library, however, the basename procedure has not behaved =
the=20
same as its namesake Unix utility. If the Unix basename utility is not =
provided=20
with a suffix, or if the specified suffix is not at the end of the name, =
then=20
basename prints everything past the path info.&nbsp;Try this at a Unix =
command=20
line:</FONT></DIV>
<DIV>&nbsp;</DIV>
<BLOCKQUOTE style=3D"MARGIN-RIGHT: 0px">
  <DIV><FONT face=3D"Courier New" size=3D2>$ basename=20
  /usr/icon/ipl/procs/basename.icn .icn</FONT></DIV>
  <DIV><FONT face=3D"Courier New" size=3D2>basename</FONT></DIV>
  <DIV><FONT face=3D"Courier New" size=3D2>$ basename=20
  /usr/icon/ipl/procs/basename.icn</FONT></DIV>
  <DIV><FONT face=3D"Courier New" size=3D2>basename.icn</FONT></DIV>
  <DIV><FONT face=3D"Courier New" size=3D2>$ basename=20
  /usr/icon/ipl/procs/basename.icn .c</FONT></DIV>
  <DIV><FONT face=3D"Courier New" =
size=3D2>basename.icn</FONT></DIV></BLOCKQUOTE>
<DIV><FONT size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT size=3D2>In the previous versions of the Icon basename =
procedure, if=20
the suffix was absent or not found at the end, the procedure would lop =
off both=20
the path and everything past the first period.&nbsp;Before the 9.3.2 =
library,=20
the name of this procedure has therefore been misleading.</FONT></DIV>
<DIV>&nbsp;</DIV>
<DIV><FONT size=3D2>As a result, the right way to keep basename true to =
its name=20
is too keep its current behavior,&nbsp;that is, matching its Unix =
namesake. The=20
documentation should be updated to reflect this. Here is how =
basename.icn should=20
look:</FONT></DIV>
<DIV>&nbsp;</DIV>
<DIV><FONT face=3D"Courier New"=20
size=3D2>################################################################=
############<BR>#<BR>#&nbsp;File:&nbsp;&nbsp;&nbsp;&nbsp;=20
basename.icn<BR>#<BR>#&nbsp;Subject:&nbsp; Procedures to produce base =
name of a=20
file<BR>#<BR>#&nbsp;Author:&nbsp;&nbsp; Ralph E.=20
Griswold<BR>#<BR>#&nbsp;Date:&nbsp;&nbsp;&nbsp;&nbsp; September 22,=20
1998<BR>#<BR>############################################################=
################<BR>#<BR>#&nbsp;&nbsp;=20
This file is in the public=20
domain.<BR>#<BR>#########################################################=
###################<BR>#<BR>#&nbsp;=20
Contributor:&nbsp; Charles Shartsis<BR>#<BR>#&nbsp; Updated By: Frank J. =

Lhota<BR>#<BR>###########################################################=
#################<BR>#<BR>#&nbsp;=20
This procedure is based on the UNIX basename(1) utility.&nbsp; It strips =

off<BR>#&nbsp; any path information. If a suffix is specified, and the =
stripped=20
name<BR>#&nbsp; ends with the suffix, the suffix will be stripped off=20
too.<BR>#<BR>#&nbsp; It should work under UNIX, MS-DOS, and the=20
Macintosh.<BR>#<BR>######################################################=
######################</FONT></DIV>
<DIV>&nbsp;</DIV>
<DIV><FONT face=3D"Courier New" size=3D2>procedure basename(name, =
suffix)&nbsp; #:=20
base name of file<BR>&nbsp;&nbsp; local base</FONT></DIV>
<DIV>&nbsp;</DIV>
<DIV><FONT face=3D"Courier New" size=3D2>&nbsp;&nbsp; name ?=20
{<BR>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; while tab ( upto('/\\:') + 1 )&nbsp; =
# get=20
rid of path, if any<BR>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; if ( base :=3D tab =
(=20
-*\suffix ) ) &amp; =
=3Dsuffix<BR>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;=20
then return base<BR>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; =
else return=20
tab ( 0 )<BR>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; }<BR>end<BR></FONT></DIV>
<DIV><FONT size=3D2></FONT>&nbsp;</DIV></BODY></HTML>

------=_NextPart_000_0022_01BED82B.EC3433B0--



From icon-group-sender  Tue Jul 27 16:55:25 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id QAA01509
	for icon-group-addresses; Tue, 27 Jul 1999 16:54:55 -0700 (MST)
Message-Id: <199907272354.QAA01509@baskerville.CS.Arizona.EDU>
From: "Nevin Liber" <nevin@eviloverlord.com>
To: <icon-group@optima.CS.Arizona.EDU>
Cc: "Frank Lhota" <lhotaf@lexma.meitech.com>
Subject: Re: The Correct Basename Documentation
Date: Tue, 27 Jul 1999 18:32:31 -0500
X-Priority: 3 (Normal)
X-MSMail-Priority: Normal
Importance: Normal
X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2014.211
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

>>   A wise man once said, ``If you change the specification of a =
>> procedure, also change its name.''

> As a result, the right way to keep basename true to its name is too keep =
> its current behavior, that is, matching its Unix namesake.

I'm not sure this is the correct behavior (I'm not sure it isn't either).

Whenever someone makes a change to a public library or API, what that person
really has to weigh is how much of the installed base of code depends on the
previous buggy or unspecified behavior.  While most would agree that it is a
bad idea to code to an implementation instead of an interface (can anyone
tell that I'm in the middle of reading Design Patterns? :-)), library
writers, OS implementers, etc., have to worry about the people who do code
to a specific, concrete implementation.

Just changing the documentation isn't enough; what you really have to do is
regression test any code that is dependent upon the particular routine you
are "improving".  In practice, this can't be accomplished, because you
really don't know the entire user base of your library, so you have to
guess.

It isn't the new users of the function that worry me nearly as much as the
existing users of the function.  Are people dependent on the previous
behavior?
__
 Nevin ":-)" Liber  <mailto:nevin@eviloverlord.com>  (312) 855-1000 x199



From icon-group-sender  Thu Jul 29 12:28:22 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id MAA02945
	for icon-group-addresses; Thu, 29 Jul 1999 12:26:12 -0700 (MST)
Message-Id: <199907291926.MAA02945@baskerville.CS.Arizona.EDU>
From: "Frank Lhota" <lhotaf@lexma.meitech.com>
To: <icon-group@optima.CS.Arizona.EDU>
Subject: How to Bring Back the Old basename
Date: Thu, 29 Jul 1999 12:10:19 -0400
X-Priority: 3
X-MSMail-Priority: Normal
X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2314.1300
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Nevin Liber has a good point regarding basename. Even if the latest version
is closer to the original intentions, as well as the closer to what the name
suggests, there may be many applications dependant on the old behavior. To
keep these applications from breaking, we need to retain the old, quirky
behavior. This is unfortunate for new developers who would rather have
something closer to the Unix utility.

Has anyone else noticed how superfluous the second parameter to the old
basename is? The old basename chops off the extension whether or not the
second parameter is specified, and whether or not it is present, so when
does specifying an extension make a difference?

The best option for now may be to fix basename to return the old behavior,
and also update the documentation to clarify the quirk. Here is my proposal:

############################################################################
#
# File:     basename.icn
#
# Subject:  Procedures to produce base name of a file
#
# Author:   Ralph E. Griswold
#
# Date:     September 22, 1998
#
############################################################################
#
#   This file is in the public domain.
#
############################################################################
#
#  Contributor:  Charles Shartsis
#
############################################################################
#
#  This procedure is based on the UNIX basename(1) utility.  It strips off
#  any path information and removes the specified suffix, if present.
#
#  If no suffix is provided, or if the suffix is not at the end of the name,
#  the portion of the name up to the first "." is returned.
#
#  It should work under UNIX, MS-DOS, and the Macintosh.
#
############################################################################

procedure basename(name, suffix) #: base name of file
   local base

   name ? {
      while tab ( upto('/\\:') + 1 )  # get rid of path, if any
      if ( base := tab ( -*\suffix ) ) & =suffix
         then return base
         else return tab ( find ( "." ) | 0 )
      }

end





From icon-group-sender  Thu Jul 29 16:26:22 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id QAA11212
	for icon-group-addresses; Thu, 29 Jul 1999 16:25:50 -0700 (MST)
Message-Id: <199907292325.QAA11212@baskerville.CS.Arizona.EDU>
Date: Thu, 29 Jul 1999 13:31:49 -0700 (PDT)
From: "Glenn E. Thobe" <thobe@lafn.org>
To: icon-group@optima.CS.Arizona.EDU
Subject: coswitch for Sun3 SunOS 4.1.1
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

In the new icon 9.3.2 source distribution, I cannot find the coswitch
routine for the Sun 3 (Motorola 680x0) architecture.  This was included
in the 9.3.1 source code although apparently not by intention.  I no
longer have the 9.3.1 source, but I believe it was in
config/unix/*/rswitch.c where * was NeXT or Mac or some other Motorola
based machine.  The old source distribution has been removed from
www.cs.arizona.edu.  Is there any way to get this code?

I am trying to build icon on my Sun 3/80, SunOS 4.1.1_U1 with gcc and
X11R6.3.  I almost had 9.3.1 working, but 9.3.2 came out while before I
had solved all the problems and I deleted the 9.3.1 source code before
I realized this vital bit of code was missing from the newer edition.

Thanks.

-Glenn Thobe <thobe@lafn.org>


From icon-group-sender  Fri Jul 30 07:26:11 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id HAA04723
	for icon-group-addresses; Fri, 30 Jul 1999 07:25:27 -0700 (MST)
Message-Id: <199907301425.HAA04723@baskerville.CS.Arizona.EDU>
From: gep2@terabites.com
Date: Thu, 29 Jul 1999 18:48:11 -0500
Subject: How to Bring Back the Old basename
To: icon-group@optima.CS.Arizona.EDU
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

> Nevin Liber has a good point regarding basename. Even if the latest version
is closer to the original intentions, as well as the closer to what the name
suggests, there may be many applications dependant on the old behavior. 

I'm going to make several observations here.

> To keep these applications from breaking, we need to retain the old, quirky
behavior. This is unfortunate for new developers who would rather have
something closer to the Unix utility.

Personally, I have no vested interest (or other interest, for that matter) in 
making ANYTHING work like some Unix version does... I consider Unix in general 
to be an anachronistic relic, with crappy documentation standards and inherently 
anarchistic ongoing development efforts.  And I also don't consider it 
particularly compelling to argue that any given name (especially something as 
generic as "basename") has any kind of copyright on the name, giving them the 
right to define what said named function does (or doesn't do) forever and ever 
amen.

And that's especially true when the function's behavior can be argued to be 
"broken" in the way an old, stupid, anachronistic implementation did things.

> Has anyone else noticed how superfluous the second parameter to the old
basename is? The old basename chops off the extension whether or not the
second parameter is specified, and whether or not it is present, so when
does specifying an extension make a difference?

> The best option for now may be to fix basename to return the old behavior,
and also update the documentation to clarify the quirk. Here is my proposal:

I think my recommendation would be to specify in the documentation and/or 
routine comments that this routine has a number of quirks and odd behavior, and 
give some specific examples showing cases that might yield unexpected results.  
It would be perfectly reasonable to compare (in the comments) how the 
old-fashioned Unix-y "basename", the "old Icon" basename, and the "new Icon" 
basename would behave in the same situations.     

Yet another approach might be to add a third "mode" parameter, which could be 
set to different values specifying which type of behavior one wanted from the 
routine.  

To eliminate unwanted "default" behavior, one might purposely "break" old 
programs by requiring that the third mode parameter be explicitly provided... 
this would attract attention to the comments in the revised routine and force 
the programmer to decide (and specify) which behavior he (or she) wants.



Gordon Peterson
http://web2.airmail.net/gep2/
Support the Anti-SPAM Amendment!  Join at http://www.cauce.org/
12/19/98: the day the Conservatives demonstrated their scorn for their
   fraudulent sham of representative government.  Voters, remember it!



From icon-group-sender  Fri Jul 30 12:37:01 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id MAA13972
	for icon-group-addresses; Fri, 30 Jul 1999 12:35:39 -0700 (MST)
Message-Id: <199907301935.MAA13972@baskerville.CS.Arizona.EDU>
From: "Frank Lhota" <lhotaf@lexma.meitech.com>
To: <icon-group@optima.CS.Arizona.EDU>
Subject: When Do You Keep A Quirk?
Date: Fri, 30 Jul 1999 11:39:17 -0400
X-Priority: 3
X-MSMail-Priority: Normal
X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2314.1300
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Gordon Peterson makes an excellent argument for moving to a less quirky
definition of basename, going so far as to purposely break older programs
that might depend on the quirks. I am totally sympathetic to this point of
view. I had a friend that worked on the ANSI C standards committee. He
frequently lamented that there were proposed changes to C that every
committee member agreed would make a for a better language, but were voted
down because they would break large, existing applications. One has to weigh
the cost of revising old programs with the possible gains in new programs.

In the case of basename, there are over a dozen IPL sources that use
basename. If these IPL programs have not been tested with the new version of
basename, I would be very concerned about this change, for this can effect
programs that are not even using basename directly. Of course, it is very
hard to determine how much non-IPL code out there uses basename, or how many
programs depend on the quirks. As much as I prefer the new version of
basename, making this change in the short run is too risky.

If and when we do wish to change the behavior of basename, there is a way to
ease the migration path. We can write a procedure old_basename that has the
old behavior, then add the following line to old source files:

    $define basename old_basename

One final point: could we please cut down on ending Icon group mailings with
political or religious messages? I say this not because I disagree (or
agree) with the religious / political / aesthetic views of the members of
the group. It's just that the purpose of the group is to exchange
information on the Icon programming language. This is not the place for
discussion of these other topics.



From icon-group-sender  Fri Jul 30 17:49:41 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id RAA24130
	for icon-group-addresses; Fri, 30 Jul 1999 17:44:44 -0700 (MST)
Message-Id: <199907310044.RAA24130@baskerville.CS.Arizona.EDU>
From: gep2@terabites.com
Date: Fri, 30 Jul 1999 18:16:13 -0500
Subject: When Do You Keep A Quirk?
To: icon-group@optima.CS.Arizona.EDU
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

> Gordon Peterson makes an excellent argument for moving to a less quirky
definition of basename, 

...thank you...

> ...going so far as to purposely break older programs that might depend on the 
quirks. 

Right.

Some years ago, I was the programmer in charge of Datapoint's disk operating 
system.  Although the system was really quite reliable, we'd (only rarely) 
occasionally seen problems of the Cluster Allocation Table (the bitmap of 
available clusters on the disk) becoming corrupted.  This would eventually 
result in clusters being cross-linked to more than one file, obviously resulting 
in serious problems.  Other times, we'd sometimes simply "lose" space on the 
disk that was no longer allocated to any file.

One day I was wondering about how it was that file close (and space 
deallocation, including when a file was being totally deleted) function was able 
to run so quickly, even for large files.  

In our system, we had a fixed number of disk controller buffers, and a fixed 
number of Logical File Table entries (think of these as being similar the the 
data referenced by a "file handle").  When deleting a file (or the end, unused 
part of a file that had been extended) the system loaded the Cluster Allocation 
Table into disk controller buffer zero, and the RIB (Retrieval Information 
Block) for the file (detailing the locations and sizes of the allocated pieces 
for the file) into the disk controller buffer corresponding to the LFT entry in 
use when the file was being closed.  It then ran down through the appropriate 
entries in the RIB, clearing bits in the CAT and rewriting the CAT and its 
backup copy to disk when done.

As I thought about this, I realized that if one were using LFT entry zero to 
close a file and deallocate space, the two blocks would be loaded into the same 
buffer, resulting in (sometimes) extraneous entries in the CAT being cleared 
(and for a variety of technical reasons, probably not worth going into here, 
this tended to occur most often when a file that had been on the disk for a long 
time had been moved or replaced) since now the CAT was being also used as if it 
were the RIB of the file being closed as well.  

I couldn't imagine a program actually doing such a thing as closing a file using 
LFT entry zero, but just as a precaution built into the system a feature that 
would prevent the incorrect behavior by not allowing files to be closed and 
their space deallocated when using that LFT entry.

It wasn't too long after that when one afternoon the programmer in charge of our 
BASIC compiler came in, angry as all hell, pissed off at the change I've made 
and complaining that BASIC needed to use LFT entry zero when deleting files... 
and BINGO, we'd found the cause of the mysterious CAT corruption.  :-)

Anyhow, I tend to think that (if this BASENAME routine is actually used at all) 
it's PROBABLY being used by people who are unaware of the bug, and using it 
presuming that it works the way it sounds like it's supposed to work.  The 
"quirky" behavior (i.e. LONGSTANDING BUG) probably results in a lot of "quirky 
behavior" (i.e. LONGSTANDING BUGS) in the programs which use BASENAME, too.

When changing such an implementation detail, I'd think that a reasonable way to 
handle that is to carefully document (preferably in a source file comment, 
perhaps also in a more global README file or something) the "bug" that was 
fixed, examples of previous and new behavior, and let it go at that.  The other 
idea (about adding a mandatory third parameter) would be one way to make SURE 
that programs using the routine had been made aware of the fix, and that someone 
responsible for each such program had (hopefully) considered the behavior and 
what mode they genuinely wanted.  (Leaving each of the perhaps-desired modes in 
the routine, in this case, is no big deal and results in a routine only slightly 
larger than it would be otherwise... therefore I don't see this as being a huge 
issue).

> I am totally sympathetic to this point of view. I had a friend that worked on 
the ANSI C standards committee. He frequently lamented that there were proposed 
changes to C that every committee member agreed would make a for a better 
language, but were voted down because they would break large, existing 
applications. One has to weigh the cost of revising old programs with the 
possible gains in new programs.

That's perhaps a very different issue, since there it may be much more difficult 
to provide BOTH functionalities as options.  And here we're not talking about a 
core language issue anyhow... BASENAME is only just an IPL routine (not even 
part of the standard compiler runtime or anything) and the IPL is primarily used 
as EXAMPLES of how stuff is done in Icon.  Although certainly some of the other 
IPL routines use it (and in fact, that might be an interesting exercise... to 
see if any of those programs count on the quirky behavior or leave it as just 
another latent bug) it's not clear that the routine is hugely widely used 
otherwise.  (And if, indeed, the correction would FIX a latent bug in each of 
the IPL sources which call BASENAME, then that would reinforce the decision to 
fix it, and reduce the worry about programs counting on the anachronistic, 
buggy, Unix-y behavior).

> In the case of basename, there are over a dozen IPL sources that use
basename. If these IPL programs have not been tested with the new version of
basename, I would be very concerned about this change, for this can effect
programs that are not even using basename directly. 

Are those IPL sources complete stand-alone programs, or lower-level functions 
likely to have been designed into larger programs or systems?

> Of course, it is very hard to determine how much non-IPL code out there uses 
basename, or how many programs depend on the quirks. 

Yes.  But there are probably at least as many programs which have latent bugs as 
a result of the quirks... and I'd tend to believe that a program would count on 
the routine behaving the way one would EXPECT it to behave (i.e. without bugs) 
based on the description of what the routine is supposed to do.

Again, I'll point out that I'm not aware that the routine has ever been 
documented as behaving with all the same bugs and flaws as the anachronistic 
Unix routine historically called "basename" (and that's a VERY generic name, not 
something like "fsprintx" or the like), and that in this business we don't have 
eternal copyrights or trademarks on function names and ancient (buggy!) 
implementations based on those names.

> As much as I prefer the new version of basename, making this change in the 
short run is too risky.

You're entitled to your opinion... as are all the rest of us too.

> If and when we do wish to change the behavior of basename, there is a way to
ease the migration path. We can write a procedure old_basename that has the
old behavior, then add the following line to old source files:

>    $define basename old_basename

Personally, I don't see that that makes a whole lot of sense.  The change is 
cosmetic at best, and more confusing to programmers who (forever) look at a 
source file being edited and see 'basename' without being made aware of the 
macro substitution.  If you're gonna change the routine, then let's change it 
and be done with it.  At least adding a third (required) parameter would help 
make it more obvious to programmers that this routine is NOT an Icon-implemented 
clone of the archaic Unix "basename".

> One final point: could we please cut down on ending Icon group mailings with
political or religious messages? I say this not because I disagree (or
agree) with the religious / political / aesthetic views of the members of
the group. It's just that the purpose of the group is to exchange
information on the Icon programming language. This is not the place for
discussion of these other topics.

I presume the writer is commenting on my sig file, which makes a statement 
against SPAM E-mails, as well as reminding people of the **shameful** way in 
which our congresscritters, sent to Washington promising to represent their 
constituents, instead brazenly participated in the most outrageous orgy of 
partisan, politically motivated bullshit (and CLEARLY contrary to the expressed 
will of the majority of the voters) that this country has seen in as far back as 
*I* (at almost 50) can remember.  While I do agree that this is NOT the forum to 
dwell on discussions of that occurrence, I'll point out too that sig files are a 
VERY longstanding way for people on the Internet to voice their opinions on a 
variety of subjects (sort of a "bumpersticker for E-mail") and as such are an 
important element of free speech.  If you don't like to put your stand on such 
issues in your SIG file, then feel free not to.  But don't bitch about other 
people who do so, just because it makes you uncomfortable to be reminded about 
the kind of shameful bullshit that some politicos in this country seem to think 
the public will forget about prior to the next elections.  Some of us have 
longer memories than that!!!

Gordon Peterson
http://web2.airmail.net/gep2/
Support the Anti-SPAM Amendment!  Join at http://www.cauce.org/
12/19/98: the day the Conservatives demonstrated their scorn for their
   fraudulent sham of representative government.  Voters, remember it!



From icon-group-sender  Mon Aug  2 16:52:39 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id QAA20277
	for icon-group-addresses; Mon, 2 Aug 1999 16:50:42 -0700 (MST)
Message-Id: <199908022350.QAA20277@baskerville.CS.Arizona.EDU>
Date: Tue, 3 Aug 1999 10:50:09 +1200 (NZST)
From: "Richard A. O'Keefe" <ok@hermes.otago.ac.nz>
To: gep2@terabites.com, icon-group@optima.CS.Arizona.EDU
Subject: Re:  When Do You Keep A Quirk?
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

It sounds as though the right thing to do is to withdraw the
basename function *entirely*.

I'd like to point out that the situation in UNIX is actually
rather confused, thanks to some people adding 'neat hacks'.

Solaris:
    basename string [suffix]

    deletes any prefix ending in / (does that mean '/' or does it
    mean _any_ directory separator?  matters on DOS/Windows/OS/2)

    suffix, if present, is a REGULAR EXPRESSION; if there is a
    tail of string matching suffix, it will be removed (but if
    there are _several_ such tails, _which_?)

XPG4, BSD
    basename string [suffix]

    same rule and problem with string; but suffix is a LITERAL
    and NONE of its characters are significant.  There is for
    example nothing special about '.'.

C library:
    basename(path)

    does NOT take a suffix argument; removes leading directory
    prefixes by smashing its argument.  It also deletes any
    trailing slashes, except that a path that is one or more
    slashes and nothing else yields "/".  A null pointer or an
    empty string yields ".".

In particular,
    basename latex .tex
could give you 'l' or 'latex' depending on which of
/usr/bin /usr/xpg4/bin comes first in your $PATH.

The basename command combines two operations:
    remove_directory_part
which is what the basename() function does, and
    remove_matching_extension
which has no corresponding library function.  Nor is the function
I've _really_ wanted often enough,
    remove_any_extension
provided in any form.

I firmly believe that the _best_ thing to do is to junk basename and
its relatives completely and to provide a 'pathname' library modelled
on Common Lisp.  The second best thing to do is to throw 'basename'
out as being just too confusing, and to provide two _new_
library functions

    remove_directory_part(FileName) ->
        FileName without any (drive or) directory part.
        name (if present), extension (if present), and
        version number (if present) are retained.

    remove_extension(FileName[, Extension]) ->
	if Extension is provided,
	   ignore any leading dot to get Extension',
	   remove trailing . Extension' from FileName if present,
	   no change to FileName if . Extension' not present
	if Extension not provided,
	    remove trailing . ext from FileName if present,
	    where ext is anything host file system allows in an
	    extension.
        Leave any directory part intact.
	return modified FileName.

When writing new code, you can use these without wondering which
version of `basename', if any, Icon `basename' corresponds to.

Old code will promptly break.  If it assumed the documented
behavior of basename, just plug in
    basename(F, E) = remove_extension(remove_directory_part(F), E)
    basename(F)    = remove_directory_part(F)

If it assumed the actual behaviour, just plug in
    basename(F) = remove_extension(remove_directory_part(F))

One of the best things about Java is @deprecated, obviously swiped
from Eiffel's 'obsolete' keyword.  In Eiffel, you can write
    <procedure heading>
	obsolete "<string>"
	<procedure body>
and the compiler writes the string as a warning if your program uses
that procedure.  Icon could do with something similar for cases like this.



From icon-group-sender  Tue Aug  3 07:45:33 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id HAA12956
	for icon-group-addresses; Tue, 3 Aug 1999 07:41:54 -0700 (MST)
Message-Id: <199908031441.HAA12956@baskerville.CS.Arizona.EDU>
From: gep2@terabites.com
Date: Mon, 02 Aug 1999 23:08:37 -0500
Subject: Re:  When Do You Keep A Quirk?
To: "Richard A. O'Keefe" <ok@hermes.otago.ac.nz>,
        icon-group@optima.CS.Arizona.EDU
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

The fact, as Richard points out, that Unix systems aren't even consistent within 
themselves gives even one more reason to not worry about trying to recreate the 
mythical "standard" basename()'s routine's anachronistic behavior.

His suggestions for new routines (which would hopefully be even more useful and 
practical) sound fine, and I'm sure would be appreciated by many programmers.

The idea, though, that we have to retire the generic function name "basename" 
for all time still strikes me as being pretty ridiculous.

Again, don't people understand that a lot of the routines in IPL are there 
principally AS EXAMPLES!!!???  

Gordon Peterson
http://web2.airmail.net/gep2/
Support the Anti-SPAM Amendment!  Join at http://www.cauce.org/
12/19/98: the day the Conservatives demonstrated their scorn for their
   fraudulent sham of representative government.  Voters, remember it!



From icon-group-sender  Tue Aug  3 16:45:48 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id QAA22281
	for icon-group-addresses; Tue, 3 Aug 1999 16:41:17 -0700 (MST)
Message-Id: <199908032341.QAA22281@baskerville.CS.Arizona.EDU>
X-Sender: memphis@popmail.macconnect.com
Date: Tue, 3 Aug 1999 15:08:17 -0500
To: gep2@terabites.com, icon-group@optima.CS.Arizona.EDU
From: Garry <memphis@macconnect.com>
Subject: Learning by Example (was:  When Do You Keep A Quirk?)
X-MIME-Autoconverted: from quoted-printable to 8bit by baskerville.CS.Arizona.EDU id NAA09706
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Gordon Peterson wrote:
>The fact, as Richard points out, that Unix systems aren't even consistent within 
>themselves gives even one more reason to not worry about trying to recreate the 
>mythical "standard" basename()'s routine's anachronistic behavior.
>
>His suggestions for new routines (which would hopefully be even more useful and 
>practical) sound fine, and I'm sure would be appreciated by many programmers.
>
>The idea, though, that we have to retire the generic function name "basename" 
>for all time still strikes me as being pretty ridiculous.
>
>Again, don't people understand that a lot of the routines in IPL are there 
>principally AS EXAMPLES!!!???  

The IPL has quite a large number of good examples, as Gordon has pointed out (some of the IPL code I don't understand yet but I'm making good progress on my first set of programs, for maintaining my programming notes and extracting reports from them -- stuff that I used to with big ugly shell scripts).  Personally I can't imagine blindly incorporating anything from any library into my work without at least considering checking out (I keep everything under version control using MPW Projector) a branch version that I will modify for the specific needs of the project.  That's why I was looking at the implementation of Basename in the first place; I was considering whether it was suitable as a replacement for my own quirky (accepts regex for the suffix) implementation of Basename in a current project.

As a newbie I can tell you that I have found three great resources for learning Icon:
1) The Icon Programming Language by Griswold and Griswold.  This book is worth every penny; it is a labor of love and has obviously been honed and polished to explain the Deep Concepts of Icon in crystal clear fashion.
2) The Icon Programming Language Handbook by Thomas W. Christopher.  A gem of almost-mathematical precision. Explains some idioms and gives a viewpoint that complements the Griswold book.  This is a "must have" and that's not a problem because it's FREE.  <http://www.iit.edu/~tc/iconplh.htm>  This handbook has lots of very helpful examples.  Some problems are solved repeatedly using different techniques to show how you would do it using the different features of Icon.
3) The IPL.  May be easy to overlook as a teaching resource because it is so large.  It would be almost impossible to read all of the good stuff in the Icon Library.  Don't neglect reading some of the Library just because you can't read all of the library.  (Some of it is discouraging to newbies because it is incomprehensible even after a once-through reading of the Griswold book.  Man! You can do some wild things with Icon. ) 

The golden rule to writing Icon seems to be "tell what you want to do".  I could write in a similar style in Forth, but then I had to invent all of the constructs that I used.  Icon, as William Mitchell says, "provides a great set of abstract data types right out of the box."  An example of "telling what you want to do" occured last night when I wanted to step through a file to the first line that starts with a section mark character (some of you can see the character here: §)  I had the darndest time until I wrote simply
	if not while line := read( file ) do
		if match( "§ ", line) then break then fail
which is close to a direct statement of what I was trying to do and was possible because Icon, like Scheme, has only expressions, even "while e1 do e2" is an expression (that ultimately fails unless exited by a "break").


--
  Garry Roseman  <mailto:memphis@macconnect.com>
  Tech Writer & Freelance Programmer




From icon-group-sender  Wed Aug  4 17:34:58 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id RAA11646
	for icon-group-addresses; Wed, 4 Aug 1999 17:33:28 -0700 (MST)
Message-Id: <199908050033.RAA11646@baskerville.CS.Arizona.EDU>
Date: Wed, 04 Aug 1999 16:16:42 -0500
From: "Thomas W. Christopher" <tc@charlie.cns.iit.edu>
X-Accept-Language: en
To: Garry <memphis@macconnect.com>
CC: gep2@terabites.com, icon-group@optima.CS.Arizona.EDU
Subject: Re: Learning by Example (was:  When Do You Keep A Quirk?)
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Garry Roseman wrote:
... As a newbie I can tell you that I have found three great resources for learning Icon: ...
2) The Icon Programming Language Handbook by Thomas W. Christopher.  A gem of almost-mathematical precision. Explains some idioms and gives a viewpoint that complements the Griswold book.  This is a "must have" and that's not a problem because it's FREE.  <http://www.iit.edu/~tc/iconplh.htm>  This handbook has lots of very helpful examples.  Some problems are solved repeatedly using different techniques to show how you would do it using the different features of Icon. ...

Thanks for the kind words, Garry.  People can get to it with one fewer click from
     http://www.toolsofcomputing.com/freesoftware.htm
or directly from
     http://www.toolsofcomputing.com/iconprog.pdf

The IIT page may vanish after a while, since I've resigned from there effective Aug. 5th, 1999.

--
Thomas W. Christopher -- tc@toolsofcomputing.com
Operating Manager, Tools of Computing LLC. -- http://www.toolsofcomputing.com




From icon-group-sender  Mon Aug  9 08:00:40 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id HAA04946
	for icon-group-addresses; Mon, 9 Aug 1999 07:59:10 -0700 (MST)
Message-Id: <199908091459.HAA04946@baskerville.CS.Arizona.EDU>
Date: Sun, 08 Aug 1999 14:43:58 +0100
From: "C.R. Smith" <m7crs@csc.liv.ac.uk>
To: icon-group@optima.CS.Arizona.EDU
CC: m8hz@csc.liv.ac.uk
Subject: newbie question 
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Please can anyone help novice programmer re highligting with ICON

I am reading an array of student records into a "display" window when I
click on a record the record is selected and displayed in a "selection
window". I would like the selected record to be highlighted in the
display record - How do I go about this.


From icon-group-sender  Tue Aug 10 12:39:24 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id MAA00681
	for icon-group-addresses; Tue, 10 Aug 1999 12:35:31 -0700 (MST)
Message-Id: <199908101935.MAA00681@baskerville.CS.Arizona.EDU>
X-Authentication-Warning: nntprelay.berkeley.edu: news set sender to <news> using -f
From: espie@liafa.jussieu.fr (Marc Espie)
Subject: Re: When Do You Keep A Quirk?
Date: 10 Aug 1999 18:59:44 GMT
To: icon-group@optima.CS.Arizona.EDU
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

If you want useful directory/filename handling functions in Icon,
why not take perl as a prime example ?

File::Spec and friends, in recent perl functions, provide everything
that's needed to make filename analysis/reconstruction a breeze...
independently of the OS involved, which is the whole point, since 
any of those functions is trivial for one given OS.
-- 
	Marc Espie		
|anime, sf, juggling, unicycle, acrobatics, comics...
|AmigaOS, OpenBSD, C++, perl, Icon, PostScript...
| `real programmers don't die, they just get out of beta'


From icon-group-sender  Wed Sep  1 07:38:48 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id HAA28389
	for icon-group-addresses; Wed, 1 Sep 1999 07:37:10 -0700 (MST)
Message-Id: <199909011437.HAA28389@baskerville.CS.Arizona.EDU>
Date: Wed, 01 Sep 1999 13:55:22 +0100
From: J D Burnley <D.Burnley@sheffield.ac.uk>
To: icon-group@optima.CS.Arizona.EDU
Subject: The remove() function
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

I'm no programmer, although I do manage to do useful things with Icon
which I couldn't do with any other language (esp. text manipulation).
In most of the programs I write (using DOS), the remove() function
refuses to delete the files named. Although of course it works from a
stand-alone procedure devoted to the purpose, it won't work predictably
from the same procedure called within a larger program -- even if the
file has been explicitly closed.

There may be no simple answer to this, but if there is, I'd be grateful
to know it.

Regards,

David Burnley.


From icon-group-sender  Wed Sep  1 12:53:30 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id MAA19536
	for icon-group-addresses; Wed, 1 Sep 1999 12:52:52 -0700 (MST)
Message-Id: <199909011952.MAA19536@baskerville.CS.Arizona.EDU>
From: "Frank Lhota" <lhotaf@lexma.meitech.com>
To: "J D Burnley" <D.Burnley@sheffield.ac.uk>,
        <icon-group@optima.CS.Arizona.EDU>
Subject: Re: The remove() function
Date: Wed, 1 Sep 1999 12:15:01 -0400
X-Priority: 3
X-MSMail-Priority: Normal
X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2314.1300
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

First of all, you should be aware that DOS has problems with deleting a file
that is currently open. Depending on how the file is opened, the version of
DOS, and whether there is any buffered data, deleting a file that is open
can succeed, fail, or result in "lost clusters". My advice is that your
program be modified to make absolutely sure that the file is closed before
remove is called. Icon cannot make any promises that the host OS cannot
keep.

My advice is that you try to come up with a small Icon program that
reproduces the problem by paring down your current program. This might
clarify the problem, and if it does not, you can post the small program to
this newsgroup for evaluation.

----- Original Message -----
From: J D Burnley <D.Burnley@sheffield.ac.uk>
To: <icon-group@optima.CS.Arizona.EDU>
Sent: Wednesday, September 01, 1999 8:55 AM
Subject: The remove() function


> I'm no programmer, although I do manage to do useful things with Icon
> which I couldn't do with any other language (esp. text manipulation).
> In most of the programs I write (using DOS), the remove() function
> refuses to delete the files named. Although of course it works from a
> stand-alone procedure devoted to the purpose, it won't work predictably
> from the same procedure called within a larger program -- even if the
> file has been explicitly closed.
>
> There may be no simple answer to this, but if there is, I'd be grateful
> to know it.
>
> Regards,
>
> David Burnley.



From icon-group-sender  Thu Sep  2 07:52:01 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id HAA16293
	for icon-group-addresses; Thu, 2 Sep 1999 07:49:59 -0700 (MST)
Message-Id: <199909021449.HAA16293@baskerville.CS.Arizona.EDU>
X-Authentication-Warning: agate-ether.berkeley.edu: news set sender to <news> using -f
From: memphis@macconnect.com (Garry)
X-Newsgroups: comp.lang.icon
Subject: Is open(..,"b") broken in MPW Icon 9.0?
Date: Wed, 1 Sep 1999 22:32:14 -0500
User-Agent: MacSOUP/2.4
To: icon-group@optima.CS.Arizona.EDU
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

The most up-to-date Icon available for the Macintosh MPW environment is
version 9.0, which I think may have been implemented by Bob Alexander
(email address?). I've implemented most of an outline processor / note
organizer for MPW using Icon; maybe I'll proceed to try using it for
keeping source code in outline form, BUT after more than a month of fun
with Icon I've run into a problem!!  I can't open a file for update
(read/write).  When I follow the Icon practice of opening my file with
"b" mode I find that I can read it but I can not write to the file.  For
example if I execute
   f := open( "Testfile", "b" )
   seek( f )
   writes( f, "**" )
where "Testfile" is a small test file created previously, then I get the
message
   Run-time error 214
   input/output error

I've tried specifying "r+" mode or "b+r" or "ub" mode without success.
As a test I opened the same file in "w" mode and proved that I was able
to write it (destroying the previous contents) but I can not find a way
to write to random locations.

BTW, is any progress being made on Icon 9.3 for Macintosh?

Advice, or just sympathy, will be appreciated.  Right now my fall-back
solution is to rewrite the entire file just to update the 8 bytes which
holds my note ID counter.

-- 
Garry
Tech Writer & Freelance Programmer


From icon-group-sender  Thu Sep  2 13:18:14 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id NAA00177
	for icon-group-addresses; Thu, 2 Sep 1999 13:17:34 -0700 (MST)
Message-Id: <199909022017.NAA00177@baskerville.CS.Arizona.EDU>
Date: Thu, 2 Sep 1999 09:21:02 -0700 (MST)
From: Gregg Townsend <gmt@baskerville.CS.Arizona.EDU>
To: icon-group@optima.CS.Arizona.EDU, memphis@macconnect.com
Subject: Re: Mac Icon
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

    From: memphis@macconnect.com (Garry)
    Date: Wed, 1 Sep 1999 22:32:14 -0500
    
    BTW, is any progress being made on Icon 9.3 for Macintosh?

No.  Nobody that we know of is working on a Macintosh port.
One volunteer made a start in 1998, but that project stalled.

The 9.3.2 portability cleanup might make it a little less difficult
if somebody wants to give it a shot.  Another possibility would be
to get Jcon (the Java implementation of Icon) working on a Macintosh.

The work at Arizona, however, is restricted to maintenance of the
Unix version.

   ---------------------------------------------------------------------------
   Gregg Townsend         Staff Scientist      The University of Arizona
   gmt@cs.arizona.edu     Computer Science     Tucson, Arizona, USA


From icon-group-sender  Thu Sep  2 13:19:32 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id NAA00261
	for icon-group-addresses; Thu, 2 Sep 1999 13:19:13 -0700 (MST)
Message-Id: <199909022019.NAA00261@baskerville.CS.Arizona.EDU>
From: "Frank Lhota" <lhotaf@lexma.meitech.com>
To: "Garry" <memphis@macconnect.com>, <icon-group@optima.CS.Arizona.EDU>
Subject: Re: Is open(..,"b") broken in MPW Icon 9.0?
Date: Thu, 2 Sep 1999 12:32:19 -0400
X-Priority: 3
X-MSMail-Priority: Normal
X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2314.1300
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

The Icon file management routines are based on the C <stdio.h> facilitates.
The Icon open mode specifications are mapped to the C fopen mode parameter,
and unfortunately the available fopen modes vary from platform to platform.

Do you have a C compiler for the Macintosh MPW environment? If so, try to
see which C fopen modes will give you the desired results. From this, we can
figure out the how to fix the Icon run time code.

----- Original Message -----
From: Garry <memphis@macconnect.com>
To: <icon-group@optima.CS.Arizona.EDU>
Sent: Wednesday, September 01, 1999 11:32 PM
Subject: Is open(..,"b") broken in MPW Icon 9.0?


> The most up-to-date Icon available for the Macintosh MPW environment is
> version 9.0, which I think may have been implemented by Bob Alexander
> (email address?). I've implemented most of an outline processor / note
> organizer for MPW using Icon; maybe I'll proceed to try using it for
> keeping source code in outline form, BUT after more than a month of fun
> with Icon I've run into a problem!!  I can't open a file for update
> (read/write).  When I follow the Icon practice of opening my file with
> "b" mode I find that I can read it but I can not write to the file.  For
> example if I execute
>    f := open( "Testfile", "b" )
>    seek( f )
>    writes( f, "**" )
> where "Testfile" is a small test file created previously, then I get the
> message
>    Run-time error 214
>    input/output error
>
> I've tried specifying "r+" mode or "b+r" or "ub" mode without success.
> As a test I opened the same file in "w" mode and proved that I was able
> to write it (destroying the previous contents) but I can not find a way
> to write to random locations.
>
> BTW, is any progress being made on Icon 9.3 for Macintosh?
>
> Advice, or just sympathy, will be appreciated.  Right now my fall-back
> solution is to rewrite the entire file just to update the 8 bytes which
> holds my note ID counter.
>
> --
> Garry
> Tech Writer & Freelance Programmer



From icon-group-sender  Thu Sep  2 17:06:11 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id RAA05094
	for icon-group-addresses; Thu, 2 Sep 1999 17:05:54 -0700 (MST)
Message-Id: <199909030005.RAA05094@baskerville.CS.Arizona.EDU>
X-Authentication-Warning: agate-ether.berkeley.edu: news set sender to <news> using -f
From: Robert Valliant <valliant@hawaii.edu>
X-Newsgroups: comp.lang.icon
Subject: Re: Is open(..,"b") broken in MPW Icon 9.0?
Date: 02 Sep 1999 11:55:28 -1000
X-Trace: news.hawaii.edu 936309455 27487 128.171.23.89 (2 Sep 1999 21:57:35 GMT)
X-Complaints-To: usenet@hawaii.edu
X-Newsreader: Gnus v5.5/Emacs 20.3
To: icon-group@optima.CS.Arizona.EDU
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

In my icon manual, seek says:

seek(f,i)

where f is the filename and "i" is the position (1=beginning, 0=end)

Did you try that, but leave it out of your post?

-- 
Robert Valliant
Center for Russia in Asia, University of Hawaii at Manoa
www.hawaii.edu/shaps/russia
valliant@hawaii.edu Fax: 808.956.2682 Tel: 808.956.7814



From icon-group-sender  Thu Sep  2 17:07:22 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id RAA05171
	for icon-group-addresses; Thu, 2 Sep 1999 17:07:01 -0700 (MST)
Message-Id: <199909030007.RAA05171@baskerville.CS.Arizona.EDU>
X-Sender: memphis@popmail.macconnect.com
Date: Thu, 2 Sep 1999 17:17:45 -0500
To: "Frank Lhota" <lhotaf@lexma.meitech.com>,
        <icon-group@optima.CS.Arizona.EDU>
From: Garry <memphis@macconnect.com>
Subject: Re: Is open(..,"b") broken in MPW Icon 9.0?
X-MIME-Autoconverted: from quoted-printable to 8bit by baskerville.CS.Arizona.EDU id PAA20871
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

At 12:32 -0400 1999-09-02, Frank Lhota wrote:
>The Icon file management routines are based on the C <stdio.h> facilitates.
>The Icon open mode specifications are mapped to the C fopen mode parameter,
>and unfortunately the available fopen modes vary from platform to platform.
>
>Do you have a C compiler for the Macintosh MPW environment? If so, try to
>see which C fopen modes will give you the desired results. From this, we can
>figure out the how to fix the Icon run time code.

Thanks for your response Frank. I don't quite understand; the mode to pass to fopen for updating a file is "r+" on every platform (the lower level Mac toolbox call is something funky but the standard C library function is "standard").  I checked the "r+" mode on my problem file using a C program.   "f = fopen( "Testfile", "r+" );" works fine.   Thereafter I can read lines with fgets() or write to random locations using fseek() and fprintf() or fputs().

Of course I'm using an up-to-date StdCLib (I'm using MacOS 8.6 and MPW / MrC with Universal Interfaces 3.2).  I'm also using the new file system, HFS+, which (I hope) is transparent at the level of standard C functions.  Maybe MPW Icon (9.0)  just needs to be recompiled against new libraries -- but that's easier said than done.  It would become somewhat larger (because of changing from 68K to PowerPC RISC instructions) and much faster.  Would anyone be willing to help me?  

(As for my immediate problem -- it is perhaps most simply solved by using Icon for the analysis and reporting functions, in which it excels, and dropping back to C for writing my files in the first place. Since what I'm working on is a notekeeper/outliner to use in my MPW programming environment, I can tie Icon and C programs together using the Shell.)


>----- Original Message -----
>From: Garry <memphis@macconnect.com>
>
>> The most up-to-date Icon available for the Macintosh MPW environment is
>> version 9.0, which I think may have been implemented by Bob Alexander
>> (email address?). I've implemented most of an outline processor / note
>> organizer for MPW using Icon; maybe I'll proceed to try using it for
>> keeping source code in outline form, BUT after more than a month of fun
>> with Icon I've run into a problem!!  I can't open a file for update
>> (read/write).  When I follow the Icon practice of opening my file with
>> "b" mode I find that I can read it but I can not write to the file.  For
>> example if I execute
>>    f := open( "Testfile", "b" )
>>    seek( f )
>>    writes( f, "**" )
>> where "Testfile" is a small test file created previously, then I get the
>> message
>>    Run-time error 214
>>    input/output error
>>
>> I've tried specifying "r+" mode or "b+r" or "ub" mode without success.

-- 
   Garry Roseman  <mailto:memphis@macconnect.com>
   Tech Writer & Freelance Programmer
   Memphis, TN USA




From icon-group-sender  Tue Sep  7 16:58:37 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id QAA07136
	for icon-group-addresses; Tue, 7 Sep 1999 16:57:41 -0700 (MST)
Message-Id: <199909072357.QAA07136@baskerville.CS.Arizona.EDU>
From: "Bob Alexander" <bobalex@ix.netcom.com>
To: "Garry" <memphis@macconnect.com>, <icon-group@optima.CS.Arizona.EDU>
Subject: Re: Is open(..,"b") broken in MPW Icon 9.0?
Date: Thu, 2 Sep 1999 20:26:54 -0700
X-Priority: 3
X-MSMail-Priority: Normal
X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2615.200
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

I'm the author of MPW Icon, and as far as I know it hasn't been maintained
since my last update. I don't remember exactly when that was, but it's been
at least, I'd guess, 3 years. That version was very solid, although I don't
know what's happened with MPW since then. Last I knew they were talking
about dropping official support and releasing the source code. Currently,
I'm more involved with the dark side and seem to only have Windows boxes
around, so I can't really even check your bug. So... all I'm offering is
sympathy.

I really enjoyed using MPW and Icon together -- Icon's a great language and
they are a natural pair!

I personally never used the "b" option. There weren't any any complaints
about it (until now), but then the intersection of MPW users and "b" option
users is probably pretty sparse  :-)

The MPW Icon source code is available, and the Icon open() function
translates to a variant of fopen(). Check the source code to see what
fopen() option it uses, and then either:

  - change it to something that works with MPW (if possible) and rebuild.
  - stick with the copy method you're using now.

I would guess that it is trying to use an fopen() option that works under
Unix but not under MPW.

Sorry I can't offer any real support on this. Feel free to ask any more
questions you may have -- I'll do my best to answer.

-- Bob


----- Original Message -----
From: Garry <memphis@macconnect.com>
To: <icon-group@optima.CS.Arizona.EDU>
Sent: Thursday, September 02, 1999 7:53 AM
Subject: Is open(..,"b") broken in MPW Icon 9.0?


> The most up-to-date Icon available for the Macintosh MPW environment is
> version 9.0, which I think may have been implemented by Bob Alexander
> (email address?). I've implemented most of an outline processor / note
> organizer for MPW using Icon; maybe I'll proceed to try using it for
> keeping source code in outline form, BUT after more than a month of fun
> with Icon I've run into a problem!!  I can't open a file for update
> (read/write).  When I follow the Icon practice of opening my file with
> "b" mode I find that I can read it but I can not write to the file.  For
> example if I execute
>    f := open( "Testfile", "b" )
>    seek( f )
>    writes( f, "**" )
> where "Testfile" is a small test file created previously, then I get the
> message
>    Run-time error 214
>    input/output error
>
> I've tried specifying "r+" mode or "b+r" or "ub" mode without success.
> As a test I opened the same file in "w" mode and proved that I was able
> to write it (destroying the previous contents) but I can not find a way
> to write to random locations.
>
> BTW, is any progress being made on Icon 9.3 for Macintosh?
>
> Advice, or just sympathy, will be appreciated.  Right now my fall-back
> solution is to rewrite the entire file just to update the 8 bytes which
> holds my note ID counter.
>
> --
> Garry
> Tech Writer & Freelance Programmer
>
>



From icon-group-sender  Tue Sep  7 16:59:11 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id QAA07190
	for icon-group-addresses; Tue, 7 Sep 1999 16:58:51 -0700 (MST)
Message-Id: <199909072358.QAA07190@baskerville.CS.Arizona.EDU>
From: "Frank Lhota" <lhotaf@lexma.meitech.com>
To: <icon-group@optima.CS.Arizona.EDU>, "Garry" <memphis@macconnect.com>
Subject: Re: Is open(..,"b") broken in MPW Icon 9.0?
Date: Fri, 3 Sep 1999 09:04:21 -0400
X-Priority: 3
X-MSMail-Priority: Normal
X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2314.1300
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Amazingly enough, it looks like the code for the Icon open function does not
include the right code for opening a file in bi-directional mode. I think I
can help you patch the source code, however, so that it will work correctly

In the file RUNTIME\FSYS.R, search for the text "fsys.05". This search
should bring you to the following comment:

/*
 * The following code is operating-system dependent [@fsys.05].  Handle open
 *  modes.
 */

Previous to this section, the bits of the variable status have been set to
indicate the desired open mode. The bits in status are translated into a
mode string to be passed on to the C fopen function. Right before this
comment, the first mode character is set to one of 'r', 'w', or 'a'. The
possible mode parameter characters are, unfortunately, not that well
standardized. The code after this comment should have the OS dependent code
for handling the addition open modes. For the Macintosh, we see

#if MACINTOSH
      /* nothing to do */
#endif     /* MACINTOSH */

This is undoubtedly wrong; there are definitely things to do here for the
Mac. At a minimum, we should add '+' to mode if both the Fs_Read and
Fs_Write flags in status are on. Also, I believe that we also need to
process the Fs_Untrans flag for the Macintosh, since the Mac does not use
the Unix text file format. I am not sure what mode characters MPW uses to
indicate translated / untranslated mode; in my fix, I assume that MPW fopen
uses 't' for translated mode, 'b' for untranslated mode. Given, this, the
MACHINTOSH section for the open modes should look like this:

#if MACINTOSH
      if ((status & (Fs_Read|Fs_Write)) == (Fs_Read|Fs_Write)) {
         mode[1] = '+';
         mode[2] = ((status & Fs_Untrans) != 0) ? 'b' : 't';
         }
      else mode[1] = ((status & Fs_Untrans) != 0) ? 'b' : 't';
#endif     /* MACINTOSH */

Try making this change and rebuild.

----- Original Message -----
From: Garry <memphis@macconnect.com>
To: Frank Lhota <lhotaf@lexma.meitech.com>;
<icon-group@optima.CS.Arizona.EDU>
Sent: Thursday, September 02, 1999 6:17 PM
Subject: Re: Is open(..,"b") broken in MPW Icon 9.0?


At 12:32 -0400 1999-09-02, Frank Lhota wrote:
>The Icon file management routines are based on the C <stdio.h> facilitates.
>The Icon open mode specifications are mapped to the C fopen mode parameter,
>and unfortunately the available fopen modes vary from platform to platform.
>
>Do you have a C compiler for the Macintosh MPW environment? If so, try to
>see which C fopen modes will give you the desired results. From this, we
can
>figure out the how to fix the Icon run time code.

Thanks for your response Frank. I don't quite understand; the mode to pass
to fopen for updating a file is "r+" on every platform (the lower level Mac
toolbox call is something funky but the standard C library function is
"standard").  I checked the "r+" mode on my problem file using a C program.
"f = fopen( "Testfile", "r+" );" works fine.   Thereafter I can read lines
with fgets() or write to random locations using fseek() and fprintf() or
fputs().

Of course I'm using an up-to-date StdCLib (I'm using MacOS 8.6 and MPW / MrC
with Universal Interfaces 3.2).  I'm also using the new file system, HFS+,
which (I hope) is transparent at the level of standard C functions.  Maybe
MPW Icon (9.0)  just needs to be recompiled against new libraries -- but
that's easier said than done.  It would become somewhat larger (because of
changing from 68K to PowerPC RISC instructions) and much faster.  Would
anyone be willing to help me?

(As for my immediate problem -- it is perhaps most simply solved by using
Icon for the analysis and reporting functions, in which it excels, and
dropping back to C for writing my files in the first place. Since what I'm
working on is a notekeeper/outliner to use in my MPW programming
environment, I can tie Icon and C programs together using the Shell.)


>----- Original Message -----
>From: Garry <memphis@macconnect.com>
>
>> The most up-to-date Icon available for the Macintosh MPW environment is
>> version 9.0, which I think may have been implemented by Bob Alexander
>> (email address?). I've implemented most of an outline processor / note
>> organizer for MPW using Icon; maybe I'll proceed to try using it for
>> keeping source code in outline form, BUT after more than a month of fun
>> with Icon I've run into a problem!!  I can't open a file for update
>> (read/write).  When I follow the Icon practice of opening my file with
>> "b" mode I find that I can read it but I can not write to the file.  For
>> example if I execute
>>    f := open( "Testfile", "b" )
>>    seek( f )
>>    writes( f, "**" )
>> where "Testfile" is a small test file created previously, then I get the
>> message
>>    Run-time error 214
>>    input/output error
>>
>> I've tried specifying "r+" mode or "b+r" or "ub" mode without success.

--
   Garry Roseman  <mailto:memphis@macconnect.com>
   Tech Writer & Freelance Programmer
   Memphis, TN USA





From icon-group-sender  Tue Sep  7 16:59:45 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id QAA07227
	for icon-group-addresses; Tue, 7 Sep 1999 16:59:24 -0700 (MST)
Message-Id: <199909072359.QAA07227@baskerville.CS.Arizona.EDU>
From: dgamey@ca.ibm.com
X-Lotus-FromDomain: IBMCA@IBMUS
To: icon-group@optima.CS.Arizona.EDU,
        "J D Burnley" <D.Burnley@sheffield.ac.uk>
Date: Mon, 6 Sep 1999 13:18:52 -0400
Subject: Re: The remove() function
Content-Disposition: inline
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO



Checkj out http://www.cs.arizona.edu/icon/library/src/procs/openchk.icn  it
shoul help you find out if you haven't closed the file.

David




From icon-group-sender  Wed Sep  8 07:48:30 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id HAA28228
	for icon-group-addresses; Wed, 8 Sep 1999 07:47:56 -0700 (MST)
Message-Id: <199909081447.HAA28228@baskerville.CS.Arizona.EDU>
Date: Wed, 8 Sep 1999 13:13:46 +1200 (NZST)
From: "Richard A. O'Keefe" <ok@hermes.otago.ac.nz>
To: icon-group@optima.CS.Arizona.EDU, lhotaf@lexma.meitech.com,
        memphis@macconnect.com
Subject: Re: Is open(..,"b") broken in MPW Icon 9.0?
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

	#if MACINTOSH
	      if ((status & (Fs_Read|Fs_Write)) == (Fs_Read|Fs_Write)) {
	         mode[1] = '+';
	         mode[2] = ((status & Fs_Untrans) != 0) ? 'b' : 't';
	         }
	      else mode[1] = ((status & Fs_Untrans) != 0) ? 'b' : 't';
	#endif     /* MACINTOSH */
	
't' is a Windows-ism that has no defined effect in the C standard,
nor is it defined in any of the Macintosh C compilers I have ready
access to.  The modes supported are

	(r|w|a)(+[b]|b[+])

Some Macintosh C compilers define '\n' to be CR, which is the only
really sensible choice, and with those compilers the 'b' option is
superfluous.  Some define '\n' to be LF, which is really really
silly, and those compilers translate between external CR and internal
LF.  So the 'b' option is not superfluous with those compilers.



From icon-group-sender  Wed Sep  8 07:49:48 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id HAA28346
	for icon-group-addresses; Wed, 8 Sep 1999 07:49:39 -0700 (MST)
Message-Id: <199909081449.HAA28346@baskerville.CS.Arizona.EDU>
From: "Frank Lhota" <lhotaf@lexma.meitech.com>
To: "Richard A. O'Keefe" <ok@hermes.otago.ac.nz>,
        <icon-group@optima.CS.Arizona.EDU>, <memphis@macconnect.com>
Subject: Re: Is open(..,"b") broken in MPW Icon 9.0?
Date: Wed, 8 Sep 1999 09:38:42 -0400
X-Priority: 3
X-MSMail-Priority: Normal
X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2314.1300
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Armed with this information, I would try this in the [@fsys.05] section for
MACINTOSH:

#if MACINTOSH
      if ((status & (Fs_Read|Fs_Write)) == (Fs_Read|Fs_Write)) {
         mode[1] = '+';
         if ((status & Fs_Untrans) != 0) mode[2] = 'b';
         }
      else if ((status & Fs_Untrans) != 0) mode[1] = 'b';
#endif     /* MACINTOSH */

Please let me know if this fixes the problem.

While we're on the subject, it might be worth our while to revisit the
[@fsys.05] section of code. Quite conceivably, we could cut this section out
entirely The raison d'jtre for this section is that in Vanilla C, the form
of the fopen mode parameter was platform specific. The only mode parameters
spelled out in the original K&R book were "r", "w", and "a", although many
pre-ANSI compilers supported more options.

With ANSI C, the form of the fopen mode has become much better standardized.
The ANSI standard actually requires support for the '+' and 'b' fopen modes,
so the code given above for the MACINTOSH could be used for all platforms.
Once this is done, the only platform specific tweaking would be if you
wanted to include a 't' for translated modes, e.g.

#if MSDOS || ( OS2 && !CSET2 )
      if ((status & Fs_Untrans) == 0) strcat(mode,"t");
endif     /* MSDOS || ( OS2 && !CSET2 ) */

Even this is not absolutely necessary, since fopen defaults to opening in
text mode if the mode parameter does not include 'b'. Since Icon now
requires an ANSI compiler, we could simply retire the [@fsys.05] section and
replace it with the code for adding 'b' and '+' to mode.

----- Original Message -----
From: Richard A. O'Keefe <ok@hermes.otago.ac.nz>
To: <icon-group@optima.CS.Arizona.EDU>; Lhota, Frank
<lhotaf@lexma.meitech.com>; <memphis@macconnect.com>
Sent: Tuesday, September 07, 1999 9:13 PM
Subject: Re: Is open(..,"b") broken in MPW Icon 9.0?


> #if MACINTOSH
>       if ((status & (Fs_Read|Fs_Write)) == (Fs_Read|Fs_Write)) {
>          mode[1] = '+';
>          mode[2] = ((status & Fs_Untrans) != 0) ? 'b' : 't';
>          }
>       else mode[1] = ((status & Fs_Untrans) != 0) ? 'b' : 't';
> #endif     /* MACINTOSH */
>
> 't' is a Windows-ism that has no defined effect in the C standard,
> nor is it defined in any of the Macintosh C compilers I have ready
> access to.  The modes supported are
>
> (r|w|a)(+[b]|b[+])
>
> Some Macintosh C compilers define '\n' to be CR, which is the only
> really sensible choice, and with those compilers the 'b' option is
> superfluous.  Some define '\n' to be LF, which is really really
> silly, and those compilers translate between external CR and internal
> LF.  So the 'b' option is not superfluous with those compilers.



From icon-group-sender  Thu Sep  9 12:45:07 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id MAA23587
	for icon-group-addresses; Thu, 9 Sep 1999 12:43:22 -0700 (MST)
Message-Id: <199909091943.MAA23587@baskerville.CS.Arizona.EDU>
From: "Frank Lhota" <lhotaf@lexma.meitech.com>
To: <icon-group@optima.CS.Arizona.EDU>
Subject: Could we tend the result of memb?
Date: Thu, 9 Sep 1999 13:46:07 -0400
X-Priority: 3
X-MSMail-Priority: Normal
X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2314.1300
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

This message deals with the implementation of Icon, and assumes familiarity
with the current implementation.

Icon implements sets and tables using a hash chain system. The structures
for sets and tables are similar enough that they share a fair number of
common utilities. One of the most commonly used utilities for set and table
work is memb. The memb function determines whether a given item is a member
of a set or table. This function has the following prototype:

    union block **memb  (union block *pb,dptr x,uword hn, int *res);

The pb parameter is a pointer to either a set or table block. The x
parameter points to the descriptor of the value that memb will search for,
and hn is the hash value for *x. After calling memb, *res receives whether
*x cam be found in the set / table *pb (*res = 1 if the search succeeded, 0
otherwise). The result of memb is a pointer to a block pointer member within
some block that is part of the set / table representation. Assume we perform

    int res;
    union block **slot;

    slot = memb(pb,x,hn,&res);

If *x was found in *pb (res == 1), *slot points to the set / table element
in *pd for *x. If *x was not found in *pb (res == 0), *slot is the location
that should receive the address of a newly created element for *x.

The handy, dandy memb function is useful for searching, modifying, adding
and deleting the elements of a set or table. There is one problem, however,
with memb: the result of memb cannot be tended. If a new block is allocated,
the existing blocks may get moved around, and a stored memb result will be
invalidated. Moreover, searches take a long time, so you really do not want
to repeat a search unless absolutely necessary. To avoid invalidating the
result of memb, several functions will perform an allocation before calling
memb, even if we need to see the results of memb to determine if the
allocation was actually needed. For example, the pseudo-code for inserting a
value X into a set S is basically this.

    Allocate a new set element block;
    Call memb to see if X is in S;
    If ( X is not in S ) {
        Fill in new set element block with data for X;
        Link new set element block into S using result of memb;
        }
    else
        Deallocate new set element block;    /* it was not really needed */

This approach is complicated, error prone and inefficient. If there was only
some way to tend the result of memb, we could do the following more natural
and efficient algorithm:

    Call memb to see if X is in S;
    If ( X is not in S ) {
        Allocate a new set element block;
        Fill in new set element block with data for X;
        Link new set element block into S using result of memb;
        }

There is no known method for tending a value of type (union block **), since
one cannot reliably find the start of the block that this value points into.
Basically, memb is returning the address of a member in some block. Could
memb return this information in another form that can be tended?

One possibility that comes to mind is that memb can return its result in a
tended variable of type struct descrip. Remember, the result is always an
address of a member of some block. We can store the start of this block in
the BlkLoc part of the descriptor. The offset from the start of the block to
the particular member can be stored in the Offset portion of the dword
member of the descriptor. The resulting descriptor is in a form quite
similar to descriptors that are currently tended. Consider an alternative
form of memb:

    int memb  (union block *pb,dptr x,uword hn, dptr slot);

In this new version, the function return value indicates whether the search
was successful (1 if search succeeded, 0 if failed). The slot parameter is
the address of a tended descriptor. Upon return, we have:

    slot.dword = (F_Nqual | F_Ptr) | ( offset in words to the member )
    BlkLoc(*s) = (Start of the block)

If we do this, should also add a macro that will turn such a descriptor into
a (union block **) value.

Before I start working on this, I pose this question: does anyone know of a
reason why this wouldn't work, or wouldn't be desirable?



From icon-group-sender  Thu Sep  9 17:07:14 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id RAA03088
	for icon-group-addresses; Thu, 9 Sep 1999 17:06:36 -0700 (MST)
Message-Id: <199909100006.RAA03088@baskerville.CS.Arizona.EDU>
Date: Thu, 9 Sep 1999 15:21:37 -0700 (MST)
From: Gregg Townsend <gmt@baskerville.CS.Arizona.EDU>
To: icon-group@optima.CS.Arizona.EDU, lhotaf@lexma.meitech.com
Subject: Re: Could we tend the result of memb?
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

    From: "Frank Lhota" <lhotaf@lexma.meitech.com>
    Date: Thu, 9 Sep 1999 13:46:07 -0400
    
    Before I start working on this, I pose this question: does anyone know
    of a reason why this wouldn't work, or wouldn't be desirable?


Allocation and deallocation are pretty cheap.  It's not obvious to me
whether this would actually gain anything.  It might, but tended
descriptors incur some cost, too.

   ---------------------------------------------------------------------------
   Gregg Townsend         Staff Scientist      The University of Arizona
   gmt@cs.arizona.edu     Computer Science     Tucson, Arizona, USA


From icon-group-sender  Tue Sep 14 12:28:18 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id MAA04598
	for icon-group-addresses; Tue, 14 Sep 1999 12:26:28 -0700 (MST)
Message-Id: <199909141926.MAA04598@baskerville.CS.Arizona.EDU>
X-Authentication-Warning: agate-ether.berkeley.edu: news set sender to <news> using -f
From: Stephen Harborne <sjharbo@sc98a.demon.co.uk>
X-Newsgroups: comp.lang.icon
Subject: Icon 9.3.2 under HP-UX
Date: Tue, 14 Sep 1999 16:53:06 +0100
X-NNTP-Posting-Host: sc98a.demon.co.uk:212.228.9.236
X-Trace: news.demon.co.uk 937324556 nnrp-06:9553 NO-IDENT sc98a.demon.co.uk:212.228.9.236
X-Complaints-To: abuse@demon.net
X-Newsreader: Turnpike (32) Version 4.01  <WZQA+nUnpI7J3MTChYGRsiYleS>
To: icon-group@optima.CS.Arizona.EDU
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

I am trying to build Icon on a Hewlett Packard machine under HP-UX 11.0.

I found that 9.3.1 of Icon would not build. I then received
9.3.2. This compiled but linked with errors, related to the
co-expression modules. Non of Icon programs would then run
and they all produce error 117 - "no main program". I have not
investigated the code as I was trying to build Icon "out-of-the-box".
This is the first OS that I have had problems with at this early
stage.

Is a HP a common platform for running Icon?

Grateful for any thoughts?

Regards


-- 
Stephen Harborne


From icon-group-sender  Tue Sep 21 09:32:57 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id JAA17244
	for icon-group-addresses; Tue, 21 Sep 1999 09:29:20 -0700 (MST)
Message-Id: <199909211629.JAA17244@baskerville.CS.Arizona.EDU>
From: jeffery@big-bill.CS.UNLV.EDU
To: icon-group@optima.CS.Arizona.EDU
Subject: "Programming with Unicon" draft manuscript now available
Date: Tue, 21 Sep 1999 04:12:06 -0700
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO


You can't download the software yet, but we are soliciting comments,
corrections, and suggestions on the manuscript for our new book "Programming
with Unicon" (formerly titled "Programming with Icon"), which is available
in PDF format on our web site (http://icon.cs.unlv.edu), thanks to help from
volunteer Kostas Oikonomou.

The book describes the core language you know and love, the portable system
and internet interface, the new object oriented facilities, packages,
parameter coercion syntax, a new OO GUI class library and interface builder,
compiler writer's tools, substantive example application areas, and the
valuable and underutilized Icon procedure and program libraries.

Its 474 pages with lots of graphics, so the book is a large download.
We are sure there will be many changes and improvements between now and
the book's publication...thanks to your help!

Clint Jeffery, jeffery@cs.unlv.edu
and my co-authors, Shamim Mohamed, Ray Pereda, and Robert Parlett


From icon-group-sender  Mon Sep 27 08:01:18 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id HAA15798
	for icon-group-addresses; Mon, 27 Sep 1999 07:58:34 -0700 (MST)
Message-Id: <199909271458.HAA15798@baskerville.CS.Arizona.EDU>
X-Authentication-Warning: agate-ether.berkeley.edu: news set sender to <news> using -f
From: "Steve Adams" <adamss@ascinet.com>
X-Newsgroups: comp.lang.icon
Subject: IDBC?
X-Newsreader: Microsoft Outlook Express 4.72.3110.5
X-MimeOLE: Produced By Microsoft MimeOLE V4.72.3110.3
Date: Wed, 22 Sep 1999 08:48:09 -0400
X-Trace: newsfeed.slurp.net 938004261 206.153.26.162 (Wed, 22 Sep 1999 07:44:21 CDT)
To: icon-group@optima.CS.Arizona.EDU
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Is there any information on the Icon IDBC project?  Thanks.





From icon-group-sender  Mon Sep 27 08:03:16 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id IAA16179
	for icon-group-addresses; Mon, 27 Sep 1999 08:03:02 -0700 (MST)
Message-Id: <199909271503.IAA16179@baskerville.CS.Arizona.EDU>
From: SungJae Yi <sung@opencon.com>
To: "'icon-group@cs.arizona.edu'" <icon-group@optima.CS.Arizona.EDU>
Subject: newbie question
Date: Thu, 23 Sep 1999 16:42:10 -0400
X-MIME-Autoconverted: from quoted-printable to 8bit by baskerville.CS.Arizona.EDU id NAA22186
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Hi:
I have a newbie question. ( Sorry, I tried to look up various documents but I could not find the info).

What I want to do is to have my C++ program, which has the main(), call the Icon interpreter and let it process the provided data based on the external configuration (Icon program).

So the system would look like the following:
[data file in]-->   [ My Main Process         ] -->      [Icon Interpreter]
                        [                                   ]              |        ^
[data file out]<-- [                                   ]<----------+       |
                                                                           [Icon program]

Textual description:
A program, My Main Process has two file I/O: input & output.

Once the input data has been read in, it is massaged and passed to the Icon interpreter.
The Icon interpreter processes the data according to the external (external to My Main Process) configuration (Icon program) and returns the data to My Main Process.
The returned data is then written out to output data file.

I noticed that I can call C functions from Icon program, but can I call Icon from a C/C++ program?

Sincerely,
SungJae Yi



From icon-group-sender  Mon Sep 27 08:04:15 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id IAA16252
	for icon-group-addresses; Mon, 27 Sep 1999 08:04:05 -0700 (MST)
Message-Id: <199909271504.IAA16252@baskerville.CS.Arizona.EDU>
Date: Thu, 23 Sep 1999 23:42:03 -0400 (EDT)
From: Mark Kot <kot@buteo.math.utk.edu>
Subject: FFT in Icon ?
To: icon-group@optima.CS.Arizona.EDU
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO


Before I spend too much time reinventing the wheel,
I thought I would ask if anyone has a good clean
implementation of the FFT (or of the fast Hartley
transform) in Icon.

Any leads are appreciated.  Many thanks !


					Best wishes,
					
					Mark Kot
					



From icon-group-sender  Mon Sep 27 14:06:05 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id OAA03100
	for icon-group-addresses; Mon, 27 Sep 1999 14:03:41 -0700 (MST)
Message-Id: <199909272103.OAA03100@baskerville.CS.Arizona.EDU>
To: Steve Adams <adamss@ascinet.com>
CC: icon-group@optima.CS.Arizona.EDU
Subject: Re: IDBC? 
Date: Mon, 27 Sep 1999 09:07:22 -0700
From: Clinton L Jeffery <jeffery@bo.Egr.UNLV.EDU>
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO


Steve Adams> Is there any information on the Icon IDBC project?  Thanks.

Steve,

Check out http://icon.cs.utsa.edu for a preliminary look at what we've got.
The ODBC software is in alpha testing, and some functionality may change.
In particular, what my fine student Federico Balbi has implemented
doesn't look very much like what I originally specified, and I may cajole
him into some changes before it ships.  The website is his version of the
documentation, so it corresponds to what is actually implemented.

The URL is not a typo; the project was started at UT San Antonio but will be
distributed and maintained from University of Nevada, Las Vegas, because I
moved there.  Federico is at UT San Antonio and is still employed on the
project.

Clint


From icon-group-sender  Mon Sep 27 14:07:09 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id OAA03315
	for icon-group-addresses; Mon, 27 Sep 1999 14:07:02 -0700 (MST)
Message-Id: <199909272107.OAA03315@baskerville.CS.Arizona.EDU>
Date: Mon, 27 Sep 1999 09:18:44 -0700 (MST)
From: Gregg Townsend <gmt@baskerville.CS.Arizona.EDU>
To: icon-group@optima.CS.Arizona.EDU, sung@opencon.com
Subject: Re: newbie question
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

    From: SungJae Yi <sung@opencon.com>
    Date: Thu, 23 Sep 1999 16:42:10 -0400
    
    What I want to do is to have my C++ program, which has the main(),
    call the Icon interpreter...

There's no provision for that in the Icon system.  Your best bet, if the
problem can be so structured, is to arrange a pipeline of two programs.

   ---------------------------------------------------------------------------
   Gregg Townsend         Staff Scientist      The University of Arizona
   gmt@cs.arizona.edu     Computer Science     Tucson, Arizona, USA


From icon-group-sender  Thu Sep 30 17:18:29 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id RAA28106
	for icon-group-addresses; Thu, 30 Sep 1999 17:13:55 -0700 (MST)
Message-Id: <199910010013.RAA28106@baskerville.CS.Arizona.EDU>
X-Sender: memphis@popmail.macconnect.com
Date: Thu, 30 Sep 1999 15:57:25 -0500
To: icon-group@optima.CS.Arizona.EDU
From: Garry <memphis@macconnect.com>
Subject: Compiler Design in Icon
X-MIME-Autoconverted: from quoted-printable to 8bit by baskerville.CS.Arizona.EDU id OAA21468
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

I was just re-reading chapter 18 of the Griswolds's "The Icon Programming Language"; it is a very inspiring treatment of matching procedures, recognition of grammars, and parsing in Icon.  That got me thinking about the courses I took long ago on compiler design and on finite state automata but I wonder if the availability of generators (and recursive generators) and coexpressions in the compiler implementation language doesn't affect the whole treatment of the subject.  Maybe books written from within a Scheme viewpoint would be most sympathetic.  (Scheme and Icon bear deep similarity don't they?)  If any of the Icon experts here have found books on compiler design/automata/formal language theory (or string matching for that matter) that are particularly mappable to Icon, your recommendations would be appreciated.

-- 
   Garry Roseman  <mailto:memphis@macconnect.com>
   Tech Writer & Freelance Programmer
   Memphis, TN USA




From icon-group-sender  Thu Sep 30 17:34:12 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id RAA28600
	for icon-group-addresses; Thu, 30 Sep 1999 17:33:53 -0700 (MST)
Message-Id: <199910010033.RAA28600@baskerville.CS.Arizona.EDU>
Date: Thu, 30 Sep 1999 17:25:43 -0700
From: Steve Wampler <sbw@tapestry.tucson.az.us>
X-Accept-Language: en
To: icon-group@optima.CS.Arizona.EDU
Subject: A small puzzle
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO


I haven't posted anything for a while, but thought this might be fun:

   Write an Icon program to generate the pairings in a round-robin
   tournament.  The program should accept the player names as
   command line arguments (see the example below).

In a round-robin tournament, every player plays every other player
exactly once.  If there are an odd number of players, then a
special "bye" player is added.

For example, given players Alice, Bonnie, and Carole, a sample run
of the program should produce:

$ roundrobin Alice Bonnie Carole
Round 1
        Court  1: Alice vs bye
        Court  2: Bonnie vs Carole
Round 2
        Court  1: Alice vs Bonnie
        Court  2: Carole vs bye
Round 3
        Court  1: Alice vs Carole
        Court  2: bye vs Bonnie

Naturally, there should be no limit on the number of players.  You can
assume that there are enough courts available to satisfy each round.
(As an additional challenge, implement an optional -n argument, where
n is the number of courts - this is as much an exercise in determining
a reasonable policy for handling the pairing overfow as it is an
exercise in Icon programming.)

I have a non-backtracking solution and think I know the way to do
a back-tracking one, but thought others might enjoy playing with this
problem.  Send solutions to me (or back to icon-group) - I'll be
happy to look at them and summarize.  Speed isn't much of
a concern - though randomly producing pairings and rejecting bad ones
probably isn't going to win great praise!

Finally, if you really have too much time on your hand, add code to
produce a scoresheet suitable for recording the results. I'd prefer
it if you provided this separately, to keep the size of the solutions
to the original problem down...

-- 
---
Steve Wampler     {sbw@tapestry.tucson.az.us}
The gods that smiled upon your birth are laughing now. -- fortune cookie


From icon-group-sender  Fri Oct  1 08:19:23 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id IAA23810
	for icon-group-addresses; Fri, 1 Oct 1999 08:18:27 -0700 (MST)
Message-Id: <199910011518.IAA23810@baskerville.CS.Arizona.EDU>
Date: Thu, 30 Sep 1999 23:58:02 -0500
From: "Thomas W. Christopher" <tc@charlie.cns.iit.edu>
X-Accept-Language: en
To: Garry <memphis@macconnect.com>
CC: icon-group@optima.CS.Arizona.EDU
Subject: Re: Compiler Design in Icon
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

For a couple of parser generators and an Euler compiler in Icon, see:
    http://www.toolsofcomputing.com/freesoftware.htm
For a couple of more compilers in Icon (not as well documented), see
    http://www.iit.edu/~tc/toolsfor.htm#Compilers
These are educational compilers, not production quality.

Garry wrote:

> ...If any of the Icon experts here have found books on compiler design/automata/formal language theory (or string matching for that matter) that are particularly mappable to Icon, your recommendations would be appreciated.
>

--
Thomas W. Christopher, tchristopher@cs.depaul.edu
Visiting Associate Professor, DePaul Univ., Chicago
Operating Manager, Tools of Computing LLC.:
  http://www.toolsofcomputing.com, tc@toolsofcomputing.com




From icon-group-sender  Fri Oct  1 13:38:47 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id NAA05989
	for icon-group-addresses; Fri, 1 Oct 1999 13:38:23 -0700 (MST)
Message-Id: <199910012038.NAA05989@baskerville.CS.Arizona.EDU>
Date: Fri, 01 Oct 1999 08:41:09 -0700
From: Steve Wampler <swampler@noao.edu>
X-Accept-Language: en
To: Steve Wampler <sbw@tapestry.tucson.az.us>
CC: icon-group@optima.CS.Arizona.EDU
Subject: Re: A small puzzle
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Steve Wampler wrote:
> 
>    Write an Icon program to generate the pairings in a round-robin
>    tournament.  The program should accept the player names as
>    command line arguments (see the example below).

As a postscript, here's one more little challenge you might want to
work on...

    Assuming that the order of names given as arguments is the order
    of seeding in the tournament (i.e. first name is top seed, second
    name is 2nd seed, and so on).  Arrange the rounds so that the last
    round has the first and 2nd seeds paired, 3rd and 4th paired, etc.

--
Steve Wampler-  SOLIS Project, National Solar Observatory
swampler@noao.edu


From icon-group-sender  Fri Oct  1 13:48:03 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id NAA06491
	for icon-group-addresses; Fri, 1 Oct 1999 13:47:54 -0700 (MST)
Message-Id: <199910012047.NAA06491@baskerville.CS.Arizona.EDU>
Date: Fri, 01 Oct 1999 12:34:37 -0700
From: Steve Wampler <swampler@noao.edu>
X-Accept-Language: en
To: icon-group <icon-group@optima.CS.Arizona.EDU>
Subject: An a2ps style sheet for Icon
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

This is a multi-part message in MIME format.
--------------351713C02FB9FE89DD635AF3
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit


If anyone's interested, here's the first cut at a style-sheet
for printing Icon files using the a2ps utility.  I'm hoping
that someone will take this and improve upon it (the handling
of operators could be a lot better).  If you do improve it,
please send me the improved version (and "Thanks"!).

--
Steve Wampler-  SOLIS Project, National Solar Observatory
swampler@noao.edu
--------------351713C02FB9FE89DD635AF3
Content-Type: text/plain; charset=us-ascii;
 name="icon.ssh"
Content-Disposition: inline;
 filename="icon.ssh"
Content-Transfer-Encoding: 7bit

# Style sheet for Icon
#

#
# This file is part of a2ps.
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation; either version 2, or (at your option)
# any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program; see the file COPYING.  If not, write to
# the Free Software Foundation, 59 Temple Place - Suite 330,
# Boston, MA 02111-1307, USA.
#

## 1.0 Steve Wampler

style Icon is
version is 1.0
requires a2ps version 4.10
written by "Steve Wampler <sbw@tapestry.tucson.az.us>"

first alphabet is
   "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz_#$&"
second alphabet is
   "0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz._"
case sensitive

keywords in Keyword are
   "\&allocated", "\&ascii", "\&clock", "\&collections", "\&cset",
   "\&current", "\&date", "\&dateline", "\&digits", "\&dump", "\&e",
   "\&error", "\&errornumber", "\&errortext", "\&errorvalue",
   "\&errout", "\&fail", "\&features", "\&file", "\&host", "\&input",
   "\&lcase", "\&letters", "\&level", "\&line", "\&main", "\&null",
   "\&outpout", "\&phi", "\&pi", "\&pos", "\&progname", "\&random",
   "\&regions", "\&source", "\&storage", "\&subject", "\&time",
   "\&trace", "\&ucase", "\&version"
end keywords

keywords in Keyword_strong are
   "break", "by", "case", "create", "default", "do", "else", "end",
   "every", "fail", "global", "if", "initial", "invocable", "link",
   "local", "next", "not", "of", "procedure", "record", "repeat",
   "return", "static", "suspend", "then", "to", "until", "while"
end keywords

optional operators are
   #  This needs more work!

   <- \leftarrow,
   ~= \neq,
   === \equiv,
   <<<=,
   >>>=,
   <<=,
   >>=,
   <= \leq,
   >= \geq,
   ~ \not
end operators

sequences are

    # Comments (could make this "Comment_strong" to stand out more...)
    "#" Comment,

    # Strings
    C-string,

    # Csets
    "'" Plain String "'" Plain
       exceptions are
	"\\'", "\\\\"
       end exceptions,

    # Special case procedure declarations so names stand out
    "procedure " Keyword_strong Label_strong /[ \t]*(/ Plain

end sequences

end style

--------------351713C02FB9FE89DD635AF3--



From icon-group-sender  Mon Oct  4 08:42:18 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id IAA00712
	for icon-group-addresses; Mon, 4 Oct 1999 08:42:09 -0700 (MST)
Message-Id: <199910041542.IAA00712@baskerville.CS.Arizona.EDU>
X-Authentication-Warning: agate-ether.berkeley.edu: news set sender to <news> using -f
From: luvisi@andru.sonoma.edu
X-Newsgroups: comp.lang.icon
Subject: Re: A small puzzle
Date: 02 Oct 1999 15:23:04 -0700
To: Steve Wampler <sbw@tapestry.tucson.az.us>
X-Newsreader: Gnus v5.3/Emacs 19.34
To: icon-group@optima.CS.Arizona.EDU
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Steve Wampler <sbw@tapestry.tucson.az.us> writes:
[snip]
>    Write an Icon program to generate the pairings in a round-robin
>    tournament.  The program should accept the player names as
>    command line arguments (see the example below).
[snip]

Here's my attempt:

procedure main()
 every gameset := turnament(["a","b","c","d","e"])
  do every write((left((pair := !gameset)[1], 5) || left(pair[2], 5)) | "");
 every gameset := turnament(["a","b","c","d","e","f"])
  do every write((left((pair := !gameset)[1], 5) || left(pair[2], 5)) | "");
end

procedure turnament(people)
 local i;
 local stationary_player, gameset;

 stationary_player := (*people % 2 = 0 & get(people)) | "Buy";
 every 1 to *people do {
   gameset := [ [ stationary_player, people[*people] ] ];
   every gameset |||:= [[people[ i := 1 to (*people - 1) / 2 ], people[-i-1]]];
   suspend gameset;
   push(people, pull(people));
 }
 fail;
end

[snip]
> (As an additional challenge, implement an optional -n argument, where
> n is the number of courts - this is as much an exercise in determining
> a reasonable policy for handling the pairing overfow as it is an
> exercise in Icon programming.)
[snip]

hm... it seems the hardest part would be when at the end of a virtual
match, finding the pairs in the next virtual match that don't contain
anybody who's playing in the curent physical match...

Andru
-- 
-------------------------------------------------------------------------- 
| Andru Luvisi                 | http://libweb.sonoma.edu/		 |
| Programmer/Analyst           |   Library Resources Online              | 
| Ruben Salazar Library        |-----------------------------------------| 
| Sonoma State University      | http://www.belleprovence.com/		 |
| luvisi@andru.sonoma.edu      |   Textile imports from Provence, France |
--------------------------------------------------------------------------


From icon-group-sender  Mon Oct  4 09:07:27 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id JAA01781
	for icon-group-addresses; Mon, 4 Oct 1999 09:07:12 -0700 (MST)
Message-Id: <199910041607.JAA01781@baskerville.CS.Arizona.EDU>
Date: Sun, 03 Oct 1999 14:45:55 -0700
From: Blake Chapman <blakec@pacbell.net>
Subject: Any File System for Icon Yet?
To: icon-group@optima.CS.Arizona.EDU
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Icon Project,

I have downloaded, installed, and done some little
programs in Icon for Windows.  I find that the documentation
that comes with the package is missing library routines
that are actually present; that is, it doesn't document them.
The free manual from
<http://www.toolsofcomputing.com/freesoftware.htm>
documents some missing routines, like xencode() and
xdecode(), and they work.  I find no routines to pluck
one element out of the middle of a list, or insert one in.
Are there such routines?   Would I get documentation
of the if I got the book?  At $35 a student like myself
has to find out whether Icon has enough library routines
to support real programming problems before I buy it.
There is no mention of any file system in downloaded
documentation.  To work on persistent data I have
to write a whole list out with xencode() and read it in
with xdecode().  I read in the Dec. 1, 1998 Icon
Newsletter the article titled "Database Connectivity":
"Work is beginning at UT San Antonio, with generous support from the
National
Library of Medicine, on a high-level database interface to Icon. The
initial
implementation will emphasize Windows Icon ...".

Is this done yet?  If not, is there even a primitive file system
available, with a B-tree index method and data file
access routines?

Icon is a nice language; very powerful!

Blake Chapman




From icon-group-sender  Mon Oct  4 12:38:36 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id MAA09976
	for icon-group-addresses; Mon, 4 Oct 1999 12:38:02 -0700 (MST)
Message-Id: <199910041938.MAA09976@baskerville.CS.Arizona.EDU>
From: eka@corp.cirrus.com (Eka Laiman)
Subject: Re: A small puzzle
To: sbw@tapestry.tucson.az.us (Steve Wampler)
Date: Mon, 4 Oct 1999 10:49:38 -0700 (PDT)
Cc: icon-group@optima.CS.Arizona.EDU
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

> I haven't posted anything for a while, but thought this might be fun:
> 
>    Write an Icon program to generate the pairings in a round-robin
>    tournament.  The program should accept the player names as
>    command line arguments (see the example below).
> 
> In a round-robin tournament, every player plays every other player
> exactly once.  If there are an odd number of players, then a
> special "bye" player is added.
> ................. edited ..................

If I understand the problem correctly, here is how I approach the
problem:

Model the problem as "graph" in the following sense:

    Each player forms a vertex of the graph. Each pair of player
    is an edge in the graph.

>From the condition that every player plays every other player exactly
once, we see the following graph property: each vertex is connected
to every vertex in the graph, forming a "complete graph".

Let N be the number of the vertices in the graph, clearly there are
	N(N-1)/2
edges.

>From this analysis we come up with the following algorithm:

    Let player[1..n] be the array of the players.

    1.  Set up the second array partner[1..n] such that:
	    partner[i] := player[n - i + 1]
        # initialize partner as the reverse of player.
    2.  Repeat the following steps (n - 1) times:
        2a. For i := 1 to n/2 do
		write(pair(player[i], partner[i])
	2b. Rotate array partner one step.

>From the rule of counting, we see that step 2 generates all edges
in our complete graph.

I think the implementation of the above algorithm is straightforward.

-eka-


From icon-group-sender  Mon Oct  4 12:39:15 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id MAA10026
	for icon-group-addresses; Mon, 4 Oct 1999 12:39:01 -0700 (MST)
Message-Id: <199910041939.MAA10026@baskerville.CS.Arizona.EDU>
From: gep2@terabites.com
Date: Mon, 04 Oct 1999 13:13:51 -0500
Subject: Any File System for Icon Yet?
To: blakec@pacbell.net, icon-group@optima.CS.Arizona.EDU
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

> I have downloaded, installed, and done some little
programs in Icon for Windows.  I find that the documentation
that comes with the package is missing library routines
that are actually present; that is, it doesn't document them.
The free manual from
<http://www.toolsofcomputing.com/freesoftware.htm>
documents some missing routines, like xencode() and
xdecode(), and they work.  

I haven't used (specifically) Icon for Windows, although I've worked with the 
command-line version to a fair degree.  Others here will certainly know more 
about it, but I'll offer at least some observations...

> I find no routines to pluck one element out of the middle of a list, or insert 
one in.  Are there such routines?   

Perhaps not, since those functions are really pretty trivial (you can use 
subscript ranges to specify sublists, making that kind of thing a simple 
expression).

> Would I get documentation of the if I got the book?  

Dunno which book in particular you're referring to.  The book I have and use, 
'The Icon Programming Language' is a very complete description of the basic 
language.

> At $35 a student like myself has to find out whether Icon has enough library 
routines to support real programming problems before I buy it.

That, of course, depends entirely upon what kind of "real programming problems" 
you intend to tackle with it.

One serious missing thing, for business type programming, is any kind of 
indexed-sequential file system.  (If there is such by now, I'm not aware of it).

On the other hand, I'm not aware of ANY general-purpose programming language 
which has ALL of every kind of support routine that anyone is likely to ever 
need, already built-in...

> There is no mention of any file system in downloaded
documentation.  To work on persistent data I have
to write a whole list out with xencode() and read it in
with xdecode().  

There are other ways to import and export lists.  In general, I think that Icon 
is better equipped for this than most other programming languages are.

In SNOBOL4/SPITBOL (predecessor languages to Icon) there is a SAVE() function 
which allows the entire workspace (program and data) to be written to disk for 
reloading later, and that's a way to instantly save (and later restore) even 
very complex data structures without having to develop a separate 
importable/exportable disk-file-based format for them.

> I read in the Dec. 1, 1998 Icon Newsletter the article titled "Database 
Connectivity":  "Work is beginning at UT San Antonio, with generous support from 
the National Library of Medicine, on a high-level database interface to Icon. 
The initial implementation will emphasize Windows Icon ...".

> Is this done yet?  If not, is there even a primitive file system available, 
with a B-tree index method and data file access routines?

With the increasing emphasis on multi-user transaction processing, it's perhaps 
true that it's becoming more appropriate and desirable to provide a more general 
interface to an external SQL (or other) database system... rather than trying to 
keep the whole database system within the individual program.

> Icon is a nice language; very powerful!

I couldn't agree more!!!  Despite the fact that for my own personal use, I find 
SNOBOL4/SPITBOL (which has many of the same concepts, although a very different 
overall look) to be a very attractive alternative.  I end up using S*BOL much 
more than Icon.  NT SPITBOL produces true .EXE files, compiles and executes VERY 
quickly, reliably handles exceedingly large amounts of data, and is very robust. 
 It also provides for the eval() and code() functions which are missing from 
Icon (allowing you to extend the program and reference variables symbolically in 
expressions at runtime).

Anyhow, Clint is quite active on this list, and will surely know more about the 
status of Icon for Windows... good luck!

Gordon Peterson
http://web2.airmail.net/gep2/
Support the Anti-SPAM Amendment!  Join at http://www.cauce.org/
12/19/98: the day the Conservatives demonstrated their scorn for their
   fraudulent sham of representative government.  Voters, remember it!



From icon-group-sender  Wed Oct  6 12:29:10 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id MAA12810
	for icon-group-addresses; Wed, 6 Oct 1999 12:26:46 -0700 (MST)
Message-Id: <199910061926.MAA12810@baskerville.CS.Arizona.EDU>
Date: Wed, 6 Oct 1999 10:19:55 -0700 (MST)
From: Gregg Townsend <gmt>
To: icon-group
Subject: Version 2.1 of Jcon
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Version 2.1 of Jcon, a Java-based implementation of Icon, is now
available for free downloading from

	http://www.cs.arizona.edu/icon/jcon/

Version 2.1 includes minor feature additions, documentation edits,
and bug fixes. Changes include the following:

   * A directory can be read by opening it as a file.
   * JPEG images can be written under Java 2 implementations.
   * Java class files can be bundled with a Jcon program for easier
     dynamic loading.
   * The runtime system can be bundled with a Jcon program to make
     it completely self-contained.
   * Large integers now work with to-by, seq(), limitation (e1 \ e2),
     and exponentiation (e1 ^ e2).
   * A zero increment value is diagnosed by seq().
   * &host no longer spawns a shell invocation of the "uname" utility.
   * The runtime package name has been changed from "rts" to "jcon".
     All programs must be recompiled.

The release notes can be viewed at
	http://www.cs.arizona.edu/icon/jcon/doc/relnotes.htm

A compiled version of the current (9.3.2) Icon Program Library
for Jcon 2.1 is also available from the Jcon web page.

   ---------------------------------------------------------------------------
   Gregg Townsend         Staff Scientist      The University of Arizona
   gmt@cs.arizona.edu     Computer Science     Tucson, Arizona, USA


From icon-group-sender  Thu Oct  7 12:26:03 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id MAA28331
	for icon-group-addresses; Thu, 7 Oct 1999 12:25:00 -0700 (MST)
Message-Id: <199910071925.MAA28331@baskerville.CS.Arizona.EDU>
Date: Thu, 07 Oct 1999 09:33:25 -0700
From: Steve Wampler <swampler@noao.edu>
X-Accept-Language: en
To: icon-group@optima.CS.Arizona.EDU
CC: Steve Wampler <sbw@tapestry.tucson.az.us>
Subject: Re: A small puzzle
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

This is a multi-part message in MIME format.
--------------D3D91CF3B7EB1333F3CB751F
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Steve Wampler wrote:
> 
>    Write an Icon program to generate the pairings in a round-robin
>    tournament.  The program should accept the player names as
>    command line arguments (see the example below).
> 

Ok, I don't think anyone else is going to submit a response, so here's
a quick summary.  Two people submitted working solutions - Gregg Townsend
and Andru Luvisi.  Andru's solution needs a (very) minor modification
to meet the requirement of using names from the command-line arguments,
but that's a pretty trivial change (I've made it in the attachment).
Gregg's solution is pretty interesting, it's crafted from scratch and
uses a mapping to single characters to allow string scanning and cset
operations - it's also the best-commented.  The use of csets limits
the tournament to at most 255 players, which is probably ok in most
real-world situations.  Gregg's solution is also attached.

One person gave a nice graph-theory-based analysis of the problem
with an algorithm for the solution, but without actual code.  The
analysis is good, though the algorithm is mildly wasteful of space.
(I won't repost the algorithm here since it has already shown up in
this group, but you might take a look at it and see if you can
identify the wasted space...).

No one tried either of the two challenges: (a) add an option to
specify the number of 'real' courts and (b) arrange the ordering
so the last round pairs players in the same order they appear on
the command line.  It's probably just as well that no one did (a),
as Andru pointed out, the ugliness is when the number of courts
exceeds the number of players. I've included in my code a solution to
(b) - the same strategy could be applied to Andru's solution, but
would need a different initial pattern.

My code is similar to Andru's, using the same basic algorithm
(which is a varient of the graph-based algorithm already posted.
You might look to see how the use of the space is improved.
Can you figure out how to use still less space?)

After seeing how nicely Gregg's code was commented, I went back
and added some comments...

----------------------------------------------------------------
# Generate the pairings for each round of a round-robin tournament.
#
# Usage: roundrobin list-of-players...
#
procedure main(args)
    if *args = 0 then args := ["alice", "bonnie", "carole"]
    doRoundRobin(args)
end

# Do the scheduling for the tournament.
#
#   This implementation produces an 'interesting' final round.  If the
#   order of the names given on the command line corresponds to the
#   seeding order for the tournament, then the final round pairs
#   the top two seeds, the next two, and so on.  This should help
#   keep the spectators around, allowing you to make more money
#   on refreshments.  If you don't want this, comment out the line:
#
#     names := reorder(names)
#
procedure doRoundRobin(names)
    # force an even number of contestents
    if *names % 2 = 1 then put(names,"bye")

    names := reorder(names)

    every outPairs(1 to *names-1, names := adjust(names))
end 

# Adjust the list of players for the next round.
#
#   The algorithm is "leave the first alone, rotate the rest one position"
#
procedure adjust(playerList)
    return playerList[1:2] ||| playerList[3:0] ||| playerList[2:3]
end

# Display the pairings for a single round.
#
procedure outPairs(i, names)
    write("Round ",i)
    every c := 1 to (*names/2) do {   # pair from edges toward center
        write("\tCourt ", right(c,2), ": ", names[c], " vs ", names[-c])
        }
end

# Reorder the names so the first call to adjust() produces an ordering
#   that ensures the final round pairs the players in the original order
#
procedure reorder(names)
    every new := put([], names[(1 to *names by 2) | (*names to 2 by -2)])
    return new
end
-------------------------------------------------------

I hope some of you had fun thinking about this problem!
--
Steve Wampler-  SOLIS Project, National Solar Observatory
swampler@noao.edu
--------------D3D91CF3B7EB1333F3CB751F
Content-Type: text/plain; charset=us-ascii;
 name="andru.icn"
Content-Disposition: inline;
 filename="andru.icn"
Content-Transfer-Encoding: 7bit

procedure main(args)
 every gameset := turnament(args)
  do every write((left((pair := !gameset)[1], 5) || left(pair[2], 5)) | "");
end

procedure turnament(people)
 local i;
 local stationary_player, gameset;

 stationary_player := (*people % 2 = 0 & get(people)) | "Buy";
 every 1 to *people do {
   gameset := [ [ stationary_player, people[*people] ] ];
   every gameset |||:= [[people[ i := 1 to (*people - 1) / 2 ], people[-i-1]]];
   suspend gameset;
   push(people, pull(people));
 }
 fail;
end

--------------D3D91CF3B7EB1333F3CB751F
Content-Type: text/plain; charset=us-ascii;
 name="gregg.icn"
Content-Disposition: inline;
 filename="gregg.icn"
Content-Transfer-Encoding: 7bit

## rrobin.icn -- generate round-robin pairings
#
#  usage:  rrobin name...
#
#  1-Oct-1999 / gmt  (inspired by a Steve Wampler puzzle challenge)
#  
#  This program generates an optimal set of pairings for a round-robin
#  tournament.  For an even number N of players, N-1 rounds on N/2  
#  courts are needed.  For N odd, N rounds on N/2 courts are required,
#  with a "bye" on an additional "court" at each round.

procedure main(args)
   local n, ct, labels, round, name1, name2
   
   if *args > 255 then
      stop("too many names")
   labels := (&cset -- ' ')[1+:*args]
   n := 0
   every round := pairs(labels) do round ? {
      write("Round ", n +:= 1)
      ct := 0
      while =" " do {
         name1 := args[upto(move(1), labels)]
         name2 := args[upto(move(1), labels)]
         write("\tCourt  ", ct +:= 1, ": ", name1, " vs ", name2)
         }
      if name1 := args[upto(labels -- round, labels)] then
         write("\tCourt  ", ct +:= 1, ": bye vs ", name1)
      }
end


## pairs(players) -- generate pairings for round-robin tournament
#
#  The characters of the string "players" label contestants to be paired
#  in a round-robin tournament.  This procedure generates an optimal set
#  of rounds such that at the end each player is paired exactly once with
#  each other player.  Characters in "players" must be unique and must not
#  include the space character (' '); thus a maximum of 255 contestants
#  can be handled.
#
#  Output is a sequence of strings where each string represents one
#  round of disjoint pairings.  Each pairing consists of three
#  characters: a space followed by the labels of the two contestants.
#
#  1-Oct-1999 / gmt

procedure pairs(players)
   if *players < 2 then
      fail
   else if *players = 2 then
      return " " || players
   else if *players % 2 = 1 then
      suspend oddpairs(players)
   else
      suspend evenpairs(players)
end



## oddpairs(players) -- generate pairings for an odd number of contestants
#
#  With an odd number N of players, the optimum solution requires N rounds
#  on N/2 courts with one "bye" in each round.  A regular pattern produces
#  the pairings:  Each round pairs 1 vs N, 2 vs N-1, 3 vs N-2, etc.,
#  designating a different player as #1 for each of N rounds.

procedure oddpairs(players)
   local n, i, j, round

   n := *players
   every i := 1 to n do {
      round := ""
      every j := 1 to n / 2 do
         round ||:= " " || players[j] || players[n - j]
      players := players[2:0] || players[1]
      suspend round
      }
end



## evenpairs(players) -- generate pairings for an even number of contestants
#
#  With an even number N of players, N-1 rounds on N/2 courts are needed.
#  There are no byes.  First, the players are divided into two groups,
#  and the pairings within each group are enumerated in parallel.  If the
#  groups are odd-sized, there is a bye within each group at each round;
#  the two unoccupied players play each other across group boundaries.
#  After the intra-group competition, the remaining combinations of
#  cross-group pairings are enumerated.

procedure evenpairs(players)
   local half, odd, grp1, grp2, p1, p2, round, step, i

   half := *players / 2
   odd := half % 2
   grp1 := players[1+:half]
   grp2 := players[0-:half]

   every put(p1 := [], pairs(grp1))
   every put(p2 := [], pairs(grp2))
   while round := get(p1) || get(p2) do {
      if odd = 1 then
         round ||:= " " || (players -- round)
      suspend round
   }

   grp2 := grp2 || grp2
   every step := odd to half - 1 do {
      round := ""
      every i := 1 to half do
         round ||:= " " || grp1[i] || grp2[i + step]
      suspend round
      }
end   


--------------D3D91CF3B7EB1333F3CB751F--



From icon-group-sender  Thu Oct  7 16:24:27 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id QAA08802
	for icon-group-addresses; Thu, 7 Oct 1999 16:22:41 -0700 (MST)
Message-Id: <199910072322.QAA08802@baskerville.CS.Arizona.EDU>
Date: Thu, 7 Oct 1999 16:19:32 -0700 (MST)
From: Gregg Townsend <gmt>
To: icon-group
Subject: Technical Report on Jcon
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

A technical report describing Jcon is now available:

    Todd A. Proebsting and Gregg M. Townsend 
    A New Implementation of the Icon Language 

    We describe Jcon, a new, Java-based implementation of the
    Icon programming language. The implementation includes a
    compiler and runtime system. The runtime system is novel
    in its concise and efficient object-oriented implementation
    of a dynamically typed language, as well as its simple
    mechanism for realizing Icon generators.

245KB PostScript:    ftp://ftp.cs.arizona.edu/reports/1999/TR99-13.ps
192KB PDF:           http://www.cs.arizona.edu/icon/jcon/impl.pdf

   ---------------------------------------------------------------------------
   Gregg Townsend         Staff Scientist      The University of Arizona
   gmt@cs.arizona.edu     Computer Science     Tucson, Arizona, USA


From icon-group-sender  Thu Oct 14 07:53:31 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id HAA17829
	for icon-group-addresses; Thu, 14 Oct 1999 07:51:22 -0700 (MST)
Message-Id: <199910141451.HAA17829@baskerville.CS.Arizona.EDU>
Date: 14 Oct 99  09:53:05 BST
From: rjhare@ed.ac.uk
Subject: &time keyword
To: icon-group@optima.CS.Arizona.EDU
Organisation:  Edinburgh Parallel Computing Centre
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Anyone know where Icon gets the value it uses for the &time keyword
on Unix systems? My time values on my local machine (which was
rebuilt last weekend) are suddenly out by 1 hour...

Thanks.

Roger Hare


From icon-group-sender  Fri Oct 15 07:24:40 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id HAA18956
	for icon-group-addresses; Fri, 15 Oct 1999 07:24:00 -0700 (MST)
Message-Id: <199910151424.HAA18956@baskerville.CS.Arizona.EDU>
Date: Thu, 14 Oct 1999 19:09:34 -0700 (MST)
From: Gregg Townsend <gmt@baskerville.CS.Arizona.EDU>
To: icon-group@optima.CS.Arizona.EDU, rjhare@ed.ac.uk
Subject: Re: &time keyword
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

    Date: 14 Oct 99  09:53:05 BST
    From: rjhare@ed.ac.uk
    Subject: &time keyword
    
    Anyone know where Icon gets the value it uses for the &time keyword
    on Unix systems? My time values on my local machine (which was
    rebuilt last weekend) are suddenly out by 1 hour...

Do you mean &clock?  &time measures elapsed processor time.

Icon v9.3.2 calls the POSIX localtime() function for &clock, &date, and
&dateline.  This is influenced, among other things, by the setting of the
TZ environment variable.

What's your machine architecture and OS?  What does &version report?

   ---------------------------------------------------------------------------
   Gregg Townsend         Staff Scientist      The University of Arizona
   gmt@cs.arizona.edu     Computer Science     Tucson, Arizona, USA
    


From icon-group-sender  Fri Oct 15 12:49:00 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id MAA00812
	for icon-group-addresses; Fri, 15 Oct 1999 12:48:08 -0700 (MST)
Message-Id: <199910151948.MAA00812@baskerville.CS.Arizona.EDU>
Date: Fri, 15 Oct 1999 09:08:44 -0700 (MST)
From: Gregg Townsend <gmt>
To: icon-group, rjhare@ed.ac.uk
Subject: Re: &time keyword
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

    From rjhare@holyrood.ed.ac.uk  Fri Oct 15 01:09:42 1999

    Icon Version 9.3.  November 25, 1996 - I haven't updated...

None of my comments apply to that older version; you'd have to check the
source code to see how it did it.  The clock and time code was rewritten
for Icon 9.3.2 and we found out afterward that the rewrite fixed at least
two previously unknown problems in that area.

   ---------------------------------------------------------------------------
   Gregg Townsend         Staff Scientist      The University of Arizona
   gmt@cs.arizona.edu     Computer Science     Tucson, Arizona, USA


From icon-group-sender  Mon Oct 18 08:55:28 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id IAA04021
	for icon-group-addresses; Mon, 18 Oct 1999 08:53:29 -0700 (MST)
Message-Id: <199910181553.IAA04021@baskerville.CS.Arizona.EDU>
Date: 16 Oct 99  06:50:59 BST
From: rjhare@ed.ac.uk
Subject: Re: &time keyword
To: Gregg Townsend <gmt@baskerville.CS.Arizona.EDU>
cc: icon-group@optima.CS.Arizona.EDU
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

> Do you mean &clock?  &time measures elapsed processor time.

Yes, I did indeed mean &clock. A slip of the digit! My apologies.

The problem is now sorted out - for some reason, the TZ environment variable
was not being set on my login - only on my login, and no-one knows why! It
is said locally that if you want some-one to break the system, send for
Roger. I appear to have done it again...

Thanks for the replies.

Roger Hare


From icon-group-sender  Thu Oct 21 08:00:13 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id HAA16777
	for icon-group-addresses; Thu, 21 Oct 1999 07:58:13 -0700 (MST)
Message-Id: <199910211458.HAA16777@baskerville.CS.Arizona.EDU>
Date: Wed, 20 Oct 1999 19:21:24 -0700
From: Blake Chapman <blakec@pacbell.net>
Subject: List subscription insert/removal
To: icon-group@optima.CS.Arizona.EDU
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Icon Project,

I am trying to figure out how to insert a new element into a list,
and remove an element from the middle of a list (not the end).
In answer to this question previously Gordon Peterson sent
a reply.  I tried what he said and it doesn't work.

My original question, that he answered, was
"I find no routines to pluck one element out of the middle of a list, or insert
one in.  Are there such routines?".

His answer was:
"Perhaps not, since those functions are really pretty trivial (you can use
subscript ranges to specify sublists, making that kind of thing a simple
expression).".

I assume he meant that I can insert an element into a list the same way
I can insert a character into a string variable.  So I tried it, 4 ways; none
of them worked, they all produce a run-time error and abort.  Here is
the program.  Though I use mere charactes or 1-char strings as list
elements, it is just to try out list manipulation...

procedure main()
 local s  # it will be a list

 s:= ["a", "b", "c", "d"]
 WriteL(s)   # this writes "abcd" as it should
  # Next, I want to insert "x" as a list element between "b" and "c" ....
  # didn't work:  s[3:3]:="x"
  # didn't work:  s[3:3] := ["x"]
  # didn't work:  s[3+:0] := ["x"]
  s[3+:0] := "x"   # didn't work; I get a run-time error and program abort
 WriteL(s)
end

procedure WriteL(L)
  local i, len
  len := *L
  i := 1
  while  i <= len do  { writes(L[i])  ;  i +:= 1 }
  write()
end

I get the following error message when I run the above:

Run-time error 111
File E:\icon\wsource\sublist.icn; Line 28   [the line:   s[3+:0] := "x" ]
variable expected
offending value: list_3 = []
Traceback:
   main()
   {list_3 = [] := "x"} from line 28 in E:\icon\wsource\sublist.icn

Can you tell me how to accomplish what I want?

Thank You,
Blake Chapman




From icon-group-sender  Thu Oct 21 15:23:09 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id PAA04497
	for icon-group-addresses; Thu, 21 Oct 1999 15:17:38 -0700 (MST)
Message-Id: <199910212217.PAA04497@baskerville.CS.Arizona.EDU>
From: "Frank Lhota" <lhotaf@lexma.meitech.com>
To: "Blake Chapman" <blakec@pacbell.net>, <icon-group@optima.CS.Arizona.EDU>
Subject: Re: List subscription insert/removal
Date: Thu, 21 Oct 1999 11:20:28 -0000
X-Priority: 3
X-MSMail-Priority: Normal
X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2314.1300
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

I think what Gordon Peterson had in mind was using subscript ranges along
with list catenation, e.g.,

    s := s[1:3] ||| [ "x" ] ||| s[3:0]

There is probably a more efficient way to do this, but this approach is
certainly straightforward.

----- Original Message -----
From: Blake Chapman <blakec@pacbell.net>
To: <icon-group@optima.CS.Arizona.EDU>
Sent: Thursday, October 21, 1999 2:21 AM
Subject: List subscription insert/removal


> Icon Project,
>
> I am trying to figure out how to insert a new element into a list,
> and remove an element from the middle of a list (not the end).
> In answer to this question previously Gordon Peterson sent
> a reply.  I tried what he said and it doesn't work.
>
> My original question, that he answered, was
> "I find no routines to pluck one element out of the middle of a list, or
insert
> one in.  Are there such routines?".
>
> His answer was:
> "Perhaps not, since those functions are really pretty trivial (you can use
> subscript ranges to specify sublists, making that kind of thing a simple
> expression).".
>
> I assume he meant that I can insert an element into a list the same way
> I can insert a character into a string variable.  So I tried it, 4 ways;
none
> of them worked, they all produce a run-time error and abort.  Here is
> the program.  Though I use mere charactes or 1-char strings as list
> elements, it is just to try out list manipulation...
>
> procedure main()
>  local s  # it will be a list
>
>  s:= ["a", "b", "c", "d"]
>  WriteL(s)   # this writes "abcd" as it should
>   # Next, I want to insert "x" as a list element between "b" and "c" ....
>   # didn't work:  s[3:3]:="x"
>   # didn't work:  s[3:3] := ["x"]
>   # didn't work:  s[3+:0] := ["x"]
>   s[3+:0] := "x"   # didn't work; I get a run-time error and program abort
>  WriteL(s)
> end
>
> procedure WriteL(L)
>   local i, len
>   len := *L
>   i := 1
>   while  i <= len do  { writes(L[i])  ;  i +:= 1 }
>   write()
> end
>
> I get the following error message when I run the above:
>
> Run-time error 111
> File E:\icon\wsource\sublist.icn; Line 28   [the line:   s[3+:0] := "x" ]
> variable expected
> offending value: list_3 = []
> Traceback:
>    main()
>    {list_3 = [] := "x"} from line 28 in E:\icon\wsource\sublist.icn
>
> Can you tell me how to accomplish what I want?
>
> Thank You,
> Blake Chapman
>
>



From icon-group-sender  Thu Oct 21 15:43:50 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id PAA05509
	for icon-group-addresses; Thu, 21 Oct 1999 15:42:36 -0700 (MST)
Message-Id: <199910212242.PAA05509@baskerville.CS.Arizona.EDU>
Date: Thu, 21 Oct 1999 09:01:50 -0700
From: Steve Wampler <swampler@noao.edu>
X-Accept-Language: en
CC: icon-group@optima.CS.Arizona.EDU
Subject: Re: List subscription insert/removal
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Blake Chapman wrote:
> 
> Icon Project,
> 
> I am trying to figure out how to insert a new element into a list,
> and remove an element from the middle of a list (not the end).
> In answer to this question previously Gordon Peterson sent
> a reply.  I tried what he said and it doesn't work.
> 
> My original question, that he answered, was
> "I find no routines to pluck one element out of the middle of a list, or insert
> one in.  Are there such routines?".
...
> procedure main()
>  local s  # it will be a list
> 
>  s:= ["a", "b", "c", "d"]
>  WriteL(s)   # this writes "abcd" as it should
>   # Next, I want to insert "x" as a list element between "b" and "c" ....
>   # didn't work:  s[3:3]:="x"
>   # didn't work:  s[3:3] := ["x"]
>   # didn't work:  s[3+:0] := ["x"]
>   s[3+:0] := "x"   # didn't work; I get a run-time error and program abort
>  WriteL(s)
> end

Unfortunately, sublists are not variables, as you've seen with your test
cases.  So trying to assign to a sublist isn't going to work, no matter
how you try and construct it.  I'm not sure why sublists are implemented
this way, but can imagine all sorts of difficulties that might arise from
implementing it otherwise.  (I've often thought that it would be nice
to implement lists so they are more consistent with strings, but the result,
while perhaps quite nice, would no longer be Icon...  Perhaps someone with
more time on their hands would like to think about this more?  (List scanning
would become easier to implement, among other things.)

Anyway, you have to implement the operations you want by constructing
new lists:

procedure replace(a, start, stop, value)
    return a[1:start] ||| value ||| a[stop:0]
end

where the call would be:

    s := replace(s, 3, 3, ["x"])

I agree that this is not as clean as one might hope for, but I don't
think there's any way around this.  

Because of the call-by-value semantics of Icon, you have
to do the "pluck" (where I assume you mean to remove the elements
from the list when you pluck them) as a two step process:

    a := s[3:5]
    s := replace(s, 3, 5, [])

--
Steve Wampler-  SOLIS Project, National Solar Observatory
swampler@noao.edu


From icon-group-sender  Thu Oct 21 15:50:06 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id PAA05811
	for icon-group-addresses; Thu, 21 Oct 1999 15:49:56 -0700 (MST)
Message-Id: <199910212249.PAA05811@baskerville.CS.Arizona.EDU>
From: gep2@terabites.com
Date: Thu, 21 Oct 1999 13:52:24 -0500
Subject: List subscription insert/removal
To: icon-group@optima.CS.Arizona.EDU
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

> I am trying to figure out how to insert a new element into a list,
and remove an element from the middle of a list (not the end).
In answer to this question previously Gordon Peterson sent
a reply.  I tried what he said and it doesn't work.

It does if you do it right.  :-)

> My original question, that he answered, was
"I find no routines to pluck one element out of the middle of a list, or insert
one in.  Are there such routines?".

> His answer was:
"Perhaps not, since those functions are really pretty trivial (you can use
subscript ranges to specify sublists, making that kind of thing a simple
expression).".

> I assume he meant that I can insert an element into a list the same way
I can insert a character into a string variable.  So I tried it, 4 ways; none
of them worked, they all produce a run-time error and abort.  Here is
the program.  Though I use mere charactes or 1-char strings as list
elements, it is just to try out list manipulation...

> procedure main()
 local s  # it will be a list

>  s:= ["a", "b", "c", "d"]
 WriteL(s)   # this writes "abcd" as it should
  # Next, I want to insert "x" as a list element between "b" and "c" ....
  # didn't work:  s[3:3]:="x"
  # didn't work:  s[3:3] := ["x"]
  # didn't work:  s[3+:0] := ["x"]
  s[3+:0] := "x"   # didn't work; I get a run-time error and program abort
 WriteL(s)
end

> procedure WriteL(L)
  local i, len
  len := *L
  i := 1
  while  i <= len do  { writes(L[i])  ;  i +:= 1 }
  write()
end

> I get the following error message when I run the above:
Run-time error 111
File E:\icon\wsource\sublist.icn; Line 28   [the line:   s[3+:0] := "x" ]
variable expected
offending value: list_3 = []
Traceback:
   main()
   {list_3 = [] := "x"} from line 28 in E:\icon\wsource\sublist.icn

> Can you tell me how to accomplish what I want?

Try sectioning the list and concatenating (if you have the Icon book, 1st 
edition, this is on pages 52 and 53).  In particular, for your test example 
above, the code might look like:

   s := s[1:2] ||| ["x"] ||| s[3:0]

The book states (page 53, but anyhow in the description of List Sections):  "In 
particular, a[i:j] is a list that is distinct from 'a', and assignment cannot be 
made to it to change 'a'."

If you don't have a copy of the Icon book, you *really* ought to get a copy.  It 
really is VERY important if you plan to make effective use of the language, and 
will pretty instantly answer most of this kind of question.

Gordon Peterson
http://web2.airmail.net/gep2/
Support the Anti-SPAM Amendment!  Join at http://www.cauce.org/
12/19/98: the day the Conservatives demonstrated their scorn for their
   fraudulent sham of representative government.  Voters, remember it!



From icon-group-sender  Tue Oct 26 14:52:47 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id OAA11594
	for icon-group-addresses; Tue, 26 Oct 1999 14:50:33 -0700 (MST)
Message-Id: <199910262150.OAA11594@baskerville.CS.Arizona.EDU>
X-Authentication-Warning: agate-ether.berkeley.edu: news set sender to <news> using -f
From: anyone@No-Such-Domain.anywhere (see.signature)
X-Newsgroups: comp.lang.icon
Subject: differences in Icon versions
X-Newsreader: slrn (0.9.4.3 UNIX)
Date: 26 Oct 1999 17:28:19 +0100
X-Trace: 26 Oct 1999 17:28:19 +0100, flex111.dnwl.wau.nl
To: icon-group@optima.CS.Arizona.EDU
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO


Hi,

my question relates to differences in the basic versions of ICON. I do
have the book "The Icon Progr. Language" (1983). It is quite old, but i
am wondering if the stuff relating to generators, string scanning,
co-expressions and list processing changed substantially.

thanks,

Marc

-- 
------------------------------------------------------------------------------
email: marc dot hoffmann at users dot whh dot wau dot nl
------------------------------------------------------------------------------


From icon-group-sender  Tue Oct 26 16:31:01 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id QAA15379
	for icon-group-addresses; Tue, 26 Oct 1999 16:30:46 -0700 (MST)
Message-Id: <199910262330.QAA15379@baskerville.CS.Arizona.EDU>
X-Authentication-Warning: agate-ether.berkeley.edu: news set sender to <news> using -f
From: "Sean R. McCorkle" <mccorkle@bnl.gov>
X-Newsgroups: comp.lang.icon
Subject: csets and sets of characters
Date: Tue, 26 Oct 1999 16:20:41 -0400
To: icon-group@optima.CS.Arizona.EDU
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO


In writing an Icon version of a command line options parser,
like getopt() in C & Perl, I decided to first convert the input
string of option characters (ie. "abc:12:") into two sets of
characters, one for options that take arguments, and one for
options which take no arguments.  This is easy using Icon sets,
but it seems more natural somehow to use the built-in cset
datatype.  But I don't see how to test for members in csets
(member( 'abc', "b" ) generates a run time error).  Any sage
advice out there about when to use sets and/or csets?

Also, if anybody has any examples of command line options parsers,
or if there's one I've overlooked the standard libs or literature,
I'd sure be grateful if someone could point 'em out to me.  (I'm a
real novice at Icon)

Thank you all,

--
------------------------------------------------------------------------------
Sean R. McCorkle                          mccorkle@bnl.gov
Biology Department, Genome Group          (516) 344-4270 phone
Brookhaven National Laboratory            (516) 344-3407 fax
Upton, New York  11973
   "Oh plunge your hands in water,  Plunge them in up to the wrist,
      Stare, stare in the basin,  And wonder what you've missed." - W.H. Auden





From icon-group-sender  Tue Oct 26 16:31:24 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id QAA15432
	for icon-group-addresses; Tue, 26 Oct 1999 16:31:12 -0700 (MST)
Message-Id: <199910262331.QAA15432@baskerville.CS.Arizona.EDU>
X-Authentication-Warning: agate-ether.berkeley.edu: news set sender to <news> using -f
From: gmt@optima.CS.Arizona.EDU (Gregg Townsend)
X-Newsgroups: comp.lang.icon
Subject: Re: csets and sets of characters
Date: 26 Oct 1999 15:16:05 -0700
To: icon-group@optima.CS.Arizona.EDU
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Sean R. McCorkle <mccorkle@bnl.gov> wrote:
>
> ...But I don't see how to test for members in csets

any(c,s) succeeds if the character s is in the cset c.

> Also, if anybody has any examples of command line options parsers,
> or if there's one I've overlooked the standard libs or literature,
> I'd sure be grateful if someone could point 'em out to me.

The "standard" parser from the Icon Program Library can be found at

	http://www.cs.arizona.edu/icon/library/procs/options.htm

   ---------------------------------------------------------------------------
   Gregg Townsend         Staff Scientist      The University of Arizona
   gmt@cs.arizona.edu     Computer Science     Tucson, Arizona, USA


From icon-group-sender  Wed Oct 27 07:56:18 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id HAA09400
	for icon-group-addresses; Wed, 27 Oct 1999 07:55:41 -0700 (MST)
Message-Id: <199910271455.HAA09400@baskerville.CS.Arizona.EDU>
X-Sender: whm@pop.primenet.com
Date: Tue, 26 Oct 1999 21:38:01 -0700
To: icon-group@optima.CS.Arizona.EDU
From: "William H. Mitchell" <whm@mse.com>
Subject: Re: csets and sets of characters
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

At 03:16 PM 10/26/99 -0700, Gregg Townsend wrote:
>Sean R. McCorkle <mccorkle@bnl.gov> wrote:
>>
>> ...But I don't see how to test for members in csets
>
>any(c,s) succeeds if the character s is in the cset c.

For those who prefer the scenic route:

   *s = *(c++s)



From icon-group-sender  Wed Oct 27 07:57:44 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id HAA09494
	for icon-group-addresses; Wed, 27 Oct 1999 07:57:38 -0700 (MST)
Message-Id: <199910271457.HAA09494@baskerville.CS.Arizona.EDU>
From: gep2@terabites.com
Date: Wed, 27 Oct 1999 01:43:11 -0500
Subject: csets and sets of characters
To: icon-group@optima.CS.Arizona.EDU
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

> In writing an Icon version of a command line options parser,
like getopt() in C & Perl, I decided to first convert the input
string of option characters (ie. "abc:12:") into two sets of
characters, one for options that take arguments, and one for
options which take no arguments.  This is easy using Icon sets,
but it seems more natural somehow to use the built-in cset
datatype.  But I don't see how to test for members in csets
(member( 'abc', "b" ) generates a run time error).  Any sage
advice out there about when to use sets and/or csets?

It SOUNDS like you're trying to write it the way you'd write it in C or Perl, 
and that's usually a bad sign.

Usually what I think you really want to do is instead to use string scanning to 
process your options.  This might be very different of course depending on just 
what you're trying to do, but I'd think you'd want something more based on 
something like 

   tab(any(noargparms) | (any(argparms) many("0123456789")))

Sorry to not provide a more concrete example, but it really does depend on just 
what you're trying to accomplish... and what you want your command line to look 
like.

Gordon Peterson
http://web2.airmail.net/gep2/
Support the Anti-SPAM Amendment!  Join at http://www.cauce.org/
12/19/98: the day the Conservatives demonstrated their scorn for their
   fraudulent sham of representative government.  Voters, remember it!



From icon-group-sender  Fri Nov  5 10:56:47 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id KAA21852
	for icon-group-addresses; Fri, 5 Nov 1999 10:53:23 -0700 (MST)
Message-Id: <199911051753.KAA21852@baskerville.CS.Arizona.EDU>
X-Authentication-Warning: agate.berkeley.edu: news set sender to <news> using -f
From: Andrew McGhie <ceeam8@cee.hw.ac.uk>
X-Newsgroups: comp.lang.icon
Subject: How do I create an array of records??
Date: Mon, 01 Nov 1999 10:48:44 +0000
X-Trace: glencoe.hw.ac.uk 941453140 29905 137.195.187.207 (1 Nov 1999 10:45:40 GMT)
X-Complaints-To: news
To: icon-group@optima.CS.Arizona.EDU
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

I've just started programming with Icon and was wondering how to create
an array of records, for instance a database of people.
I've declared a record as follows:
   record Person(surname, forename, age, height)
but what do I do after this?  Or is there a better was to do it?

Thanks for your time
ANDY.



From icon-group-sender  Fri Nov  5 11:00:16 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id KAA22015
	for icon-group-addresses; Fri, 5 Nov 1999 10:58:32 -0700 (MST)
Message-Id: <199911051758.KAA22015@baskerville.CS.Arizona.EDU>
X-Authentication-Warning: agate.berkeley.edu: news set sender to <news> using -f
From: Steve Wampler <swampler@noao.edu>
X-Newsgroups: comp.lang.icon
Subject: Re: How do I create an array of records??
Date: Fri, 05 Nov 1999 08:17:34 -0700
X-Trace: noao.edu 941815056 33041 140.252.38.6 (5 Nov 1999 15:17:36 GMT)
X-Complaints-To: abuse@noao.edu
X-Accept-Language: en
To: icon-group@optima.CS.Arizona.EDU
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Andrew McGhie wrote:
> 
> I've just started programming with Icon and was wondering how to create
> an array of records, for instance a database of people.
> I've declared a record as follows:
>    record Person(surname, forename, age, height)
> but what do I do after this?  Or is there a better was to do it?

Hmmm, you have lots of ways to go at this point.  For example,
if you really just want an array of records, use a list:

    db := list()		# create the 'array' (db declared global)

    procedure addPerson(surname, forename, age, height)
        put(db, Person(surname, forename, age, height))
    end

However, it's more likely you'll want to have some way to quickly
look up entries in the database, so you might prefer to have
several tables for quickly locating items using different keys:

    dbNames  := table()
    dbAge    := table()
    dbHeight := table()

(Note for the more Icon literate - this could be written as:

    every (dbNames | dbAge | dbHeight) := table()

but not as:

    dbNames := dbAge := dbHeight := table()

do you see why?)

    procedure addPerson(surname, forename, age, height)
        person := Person(surname, forename, age, height)
        /(dbNames[surname]) := []     # Use lists to allow
        /(dbAge[age]) := []           #    for collisions
        /(dbHeight[height]) := []
        put(dbNames[surname], person)
        put(dbAge[age], person)
        put(dbHeight[height], person)
    end

Of course, if you're writing a "real" application, you would want
something considerably more sophisticated and with some sanity
checks along the way - and probably want to separate the
database 'operations' [store, search, retrieve, etc] from the
actual items being stored, as in:

    procedure addPerson(surname, forename, age, height)
        person := Person(surname, forename, age, height)
        dbStore(person)
    end

  

--
Steve Wampler-  SOLIS Project, National Solar Observatory
swampler@noao.edu


From icon-group-sender  Fri Nov  5 17:06:17 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id RAA04185
	for icon-group-addresses; Fri, 5 Nov 1999 17:04:16 -0700 (MST)
Message-Id: <199911060004.RAA04185@baskerville.CS.Arizona.EDU>
X-Authentication-Warning: agate.berkeley.edu: news set sender to <news> using -f
X-Newsgroups: comp.lang.icon
From: jrinkel@no.email.com (Jay Rinkel)
Subject: Imbeddable Icon?
X-Newsreader: News Xpress 2.01
X-Trace: typ11.nn.bcandid.com 941841136 216.180.15.69 (Fri, 05 Nov 1999 17:32:16 EST)
Date: Fri, 05 Nov 1999 22:31:51 GMT
To: icon-group@optima.CS.Arizona.EDU
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Hi!  I was wondering if Icon was designed to be set up as a control/scripting 
language to run inside an existing C/C++/Java program?  Seems like I read that 
Icon had some limited support for calling external routines but the Icon 
interpreter was considered to be the main executable while anything it calls 
was considered to be "support" type routines.  What would be nice is to have 
the ability to imbed an Icon interpreter inside the desired application and 
the application can call what ever procedure of the icon "program" it wanted.  
Then the ability to map Icon procedure names to functions/methods of the host 
application would also be nice.  Is this currently possible?  If not, has 
there been any thought in making a version of the interpreter which could be 
imbedded as I mentioned?

I have been programming in Icon (mostly from a casual standpoint to build 
little "helper" programs)  off and on for about 5 years and I personally see a 
great possibility in allowing Icon to be used as a scripting language for 
other applications.

Jay Rinkel


From icon-group-sender  Wed Nov 10 09:44:51 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id JAA29118
	for icon-group-addresses; Wed, 10 Nov 1999 09:42:51 -0700 (MST)
Message-Id: <199911101642.JAA29118@baskerville.CS.Arizona.EDU>
X-Authentication-Warning: agate.berkeley.edu: news set sender to <news> using -f
From: Anders Holtsberg <andersh@maths.lth.se>
X-Newsgroups: comp.lang.icon
Subject: List question
Date: Wed, 10 Nov 1999 10:17:23 +0100
X-Accept-Language: en
To: icon-group@optima.CS.Arizona.EDU
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO


--------------2D064CE7C06019CDC4A897F9
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit


A string can be changed in place

   s[4:6] = " hello "

A list can be appended at the end and at the beginning

  every put(listA, !listB)

But does the language definition forbid

  listA[5+:0] := listB

or what is the handy way to delete and insert elements
in a list supplied as an argument? If I have

  procedure f(mylist)

     put(mylist, "hello") # This is OK
     mylist[5+:0] := ["hello"] # Run time error. The book states
    #  that run time error is indeed expected.
     mylist := mylist[1:5] ||| ["hello"] ||| mylist[5:0]  # Here the
caller is not affected!

1) Have I understood? Is it impossible to write a routine
listinsert(A,B,i)
     and that we have to write A:=listinsert(A,B,i)
2) What was the language design objectives to rule out
            listA[i:j] := listB
      There is no semantic problem , listA[i] is an element and
listA[i+:1] is a
      sublist (or rather today only a list of its own since it cannot be
a left-value).
3) I figure there are dark corners here since I saw in a list library
that
     mr Griswold himself thought about the difficulties in
similarity/differences
     between string filosophy and list filosophy.

Comments anybody?

(the code is from top of my head, I only have Icon at home just now)

Andy

--
__________________________________________________________________
Anders Holtsberg        Matematisk statistik   Telefon 046-2224953
andersh@maths.lth.se    Matematikcentrum       Telefax 046-2224998
www.maths.lth.se        Lunds universitet      Box 118, 22100 Lund



--------------2D064CE7C06019CDC4A897F9
Content-Type: text/html; charset=us-ascii
Content-Transfer-Encoding: 7bit

<!doctype html public "-//w3c//dtd html 4.0 transitional//en">
<html>
&nbsp;
<br>A string can be changed in place
<p>&nbsp;&nbsp; s[4:6] = " hello "
<p>A list can be appended at the end and at the beginning
<p>&nbsp; every put(listA, !listB)
<p>But does the language definition forbid
<p>&nbsp; listA[5+:0] := listB
<p>or what is the handy way to delete and insert elements
<br>in a list supplied as an argument? If I have
<p>&nbsp; procedure f(mylist)
<p>&nbsp;&nbsp;&nbsp;&nbsp; put(mylist, "hello") # This is OK
<br>&nbsp;&nbsp;&nbsp;&nbsp; mylist[5+:0] := ["hello"] # Run time error.
The book states
<br>&nbsp;&nbsp;&nbsp; #&nbsp; that run time error is indeed expected.
<br>&nbsp;&nbsp;&nbsp;&nbsp; mylist := mylist[1:5] ||| ["hello"] ||| mylist[5:0]&nbsp;
# Here the caller is not affected!
<p>1) Have I understood? Is it impossible to write a routine listinsert(A,B,i)
<br>&nbsp;&nbsp;&nbsp;&nbsp; and that we have to write A:=listinsert(A,B,i)
<br>2) What was the language design objectives to rule out
<br>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
listA[i:j] := listB
<br>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; There is no semantic problem , listA[i]
is an element and listA[i+:1] is a
<br>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; sublist (or rather today only a list
of its own since it cannot be a left-value).
<br>3) I figure there are dark corners here since I saw in a list library
that
<br>&nbsp;&nbsp;&nbsp;&nbsp; mr Griswold himself thought about the difficulties
in similarity/differences
<br>&nbsp;&nbsp;&nbsp;&nbsp; between string filosophy and list filosophy.
<p>Comments anybody?
<p>(the code is from top of my head, I only have Icon at home just now)
<p>Andy
<pre>--&nbsp;
__________________________________________________________________
Anders Holtsberg&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Matematisk statistik&nbsp;&nbsp; Telefon 046-2224953&nbsp;
andersh@maths.lth.se&nbsp;&nbsp;&nbsp; Matematikcentrum&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Telefax 046-2224998
www.maths.lth.se&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Lunds universitet&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Box 118, 22100 Lund</pre>
&nbsp;</html>

--------------2D064CE7C06019CDC4A897F9--



From icon-group-sender  Wed Nov 10 09:46:47 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id JAA29298
	for icon-group-addresses; Wed, 10 Nov 1999 09:46:36 -0700 (MST)
Message-Id: <199911101646.JAA29298@baskerville.CS.Arizona.EDU>
X-Authentication-Warning: agate.berkeley.edu: news set sender to <news> using -f
From: Anders Holtsberg <andersh@maths.lth.se>
X-Newsgroups: comp.lang.icon
Subject: Re: List question
Date: Wed, 10 Nov 1999 11:53:34 +0100
To: icon-group@optima.CS.Arizona.EDU
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Sorry folks. I just found the very recent discussion on 
this topic on Dejanews. But still: why not let

    mylist[5+:0] := anotherlist

be a legal construct? I e the only change would be to 
let mylist[i:j] produce an l-value instead of an r-value,
it would break no code at all, would it?

Is the implementation difficult? Are there filosophical
or implementation problems? (I'm not nagging you guys: I can
try to do the work if you think the change in semantics is OK).


Anders Holtsberg

-- 
__________________________________________________________________
Anders Holtsberg        Matematisk statistik   Telefon 046-2224953 
andersh@maths.lth.se    Matematikcentrum       Telefax 046-2224998
www.maths.lth.se        Lunds universitet      Box 118, 22100 Lund


From icon-group-sender  Wed Nov 10 09:48:02 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id JAA29359
	for icon-group-addresses; Wed, 10 Nov 1999 09:47:40 -0700 (MST)
Message-Id: <199911101647.JAA29359@baskerville.CS.Arizona.EDU>
X-Authentication-Warning: agate.berkeley.edu: news set sender to <news> using -f
From: Steve Wampler <swampler@noao.edu>
X-Newsgroups: comp.lang.icon
Subject: Re: List question
Date: Wed, 10 Nov 1999 08:44:04 -0700
X-Trace: noao.edu 942248646 36323 140.252.38.6 (10 Nov 1999 15:44:06 GMT)
X-Complaints-To: abuse@noao.edu
X-Accept-Language: en
To: icon-group@optima.CS.Arizona.EDU
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Anders Holtsberg wrote:
> 
> Sorry folks. I just found the very recent discussion on
> this topic on Dejanews. But still: why not let
> 
>     mylist[5+:0] := anotherlist
> 
> be a legal construct? I e the only change would be to
> let mylist[i:j] produce an l-value instead of an r-value,
> it would break no code at all, would it?

Correct, this change should have no impact on existing code.

> Is the implementation difficult? Are there filosophical
> or implementation problems? (I'm not nagging you guys: I can
> try to do the work if you think the change in semantics is OK).

I imagine others will disagree, but I'd like to see at least
an experimental version of Icon with this change in it.  I
assume Anders means to differentiate list sectioning from
list subscripting and have any list section produce an l-value.
This change would bring us one step closer to providing list-scanning.
In fact, it would be possible to then produce a PDCO to experiment
with list-scanning.

However, it does require a non-trivial implementation - you'd have
to essentially reproduce internally the same types of operations
that people now have to use externally (though you could apply some
heuristics).

For example, implementing:

   a[3:5] := [a,b,c,d]

would require producing a new list, as if the user had done:

   a := a[1:3] ||| [a,b,c,d] ||| a[5:0]

(because list element storage is sequential in memory) thus adding
a single element in the middle of a list could have surprisingly
high runtime cost. It has this cost currently, but the fact that
the user has to explicitly code it means it's not a surprise...

On the other hand:

   a[3:5] := []

could (possibly, depending on the desired semantics, see below) be done in
place on the original list, as could a[3:5] := [1] and a[3:5] := [1,2].

There are also a few major semantic issues to resolve.  How close to the
behavior
of strings should list section assignment go?  Strings and lists have some
fundamental differences that should be considered carefully.  In particular,
note that, depending on the implementation, there could be a major difference
between

   a[3:5] := [a,b,c,d]

and

   a := a[1:3] ||| [a,b,c,d] ||| a[5:0]

(Suppose the assignment b := a is done prior to each of the above
statements. What should the value of b be?).  So list section
assignment does mean that programmers would have to think very carefully
about how they are using lists.  Of course, this is true currently; the
following code catches people sometimes...

    a := [1,2,3,4]
    b := a
    a[2] := 5

Extending the above behavior to list sectioning is where the direct
implementation becomes complicated.  You'd have to change how lists are
represented and provide "list qualifiers" much as strings are
implemented using string qualifiers.

Altering the semantics of list processing to provide even closer semantics
to string processing would be very difficult to implement and would
break existing code (because if you're going to "fix" list sectioning
this way, you'd want to also "fix" list element assignment as well).
I don't think I would be very keen on that drastic of a change.

--
Steve Wampler-  SOLIS Project, National Solar Observatory
swampler@noao.edu


From icon-group-sender  Wed Nov 17 16:30:51 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id QAA03599
	for icon-group-addresses; Wed, 17 Nov 1999 16:28:53 -0700 (MST)
Message-Id: <199911172328.QAA03599@baskerville.CS.Arizona.EDU>
X-Authentication-Warning: agate.berkeley.edu: news set sender to <news> using -f
From: Kostas Oikonomou <oikonomou@att.com>
X-Newsgroups: comp.lang.icon
Subject: Strange table behavior
Date: Wed, 17 Nov 1999 17:48:20 -0500
To: icon-group@optima.CS.Arizona.EDU
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

When I run this program

procedure main()
  local T
  T := table([])
  put(T[1], "A")
  write("*T = ", *T)
  write(T[1][1])
end

I get

*T = 0
A


Can anyone explain to me what's happening here?  Does it have to do with
T being a table of lists?

				Kostas


From icon-group-sender  Thu Nov 18 08:16:37 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id IAA26926
	for icon-group-addresses; Thu, 18 Nov 1999 08:15:55 -0700 (MST)
Message-Id: <199911181515.IAA26926@baskerville.CS.Arizona.EDU>
X-Authentication-Warning: agate.berkeley.edu: news set sender to <news> using -f
X-Newsgroups: comp.lang.icon
From: jrinkel@no.email.com (Jay Rinkel)
Subject: Re: Strange table behavior
X-Newsreader: News Xpress 2.01
X-Trace: typ11.nn.bcandid.com 942884478 216.180.15.233 (Wed, 17 Nov 1999 19:21:18 EST)
Date: Thu, 18 Nov 1999 00:20:08 GMT
To: icon-group@optima.CS.Arizona.EDU
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

In article <383330B4.58AE@att.com>, Kostas Oikonomou <oikonomou@att.com> wrote:
>When I run this program
>
>procedure main()
>  local T
>  T := table([])
>  put(T[1], "A")
>  write("*T = ", *T)
>  write(T[1][1])
>end
>
>I get
>
>*T = 0
>A
>
>
>Can anyone explain to me what's happening here?  Does it have to do with
>T being a table of lists?
>
>                                Kostas

Please see the article at the following web page:  it is an article on the 
Icon Programming language web site and specifically addresses this issue.  It 
explains the situation much better than I can here!

http://www.cs.arizona.edu/icon/progcorn/pc_inl21.htm

Jay Rinkel


From icon-group-sender  Thu Nov 18 08:17:23 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id IAA26980
	for icon-group-addresses; Thu, 18 Nov 1999 08:17:13 -0700 (MST)
Message-Id: <199911181517.IAA26980@baskerville.CS.Arizona.EDU>
From: swampler@noao.edu (Steve Wampler)
Date: Wed, 17 Nov 1999 18:03:59 MST
To: Kostas Oikonomou <oikonomou@att.com>
Subject: Re: Strange table behavior
Cc: icon-group@optima.CS.Arizona.EDU
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

On Nov 17 at  5:48pm, Kostas Oikonomou writes:
} When I run this program
} 
} procedure main()
}   local T
}   T := table([])
}   put(T[1], "A")
}   write("*T = ", *T)
}   write(T[1][1])
} end
} 
} I get
} 
} *T = 0
} A
} 
} Can anyone explain to me what's happening here?  Does it have to do with
} T being a table of lists?

I can explain it, but I can't decide if it should be classified as a bug
or a "feature"...(right now I'm leaning toward "feature")

First:

   T := table([])

This creates a table with a specific list as the default value (note,
you probably didn't mean this, see below).  Let's call this particular
list L_001, so we can see what happens later.

If you were to ask for the size of the table at this time, you'd
expect to see the value 0, because you haven't assigned anything to
the table yet.

Now:

   put(T[1], "A")

This asks for the value of T[1].  Since you haven't assigned anything
to the table yet, you get back the default value, which is L_001 in
this case.  The string "A" is put into the list L_001, increasing the
size of this list by 1.  *Note that you have not yet assigned anything
to the table!*

So:

   write("*T = ", *T)

outputs the size of the table.  Since nothing has been assigned yet,
it's still 0.

Finally:

   write(T[1][1])

this first gets the value of the element T[1].  Since nothing has been
assigned to the table yet, you get the default value, good old list
L_001.  Then the first element of L_001 is printed, producing the "A".

The output is correct, the size of the table has never changed, because
nothing has ever been assigned to the table!  The fact that something
was assigned to the list that was created and used as the default value
for the table doesn't change this fact.

Ok, now - why isn't T := table([]) what you want.  Again, breaking it
down:

    []          creates a list (L_001, say).

table(L_001) says to create a table and make list L_001 the default
value.  Note that it doesn't say "create a new list each time a
new table element is needed".  It says "use the list (L_001)" whenever
you need the default value".  So, for example:

    put(T[1], "A")
    put(T[2], "B")
    put(T[3], "C")
    write(T[4][2])

would output "B" and

    write(*(T["aslkdjfsa"])

would output "3"!  (It's the same list being produced as the
default value, regardless of the reference key.)

What you want to do is something like:

   T := table()

   /T[1] := []		# create a brand new list and assign it only to T[1]
                        #     if nothing has ever been assigned to T[1].
   put(T[1], "A")       # put the A into this brand new list

I hope this helps!

-- 
--
Steve Wampler-  SOLIS Project, National Solar Observatory
swampler@noao.edu


From icon-group-sender  Thu Nov 18 08:18:01 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id IAA27016
	for icon-group-addresses; Thu, 18 Nov 1999 08:17:41 -0700 (MST)
Message-Id: <199911181517.IAA27016@baskerville.CS.Arizona.EDU>
From: Chris.D.Tenaglia@jci.com
Subject: Re: Strange table behavior
To: Kostas Oikonomou <oikonomou@att.com>
Cc: icon-group@optima.CS.Arizona.EDU
Date: Thu, 18 Nov 1999 08:25:52 -0600
X-Priority: 3 (Normal)
X-MIMETrack: Serialize by Router on jwimkrs1.na.jci.com/NA/Johnson_Controls(Release 5.0.1a
 (Intl)|17 August 1999) at 11/18/99 08:29:31 AM
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO


T is a table. The default value is an empty list.
T[1] is a table entry with the value A
*T does not reference a specific entry, so it picks
the default empty list, returning 0 for it's length.
*T[1] would return 1

Chris Tenaglia.





Kostas Oikonomou <oikonomou@att.com> on 11/17/99 04:48:20 PM

To:   icon-group@optima.CS.Arizona.EDU
cc:

Subject:  Strange table behavior


When I run this program

procedure main()
  local T
  T := table([])
  put(T[1], "A")
  write("*T = ", *T)
  write(T[1][1])
end

I get

*T = 0
A


Can anyone explain to me what's happening here?  Does it have to do with
T being a table of lists?

                    Kostas







From icon-group-sender  Fri Nov 19 08:17:56 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id IAA25791
	for icon-group-addresses; Fri, 19 Nov 1999 08:16:16 -0700 (MST)
Message-Id: <199911191516.IAA25791@baskerville.CS.Arizona.EDU>
From: swampler@noao.edu (Steve Wampler)
Date: Thu, 18 Nov 1999 19:14:49 MST
To: icon-project@optima.CS.Arizona.EDU, icon-group@optima.CS.Arizona.EDU
Subject: Uh, nevermind...(re: proposed change to Icon table reference semantics)
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Earlier today I posted a suggested change to the way default-valued
table references would work.  Clint Jeffery was kind enough to
gently point out that the suggestion wasn't, uh, top-quality.
In fact, it wouldn't even solve the original problem it was
addressing.  (I'd pass Clint's explanation on, but it might
be more fun for people to see if they can figure out why it
really wouldn't help after all!)

If you're curious, send me email and I'll explain the error
of my ways.

Anyway, I withdraw the proposal!

-- 
--
Steve Wampler-  SOLIS Project, National Solar Observatory
swampler@noao.edu


From icon-group-sender  Mon Dec  6 08:19:56 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id IAA16800
	for icon-group-addresses; Mon, 6 Dec 1999 08:18:18 -0700 (MST)
Message-Id: <199912061518.IAA16800@baskerville.CS.Arizona.EDU>
X-Authentication-Warning: agate.berkeley.edu: news set sender to <news> using -f
From: glass <msglass@mcs.com>
X-Newsgroups: comp.lang.icon
Subject: Text input vidget on Windows 98
Date: Sun, 05 Dec 1999 04:28:04 -0600
X-Trace: condor.cns.iit.edu 944389371 214753 255.255.255.255 (5 Dec 1999 10:22:51 GMT)
X-Complaints-To: postmaster@iit.edu
X-Accept-Language: en
To: icon-group@optima.CS.Arizona.EDU
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO


I am unable to persuade the text input vidget to work.  

The environment is Windows Icon 9.3.1 running on a W98 computer.

All the keyboard events while the mouse cursor is in the text input
vidget are ignored.  I found them in a "missed" handler in
ProcessEvent() or GetEvents(). The text in the vidget is not updated
in response to key presses.

Pressing return or enter results in a missed event. The text input call
back function is not called.

Mouse clicks within the text input vidget are *not* "missed", only the
keyboard events. With the mouse, I can move the I-beam cursor and
select text.

An experiment I tried was to use VSetState() to change the text.
That works.  The call back function still is not called, however.

I'm new at this--this is the first time I am using VIB--so I
may be missing something obvious.  But the other vidgets I am using seem
to work, and I can't persuade the text input vidget to work in even
a tiny one- or two-vidget screen.

-- Michael Glass

   (Yes, I read the caveat about the text input vidget confusing
    users by not providing enough visual feedback, and calling the
call       back function only on <enter>)


From icon-group-sender  Mon Dec  6 08:23:20 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id IAA16906
	for icon-group-addresses; Mon, 6 Dec 1999 08:23:08 -0700 (MST)
Message-Id: <199912061523.IAA16906@baskerville.CS.Arizona.EDU>
X-Authentication-Warning: agate.berkeley.edu: news set sender to <news> using -f
From: Charles Shirley <cgshirley@worldnet.att.net>
X-Newsgroups: comp.lang.icon
Subject: Low GDI resources while running Win Icon
Date: Sun, 05 Dec 1999 07:35:58 -0700
X-Trace: bgtnsc01.worldnet.att.net 944404552 10692 12.74.0.49 (5 Dec 1999 14:35:52 GMT)
X-Complaints-To: abuse@worldnet.att.net
X-Accept-Language: en
To: icon-group@optima.CS.Arizona.EDU
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

While running a program in Windows Icon 9.3.1 under Windows 98, I was
warned by Norton System Doctor that GDI (Graphics Device Interface)
resources--that is, memory-- were getting low. When I stopped the
program and checked GDI resources, they were 60-70% free. When I
re-started the program and worked with it for a while, resources dropped
until they hit the 20% mark where Norton complained and Win 98 started
responding slowly.

The program uses dialog boxes to do vocabulary tagging. For example, at
every occurrence of "to" in a text, a dialog box displays a segment of
the text and provides buttons to indicate whether "to" is a preposition
or part of an infinitive. The low-GDI warning appears after 200 or so
repetitions.

This behavior seems  like a memory leak. Could there be something I am
doing or not doing to cause the problem? For example, might it help to
force a garbage collection every 100-200 repetitions?

Charles Shirley, cgshirley@worldnet.att.net



From icon-group-sender  Fri Dec 10 09:51:29 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id JAA04487
	for icon-group-addresses; Fri, 10 Dec 1999 09:50:48 -0700 (MST)
Message-Id: <199912101650.JAA04487@baskerville.CS.Arizona.EDU>
From: jeffery@big-bill.CS.UNLV.EDU
To: icon-group@optima.CS.Arizona.EDU
Subject: simple string compression, anyone?
Date: Thu, 09 Dec 1999 18:59:57 -0800
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO


Does anyone have Icon procedures that compress and decompress strings?
Richard Goerwitz contributed a Huffman program to the Icon Program Library
that might be adaptable for my purpose, but perhaps there are some other
routines out there that would make good library procedures.  I grepped
around in the Icon Program Library and didn't see anything else.

Cheers,
Clint Jeffery, jeffery@cs.unlv.edu


From icon-group-sender  Wed Dec 15 12:35:02 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id MAA23339
	for icon-group-addresses; Wed, 15 Dec 1999 12:33:48 -0700 (MST)
Message-Id: <199912151933.MAA23339@baskerville.CS.Arizona.EDU>
X-Authentication-Warning: agate.berkeley.edu: news set sender to <news> using -f
From: Chuck Meyers <chuck.meyers@lmco.com>
X-Newsgroups: comp.lang.icon
Subject: icon to c?
Date: Wed, 15 Dec 1999 06:54:02 -0800
X-Accept-Language: en
To: icon-group@optima.CS.Arizona.EDU
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Is there a tranlation program that will convert icon to c?


chuck.meyers@lmco.com





From icon-group-sender  Wed Dec 15 16:36:04 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id QAA01394
	for icon-group-addresses; Wed, 15 Dec 1999 16:34:12 -0700 (MST)
Message-Id: <199912152334.QAA01394@baskerville.CS.Arizona.EDU>
Date: Wed, 15 Dec 1999 13:01:21 -0700 (MST)
From: Gregg Townsend <gmt@baskerville.CS.Arizona.EDU>
To: chuck.meyers@lmco.com, icon-group@optima.CS.Arizona.EDU
Subject: Re: icon to c?
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

    From: Chuck Meyers <chuck.meyers@lmco.com>
    Is there a tranlation program that will convert icon to c?
    
Not to *readable* C.  There was once a compiler (no longer supported)
that used C as an intermediate step, but you couldn't do anything
with that C code and it worked only in its specialized environment.

   ---------------------------------------------------------------------------
   Gregg Townsend         Staff Scientist      The University of Arizona
   gmt@cs.arizona.edu     Computer Science     Tucson, Arizona, USA


From icon-group-sender  Wed Dec 15 16:38:19 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id QAA01600
	for icon-group-addresses; Wed, 15 Dec 1999 16:38:01 -0700 (MST)
Message-Id: <199912152338.QAA01600@baskerville.CS.Arizona.EDU>
From: "Frank Lhota" <lhotaf@lexma.meitech.com>
To: "Chuck Meyers" <chuck.meyers@lmco.com>, <icon-group@optima.CS.Arizona.EDU>
Subject: Re: icon to c?
Date: Wed, 15 Dec 1999 15:19:39 -0000
X-Priority: 3
X-MSMail-Priority: Normal
X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2314.1300
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Yes, the Icon native compiler (iconc) converts an Icon program into C, then
uses the local C compiler to finish the job. The idea is to use C as a
universal back end. Be forewarned that the C program produced by iconc is
not very readable.

----- Original Message -----
From: Chuck Meyers <chuck.meyers@lmco.com>
To: <icon-group@optima.CS.Arizona.EDU>
Sent: Wednesday, December 15, 1999 2:54 PM
Subject: icon to c?


> Is there a tranlation program that will convert icon to c?
>
>
> chuck.meyers@lmco.com
>
>
>



From icon-group-sender  Thu Dec 16 08:04:48 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id IAA01344
	for icon-group-addresses; Thu, 16 Dec 1999 08:03:52 -0700 (MST)
Message-Id: <199912161503.IAA01344@baskerville.CS.Arizona.EDU>
X-Authentication-Warning: agate.berkeley.edu: news set sender to <news> using -f
From: Federico Balbi <fbalbi@lonestar.jpl.utsa.edu>
X-Newsgroups: comp.lang.icon
Subject: Re: icon to c?
Date: Wed, 15 Dec 1999 17:49:12 -0600
X-Trace: ringer.cs.utsa.edu 945301747 23582 129.115.120.1 (15 Dec 1999 23:49:07 GMT)
X-Complaints-To: usenet@ringer.cs.utsa.edu
To: icon-group@optima.CS.Arizona.EDU
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Hi,
  there's iconc that is supposed to be the icon compiler. It translates a 
.icn file in C and produce an executable.

----------------------------------------------------------------------------
 Federico Balbi - University of Texas at San Antonio 
 e-mail: fbalbi@lonestar.utsa.edu - URL: http://www.geocities.com/fedebalbi 

On 15 Dec 1999, Chuck Meyers wrote:

> Is there a tranlation program that will convert icon to c?
> 
> 
> chuck.meyers@lmco.com


From icon-group-sender  Fri Dec 17 08:40:46 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id IAA25895
	for icon-group-addresses; Fri, 17 Dec 1999 08:39:30 -0700 (MST)
Message-Id: <199912171539.IAA25895@baskerville.CS.Arizona.EDU>
Date: Thu, 16 Dec 1999 22:07:13 -0700
From: Anthony Hewitt <anthony.hewitt@attglobal.net>
X-Accept-Language: en
To: icon group <icon-group@optima.CS.Arizona.EDU>
Subject: icon to c
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

How about icn2c in the IPL?

--
Anthony V. Hewitt        anthony.hewitt@ibm.net
===============================================
not -------------------------- (invert failure)
not expr produces the null value if expr fails,
    but fails if expr succeeds




From icon-group-sender  Tue Dec 21 08:16:02 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id IAA17126
	for icon-group-addresses; Tue, 21 Dec 1999 08:14:13 -0700 (MST)
Message-Id: <199912211514.IAA17126@baskerville.CS.Arizona.EDU>
X-Authentication-Warning: agate.berkeley.edu: news set sender to <news> using -f
From: "Walther Senders" <Walther.Senders@daidalos.nl>
X-Newsgroups: comp.lang.icon
Subject: SGML support in ICON
Date: Tue, 21 Dec 1999 09:52:26 +0100
X-Trace: buty.wanadoo.nl 945766126 55094 194.134.112.99 (21 Dec 1999 08:48:46 GMT)
X-Complaints-To: The appropriate party according to NNTP-Posting-Host:
X-Priority: 3
X-MSMail-Priority: Normal
X-Newsreader: Microsoft Outlook Express 5.00.2314.1300
X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2314.1300
To: icon-group@optima.CS.Arizona.EDU
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Hi all,

does somebody know of a library with routines for SGML-support in ICON?

Regards, Walther Senders
Daidalos BV
The Netherlands.




From icon-group-sender  Wed Dec 22 14:16:16 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id OAA15934
	for icon-group-addresses; Wed, 22 Dec 1999 14:11:17 -0700 (MST)
Message-Id: <199912222111.OAA15934@baskerville.CS.Arizona.EDU>
From: Art Eschenlauer <art.eschenlauer@sufsys.com>
To: "'dgamey@ca.ibm.com'" <dgamey@ca.ibm.com>,
        icon-group@optima.CS.Arizona.EDU
Subject: RE: Y2k and Icon?
Date: Wed, 22 Dec 1999 12:04:03 -0600
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

I saw no reply to this question. What is the Y2K readiness of Icon
implementations? I'm particularly interested in the DOS Command Line
implementation.

-----Original Message-----
From: dgamey@ca.ibm.com [mailto:dgamey@ca.ibm.com]
Sent: Tuesday, May 18, 1999 3:13 PM
To: icon-group@optima.CS.Arizona.EDU
Subject: Y2k and Icon?




Hi,

I haven't seen anything on Icon's Y2K readiness.   Did I miss it?  What's
the
status?

I expect I'll have to check all of my IPL contributions, is there a target
date
for this kind of activity?


David





From icon-group-sender  Thu Dec 23 16:21:54 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id QAA00893
	for icon-group-addresses; Thu, 23 Dec 1999 16:20:54 -0700 (MST)
Message-Id: <199912232320.QAA00893@baskerville.CS.Arizona.EDU>
Date: Thu, 23 Dec 1999 09:58:31 -0700 (MST)
From: Gregg Townsend <gmt>
To: icon-group
Cc: art.eschenlauer@sufsys.com, dgamey@ca.ibm.com
Subject: RE: Y2k and Icon?
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

    From: Art Eschenlauer <art.eschenlauer@sufsys.com>
    
    I saw no reply to this question. What is the Y2K readiness of Icon
    implementations? I'm particularly interested in the DOS Command Line
    implementation.


We have not performed a systematic evaluation of any Icon implementation
with respect to Y2K bugs, so we can't make a definitive statement.

I am not aware of any problems in the current release (9.3.2), but that
doesn't mean there aren't any, or that something won't turn up.

As with most free software, Icon is distributed as-is, without warranty,
and we cannot make any guarantees.

   ---------------------------------------------------------------------------
   Gregg Townsend         Staff Scientist      The University of Arizona
   gmt@cs.arizona.edu     Computer Science     Tucson, Arizona, USA



From icon-group-sender  Fri Dec 24 15:37:00 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id PAA23876
	for icon-group-addresses; Fri, 24 Dec 1999 15:36:03 -0700 (MST)
Message-Id: <199912242236.PAA23876@baskerville.CS.Arizona.EDU>
From: "Evans" <evans@gte.net>
To: <icon-group@optima.CS.Arizona.EDU>
Subject: REBOL
Date: Fri, 24 Dec 1999 12:40:37 -0600
X-Priority: 3 (Normal)
X-MSMail-Priority: Normal
Importance: Normal
X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2314.1300
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Dear Icon Group,

Dr. Dobb's just ran an article on the new REBOL language.  I am curious to know
what the Icon Consortium thinks of this language in comparison to Icon.

   http://www.rebol.com/inthenews.html
   http://www.rebol.com/quotes.html

REBOL is an interpreted language with an English-like syntax and is geared
mostly toward Internet functionality.  It also has strong data manipulation
features akin to Icon's.  That is the most I can say at first glance, and leads
to my question:  Where do the philosophies of REBOL and Icon diverge?

While I grasp the key underlying aspects of Icon, I can't get my head around
REBOL.  It appears to me that in some vague sense the philosophies behind the
two languages have much in common.

To take one concrete example, Unicon/Icon2 claims to be a worthy replacement for
Perl for a variety of reasons.  However REBOL makes exactly the same claim, and
is here now for all platforms while Unicon/Icon2 sits in perpetual beta state
and only for Unix.

The author of REBOL is a well-known character in Silicon Valley with a lot of
experience.  He researched over 50 languages before starting the REBOL project.
The REBOL project has attracted venture capital, but the core version is
freeware.  The commercial vision is to have REBOL running every Internet-capable
device as the core messaging engine.  Your Internet Toaster will speak REBOL to
your watch.

The major claimed benefit of REBOL is "dialects" but I fail to grasp how these
work.  If you can explain them to me in addition to the comparison with Icon,
double thanks are yours.  The REBOL literature seems a little sparse about this
feature.  Evidently it is connected with the REBOL "parse" function.

Somewhere on the REBOL pages I also read that REBOL serves as its own
meta-language, or something equally cryptic.  I need a computer scientist to
hold my hand at this point because I lost my beanie cap.

An interesting aspect of REBOL as it touches on Icon is that REBOL already
incorporates the port/socket/IO features that were supposed to be the
new-and-improved aspects of Icon2/Unicon.  This overlap is another area for
comparison and contrast.  Internet functionality is a major theme of this
language.

In summary, my questions are
1) Do you think the driving computer-science philosophies behind REBOL and Icon
are very similar?
2) Where do they differ?
3) Is Icon's destiny pointing in the direction of REBOL?

Sincerely,

Mark Evans



From icon-group-sender  Thu Dec 30 15:35:44 1999
Return-Path: <icon-group-sender>
Received: (from root@localhost)
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) id PAA25047
	for icon-group-addresses; Thu, 30 Dec 1999 15:31:24 -0700 (MST)
Message-Id: <199912302231.PAA25047@baskerville.CS.Arizona.EDU>
X-Authentication-Warning: agate.berkeley.edu: news set sender to <news> using -f
From: Federico Balbi <fbalbi@lonestar.jpl.utsa.edu>
X-Newsgroups: comp.lang.icon
Subject: sleep
Date: Wed, 29 Dec 1999 16:46:39 -0600
X-Trace: ringer.cs.utsa.edu 946507553 7812 129.115.120.1 (29 Dec 1999 22:45:53 GMT)
X-Complaints-To: usenet@ringer.cs.utsa.edu
To: icon-group@optima.CS.Arizona.EDU
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO

Does Icon have a sleep(<seconds>) function to avoid busy-waiting? 

Thanks

----------------------------------------------------------------------------
 Federico Balbi - University of Texas at San Antonio 
 e-mail: fbalbi@lonestar.utsa.edu - URL: http://www.geocities.com/fedebalbi