From icon-group-sender  Mon Apr 27 08:09:24 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id IAA18112
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Mon, 27 Apr 1998 08:09:23 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA06896; Mon, 27 Apr 1998 08:09:20 -0700
Date: Fri, 24 Apr 1998 13:14:00 -0700
From: Gregg Townsend <gmt>
Message-Id: <9804242014.AA14229@hawk.CS.Arizona.EDU>
To: dpfj@globalnet.co.uk, icon-group
Subject: Re:  Using icon pipe features
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 1078

Most Unix utilities will buffer their output if it's connected to a pipe.
That means that even if you were able to open a two-way pipe to something
like "sed", you wouldn't be able to just write it a line and then read
the filtered result.  You'd need to write enough lines to make sed flush
its buffer before you'd be able to read anything back.  This, of course,
gets somewhat tricky, because sed can block if too much accumulates in 
its output pipe.  The tough part is that the buffering is going on
inside the forked process where you can't generally control it.

I've speculated that it might be possible to make something work using
pseudo-terminals (pttys), but I haven't tried it.

   ---------------------------------------------------------------------------
   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  Mon Apr 27 08:07:35 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id IAA17994
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Mon, 27 Apr 1998 08:07:35 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA07928; Mon, 27 Apr 1998 08:07:32 -0700
Message-Id: <3.0.5.32.19980424030719.007b7960@doors>
X-Sender: bgbauer#alpha.delta.edu@doors
X-Mailer: QUALCOMM Windows Eudora Light Version 3.0.5 (32)
Date: Fri, 24 Apr 1998 03:07:19 -0700
To: "Icon Group"<icon-group@optima.CS.Arizona.EDU>
From: Scissors <bgbauer@alpha.delta.edu>
Subject: Re: Win32 API support for Icon
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 4187

Hello,

   Thanks for all the replys.  Im new to the group, so I hope I havent
brought up a topic that has been over talked.  Im glad people have shown
interest though.  I've been feeling quite stranded since I began looking
for a compiler.  (Is Icon considered a full fledged compiler? Or is the
interpreter just stacked into that 175K that my tiny programs seem to take
up?)  Once I decided to try out Icon I felt even more alone.  Which is sad,
because as I said previously, I like the language.  If I keep up with Icon.
 (I plan to for atleast a few small programs, or until I hit a real
'Limitation'.)  

   
So let me clear up my original post.  Im not really looking for the full
win32 api.  Just a few 'key' elements (explained further down).

   Anyway, all I was really wondering was if there were ways to do what
C/C++ programs can do in windows.  I cant afford Borland or VC++ and the
books it would take to delve into C/C++.  Im not even sure if I want to
invest that amount of time.  So I cant really learn a few of the
suggestions that were made.  It would be too frustrating and confusing at
this point.

   All I was "hoping" there was, was a few ways to interface with (Im sure
every one has a different opinion on this) but 'key' elements of the
windows interface.  The registry, where my current project needs to obtain
information. (As will a future project I am planning).  The registry is
actually one of my biggest limits.  I feel like its 'key' because much of
windows 'look and feel' is stored there in settings.  If I could read the
registry, I could find the users preferred 'edit area font' and 'text
color', all of the users preferences I could know without the user needing
to set them.  I could read Ole settings to know what programs open what
types of documents.  (Useful if say I wrote an HTML editor, and wanted to
be able to allow them to browse the document. (of course I can just do this
with system("Start newfile.html").

  My current project involves reading in a serialized index file.  Now, if
I could read the registry, I could easily know exactly where the program
that created it, and the file itself, is located on disk, without needing
to prompt the user.  The path to the file, is already stored in the
registry.  This doesnt apply so much to the older win3.1, but both win95/NT
maintain a registry.  Its a key element to the O/S.

  Im guessing this ability is already included in Icon, because of Icons
use of the File Open, Font Selection, Color Selection dialogues matching
the 'look and feel' of Windows.  The ability to interface Windows API, I
mean.  Maybe Im confused, but couldnt Icon offer a 'pass through' type of
call which allows calling most WinAPI functions?

Like I said, Im new to this, so maybe that is harder to implement than I am
realising. But that is what I was really seeking.  

  This may be what one of the replys what trying to point out, when it
mentioned a need for a DLL Guru.  An MFC*.dll guru.  Who could create the
interface.

  I realise Icon is a free research langauge, and that its portability is
of concern. I just was interested in knowing how far Icon can stretch.  

  So...  Ive written a ReadIni(), and WriteIni() procedure.  Im putting
together an iowin.icn Procedure file, for anyone interested.  Mostly its
stuff that I needed, that wasnt in io.icn, or that the io.icn version didnt
seem to handle correctly for windows based systems.  Im currently creating
a Registry procedure also.  Ive got to rework it though because of limits I
found where calls to system() which create files (or atleast a call to
regedit) dont appear to be releasing the files before system() returns.
It's just a "work-around", and im not sure writing to the registry in this
manner would be %100 safe, which is why I wish I could do it correctly.
I'll find out in implementing it though.  Other than the bug with 'repeated
reads' I Can read the registry now.  If you're interested in this, let me
know.  I'd like to know how many more of you out there are working on
win95/nt.  And what procedures you may have created which are win95/nt based.

bgbauer

(My originally post was via usenet 'AKA' Mr Plow).








From icon-group-sender  Mon Apr 27 08:10:29 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id IAA18203
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Mon, 27 Apr 1998 08:10:28 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA09125; Mon, 27 Apr 1998 08:10:26 -0700
Message-Id: <3.0.5.32.19980426052832.007a9be0@doors>
X-Sender: bgbauer#alpha.delta.edu@doors
X-Mailer: QUALCOMM Windows Eudora Light Version 3.0.5 (32)
Date: Sun, 26 Apr 1998 05:28:32 -0700
To: "Icon Group"<icon-group@optima.CS.Arizona.EDU>
From: Scissors <bgbauer@alpha.delta.edu>
Subject: A few basic questions
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 741


Hi,

Im having a little trouble understanding a few of what are probably basic
Icon programming methods.

Is it possible to evaluate expression and assign at the same time in this
manner?

if \var := expr then fail

meaning :
if /expression then fail
else var := expression

It may seem trivial, but for some reason I find myself wanting to use the
first version by default.  Its easier to type and understand, takes up less
lines on the screen, but when I cant find the correct format to get it to
work.  

Also is it acceptable to use "if expr then expr else expr" all on the same
line?  Its getting harder to move around in my code as it gets longer with it
all '{' and '}'d on multiple lines.

Also what are string images used for?





From icon-group-sender  Mon Apr 27 08:10:50 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id IAA18221
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Mon, 27 Apr 1998 08:10:50 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA07008; Mon, 27 Apr 1998 08:10:47 -0700
From: Otfried Geffert <og@stardivision.de>
Date: Mon, 27 Apr 1998 08:51:20 GMT
Message-Id: <19980427.8512058@og-1042.stardiv.de>
Subject: Re: Using icon pipe features
To: Derek Paul Freeman-Jones <dpfj@globalnet.co.uk>
Cc: icon-group@optima.CS.Arizona.EDU
In-Reply-To: <353F9415.83BD5C5B@globalnet.co.uk>
References: <353F9415.83BD5C5B@globalnet.co.uk>
X-Mailer: Mozilla/3.0 (compatible; StarOffice/4.0; WinNT/Win95)
X-Priority: 3 (Normal)
Mime-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 966

As I think a _pipe_ is a unidirectional thing.
So I think you can't open it for reading and writing from the
same process.

For a similar problem I used:
- opened the real file for reading
- wrote the processed file to a temporary file
- renamed the temporary file to the real filename

Sorry, I see this is not what you wanted.

Otfried

>>>>>>>>>>>>>>>>>><<<<<<<<<<<<<<<<<<

Derek Paul Freeman-Jones <dpfj@globalnet.co.uk> on Using icon pipe=20
features:

> icon-group,
>=20
> I need to extend the functionality of the procedure 'open' when=20
running
> a process eg.
>=20
> pipe_in :=3D open ( "/usr/bin/ls", "p" )
>=20
> at present the process can only be opened for reading OR writing. Has=

> anyone done any development work on enabling the process to be BOTH
> written to and read from?
>=20
> I don't if I have to write a new procedure to plug in to Icon 'popen'=

> for example.
>=20
> Many thanks in advance.
>=20
>=20
> Derek Freeman-Jones
>=20
>=20





From icon-group-sender  Mon Apr 27 12:42:24 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id MAA29629
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Mon, 27 Apr 1998 12:42:18 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA08647; Mon, 27 Apr 1998 12:42:15 -0700
Date: Mon, 27 Apr 98 09:55:24 PDT
Message-Id: <9804271655.AA02872@bermuda.synopsys.com>
From: Shamim Mohamed <shamim@Synopsys.COM>
To: Gregg Townsend <gmt@baskerville.CS.Arizona.EDU>
Cc: dpfj@globalnet.co.uk, icon-group@baskerville.CS.Arizona.EDU
Subject: Re:  Using icon pipe features
In-Reply-To: <9804242014.AA14229@hawk.CS.Arizona.EDU>
References: <9804242014.AA14229@hawk.CS.Arizona.EDU>
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 548

[plug follows]

Gregg writes:
  > Most Unix utilities will buffer their output if it's connected to a
  > pipe.  That means that even if you were able to open a two-way pipe to
  > something like "sed", you wouldn't be able to just write it a line and
  > then read the filtered result.

One way (the right way?) to do this is by forking and having one copy write
to sed and the other reading the output. Support for all this is in Unicon;
an example (an implementation of script(1)) is in the documentation:

   http://www.drones.com/unicon/

-s


From icon-group-sender  Mon Apr 27 20:37:56 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id UAA27960
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Mon, 27 Apr 1998 20:37:45 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA07462; Mon, 27 Apr 1998 20:37:42 -0700
Message-Id: <3544EF25.E0A905CB@globalnet.co.uk>
Date: Mon, 27 Apr 1998 21:48:37 +0100
From: Derek Paul Freeman-Jones <dpfj@globalnet.co.uk>
X-Mailer: Mozilla 4.01 [en] (Win95; I)
Mime-Version: 1.0
To: Shamim Mohamed <shamim@Synopsys.COM>
Cc: Gregg Townsend <gmt@baskerville.CS.Arizona.EDU>,
        icon-group@baskerville.CS.Arizona.EDU,
        Shamim Mohamed <shamim@Synopsys.COM>
Subject: Re: Using icon pipe features
X-Priority: 3 (Normal)
References: <9804242014.AA14229@hawk.CS.Arizona.EDU> <9804271655.AA02872@bermuda.synopsys.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 155

All,

Many thanks for all your help, Shamim has provided me with the source
code I needed to read from and write to an exec'ed process.

Thanks

Derek





From icon-group-sender  Mon Apr 27 20:38:31 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id UAA27967
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Mon, 27 Apr 1998 20:38:31 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA08915; Mon, 27 Apr 1998 20:38:28 -0700
Message-Id: <3.0.5.32.19980427182410.007a92d0@doors>
X-Sender: bgbauer#alpha.delta.edu@doors
X-Mailer: QUALCOMM Windows Eudora Light Version 3.0.5 (32)
Date: Mon, 27 Apr 1998 18:24:10 -0700
To: swampler@noao.edu (Steve Wampler)
From: Scissors <bgbauer@alpha.delta.edu>
Subject: Re: A few basic questions
Cc: "Icon Group"<icon-group@optima.CS.Arizona.EDU>
In-Reply-To: <swampler-9803271524.AA001226837@orpheus.gemini.edu>
References: <3.0.5.32.19980426052832.007a9be0@doors>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 1168

At 08:24 AM 4/27/98 -0700, you wrote:
>Try any of the following:
>
>var := expr | fail
>
>if not (var := expr) then fail
>
>> Also is it acceptable to use "if expr then expr else expr" all on the same
>> line?  Its getting harder to move around in my code as it gets longer
with >> it all '{' and '}'d on multiple lines.
>
>Acceptable, but not considered good practice.

Well, thats what I expected, but I'd hoped I was using an older form. : /

What about this example:

if expression then {
   expression 
   expression }

Saving one line with the last '}' being moved up?


>
>> Also what are string images used for?
>
>Uh, string image of what?  String images of different things have different
>uses.

Well, for example one of the functions in icon is:

image (x) : s -- produce string image 
Produces a string image of x. 

   Under what circumstances would this be used?  I realise (I think) what
an image is, but im not sure why I wouldnt need an image of an object, if I
have the object itself?

There is also an IPL procedure:
fullimag.icn 
Procedures to produce complete image of structured data 

Or am I better of not worrying about that for now? : )





From icon-group-sender  Wed Apr 29 12:22:14 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id MAA21969
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Wed, 29 Apr 1998 12:22:14 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA10283; Wed, 29 Apr 1998 12:22:11 -0700
Message-Id: <3.0.5.32.19980429075520.007a38f0@doors>
X-Sender: bgbauer#alpha.delta.edu@doors
X-Mailer: QUALCOMM Windows Eudora Light Version 3.0.5 (32)
Date: Wed, 29 Apr 1998 07:55:20 -0700
To: "Icon Group"<icon-group@optima.CS.Arizona.EDU>
From: Scissors <jebediah@bigfoot.com>
Subject: EOL Termination Sequences.
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 1599


Is anyone out there using Win95/dos?

Im having a terrible time with eol termination sequences.

Two problems mostly.  Ive set the WI Line termination option to

linebreak=CRLF

but for some reason, every so often. (seems to be random as far as I can
tell.)
I will save a file, and *BANG* all the 100+ lines of text lose their EOL's and
end up in four lines at the top of my window.  For the most part this is
happening inside of notepad.  If I edit a file that was saved using WI, it
appears to be using strange EOL's which Notepad zaps on saving the file.
Ive never encountered this before using Icon, and it doesnt happen when Im 
editing anything but Icon source.  Ive checked the settings, all seems fine.

: /

Help!?!

Secondly, im loading a file using:

lines := []
lines := filetext(fname)  

when I find the line I wish to insert text after, I have been using

lines[i] := lines[i] || "\r\l" || "what I want to insert"

(The reason Im 'appending it' is because when I tried it before, it would
allow me to add any new lines to 'lines[]'.  When I first created the
procedure, I was attempting to slide all the remaining down a notch, thus
freeing up one of list[], so I could then inserting the new line.  But it
would not add anything beyond the original *line.)

ok, so this all works ok... until I open the file in notepad, and save it.
then the line I inserted last shows up as

Previous TextNew Text

is "\r\l" the incorrect line termination sequence?  I tried "\r\n" and it
does the same.  What is the correct EOL for dos/win?

Thanks for any help in advance

jebediah@bigfoot.com



From icon-group-sender  Wed Apr 29 12:22:53 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id MAA21995
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Wed, 29 Apr 1998 12:22:53 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA10611; Wed, 29 Apr 1998 12:22:50 -0700
Message-Id: <3.0.5.32.19980429091743.00798100@doors>
X-Sender: bgbauer#alpha.delta.edu@doors
X-Mailer: QUALCOMM Windows Eudora Light Version 3.0.5 (32)
Date: Wed, 29 Apr 1998 09:17:43 -0700
To: "Icon Group"<icon-group@optima.CS.Arizona.EDU>
From: Scissors <jebediah@bigfoot.com>
Subject: RE: EOL Termination Sequences.
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 884

Ok, I got into thinking about this.  Here's whats happening.

I realised I Can open a file in DOS edit, and it shows the IBM/ASCII
characters.  So I opened the file I was trying to insert the line into
and at the end of that line, and only that line, there is just a carriage
return.

Any Idea why when I inserted "\r\l" it is only placing "\r"?

So I also opened up files Ive been editing using WI, and every line ends in
a carriage return.  The little musical symbol in dos edit.  When a normal text
file Ive edited using notepad, or even dos edit shows no end of line
characters
at all.  Even if I add one right then, it isnt shown?

Has anyone else noticed a problem like this?  Maybe something is wrong just
on my setup.  Anyone have any suggestions what I might do to fix this?

Thanks,
(and sorry for writing yet again... still getting used to Icon.)

jebediah@bigfoot.com





From icon-group-sender  Thu Apr 30 08:10:43 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id IAA05998
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Thu, 30 Apr 1998 08:10:42 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA11574; Thu, 30 Apr 1998 08:10:39 -0700
Message-Id: <s5484c8b.053@housmtp.oceaneering.com>
X-Mailer: Novell GroupWise 4.1
Date: Thu, 30 Apr 1998 10:02:52 -0500
From: Charles Hethcoat <CHETHCOA@oss.oceaneering.com>
To: icon-group@optima.CS.Arizona.EDU
Subject: RE: EOL Termination Sequences. -Reply
Mime-Version: 1.0
Content-Type: text/plain
Content-Disposition: inline
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 1880

I use NT 4--not that I'm so very fond of it--and I had so many odd little
problems with Notepad that I banished it from my machine.  I found PFE
(Programmer's File Editor) to be a far superior choice across the board. 
Pertinent to your problem is that PFE is knowledgable of line termination
sequences in both DOS and Unix.  I haven't used Wi that much yet, so I
cannot comment on any possible problems there.

I would suggest trying the same things you were doing, only with PFE
instead of Notepad, and see if it helps.  If not, well, you've at least
eliminated one possible source of the problem.

You can get PFE for free from:
Alan Phillips	( A.Phillips@lancaster.ac.uk )
( http://www.lancs.ac.uk/people/cpaap/pfe )

Charles Hethcoat		
Senior Engineer, Analysis Department
Oceaneering Space Systems, Inc.
16665 Space Center Blvd., Houston, Texas 77058 USA
chethcoa@oss.oceaneering.com
Voice:  281-228-5568 Fax:  281-488-2027



>>> Scissors <jebediah@bigfoot.com> 1998 Apr 29, 11:17am >>>
Ok, I got into thinking about this.  Here's whats happening.

I realised I Can open a file in DOS edit, and it shows the IBM/ASCII
characters.  So I opened the file I was trying to insert the line into
and at the end of that line, and only that line, there is just a carriage
return.

Any Idea why when I inserted "\r\l" it is only placing "\r"?

So I also opened up files Ive been editing using WI, and every line ends in
a carriage return.  The little musical symbol in dos edit.  When a normal
text
file Ive edited using notepad, or even dos edit shows no end of line
characters
at all.  Even if I add one right then, it isnt shown?

Has anyone else noticed a problem like this?  Maybe something is wrong
just
on my setup.  Anyone have any suggestions what I might do to fix this?

Thanks,
(and sorry for writing yet again... still getting used to Icon.)

jebediah@bigfoot.com







From icon-group-sender  Thu Apr 30 08:10:10 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id IAA05988
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Thu, 30 Apr 1998 08:10:09 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA11144; Thu, 30 Apr 1998 08:10:06 -0700
Message-Id: <3.0.5.32.19980430041203.007b14b0@doors>
X-Sender: bgbauer#alpha.delta.edu@doors
X-Mailer: QUALCOMM Windows Eudora Light Version 3.0.5 (32)
Date: Thu, 30 Apr 1998 04:12:03 -0700
To: "Icon Group"<icon-group@optima.CS.Arizona.EDU>
From: Scissors <jebediah@bigfoot.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 3513


Thanks for the responses.  I should get around to trying out just '\n'
tonight.
And see if that solves the problem.  

At 11:10 PM 4/29/98 -0500, you wrote:
>Hi,
>
>I am not sure quite what is happening with your EOL terminations.  Wi's
>linebreak=CRLF feature is not supposed to turn itself off accidentally and
>at random.

After I opened up the files and got to looking at it I noticed it isnt random.
All of the files I've edited using WI appear to have the same 'musical note'
ending for lines.  So the files are appearing fine, until I open, and then
save them in notepad, which loses all of the EOL marks.

Im using WI, and the Win32 version of Icon on win 95. (But only working so
far with DOS Console programming.  (Ive only been using WI so far because I
like having a project file, so I can switch back and forth.  I cant figure
out how to compile with it yet, so I do this on the command line.)

  Wi could be writing out the files in a way that notepad doesn't
>like, and it shouldn't be that hard to identify the problem if that's the
>case.  I'll try to reproduce your bug, and get back to you.

It may just be a win 95 problem, as you mentioned NT uses differnt EOL's.

>Regarding your list of lines, there is no easy way in Icon to insert an
>element into the middle of the list, but it is easy to grow the list by
>calling put(L, "a new element") and thenyou can slide the elements down
>a notch as you suggested.
>
>>  is "\r\l" the incorrect line termination sequence?  I tried "\r\n" and it
>>  does the same.  What is the correct EOL for dos/win?
>
>Under DOS Icon, you just use "\n" as your newline character and the "\n" is
>converted to "\r\n".

Will this work with Win 95 version of Icon?  Im assuming so...

> Windows NT uses the UNIX convention that \n is an EOL
>so again the \r is generally not needed.  Win32 applications like Windows
Icon
>are native on NT, but legacy programs like notepad may still require \r\n.

>To write old DOS-style lines you might say \r\n.  Note that write() appends
>its own \n, so to write hello in a DOS file you would say write("hello\r").
>There is no \l character in Icon as far as I know.
>
>I'm happy to help work through the newline problems you are experiencing.
>First thing will be for me to try going back and forth between notepad
>and wi.

I tinkered with it a bit today, and I got it so that I Can insert the line
correctly.  It appears as two lines now.  I believe I switched the EOL's I
was trying to use, to "\l".  (Which I got from the help file as below under
Escape sequences).

\f 	formfeed 		\\ 	backslash 
\l 	linefeed (newline) 	\ddd 	octal code 
\n 	newline (linefeed) 	\xdd 	hexadecimal code 
\r 	carriage return 		\^c 	control code 

But now when I attempt to tab(match()) the 'most recently' inserted line I
get no match.  Apparantly it is being interpreted in my Read() as being
part of the previous line still.  This only affects the last line I
inserted this way, because after I read() the file in again, and write() it
out, it corrects the problem, due to write() adding the correct EOL's.  

Hopefully using "\n" will fix this for me : )!

I looked through WI's code, but all I saw for an EOL Seq. assignment was it
being set to CRLF, which I couldnt determine where it knew what the value
of CRLF was.  If you do find some way to overcome this, please let me know
so I Cant edit it into WI.  Until then I will remember not to save in
notepad! Youch!  

Thanks to all!

>
>Clint Jeffery, jeffery@cs.utsa.edu
>
>
>
>



From icon-group-sender  Fri May  1 07:57:42 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id HAA25027
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Fri, 1 May 1998 07:57:41 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA13255; Fri, 1 May 1998 07:57:38 -0700
Message-Id: <3.0.5.32.19980430163932.007a5370@alpha.delta.edu>
X-Sender: bgbauer@alpha.delta.edu
X-Mailer: QUALCOMM Windows Eudora Light Version 3.0.5 (32)
Date: Thu, 30 Apr 1998 16:39:32 -0700
To: "Icon Group"<icon-group@optima.CS.Arizona.EDU>
From: Scissors <bgbauer@alpha.delta.edu>
Subject: Re: EOL Termination Sequences.
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 465


Thanks Clinton, and all who replied.

Using "\n" fixed the problems I was having with my program.

Is there a way for me to edit this into the WI source?  (Meaning which line
no.?)  The only line I saw regarding this was in the mainloop() I believe,
where the menu options are handled.  it just assigns linebreak="CRLF" if I
remember right.  Where is it getting the meaning of CRLF?  (I guess I
Should look myself and figure that out. : P)   

Thanks again,
B





From icon-group-sender  Tue May  5 08:53:49 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id IAA14721
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Tue, 5 May 1998 08:53:32 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA16453; Tue, 5 May 1998 08:53:29 -0700
Date: Mon, 4 May 1998 22:09:13 -0500
Message-Id: <199805050309.WAA30479@segfault.cs.utsa.edu>
From: Clinton Jeffery <jeffery@segfault.cs.utsa.edu>
To: icon-group@optima.CS.Arizona.EDU
Subject: Windows Icon update
Reply-To: jeffery@cs.utsa.edu
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 1087


The Windows Icon distribution at ringer.cs.utsa.edu/pub/icon/nt/binaries/
has been updated.  This is a maintenance release that includes no new
features, but incorporates fixes for the following problems reported by
users, mainly from reports posted to the icon-group mailing list:

        Hardwired paths removed from generated .EXE's
            (fix suggested by Mark Evans -- thanks Mark!)
	Improvements to system() under Win95 and NT
            (fix suggested by Mike Stradley -- thanks Mike!)
        Elimination of most "junk" logfiles left in root directory after
            abnormal program termination.
	WildCard expansion support added to nticonx and generated console apps.
        Bugfix: WinButton(), Windows native buttons
	Bugfix: FillRectangle when using colored patterns
	Bugfix: Wi should no longer write extra carriage returns into files.

My thanks to everyone who sent me comments or reported a problem.

Clint Jeffery, jeffery@cs.utsa.edu
Division of Computer Science, The University of Texas at San Antonio
Research http://www.cs.utsa.edu/research/plss.html


From icon-group-sender  Tue May  5 17:13:05 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id RAA21103
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Tue, 5 May 1998 17:13:05 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA16917; Tue, 5 May 1998 17:13:01 -0700
Message-Id: <s54f5079.065@housmtp.oceaneering.com>
X-Mailer: Novell GroupWise 4.1
Date: Tue, 05 May 1998 17:45:32 -0500
From: Charles Hethcoat <CHETHCOA@oss.oceaneering.com>
To: icon-group@optima.CS.Arizona.EDU
Subject: Followup to PDF problem
Mime-Version: 1.0
Content-Type: text/plain
Content-Disposition: inline
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 351

I just downloaded IPD284 perfectly in PDF format...no problems.  But the
PostScript version scrambles GhostScript!  What gives?

Good luck.

Charles Hethcoat		
Senior Engineer, Analysis Department
Oceaneering Space Systems, Inc.
16665 Space Center Blvd., Houston, Texas 77058 USA
chethcoa@oss.oceaneering.com
Voice:  281-228-5568 Fax:  281-488-2027



From icon-group-sender  Wed May  6 08:50:05 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id IAA22328
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Wed, 6 May 1998 08:50:04 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA17324; Wed, 6 May 1998 08:50:01 -0700
To: icon-group@optima.CS.Arizona.EDU
Date: Tue, 05 May 1998 20:04:44 -0600
From: Terry Murray <tmurray@primary.net>
Message-Id: <354FC528.4D86@kc-primary.net>
Organization: Primary Network. http://www.primary.net
Sender: icon-group-request@optima.CS.Arizona.EDU
Reply-To: tmurray@primary.net
Subject: AI use for Icon
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 1096

I'm writing a program with icon that reads an email message and
determines the who, what, when, where and by whom. At least that is what
I am trying to do.  This message would come form a supervisor to the
human resources department asking them to promote Mr. Any Body to a New
Title on a certain date.  By having all the employees names prior to
running the program I weight each name based on its reletive position to
the other name within the organization to get a fuzzy match.  One of
which must be a supervisor.  I've never tried using AI technology before
so I looking for someone who has done something similar that would share
with me their text parse-ing process for ideas.  I'm currently looking
for words that match my list as for the 'what' the matching terms would
be 'promote, demote, reassign, realign, fire...'.  Likewise, 'who' is
the list of employees.

I'm now working on the 'when' to search for dates.  I need a different
method than giving the program every possible date in every possible
way.  Does someone have a program that would pick a date out of a
sentence?

Terry


From icon-group-sender  Wed May  6 12:36:51 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id MAA07362
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Wed, 6 May 1998 12:36:51 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA15209; Wed, 6 May 1998 12:36:47 -0700
From: gep2@computek.net
Date: Wed, 6 May 1998 12:04:42 -0500
Message-Id: <199805061704.MAA14525@axp.cmpu.net>
Mime-Version: 1.0
Content-Type: text/plain
Content-Transfer-Encoding: 7bit
Subject: AI use for Icon
To: icon-group@optima.CS.Arizona.EDU
X-Mailer: SPRY Mail Version: 04.00.06.17
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 2300

> I'm writing a program with icon that reads an email message and
determines the who, what, when, where and by whom. At least that is what
I am trying to do.  This message would come form a supervisor to the
human resources department asking them to promote Mr. Any Body to a New
Title on a certain date.  By having all the employees names prior to
running the program I weight each name based on its reletive position to
the other name within the organization to get a fuzzy match.  One of
which must be a supervisor.  I've never tried using AI technology before
so I looking for someone who has done something similar that would share
with me their text parse-ing process for ideas.  I'm currently looking
for words that match my list as for the 'what' the matching terms would
be 'promote, demote, reassign, realign, fire...'.  Likewise, 'who' is
the list of employees.

Sounds like an interesting problem!  

For the "fuzzy match" I think that one interesting way to at least help winnow 
down the possibilities would be to examine the intersection of the character 
sets of the different names.  Those which have a high intersection (all but a 
"few" characters) can be then examined more closely.

For a better "fuzzy compare" function I've liked the use of overlapping 
character pairs (including a blank added to the start and end of each name).

I think that character ADJACENCY is an important aspect of the "visual 
similarity" of two words or names.

> I'm now working on the 'when' to search for dates.  I need a different
method than giving the program every possible date in every possible
way.  Does someone have a program that would pick a date out of a
sentence?

Certainly it ought to be very easy to parse NUMERIC dates.  And since most 
formats of dates (with the exception of things like "the first of May") have 
numeric characters associated with them, I wouldn't think it would be so 
difficult to zero in on a very small number of possibilities, which could then 
be processed by more exhaustive "multi-alternative" patterns.

It sounds like a fun project, in any case... if you decide you'd like to farm it 
out, I would enjoy taking a stab at it.  :-)

Gordon Peterson
http://www.computek.net/public/gep2/
Support the Anti-SPAM Amendment!  Join at http://www.cauce.org/



From icon-group-sender  Wed May  6 12:36:37 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id MAA07359
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Wed, 6 May 1998 12:36:36 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA18434; Wed, 6 May 1998 12:36:33 -0700
From: eka@corp.cirrus.com (Eka Laiman)
Message-Id: <199805061638.JAA29265@sims-rd.corp.cirrus.com>
Subject: Re: AI use for Icon
To: tmurray@primary.net
Date: Wed, 6 May 1998 09:35:07 -0700 (PDT)
Cc: icon-group@optima.CS.Arizona.EDU
In-Reply-To: <354FC528.4D86@kc-primary.net> from "Terry Murray" at May 5, 98 08:04:44 pm
X-Mailer: ELM [version 2.4 PL24alpha3]
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 1470

Terry wrote:
> I'm writing a program with icon that reads an email message
> and determines the who, what, when, where and by whom. At least
> that is what I am trying to do.
> ...... edited ....

I have to do "incoming mail filtering" myself. In my case, due to
my belonging to certain kind of SIG (Special Interest Group), it
is quite common for me to receive 50+ mails a day. This large number
of incoming mail can potentially shield me from "business" related
mails which require my attention. To avoid this, every incoming mail
will be automatically be passed through a "mail filter". If the 
mail comes the mentioned SIG, they will automatically be accumulated
in a "special mail folder" without notifying me of their arrivals;
otherwise, the mail will be passed through just like normal mails
that do not go through the filter.

If you are working with UNIX or Linux, the best way to do this kind
of filtering is through "procmail" which is well documented. You can
do a search in the internet for various pages explaining how this
works, a tutorial of using this facility is also available.

Since procmail pass the mail as a "stream of text" and the blank line
separate the mail header from the body of the mail, you can easily
do filtering based on "content" or "mail header". In fact the 
list server program known as "majordomo" works this way. The method
used is to "hash" (icon's implementation is through "table") each
field in the mail header. 

-eka-



From icon-group-sender  Wed May  6 16:37:16 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id QAA17698
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Wed, 6 May 1998 16:37:16 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA14346; Wed, 6 May 1998 16:37:12 -0700
Message-Id: <s550a4ba.077@housmtp.oceaneering.com>
X-Mailer: Novell GroupWise 4.1
Date: Wed, 06 May 1998 17:55:54 -0500
From: Charles Hethcoat <CHETHCOA@oss.oceaneering.com>
To: icon-group@optima.CS.Arizona.EDU
Subject: AI use for Icon -Reply
Mime-Version: 1.0
Content-Type: text/plain
Content-Disposition: inline
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 1963

Terry Murray <tmurray@primary.net> wrote:

>...message would come form a supervisor to the human resources
department asking them to promote Mr. Any Body to a New Title on a
certain date.  By having all the employees names prior to running the
program I weight each name based on its reletive position to the other
name within the organization to get a fuzzy match.  

If you get a fuzzy name match algorithm together, it sure would be a nice
contribution to the Icon library...I don't know much about the methods
used in these types of searches, but I'll give it some thought also.

>I'm now working on the 'when' to search for dates.  I need a different
method than giving the program every possible date in every possible
way.  

My first impulse says to try to find candidates for the elements first: year,
month, day of week, and day written in every possible way.  Then dope
out month from day separately, based partly on proximity, and then put
the date in a canonical form such as yyyy-mm-dd.

December could be written (in English) as Dec or December (possibly
misspelled), or as 12.  Years could be 1998 or 98 or '98.  Days could be
1..N where N is a well-known function of month (28..31).

By "proximity" I mean, for example, that a numeric month should be
followed very shortly afterward by viable candidates for year and day,
or it's probably not a date.

A date like 7/9 is ambiguous internationally:  July 9 in the U.S. but
September 7 in Europe.  This can usually be resolved by knowing the
origin of the writer.

Delimiters seem only to be slashes, dashes or spaces.

Seems like a job tailormade for Icon string scanning and tables.

>Does someone have a program that would pick a date out of a
sentence?

Sorry, I don't.  It would make a another great library contribution.

Charles Hethcoat		
Senior Engineer, Analysis Department
Oceaneering Space Systems, Inc.
16665 Space Center Blvd., Houston, Texas 77058 USA
chethcoa@oss.oceaneering.com



From icon-group-sender  Thu May  7 08:43:42 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id IAA18770
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Thu, 7 May 1998 08:43:42 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA16115; Thu, 7 May 1998 08:43:38 -0700
Message-Id: <3.0.5.32.19980506220526.007b0e20@doors>
X-Sender: bgbauer#alpha.delta.edu@doors
X-Mailer: QUALCOMM Windows Eudora Light Version 3.0.5 (32)
Date: Wed, 06 May 1998 22:05:26 -0700
To: "Icon Group"<icon-group@optima.CS.Arizona.EDU>
From: Scissors <jebediah@bigfoot.com>
Subject: AI Use for Icon
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 740


Well, this idea may not work, but the first thing that came to my mind was
to create a program which prompted the user for thier input, and then pre-
formatted it correctly, so after it is recieved, it can be easily processed
for the needed info.

I think the easiest way to do this would be a javascript web page with forms.
This would allow you to easily make sure all the required fields are filled,
and that when they are sent out they are formatted correctly, for the easiest
possible processing later.

Do all the people involved in this have access to a web browser?

(PS. Javascript/HTML would probably be MUCH easier to implement than all 
of the complex work it sounds like fuzzy logic parsing would involve.)

Good luck,
JS





From icon-group-sender  Thu May  7 08:43:50 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id IAA18789
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Thu, 7 May 1998 08:43:50 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA16697; Thu, 7 May 1998 08:43:46 -0700
Message-Id: <3.0.5.32.19980506221248.00798ce0@doors>
X-Sender: bgbauer#alpha.delta.edu@doors
X-Mailer: QUALCOMM Windows Eudora Light Version 3.0.5 (32)
Date: Wed, 06 May 1998 22:12:48 -0700
To: "Icon Group"<icon-group@optima.CS.Arizona.EDU>
From: Scissors <jebediah@bigfoot.com>
Subject: AI Use for Icon
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 494


I forgot to include it in my previous response, but one not so used but
easy to implement forms submit action is "mailto:"

Your user could fill out the form, and when they hit 'submit' it would
automatically be emailed to the correct address.  This would remove
the need for a CGI script, or any server involvement really.

Good luck again,
B Bauer

If you're interested I could toss together a sample javascript/html
page which would give an idea how easy this way of doing it could be.





From icon-group-sender  Thu May  7 12:42:47 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id MAA06155
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Thu, 7 May 1998 12:42:47 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA16998; Thu, 7 May 1998 12:42:43 -0700
From: gep2@computek.net
Date: Thu, 7 May 1998 12:19:46 -0500
Message-Id: <199805071719.MAA26136@axp.cmpu.net>
Mime-Version: 1.0
Content-Type: text/plain
Content-Transfer-Encoding: 7bit
Subject: Re: AI Use for Icon
To: icon-group@optima.CS.Arizona.EDU
In-Reply-To: <3.0.5.32.19980506221248.00798ce0@doors>
X-Mailer: SPRY Mail Version: 04.00.06.17
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 726

Certainly if they can put the stuff in a mailed form, this eliminates the main 
hassles of parsing the stuff.

Certainly **no** need for Javashit to do this, and no CGI stuff either.  Almost 
all browsers (even the really limited ones) can handle this.

If you visit my Web page and particularly the Guest Book page (download a copy 
of the HTML to your system), you can see how simple the HTML is for this... the 
resulting forms get mailed through normal E-mail and are relatively easy to 
process using either S*BOL or Icon.

If you don't HAVE to handle freeform text, of course the problem is simpler.

Gordon Peterson
http://www.computek.net/public/gep2/
Support the Anti-SPAM Amendment!  Join at http://www.cauce.org/



From icon-group-sender  Mon May 11 08:15:03 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id IAA10090
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Mon, 11 May 1998 08:15:03 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA09862; Mon, 11 May 1998 08:14:59 -0700
To: icon-group@optima.CS.Arizona.EDU
Date: Sun, 10 May 1998 06:58:41 +0100
From: David Feustel <feustel@ix.netcom.com>
Message-Id: <35554211.42AA67BA@ix.netcom.com>
Sender: icon-group-request@optima.CS.Arizona.EDU
Subject: Com Version of ICON
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 143

Is there a Com version of ICON? I'd like to try using ICON with VBScript

for an application involving Server-side Web Programming.

Thanks.



From icon-group-sender  Wed May 13 07:42:13 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id HAA17886
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Wed, 13 May 1998 07:42:13 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA22189; Wed, 13 May 1998 07:42:08 -0700
Message-Id: <3.0.5.32.19980512173306.007a7c40@doors>
X-Sender: bgbauer#alpha.delta.edu@doors
X-Mailer: QUALCOMM Windows Eudora Light Version 3.0.5 (32)
Date: Tue, 12 May 1998 17:33:06 -0700
To: "Icon Group"<icon-group@optima.CS.Arizona.EDU>
From: Scissors <jebediah@bigfoot.com>
Subject: Order of operations
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 793


Hi again,

Still trying to learn Icon's little intricasies, so I have another question.

In what order would the following code be interpreted?

if var := "string" then var2 := "string2" & var3 := "string3"

is this interpreted as:

if var := "string" then {
   var2 := "string2"
   var3 := "string3" # this line only done if previous line succeeds?
}

or is it interpreted as:

if var := "string" then {
  var2 := "string2"
  var2 := (var3 := "string3") #resulting in var2 == "string3"?
} 


Ok, im also confused about this syntax:

"abcdefg" ? var := move(1) & move(2)

Here would the end result of var == "bc"? or "a"
And what about in this situation. (sort of applies to my first question).

"abcdefg" ? move(1) & var := move(2)

would the result be var == "bc" ?



Thanks,
scissors





From icon-group-sender  Wed May 13 07:42:30 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id HAA17889
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Wed, 13 May 1998 07:42:29 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA22013; Wed, 13 May 1998 07:42:25 -0700
Message-Id: <3.0.5.32.19980512175257.007a7c40@doors>
X-Sender: bgbauer#alpha.delta.edu@doors
X-Mailer: QUALCOMM Windows Eudora Light Version 3.0.5 (32)
Date: Tue, 12 May 1998 17:52:57 -0700
To: "Icon Group"<icon-group@optima.CS.Arizona.EDU>
From: Scissors <jebediah@bigfoot.com>
Subject: Determining O/S
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 967


Oh, I thought of another question thats been bugging me.  What is the easiest
way to determine the O/S your running on?  I would have no idea how to
determine
unix, and Im uncertain the 'best' way to determine if the O/S is win3.1,
win95,
or winNT?

Would it be best to use some of the ifdef lines I've noticed in io.icn for
o/s secific code? if so what are the available definitions?

At first I thought code similar to this would work (which is from yescr.icn):

   if find("UNIX",&features) then {
      expressions
   }
   else if find("MS-DOS", &features) then {
      expressions
   }
   else if find("MS Windows NT", &features) then {
      expressions
   }

But this would lump win3.1 (win32s) with win95, and NT, and i believe it
wouldnt give any indication of if the app is console, or windowed.   Code
i'm working on now would need to act differently on win3.1 than it would on
win95/nt.

Any help, suggestions, or comments welcome.

Thanks,
scissors





From icon-group-sender  Thu May 14 07:48:19 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id HAA23358
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Thu, 14 May 1998 07:48:18 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA24525; Thu, 14 May 1998 07:48:13 -0700
Message-Id: <s559bbe6.085@housmtp.oceaneering.com>
X-Mailer: Novell GroupWise 4.1
Date: Wed, 13 May 1998 15:22:55 -0500
From: Charles Hethcoat <CHETHCOA@oss.oceaneering.com>
To: icon-group@optima.CS.Arizona.EDU
Subject: Determining O/S -reply
Mime-Version: 1.0
Content-Type: multipart/mixed; boundary="=_23774126.8FEE8660"
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 4339

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.

--=_23774126.8FEE8660
Content-Type: text/plain
Content-Disposition: inline

Scissors <jebediah@bigfoot.com> wrote:

>But this would lump win3.1 (win32s) with win95,
and NT, and i believe it
wouldnt give any indication of if the app is
console, or windowed.

I ran:

procedure main()
	every write(&features)
end

on my NT system.  The console-mode version
reports:

MS Windows NT
ASCII
co-expressions
environment variables
external functions
large integers
system function

and wi reports:

MS Windows NT
ASCII
co-expressions
environment variables
external functions
large integers
system function
file attributes *
MS Windows *
Win32 *

where items marked * do not appear in NT console
mode.  So I guess if "Win32" doesn't appear, it
must be Win(32s or 16)?  I have Icon on Win 95 at
home, so I will have wait until tonight to try
that.

A more fundamental question is, "How do I
determine whether I'm running on Windows or Unix?"
It turns out that it's not as easy as I thought!

When I run the command "ver" in a console window,
I get:

NT 4.0:  "Windows NT Version 4.0"
Win95:  "Windows 95 Version blah"

When I tried the obvious answer (put a few lines
in the program that would run the "ver" command
and write the result to a temporary file), I found
that redirecting the command output to a file just
sendt it all to the bit bucket in Windows.  I also
checked all commands for failure, but nothing
failed!  Here is the program:

procedure main()
	every write(&features)
	system("ver > tmp$$$") | stop("Could not
execute 'ver'")
	f := open("tmp$$$") | stop("Could not open
tmp$$$")
	version := read(f) | stop("Could not read
tmp$$$")
	close(f) | stop("Could not close tmp$$$")
	remove("tmp$$$") | stop("Could not remove
tmp$$$")
	write()
	write(version)
end

Something similar should work in Unix, except I'm
not sure what the Unix equivalent of 'ver' is. 
Assuming it exists, this program should work on
Unix.  So, if it fails to work, it must be running
under Windows :-/

Charles Hethcoat
--=_23774126.8FEE8660
Content-Type: message/rfc822

Received: from optima.cs.arizona.edu ([192.12.69.5])
	by oceaneering.com (GroupWise SMTP/MIME daemon 4.1 v3)
	; Wed, 13 May 98 09:43:58 CDT
Received: from baskerville.CS.Arizona.EDU (baskerville.CS.Arizona.EDU [192.12.69.35])
	by optima.cs.arizona.edu (8.9.0.Beta5/8.9.0.Beta5) with ESMTP id HAA15312;
	Wed, 13 May 1998 07:43:39 -0700 (MST)
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id HAA17889
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Wed, 13 May 1998 07:42:29 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA22013; Wed, 13 May 1998 07:42:25 -0700
Message-Id: <3.0.5.32.19980512175257.007a7c40@doors>
X-Sender: bgbauer#alpha.delta.edu@doors
X-Mailer: QUALCOMM Windows Eudora Light Version 3.0.5 (32)
Errors-To: icon-group-errors@CS.Arizona.EDU
Date: Tue, 12 May 1998 19:52:57 -0500
From: Scissors  <jebediah@bigfoot.com>
To: icon-group@optima.CS.Arizona.EDU
Subject: Determining O/S
Mime-Version: 1.0
Content-Type: text/plain
Content-Disposition: inline


Oh, I thought of another question thats been bugging me.  What is the easiest
way to determine the O/S your running on?  I would have no idea how to
determine
unix, and Im uncertain the 'best' way to determine if the O/S is win3.1,
win95,
or winNT?

Would it be best to use some of the ifdef lines I've noticed in io.icn for
o/s secific code? if so what are the available definitions?

At first I thought code similar to this would work (which is from yescr.icn):

   if find("UNIX",&features) then {
      expressions
   }
   else if find("MS-DOS", &features) then {
      expressions
   }
   else if find("MS Windows NT", &features) then {
      expressions
   }

But this would lump win3.1 (win32s) with win95, and NT, and i believe it
wouldnt give any indication of if the app is console, or windowed.   Code
i'm working on now would need to act differently on win3.1 than it would on
win95/nt.

Any help, suggestions, or comments welcome.

Thanks,
scissors






--=_23774126.8FEE8660--


From icon-group-sender  Thu May 14 07:48:46 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id HAA23365
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Thu, 14 May 1998 07:48:45 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA23981; Thu, 14 May 1998 07:48:40 -0700
Message-Id: <199805141002.MAA28619@capway.com>
From: "Vladimir Grodzenski" <grodzens@capway.com>
To: icon-group@optima.CS.Arizona.EDU
Date: Thu, 14 May 1998 11:58:42 +0000
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7BIT
Subject: Annoying things
Reply-To: Vladimir <vladimir.grodzenski@capway.com>
Priority: urgent
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 604

1. Close one of the opened windows or dialogs (by clicking the 
CLOSE button in the upper-right corner) will cause the program 
termination. This is really annoying.

2. Under Windows95 the function system() does not work properly:
    system("wordpad " || filename) 

No way to list the directory:  
       system("DIR > FOO.TXT")  or 
       system("COMMAND /C DIR > FOO.TXT")

Well, this is mentioned in IPD271b document, but still how the h*ll
can we get the list of files then!?  

I'd suggest to implement some WinOpen dialog returning a list of
files (with filter, multiple selection etc. etc.) 


From icon-group-sender  Thu May 14 12:24:44 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id MAA12478
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Thu, 14 May 1998 12:24:34 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA22110; Thu, 14 May 1998 12:24:30 -0700
Message-Id: <199805141509.RAA25422@capway.com>
From: "Vladimir Grodzenski" <grodzens@capway.com>
To: icon-group@optima.CS.Arizona.EDU
Date: Thu, 14 May 1998 17:05:33 +0000
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7BIT
Subject: Re: AI use for Icon
Reply-To: <vladimir.grodzenski@capway.com>
Priority: urgent
In-Reply-To: <199805061704.MAA14525@axp.cmpu.net>
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 1140

On  6 May 98 at 12:04, Gordon Peterson wrote:

> For the "fuzzy match" I think that one interesting way to at least
> help winnow down the possibilities would be to examine the
> intersection of the character sets of the different names.  Those
> which have a high intersection (all but a "few" characters) can be
> then examined more closely.
> 
> For a better "fuzzy compare" function I've liked the use of
> overlapping character pairs (including a blank added to the start
> and end of each name).

Another approach:  

for each employee 'name' create a table, the keys of which
will be the characters from 'name' and values - their occurence in
'name'.   Such as:

 T := table(0)  
 every T [!name] +:= 1

We can define "sort of a difference" between two tables (T1,T2):

procedure tdiff(T1, T2)
local T, weight
 weight := 0
 T := table(0)

 every k := key(T1) | key(T2) do
     T [k] := abs( T1[k] - T2[k] )

 every weight +:= !T
 return weight
end


Vladimir Grodzenski
=================================================
E-mail: vladimir.grodzenski@capway.com
CompuServe: 100700,526
=================================================


From icon-group-sender  Thu May 14 12:24:48 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id MAA12481
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Thu, 14 May 1998 12:24:48 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA22028; Thu, 14 May 1998 12:24:43 -0700
Message-Id: <3.0.5.32.19980514113425.007b1e80@doors>
X-Sender: bgbauer#alpha.delta.edu@doors
X-Mailer: QUALCOMM Windows Eudora Light Version 3.0.5 (32)
Date: Thu, 14 May 1998 11:34:25 -0700
To: "Icon Group"<icon-group@optima.CS.Arizona.EDU>
From: Scissors <jebediah@bigfoot.com>
Subject: system command in win95
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 1697


Im using win95 and have a few successful uses of the system command,
but i havent tried it with piping or redirection.   I dont doubt it
is causing a problem for you though. 

So, i've had success with it, but I find it VERY slow.  If youre 
sending the output to another file... try pausing for 10 seconds, and
then reading the file.   I use a system command

system("regedit /e etc...") which exports part of the registry to a file if
you arent familiar with it.  And I have had to implement a large pause
before I 
can find the file!

if I do:

system(prog)   # any command which creates a new? file
every 1 to 50000 do if exists(newfile) then break  # file created
if not exists(newfile) then stop("file not created")

Of course, this is just a work around.  It was working... but only if my
loop was VERY generous.  My conclusion was that system releases before the
command is totally finished.  I have no way of knowing who is to blame, dos
or icon.

As for the listing of files in a dir...

Look in io.icn.  It contains procedures for reading directories.  I believe
it uses internal commands and interupts, and should be much faster and
better than system("DIR").   I havent tried them, so I cant explain, maybe
someone else can.   One I believe is DOS_Dir_list? or something similar...
there are a few, check em out.

And finally, another work around, but it will accomplish what you need, is to
write a batch file, then run it.  Such as:

batch := open("C:\temp\ver.bat", "w")
write(batch, "VER > ver.log")
close(batch)
system("C:\temp\ver.bat")

Then try reading.  Its possible this may not work either though, if
system() is
releasing too soon.  *shrug* It's worth a try though.






From icon-group-sender  Fri May 15 16:46:38 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id QAA22344
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Fri, 15 May 1998 16:46:38 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA24168; Fri, 15 May 1998 16:46:33 -0700
Comments: Authenticated sender is <stryx@pop3.demon.co.uk>
From: Iain Alexander <ia@stryx.demon.co.uk>
To: icon-group@optima.CS.Arizona.EDU
Date: Fri, 15 May 1998 21:50:22 +0000
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7BIT
Subject: Re: Annoying things
Priority: normal
In-Reply-To: <199805141002.MAA28619@capway.com>
X-Mailer: Pegasus Mail for Win32 (v2.54)
Message-Id: <895265449.105374.0@stryx.demon.co.uk>
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 1612

On 14 May 98 at 11:58, Vladimir Grodzenski wrote:
> 2. Under Windows95 the function system() does not work properly:
>     system("wordpad " || filename) 

Unfortunately, the Windows version of system() requires the file extension to be 
specified before it will work.  In addition, wordpad is not normally on the 
search path, so you would have to give the full pathname:

system("C:\\PROGRA~1\\ACCESS~1\\WORDPAD.EXE noname.icn")

or

system("C:\\Program Files\\Accessories\\Wordpad.exe noname.icn")

On 14 May 98 at 11:34, Scissors wrote:
> if I do:
> 
> system(prog)   # any command which creates a new? file
> every 1 to 50000 do if exists(newfile) then break  # file created
> if not exists(newfile) then stop("file not created")
> 
> Of course, this is just a work around.  It was working... but only if my
> loop was VERY generous.  My conclusion was that system releases before the
> command is totally finished.  I have no way of knowing who is to blame, dos or
> icon.

The two system(wordpad)s above behave differently - the first waits until 
wordpad has exited, the second returns immediately.  IIRC system(something.bat) 
waits for the batch file to complete.  Presumably if it thinks you are a 32-bit 
application (because you specified a long filename) running another 32-bit 
application it returns immediately, otherwise it waits!?

I imagine Icon has no real choice other than to use the Windows system() 
interface directly, and all this weirdness is down to the latter.
-- 
Iain Alexander               PGP 1024-bit key id B501A0AD
ia@stryx.demon.co.uk   I.Alexander@bsn0106.wins.icl.co.uk


From icon-group-sender  Mon May 18 08:32:49 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id IAA11112
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Mon, 18 May 1998 08:32:49 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA26695; Mon, 18 May 1998 08:32:44 -0700
Message-Id: <3.0.5.32.19980517140924.007a5450@doors>
X-Sender: bgbauer#alpha.delta.edu@doors (Unverified)
X-Mailer: QUALCOMM Windows Eudora Light Version 3.0.5 (32)
Date: Sun, 17 May 1998 14:09:24 -0700
To: "Icon Group"<icon-group@optima.CS.Arizona.EDU>
From: Scissors <jebediah@bigfoot.com>
Subject: Checking if a directory exists
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 213


What is the easiest way,  aside from having to chdir(),  to check if
a directory exists in windows?

I've tried exists(dir || "\\nul") but it doesn't appear to work.

Any ideas? suggestions?

Thanks,
scissors




From icon-group-sender  Mon May 18 08:32:42 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id IAA11095
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Mon, 18 May 1998 08:32:37 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA25440; Mon, 18 May 1998 08:32:32 -0700
Message-Id: <199805161337.PAA01330@capway.com>
From: "Vladimir Grodzenski" <grodzens@capway.com>
To: icon-group@optima.CS.Arizona.EDU
Date: Sat, 16 May 1998 15:34:05 +0000
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7BIT
Subject: Re: Annoying things
Reply-To: <vladimir.grodzenski@capway.com>
Priority: urgent
References: <199805141002.MAA28619@capway.com>
In-Reply-To: <895265449.105374.0@stryx.demon.co.uk>
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 706

On 15 May 98 at 21:50, Iain Alexander wrote:

> system("C:\\PROGRA~1\\ACCESS~1\\WORDPAD.EXE noname.icn")
> or
> system("C:\\Program Files\\Accessories\\Wordpad.exe noname.icn")

Surprisingly, the second solution (i.e. with long names) does not 
work under my Windows95... The first one with DOS file names works 
fine. Weird!

OTOH, I still cannot figure out how to list the directory since the
DOS "DIR" command system("dir > foo") does nor redirect output to a 
file... Using Int86 function is not possible in WinIcon.


Vladimir Grodzenski
=================================================
E-mail: vladimir.grodzenski@capway.com
CompuServe: 100700,526
=================================================


From icon-group-sender  Mon May 18 08:32:19 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id IAA11068
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Mon, 18 May 1998 08:32:17 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA22190; Mon, 18 May 1998 08:32:12 -0700
Date: Fri, 15 May 1998 23:43:41 -0500
Message-Id: <199805160443.XAA09845@segfault.cs.utsa.edu>
From: Clinton Jeffery <jeffery@segfault.cs.utsa.edu>
To: ia@stryx.demon.co.uk
Cc: icon-group@optima.CS.Arizona.EDU
In-Reply-To: <895265449.105374.0@stryx.demon.co.uk> (message from Iain
	Alexander on Fri, 15 May 1998 21:50:22 +0000)
Subject: Re: Annoying things
Reply-To: jeffery@cs.utsa.edu
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 705

Hi Iain,

Thanks for your helpful remarks on Windows Icon's system() function.

Up until the most recent binaries, advertized a week or two ago,
system() was implemented using WinExec() in a less than perfect way.
This less-than-perfect method is still the best I know how to do
under Win32s/Windows 3.1, but the latest binaries now use spawnvp()
under Windows95 and Windows NT, giving much improved behavior on
those platforms.  Anyone who wants to help me improve system() or
implement files opened using the "pipe" modes, please drop me a line.

Clint Jeffery, jeffery@cs.utsa.edu
Division of Computer Science, The University of Texas at San Antonio
Research http://www.cs.utsa.edu/research/plss.html


From icon-group-sender  Mon May 18 12:44:11 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id MAA24536
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Mon, 18 May 1998 12:44:11 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA22953; Mon, 18 May 1998 12:44:06 -0700
Date: Mon, 18 May 1998 13:25:59 -0500
Message-Id: <199805181825.NAA29555@segfault.cs.utsa.edu>
From: Clinton Jeffery <jeffery@segfault.cs.utsa.edu>
To: vladimir.grodzenski@capway.com
Cc: icon-group@optima.CS.Arizona.EDU
In-Reply-To: <199805161337.PAA01330@capway.com> (grodzens@capway.com)
Subject: Re: Annoying things
Reply-To: jeffery@cs.utsa.edu
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 1673


   > system("C:\\PROGRA~1\\ACCESS~1\\WORDPAD.EXE noname.icn")
   > or
   > system("C:\\Program Files\\Accessories\\Wordpad.exe noname.icn")

   Surprisingly, the second solution (i.e. with long names) does not 
   work under my Windows95... The first one with DOS file names works 
   fine. Weird!

I suspect that the command-line argument processing breaks up the latter
string incorrectly at the first space character unless the command is quoted:

   > system("\"C:\\Program Files\\Accessories\\Wordpad.exe\" noname.icn")

but I haven't checked to see if that is a proper fix yet.

>   OTOH, I still cannot figure out how to list the directory since the
>   DOS "DIR" command system("dir > foo") does nor redirect output to a 
>   file... Using Int86 function is not possible in WinIcon.

Under Windows Icon, open(dirname) should return a file value that produces
the filenames in the directory, one per line.  With the present #ifdef's,
this may only be enabled for the wicont program, not with nticont, but it
should be easy to enable it for both.

Going back to your original system("dir > foo") example, I would note that
dir is an "internal" program built-in to command.com or cmd.exe and that
you may be able to get better behavior by explicitly naming the desired
invocation of your command interpreter, something like
	system("C:\\WINDOWS\\COMMAND.COM /C /Q dir >foo")
but redirection in such a case might still be a problem; I'll put it on
my bug list and check it out.  UNIX (Linux) works much better!

Clint Jeffery, jeffery@cs.utsa.edu
Division of Computer Science, The University of Texas at San Antonio
Research http://www.cs.utsa.edu/research/plss.html


From icon-group-sender  Mon May 18 12:44:00 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id MAA24522
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Mon, 18 May 1998 12:44:00 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA20657; Mon, 18 May 1998 12:43:55 -0700
Message-Id: <s5601572.016@housmtp.oceaneering.com>
X-Mailer: Novell GroupWise 4.1
Date: Mon, 18 May 1998 11:01:53 -0500
From: Charles Hethcoat <CHETHCOA@oss.oceaneering.com>
To: icon-group@optima.CS.Arizona.EDU
Subject: Re:  Annoying things
Mime-Version: 1.0
Content-Type: text/plain
Content-Disposition: inline
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 2634

Iain Alexander wrote:
>I imagine Icon has no real choice other than to use the Windows system() 
interface directly, and all this weirdness is down to the latter.

The recent announcement of Unicon has me thinking.  I believe the Icon
community should set a more explicit goal of insuring portability between
systems, including weird ones.  (Naturally, a lot of the porting headaches
involve coping with Windows.)

Right now there are two Windows systems out there:  Windows 95 and NT. 
They both are "Win32", but are very different underneath, apparently.  I
imagine most of the Windows comments made on this group pertain to Win95,
but WinNT may be more important from a corporate-visibility viewpoint.

Anyway, I don't think the portability issues for these two OSs are the
same.

Concerning NT:  There is an interesting article in the June 1998 issue of
Performance Computing (nee Unix Review):  "Architecture of POSIX and
OpenNT".  This discusses the low-level system design of the OpenNT
product, which is a port of UNIX to run over the NT kernel.  IMHO, it
raises some interesting questions about the correct strategy for assuring
software portability between Unix and Windows NT.  (Especially now that we
Icon types have Unicon.)  A quote:  "This article discusses some of the
hurdles...a narrowly focused development community may be fueling
erroneous perceptions about about portability between the two platforms." 
Their approach:  OpenNT just bypasses most of the Win32 junk, and the
related porting problems, and works directly to the Windows NT kernel.  It
seems that Unicon could run on NT with almost no changes if a similar
approach were used.

Concerning Win 95:  One solution would seem to be to implement minimal
POSIX capabilities into or alongside the Win32 layer...at least enough to
permit Unicon to work on it.  I envision a replacement for the DOS
window...call it, oh, let's just say Winix (TM ;^).  It need not be a full
UNIX system.  It would be installed like 4DOS or other DOS replacement
shells.  Once installed, the necessary hooks would be there for Unicon to
run.

Apropos of portability, does anyone remember the Software Tools Users
Group?  Portability was their thing.  In the days before C became
universally available, they (we, for I was one of them) used to run a very
consistent, very Unix-like programming layer on systems as diverse as
OS/370, Primos, VMS, and Harris Vulcan (!).  All implemented in Fortran.

Cheers,
Charles Hethcoat
Oceaneering Space Systems, Inc.
16665 Space Center Blvd., Houston, Texas 77058 USA
chethcoa@oss.oceaneering.com
Voice:  281-228-5568 Fax:  281-488-2027



From icon-group-sender  Tue May 19 08:44:19 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id IAA11196
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Tue, 19 May 1998 08:44:18 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA15726; Tue, 19 May 1998 08:44:12 -0700
Date: Mon, 18 May 1998 19:53:12 -0500
Message-Id: <199805190053.TAA31385@segfault.cs.utsa.edu>
From: Clinton Jeffery <jeffery@segfault.cs.utsa.edu>
To: CHETHCOA@oss.oceaneering.com
Cc: icon-group@optima.CS.Arizona.EDU
In-Reply-To: <s5601572.016@housmtp.oceaneering.com> (message from Charles
	Hethcoat on Mon, 18 May 1998 11:01:53 -0500)
Subject: UNIX on top of NT
Reply-To: jeffery@cs.utsa.edu
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 1144


[Charlie Hethcoat pointed us at an article on OpenNT, a commercial UNIX
package that runs on top of Windows NT.]

That article in Performance Computing sounds interesting.  UNIX on NT people
might also be interested in a paper by Korn from the 1997 USENIX conference
which describes a full UNIX implementation on NT done at AT&T that consists
of libraries and command-line programs that run on NT and Win95.  Their
package co-exists with the Win32 stuff in the same application.

In both te case of OpenNT and the AT&T thing, the main problem I have is
that the software is not quite free enough to base Windows Icon upon it.
I will work on, or work with people on, a Windows implementation of the POSIX
facilities provided by Unicon.  If any of these UNIX on NT packages are
really free software, they might make the job easier, but a large subset
of Unicon can be done on Windows fairly easily in any case.  It is mostly
a question of finding the time (or the right person) to do it.

Clint Jeffery, jeffery@cs.utsa.edu
Division of Computer Science, The University of Texas at San Antonio
Research http://www.cs.utsa.edu/research/plss.html


From icon-group-sender  Thu May 21 08:57:28 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id IAA18936
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Thu, 21 May 1998 08:57:28 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA28817; Thu, 21 May 1998 08:57:22 -0700
Date: Wed, 20 May 1998 18:09:03 -0700
From: Gregg Townsend <gmt>
Message-Id: <9805210109.AA27090@hawk.CS.Arizona.EDU>
To: icon-group
Subject: Jcon: A Java-based Icon
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 982

We are pleased to announce the general distribution release of version 1.0
of "Jcon," an experimental Java-based Icon implementation.
 
Jcon is a new Java-based implementation of the Icon programming language.
The Jcon translator, written in Icon, generates Java class files that
execute in conjunction with a run-time system written in Java. 

Jcon is an essentially complete implementation of the core Icon language,
lacking only a few things such as chdir() that cannot be done in Java. Large
integers are not implemented, and only rudimentary graphics are present. 

Jcon is an experimental research project and should be viewed in that light.
We are distributing Jcon because it is interesting. Its performance does not
match that of the standard Icon interpreter, which is still preferred for
production and everyday use. 

Please visit 
	http://www.cs.arizona.edu/icon/jcon/
for more information about Jcon's status and distribution.

    Gregg Townsend
    Todd Proebsting


From icon-group-sender  Tue May 26 12:40:28 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id MAA27051
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Tue, 26 May 1998 12:40:27 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA06721; Tue, 26 May 1998 12:40:21 -0700
From: F.G.van.Dorp@digimedia.nl (F.G. van DORP)
To: icon-group@optima.CS.Arizona.EDU
Subject: 1-st impressions
Date: Sat, 23 May 1998 12:56:24 GMT
Message-Id: <3568c329.5170584@smtp.NL.net>
X-Mailer: Forte Agent 1.01/16.397
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 8BIT
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 1764


As the subject says, here are some Icon phenomena which,
after experimenting for a couple of weeks, struck me as interesting,
peculiar, awkward, etc. :

0) Scope restriction inside procedures automatic and no
    special syntax for variables (v. PERL e.g.)

1)  WHILE - you can't use the "{" without the DO in front.

2) MUTUAL EVAL. - doesn't backtrack, only difference
    with the COMPOUND EXPR. seems to be the option
    of selecting the outcome expression.

3) INDEXING very consistent (strings, lists, sets, even records)

4) SETS can't be concatenated (no automatic conversion to
    lists and back)

5)  &FAIL   equivalent to    NOT &NULL    and    \&NULL

6) PROC() doesn't work unless the procedure has been called at
    least once.

7) can't get RE_FIND (from IPL) to match greedily

8) STRING SCANNING is rather low-level (position-centered)
   and so a bit awkward (had to write most high-level functions
   myself), same applies to....

9) CO-EXPRESSIONS: powerful (possibilities for declarative
    programming, "lazy" evaluation, etc.) but not very intuitive 
   and (low-level) behavior sometimes puzzling

   - at times ^CE or @^CE won't work, while
     CE:=^CE seems to do the trick.

   - below are 3 expressions in order of evaluation time
     (approx. 1x, 4x, 8x)

     every write(lazy{z:=(1 to 50000), (5 > z)})
     every write(5 >       (1 to 50000))
     every write(      (z:=(1 to 50000),(5 > z)))

     procedure lazy(p)
     local a,b,c
     a:=get(p); b:=get(p)
     while (c:=(@a  @  ^b)) do {suspend c}
     end
-------------------------------------------------------------------------------------------------
Great concise language overall, but generators and especially
co-expressions not easy to get into

Bob.
======





From icon-group-sender  Tue May 26 12:41:06 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id MAA27062
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Tue, 26 May 1998 12:41:06 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA01479; Tue, 26 May 1998 12:41:00 -0700
Date: Mon, 25 May 1998 08:18:17 -0700
From: Ralph Griswold <ralph>
Message-Id: <9805251518.AA02290@jupiter.CS.Arizona.EDU>
To: icon-group
Subject: forwarded message
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 2300

From: "Paul A. Sue" <Paul_A._Sue@notes.bctel.com>
To: icon-project@optima.CS.Arizona.EDU
Message-Id: <0825660C.006C0CCA.00@notes.bctel.com>
Date: Fri, 22 May 1998 11:48:10 -0800
Subject: Help Me Promote Icon!
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Status: RO

Hi,

[Sorry to bother you, but no one seems to be responding to
me post to comp.lang.icon.]

I'm a very recent convert to Icon, and thought I'd use it here at work
to write a simple translator that I had originally written in C and Perl.
I wanted to show that Icon would be an ideal language for many
of the projects at my company.

Anyhow, here's the Perl script along with sample input and output
to give you an idea of what I'm trying to do.

Thanks in advance,

Paul
=================================================
Paul A. Sue, Consultant              ISM-BC AD/M
Voice-mail: (604) 432-4316 Pager: (604) 473-2811
E-mail:  paul@wsi.ca or paul_sue@notes.bctel.com

[From comp.lang.icon]

--------------------- Start of included file ----------------------------
From: Paul A. Sue <paul_sue@notes.bctel.com>
Subject: Newbie question
Date: Thursday, May 14, 1998 1:46 PM

Hi,

I'm wondering if anyone can help me "translate" the following Perl
script to the Icon equivalent.

[Perl gurus feel free to improve on my perl script if you wish!]

Thanks in advance,

Paul

#!/usr/local/bin/perl
while (<>) {
     chomp;
     if (/[a-z]/) {
      # --- convert from Directory to M/F format
          s#$#//#;
          s#^([a-z])#\*\1#;
          s#(\s+)([a-z])# \*\2#g;
          s#([a-z]+)([A-Z]+)#\1\(\2\)#g;
          tr/a-z/A-Z/;
          @result = split;
          $index = $#result;
          if ($index > 0) {
               $result[$index - 1] .= ",";
          }
     }
     else {
      # --- convert from M/F to Directory format
          s#//$##;
          s#,##g;
          tr/A-Z/a-z/;
          s#(\()(.*)(\))#\U\2#g;
          s#(^([a-z])|(\s+[a-z]))#\U\1\E#g;
          s#\*##g;
          @result = split;
     }
 print "@result\n";
 next;
}

Sample input/output:

input of:    "MacDonald John"
gives output of:     "MAC(D)ONALD, JOHN//"
input of:     "*VAN *DER(C)ORTLAND, NICK//"
gives output of:     "van derCortland Nick"
--------------------- End of included file ----------------------------






From icon-group-sender  Tue May 26 12:39:49 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id MAA26985
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Tue, 26 May 1998 12:39:49 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA00443; Tue, 26 May 1998 12:39:43 -0700
From: Guy Laden <guy@math.tau.ac.il>
Message-Id: <199805222014.XAA02171@orion.math.tau.ac.il>
Subject: Writing compilers in Icon
To: icon-group@baskerville.CS.Arizona.EDU
Date: Fri, 22 May 1998 23:14:55 +0300 (GMT+0300)
X-Mailer: ELM [version 2.4 PL25]
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 493

I noticed the newly announced implementation of Jcon is written in Icon. 
I was wondering whether anybody can comment on how suitable they 
find Icon for writing compilers? Do any of Icon's unique features
prove especially useful for this task? I'm wondering more about
generators and goal-directed evaluation; the built-in data-structures 
are obviously useful.

btw, does anybody have an Icon grammar they are willing to share
that is suitable for recursive descent parsing?

Regards,
Guy.


From icon-group-sender  Tue May 26 12:40:28 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id MAA27049
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Tue, 26 May 1998 12:40:15 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA31487; Tue, 26 May 1998 12:40:09 -0700
Message-Id: <199805231036.MAA17216@capway.com>
From: "Vladimir Grodzenski" <grodzens@capway.com>
To: jeffery@cs.utsa.edu
Date: Sat, 23 May 1998 12:33:14 +0000
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7BIT
Subject: Re: Annoying things
Reply-To: <vladimir.grodzenski@capway.com>
Cc: icon-group@optima.CS.Arizona.EDU
Priority: urgent
In-Reply-To: <199805181825.NAA29555@segfault.cs.utsa.edu>
References: <199805161337.PAA01330@capway.com> (grodzens@capway.com)
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 1008

On 18 May 98 at 13:25, you wrote:

> Under Windows Icon, open(dirname) should return a file value that
> produces the filenames in the directory, one per line.  With the
> present #ifdef's, this may only be enabled for the wicont program,
> not with nticont, but it should be easy to enable it for both.

Thanks for the suggestion, it works! You should however mention
it in the Icon reference otherwise we can only guess... Are there 
some other "undocumented" features like with Microsoft? :-)

After opening the directory we get a list of both files and
[sub]directories so the only way I found to distinguish them is
check the failure with chdir(name) which is rather unusual...

Of course, we have to exclude the "." and ".." directories from the 
list. Anyway, it works and even allows to easily recurse directories.


Vladimir Grodzenski
=================================================
E-mail: vladimir.grodzenski@capway.com
CompuServe: 100700,526
=================================================


From icon-group-sender  Tue May 26 12:40:58 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id MAA27055
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Tue, 26 May 1998 12:40:53 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA26931; Tue, 26 May 1998 12:40:47 -0700
Date: Sat, 23 May 1998 10:24:33 -0500
Message-Id: <199805231524.KAA05320@segfault.cs.utsa.edu>
From: Clinton Jeffery <jeffery@segfault.cs.utsa.edu>
To: vladimir.grodzenski@capway.com
Cc: icon-group@optima.CS.Arizona.EDU
In-Reply-To: <199805231036.MAA17216@capway.com> (grodzens@capway.com)
Subject: Re: Annoying things
Reply-To: jeffery@cs.utsa.edu
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 1203


[open(dirname) gives a file that returns a directory's filenames, one per line]
>   Thanks for the suggestion, it works! You should however mention
>   it in the Icon reference otherwise we can only guess... Are there 
>   some other "undocumented" features like with Microsoft? :-)

Oops.  Well, at least I release the source code.  :-)

Actually, the reason that directory reading is not yet documented is that
it is likely to change.  For these kind of operating system interactions
whatever I implement and document needs to be consistent with Shamim
Mohamed's POSIX interface for Icon, Unicon.  His interface supports
open(dirname) but specifies more complete file information returned for each
directory entry, such as permissions, ownership, and timestamps.  Prior to
this I had made open(dirname) simpler, and some graduate students of mine
had implemented a separate function for such file attributes, fattrib().
He and I need to talk about it a bit more, but I am likely to adopt his
semantics for the lines returned by open(dirname).

Clint Jeffery, jeffery@cs.utsa.edu
Division of Computer Science, The University of Texas at San Antonio
Research http://www.cs.utsa.edu/research/plss.html


From icon-group-sender  Wed May 27 08:21:10 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id IAA03028
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Wed, 27 May 1998 08:21:10 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA01621; Wed, 27 May 1998 08:21:04 -0700
Message-Id: <356B9C3B.15DA6101@charlie.cns.iit.edu>
Date: Tue, 26 May 1998 23:53:15 -0500
From: "Thomas W. Christopher" <tc@charlie.cns.iit.edu>
Organization: Illinois Institute of Technology
X-Mailer: Mozilla 4.05 [en] (WinNT; I)
Mime-Version: 1.0
To: Guy Laden <guy@math.tau.ac.il>
Cc: icon-group@baskerville.CS.Arizona.EDU
Subject: Re: Writing compilers in Icon
References: <199805222014.XAA02171@orion.math.tau.ac.il>
Content-Type: multipart/mixed; boundary="------------D07FD781A7537D236717351D"
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 8242

This is a multi-part message in MIME format.
--------------D07FD781A7537D236717351D
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

I have found Icon quite pleasant for compiler writing, although not for
production quality compilers. I use them for teaching.

See:   http://www.iit.edu/~tc/toolsfor.htm

I found generators a lot more useful for writing my parser generator
than for writing compilers themselves.

I have used coexpressions for scanners, but I don't currently due to the
difficulty of explaining it to students.

My parser uses string invocation of procedures. All the grammar symbols,
including action symbols, are represented by strings in my parse tables. The

action symbols are the names of the semantic routines to call.

I'll attach a grammar for Icon that my LL(k) parser generator, tcllk, can
handle.

If you pass it through:
        iconx tcllk -p icon.grm
you might be able to use the after-translation version of the grammar for
recursive
descent, although it's a lot easier to use my parser generator and parser.

A few suggestions for scanning tokens in Icon and passing them on to the
parser:
1) queue up the tokens leaving the scanner before the parser sees them, use
the
queue to do (2) and (3);
2) remove duplicate ";" and end-of-line symbols;
3) remove end-of-line symbols preceeding or following symbols like "do",
"then", "else";
4) fix the panic-mode error recovery to respond to an unexpected end-of-line

symbol by throwing it away without comment.

Guy Laden wrote:

> I noticed the newly announced implementation of Jcon is written in Icon.
> I was wondering whether anybody can comment on how suitable they
> find Icon for writing compilers? Do any of Icon's unique features
> prove especially useful for this task? I'm wondering more about
> generators and goal-directed evaluation; the built-in data-structures
> are obviously useful.
>
> btw, does anybody have an Icon grammar they are willing to share
> that is suitable for recursive descent parsing?
>
> Regards,
> Guy.



--
Thomas W. Christopher --------------------- tc@charlie.cns.iit.edu
Associate Prof., Illinois Inst. of Tech.--- http://www.iit.edu/~tc
Principal, Tools of Computing LLC.--------- http://www.toc-press.com


--------------D07FD781A7537D236717351D
Content-Type: application/x-unknown-content-type-grm_auto_file; name="icon.grm"
Content-Transfer-Encoding: base64
Content-Disposition: inline; filename="icon.grm"
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--------------D07FD781A7537D236717351D--



From icon-group-sender  Wed May 27 08:20:59 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id IAA03009
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Wed, 27 May 1998 08:20:58 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA32250; Wed, 27 May 1998 08:20:52 -0700
Message-Id: <4FD6422BE942D111908D00805F3158DF0757B2CE@red-msg-52.dns.microsoft.com>
From: Todd Proebsting <toddpro@microsoft.com>
To: "'Guy Laden'" <guy@math.tau.ac.il>, icon-group@baskerville.CS.Arizona.EDU
Subject: RE: Writing compilers in Icon
Date: Tue, 26 May 1998 13:22:48 -0700
X-Mailer: Internet Mail Service (5.5.2328.0)
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 2106

I really enjoyed writing Jcon's translator in Icon.  (I had help from Denise
Todd, Bob Alexander and Gregg Townsend.)  I made little use of goal-directed
evaluation, although generators were great fun to use.  In fact, the whole
translator setup is based on a series of co-expressions that feed one
another as a series of filters, which would have been very painful without
generators.  For instance, to translate a file, file.icn, into ucode files
file.u1 and file.u2, the magic incantation is

% jtran preproc file.icn : yylex : parse : u_gen_File -out:file

This directs the translator to create a 4-stage pipe: preprocessing, lexing,
parsing, and generating ucode.  Each stage is a co-expression that eats the
output of the previous co-expression as input.  For fun, try the following
pipelines:

% jtran preproc file.icn : echo
% jtran preproc file.icn : yylex : echo
% jtran preproc file.icn : yylex : parse : echo

Each will give you some idea as to what each stage communicates to the next.

(NOTE:  The colons must be separated from the commands by whitespace.  Also,
"echo" filter can be given a "-verbose" flag for really chatty output.)

I don't have an Icon grammar that is suitable for recursive-descent parsing,
BUT jtran has just such a parser in it.  Check the file tran/parse.icn.
(The command above that ends in "parse : echo" will give you some idea of
the abstract syntax trees built by the parser.)

Todd

-----Original Message-----
From: Guy Laden [mailto:guy@math.tau.ac.il]
Sent: Friday, May 22, 1998 1:15 PM
To: icon-group@baskerville.CS.Arizona.EDU
Subject: Writing compilers in Icon


I noticed the newly announced implementation of Jcon is written in Icon. 
I was wondering whether anybody can comment on how suitable they 
find Icon for writing compilers? Do any of Icon's unique features
prove especially useful for this task? I'm wondering more about
generators and goal-directed evaluation; the built-in data-structures 
are obviously useful.

btw, does anybody have an Icon grammar they are willing to share
that is suitable for recursive descent parsing?

Regards,
Guy.


From icon-group-sender  Wed May 27 08:21:22 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id IAA03054
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Wed, 27 May 1998 08:21:21 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA29918; Wed, 27 May 1998 08:21:15 -0700
From: Otfried Geffert <og@stardivision.de>
Date: Wed, 27 May 1998 11:37:39 GMT
Message-Id: <19980527.11373903@og-1042.stardiv.de>
Subject: Re: Writing compilers in Icon
To: Guy Laden <guy@math.tau.ac.il>
Cc: icon-group@baskerville.CS.Arizona.EDU
In-Reply-To: <199805222014.XAA02171@orion.math.tau.ac.il>
References: <199805222014.XAA02171@orion.math.tau.ac.il>
X-Mailer: Mozilla/3.0 (compatible; StarOffice/4.0; WinNT/Win95)
X-Priority: 3 (Normal)
Mime-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 1125

Hi Guy,

in 'The Icon programming language' 3rd edition from Griswold &=20
Griswold
they describe how to use Icon for syntax analysis using

- alternation
- recursive lists=20
- generators to represent syntactical constructs

on page 220-226.

It's really worth reading.=20

I myself wrote a compiler/linker, but I used lex/yacc and C, because
I did'nt know Icon in those days. With Icon I would have implemented=20=

it much faster, especially because of the builtin data
structures.

Regards, Otfried


>>>>>>>>>>>>>>>>>>  <<<<<<<<<<<<<<<<<<

Guy Laden <guy@math.tau.ac.il> wrote:


> I noticed the newly announced implementation of Jcon is written in=20=

Icon.=20
> I was wondering whether anybody can comment on how suitable they=20
> find Icon for writing compilers? Do any of Icon's unique features
> prove especially useful for this task? I'm wondering more about
> generators and goal-directed evaluation; the built-in data-structures =

> are obviously useful.
>=20
> btw, does anybody have an Icon grammar they are willing to share
> that is suitable for recursive descent parsing?
>=20
> Regards,
> Guy.
>=20





From icon-group-sender  Fri May 29 08:30:13 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id IAA13437
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Fri, 29 May 1998 08:30:13 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA24809; Fri, 29 May 1998 08:30:06 -0700
Message-Id: <3.0.5.32.19980528174121.007b0680@doors>
X-Sender: bgbauer#alpha.delta.edu@doors
X-Mailer: QUALCOMM Windows Eudora Light Version 3.0.5 (32)
Date: Thu, 28 May 1998 17:41:21 -0700
To: gep2@computek.net
From: Scissors <jebediah@bigfoot.com>
Subject: Re: Determining O/S
Cc: "Icon Group"<icon-group@optima.CS.Arizona.EDU>
In-Reply-To: <199805271056.FAA08434@axp.cmpu.net>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 334

At 05:56 AM 5/27/98 -0500, you wrote:
>Note that under Windows (of any flavor) the program can be changed from 
>"console" to "windowed" at any time, and vice-versa, and without advance
notice.

How?  Isnt the program limited to console only, if compiled with NTICONT?
and Windowed only, if compiled with WICONT?

hmm...

B Bauer





From icon-group-sender  Mon Jun  1 09:05:23 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id JAA02631
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Mon, 1 Jun 1998 09:05:22 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA06428; Mon, 1 Jun 1998 09:05:16 -0700
Date: Fri, 29 May 98 22:28:40 -0400
Message-Id: <9805300228.AA0191@valinet.com>
From: Paul Abrahams <abrahams@acm.org>
To: jeffery@cs.utsa.edu
Cc: icon-group@optima.CS.Arizona.EDU
In-Reply-To: <199805231524.KAA05320@segfault.cs.utsa.edu> (message from
	Clinton Jeffery on Sat, 23 May 1998 10:24:33 -0500)
Subject: Re: Annoying things
Reply-To: abrahams@acm.org
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 1627

>>>>> On Sat, 23 May 1998 10:24:33 -0500, Clinton Jeffery <jeffery@segfault.cs.utsa.edu> said:

|vladimir> [open(dirname) gives a file that returns a directory's
|vladimir> filenames, one per line] Thanks for the suggestion, it works!
|vladimir> You should however mention it in the Icon reference otherwise
|vladimir> we can only guess... Are there some other "undocumented"
|vladimir> features like with Microsoft? :-)

|Clint> Oops.  Well, at least I release the source code.  :-)

|Clint> Actually, the reason that directory reading is not yet
|Clint> documented is that it is likely to change.  For these kind of
|Clint> operating system interactions whatever I implement and document
|Clint> needs to be consistent with Shamim Mohamed's POSIX interface for
|Clint> Icon, Unicon.  His interface supports open(dirname) but
|Clint> specifies more complete file information returned for each
|Clint> directory entry, such as permissions, ownership, and timestamps.
|Clint> Prior to this I had made open(dirname) simpler, and some
|Clint> graduate students of mine had implemented a separate function
|Clint> for such file attributes, fattrib().  He and I need to talk
|Clint> about it a bit more, but I am likely to adopt his semantics for
|Clint> the lines returned by open(dirname).

Given the variances among operating systems, perhaps there ought to be a
convention that the first space-delimited item in each line produced by
open(dirname) is the filename.

This does seem close to my earlier suggestion that reading a directory
should act as a generator of the filenames within the directory.

Paul (Paul Abrahams)





From icon-group-sender  Mon Jun  1 09:05:41 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id JAA02642
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Mon, 1 Jun 1998 09:05:35 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA04735; Mon, 1 Jun 1998 09:05:28 -0700
Message-Id: <3.0.5.32.19980530023718.007aa100@doors>
X-Sender: bgbauer#alpha.delta.edu@doors
X-Mailer: QUALCOMM Windows Eudora Light Version 3.0.5 (32)
Date: Sat, 30 May 1998 02:37:18 -0700
To: "Icon Group"<icon-group@optima.CS.Arizona.EDU>
From: bauerb <bauerb@geocities.com>
Subject: Re: Determining O/S
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 3679


   >At 05:56 AM 5/27/98 -0500, GEP wrote:
   >>Note that under Windows (of any flavor) the program can be changed from 
   >>"console" to "windowed" at any time, and vice-versa, and without advance
   >>notice.

>>   How?  Isnt the program limited to console only, if compiled with NTICONT?
>>   and Windowed only, if compiled with WICONT?

>As near as I can tell, you are right.  I don't know what he means here.

>Clint, jeffery@cs.utsa.edu


Ok, I guess the easiest way for me to explain, is give a few examples.

I've created a procedure DateInFormat(i, s1):s2

i     an integer of the seconds past the standard 'base year' of 1970.
(optional,       default is current time).
s1    a string template for the Date.  Example: "dddd, MMMM dd, yyyy"
      or "MM/dd/yy".  (optional, default is the users current preference, 
      as determined using 'localization' settings for windows.
s2    a string of the date in the format of s1.  Examples: (using above
templates)
      "Tuesday, March 10, 1987" or "3/10/87"

So I can do something similar to 

sdate := DateInFormat(secs)

and have the date formatted in the standard the user is used to, or if I
specify
a format, in that format (which could even be something like
DateInFormat(,"Today is dddd") : "Today is Tuesday".

Ok, in this situation, the localization (and timezone bias etc.) settings
for win3.1 are located in win.ini, but the localization settings for win95
and NT are located in
the registry.   In win95 they are 'left' in the win.ini, but arent
maintained, so daylight savings time isnt accounted for, and neither are
changes to timezones after the user first installs Win95.  (In my
situation, Ive moved from -5 to -8 hoursfromGMT since installing win95,
which means at times the win.ini settings for timezone are up to 4 hours
incorrect.)   So this is why It would matter if Im in Win3.1 or Win95/NT.
The procedure would need to gather 'defaults' from separate places
depending on what O/S it's run on (and yet another location from UNIX if
one exists).

I realise I can use system("VER > VER.TMP") but so far I've had trouble
with system() commands which create new files (as was talked about in
previous posts).  Im just wondering the 'fastest' and 'least overhead' way
to determine O/S.

Win3.1/Win32's cant count the same as Win95 in this case, so the &features
way Ive noticed people use I don't think will work.



About the "console" or "window":

The main program I am working on now, I originally coded all to work as a
DOS Command line utility.  Im ready now, to take the source, and begin
adding windows features, and compile using wincont.

I would like to not have to maintain 2 separate versions of source, and
separately compiled executables, just to offer 2 different versions of the
program. 1 for DOS, one for Windows.

Im wondering if its possible to do something similar to:

If \win then WindowWrite("My output")  # run as a windows app
else write("My output")	            # run as console app

(Forgive me but I cant remember the function to write to a window yet.)

>From testing, ive noticed I will now need to go through all my source, and
alter
any of the commands I used to give status reports, changing them from
'write()'
to the correct "Windows" command, otherwise programs compiled with WICONT
seem to create a new window just for write() output.  But if I change them
all, I lose the ability to run the app in DOS mode.  So i'll have to
maintain two separate versions.

Is there a way to run a program compiled with wicont, as a console app? and
keeps its
output to the console, and not a window?

Hopefully I havent confused you more, I appologise if I have.

scissors





From icon-group-sender  Mon Jun  1 13:51:26 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id NAA25760
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Mon, 1 Jun 1998 13:51:25 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA06925; Mon, 1 Jun 1998 13:51:19 -0700
Date: Mon, 1 Jun 1998 12:47:16 -0500 (CDT)
From: Chris Tenaglia <cdt@post.its.mcw.edu>
To: icon-group@optima.CS.Arizona.EDU
Subject: disp.icn
Message-Id: <Pine.SOL.3.96.980601124103.8745A-100000@post.its.mcw.edu>
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 1993

>From cdt@post.its.mcw.edu Mon Jun  1 12:38:06 1998
Status: O
X-Status: 
Newsgroups: 
Date: Mon, 1 Jun 1998 12:38:06 -0500 (CDT)
From: Chris Tenaglia <cdt@post.its.mcw.edu>
To: icon-group@cs.arizona.edu
Subject: disp.icn
Fcc: sentmail
Message-ID: <Pine.SOL.3.96.980601123411.6757A@post.its.mcw.edu>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII

Here's a little snippet I found useful. We use mostly PCs
but the unix servers still have apps in Xwindows. As we
log in from various PCs, the DISPLAY variable in the unix
shell needs to change. So here is disp.icn
Environment solaris/unix. c-shell. telnet session.
After I compile it onto the path I put this line in .cshrc

setenv DISPLAY `disp`

Note : If you have other telnet sessions from other computers
simultaneously, you may get their display instead. You also
must the X-server running on your PC/MAC/Xterm as appropriate.
Tested on : SunOS xxx 5.5.1 Generic_103640-14 sun4m sparc
SUNW,SPARCstation-5
Enjoy.

####################################################################
#
# file : disp.icn
# desc : determine the "DISPLAY" variable for Xwindows in a solaris
#        unix session
# use  : disp (called by .login or .cshrc)
#
# update           by             what
# 01-jun-1998      tenaglia       initial write
#
procedure main()
  com1 := "env >disp.tmp"
  system(com1)
  in := open("disp.tmp")
  every line := !in do if match("USER",line) then self := parse(line,'= ')[2]
  close(in)

  com2 := "who >disp.tmp"
  system(com2)
  in := open("disp.tmp")
  every line := !in do if match(self,line) then display := parse(line,' \t')[6]
  close(in)

  remove("disp.tmp")
  write(display[2:-1],":0.0")
  end

#
# parse a string into a list with respect to a delimiter
#
procedure parse(line,delims)
  static chars
  chars  := &cset -- delims
  tokens := []
  line ? while tab(upto(chars)) do put(tokens,tab(many(chars)))
  return tokens
  end
#################################################################  







From icon-group-sender  Mon Jun  1 16:29:35 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id QAA08263
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Mon, 1 Jun 1998 16:29:34 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA02320; Mon, 1 Jun 1998 16:29:27 -0700
Date: Mon, 1 Jun 1998 17:11:58 -0500
Message-Id: <199806012211.RAA06897@segfault.cs.utsa.edu>
From: Clinton Jeffery <jeffery@segfault.cs.utsa.edu>
To: abrahams@acm.org
Cc: icon-group@optima.CS.Arizona.EDU
In-Reply-To: <9805300228.AA0191@valinet.com> (message from Paul Abrahams on
	Fri, 29 May 98 22:28:40 -0400)
Subject: Re: Annoying things
Reply-To: jeffery@cs.utsa.edu
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 1036


   [Paul Abrahams wrote:]
   Given the variances among operating systems, perhaps there ought to be a
   convention that the first space-delimited item in each line produced by
   open(dirname) is the filename.

This is not a bad suggestion.  Certainly the format of the string should
follow conventions that preserve portability.  Some operating systems
(Macintosh, and recent versions of Windows) allow spaces in filenames!

   This does seem close to my earlier suggestion that reading a directory
   should act as a generator of the filenames within the directory.

That's because my current Windows directory reading facility was inspired
directly by your suggestion.  It was a good suggestion, Paul!  It isn't
technically a generator, although you can use the generate operator (!)
to get a generator of filenames, as in:

	f := open(dirname)
	every filename := !f do ...

Clint Jeffery, jeffery@cs.utsa.edu
Division of Computer Science, The University of Texas at San Antonio
Research http://www.cs.utsa.edu/research/plss.html


From icon-group-sender  Mon Jun  1 16:29:41 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id QAA08266
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Mon, 1 Jun 1998 16:29:40 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA07315; Mon, 1 Jun 1998 16:29:34 -0700
Date: Mon, 01 Jun 98 18:53:29 -0400
Message-Id: <9806012253.AA0156@valinet.com>
From: Paul Abrahams <abrahams@acm.org>
To: jeffery@cs.utsa.edu
Cc: abrahams@acm.org, icon-group@optima.CS.Arizona.EDU
In-Reply-To: <199806012211.RAA06897@segfault.cs.utsa.edu> (message from
	Clinton Jeffery on Mon, 1 Jun 1998 17:11:58 -0500)
Subject: Re: Annoying things
Reply-To: abrahams@acm.org
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 1062

>>>>> On Mon, 1 Jun 1998 17:11:58 -0500, Clinton Jeffery <jeffery@segfault.cs.utsa.edu> said:

|Paul> Given the variances among operating systems, perhaps there ought
|Paul> to be a convention that the first space-delimited item in each
|Paul> line produced by open(dirname) is the filename.

|Clint> This is not a bad suggestion.  Certainly the format of the
|Clint> string should follow conventions that preserve portability.
|Clint> Some operating systems (Macintosh, and recent versions of
|Clint> Windows) allow spaces in filenames!

I care less about what the convention is than that there be one.  You're
right, of course, about the space-in-file-names issue, but that can be
an issue even in making sense of the returned value within a single OS
context.  Quoting the filename is one solution, although even that isn't
adequate if filenames can contain quotes.  The two alternatives that
come to mind that are completely general are (a) using quotes, with
internal quotes doubled, and (b) using backslashes to quote the next
character.

Paul Abrahams




From icon-group-sender  Tue Jun  2 09:05:15 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id JAA11208
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Tue, 2 Jun 1998 09:05:15 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA07969; Tue, 2 Jun 1998 09:05:08 -0700
Date: Mon, 01 Jun 98 22:45:58 -0400
Message-Id: <9806020245.AA0199@valinet.com>
From: Paul Abrahams <abrahams@acm.org>
To: ok@atlas.otago.ac.nz
Cc: icon-group@optima.CS.Arizona.EDU
Subject: Retrieving directory contents
Reply-To: abrahams@acm.org
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 1236

>>>>> On Tue, 2 Jun 1998 13:29:00 +1200 (NZST), "Richard A. O'Keefe" <ok@atlas.otago.ac.nz> said:

|Richard> A directory-with-file-attributes is NOT a sequence of text
|Richard> lines; it is NOT a sequence, and its elements are NOT lines of
|Richard> text.  Icon has the notational resources to describe what a
|Richard> directory IS quite directly, and it would be a shame not to
|Richard> use them.

|Richard> What is a directory with attributes?  It is a partial function
|Richard> from name strings to attributes.

|Richard> Something like

|Richard> 	directory(DirectoryNameString [,
|Richard> FilePatternWithWildCards]) -> table with strings as keys and
|Richard> attribute records as value

|Richard> would be the perfect fit to Icon.  Generation can then be
|Richard> based on the existing key generation.

That certainly seems like a good approach - but I'd hate to see the
whole idea of system-independent directory processing lost because the
solution seems too elaborate or requires too much work.  I'm concerned
about the best being the enemy of the good here.  What I care about more
than anything else is seeing *some* system-independent method of
retrieving the names of the files in a given directory.

Paul Abrahams



From icon-group-sender  Tue Jun  2 09:05:21 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id JAA11212
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Tue, 2 Jun 1998 09:05:21 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA05080; Tue, 2 Jun 1998 09:05:14 -0700
Date: Mon, 1 Jun 1998 23:43:37 -0500
Message-Id: <199806020443.XAA08820@segfault.cs.utsa.edu>
From: Clinton Jeffery <jeffery@segfault.cs.utsa.edu>
To: ok@atlas.otago.ac.nz
Cc: abrahams@acm.org, icon-group@optima.CS.Arizona.EDU
In-Reply-To: <199806020129.NAA09045@atlas.otago.ac.nz> (ok@atlas.otago.ac.nz)
Subject: Directory access facilities
Reply-To: jeffery@cs.utsa.edu
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 996

[Richard O'Keefe points out that directories are more like tables than they
are like files, and suggests a built-in function that returns a table mapping
file names to records of attributes, like in Tcl.]

Well, this is the kind of thing that gets object-oriented modeling people
excited.  A directory *is* a file (it has persistent state, lives in the
file system name space, etc.) and it *is* a table (set of keys that map onto
values).  Your point is well-taken.  From a table it would be trivial to use
key(T) to generate the filenames for the common case that Paul wanted.  If
you push the table analogy to its logical extreme, though, you should
implement the directory/table operations as direct manipulations on the file
system, like in the language EZ.

Anyone else have ideas (or votes) on how to do directory access facilities?

Clint Jeffery, jeffery@cs.utsa.edu
Division of Computer Science, The University of Texas at San Antonio
Research http://www.cs.utsa.edu/research/plss.html


From icon-group-sender  Tue Jun  2 09:05:06 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id JAA11202
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Tue, 2 Jun 1998 09:05:05 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA03901; Tue, 2 Jun 1998 09:04:58 -0700
Date: Tue, 2 Jun 1998 13:29:00 +1200 (NZST)
From: "Richard A. O'Keefe" <ok@atlas.otago.ac.nz>
Message-Id: <199806020129.NAA09045@atlas.otago.ac.nz>
To: abrahams@acm.org, jeffery@cs.utsa.edu
Subject: Re: Annoying things
Cc: icon-group@optima.CS.Arizona.EDU
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 1620

	
	   [Paul Abrahams wrote:]
	   Given the variances among operating systems, perhaps there ought to be a
	   convention that the first space-delimited item in each line produced by
	   open(dirname) is the filename.
	
	This is not a bad suggestion.

Given that UNIX, MacOS, and Win95/WinNT filenames may contain any number
of spaces, it is in fact a pretty BAD suggestion.  For true system
portability, whatever convention is followed should be completely
independent of file name spelling.

A directory-with-file-attributes is NOT a sequence of text lines;
it is NOT a sequence, and its elements are NOT lines of text.
Icon has the notational resources to describe what a directory IS
quite directly, and it would be a shame not to use them.

What is a directory with attributes?  It is a partial function
from name strings to attributes.

Something like

	directory(DirectoryNameString [, FilePatternWithWildCards])
	-> table with strings as keys and attribute records as value

would be the perfect fit to Icon.  Generation can then be based on
the existing key generation.

It would be a good idea to look at TCL, which works quite nicely
multiplatform.  On the evidence of TCL, it would be possible to
have file attribute records with fields

	atime, ctime, dev, executable, gid, ino, mode, mtime, nlink,
	owned, readable, readonly, size, type, uid, writable

on all systems,

	group, owner

on UNIX,

	archive, hidden, system

on Windows, and

	hidden, creator, mactype

on MacOS.

Let's face it, one of the important steps in the transition from SNOBOL
to Icon was to step *away* from 'everything's a string'.



From icon-group-sender  Tue Jun  2 09:05:28 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id JAA11215
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Tue, 2 Jun 1998 09:05:27 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA09119; Tue, 2 Jun 1998 09:05:20 -0700
Message-Id: <35740ACA.4373@gte.net>
Date: Tue, 02 Jun 1998 09:23:06 -0500
From: MJE <evans@gte.net>
Reply-To: evans@gte.net
Organization: None
X-Mailer: Mozilla 3.01 (Win95; I)
Mime-Version: 1.0
To: icon-group@optima.CS.Arizona.EDU
Cc: evans@gte.net
Subject: EditPad
References: <9806012253.AA0156@valinet.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 234

Here is a freeware replacement for Windows 95/NT's NotePad:

	http://www.ping.be/jg/

The user comments on this web site say it's fabulous.  I saw it in the
May 7 1998 issue of EDN, p. 72.

Try it out on your Icon code.

Mark Evans



From icon-group-sender  Tue Jun  2 12:30:47 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id MAA18463
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Tue, 2 Jun 1998 12:30:47 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA09174; Tue, 2 Jun 1998 12:30:40 -0700
From: gep2@computek.net
Date: Tue, 2 Jun 1998 11:34:02 -0500
Message-Id: <199806021634.LAA25018@axp.cmpu.net>
Mime-Version: 1.0
Content-Type: text/plain
Content-Transfer-Encoding: 7bit
Subject: Directory access facilities
To: icon-group@optima.CS.Arizona.EDU
X-Mailer: SPRY Mail Version: 04.00.06.17
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 989

> [Richard O'Keefe points out that directories are more like tables than they
are like files, and suggests a built-in function that returns a table mapping
file names to records of attributes, like in Tcl.]

> Well, this is the kind of thing that gets object-oriented modeling people
excited.  A directory *is* a file (it has persistent state, lives in the
file system name space, etc.) and it *is* a table (set of keys that map onto
values).  Your point is well-taken.  From a table it would be trivial to use
key(T) to generate the filenames for the common case that Paul wanted.  If
you push the table analogy to its logical extreme, though, you should
implement the directory/table operations as direct manipulations on the file
system, like in the language EZ.

Good points all.  Now let's not forget the multi-level hierarchical structure of 
file directories...

Gordon Peterson
http://www.computek.net/public/gep2/
Support the Anti-SPAM Amendment!  Join at http://www.cauce.org/



From icon-group-sender  Tue Jun  2 12:30:33 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id MAA18450
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Tue, 2 Jun 1998 12:30:32 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA05039; Tue, 2 Jun 1998 12:30:25 -0700
Date: Tue, 2 Jun 1998 09:19:59 -0700
From: Gregg Townsend <gmt>
Message-Id: <9806021619.AA28016@hawk.CS.Arizona.EDU>
To: icon-group
Subject: Re:  Directory access facilities
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 194

I'd prefer a simple built-in function that would generate the names
of the files contained in a directory, e.g.
	contents(s)
with other mechanisms (if any) used for finding attributes.

	Gregg


From icon-group-sender  Tue Jun  2 12:31:01 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id MAA18473
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Tue, 2 Jun 1998 12:31:00 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA08664; Tue, 2 Jun 1998 12:30:53 -0700
From: gep2@computek.net
Date: Tue, 2 Jun 1998 11:49:24 -0500
Message-Id: <199806021649.LAA27265@axp.cmpu.net>
Mime-Version: 1.0
Content-Type: text/plain
Content-Transfer-Encoding: 7bit
Subject: Directory and file access
To: icon-group@optima.CS.Arizona.EDU
X-Mailer: SPRY Mail Version: 04.00.06.17
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 2651

Actually, another point here comes to mind, too.

More than fifteen years ago, I was involved in the design effort for a new 
operating system at a computer manufacturer (Datapoint, in fact) and I thought 
it made sense to finally rationalize the bizarre disparity in dealing with 
various hierarchies of data storage.

Consider, in a modern network:
  
    You have one or more file servers, which usually are each named, and which 
are looked up using special calls.

    You have one or more volumes per file server, which are usually each named 
(and where normally one shouldn't have to worry at the programming level as to 
which server is hosting the volume in question today) and which are looked up 
using special calls.

    You have multiple levels of named subdirectories on each volume and which 
are looked up using special calls.

    You have multiple named files within each subdirectory, and which are looked 
up using yet again special calls.

    If the file is a library, archive (ZIP for example) or PDS, you have a 
directory of members within that file which must be looked up using still other 
special calls.

    If the file is an ISAM (or other keyed file) you have to do key searches 
within the file using OTHER special calls.
   
It strikes me that a good operating system would generalize and unify ALL of 
these into a single, coherent hierarchical named storage system where the 
concept of "directory" and "file" and "member" and "volume" would all be quite 
blurred.  One would want a variety of functions perhaps including:

    Position sequential forward
    Position forward by key
    Position sequential backward
    Position backward by key
    Update last record read
    Append new record (with/without a key)
    Ascend
    Descend
    Search by key
    Seek to specified physical position
    Read current record data

Once one is positioned to the "directory" in question, one would "open a file" 
by "search by key" using the "filename", then "descend" into the file.  One cool 
thing about such a structure is that the concept of "primary and secondary 
records" comes quite naturally... 

There are no shortage of interesting approaches which could be very nice from a 
usage standpoint if one is willing to abandon some of the old concepts.

The fact of eliminating the incredible mess of all these currently competing and 
dissimilar lookup methods and replacing them by a single unified approach would 
go a long way toward a simpler, more robust, and (I think) better system.

Gordon Peterson
http://www.computek.net/public/gep2/
Support the Anti-SPAM Amendment!  Join at http://www.cauce.org/



From icon-group-sender  Tue Jun  2 17:20:14 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id RAA06773
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Tue, 2 Jun 1998 17:20:13 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA06876; Tue, 2 Jun 1998 17:20:06 -0700
Comments: Authenticated sender is <stryx@pop3.demon.co.uk>
From: Iain Alexander <ia@stryx.demon.co.uk>
To: icon-group@optima.CS.Arizona.EDU
Date: Tue, 2 Jun 1998 23:08:33 +0000
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7BIT
Subject: Re: Annoying things
Priority: normal
In-Reply-To: <9805300228.AA0191@valinet.com>
References: <199805231524.KAA05320@segfault.cs.utsa.edu> (message from	Clinton Jeffery on Sat, 23 May 1998 10:24:33 -0500)
X-Mailer: Pegasus Mail for Win32 (v2.54)
Message-Id: <896825323.1016855.0@stryx.demon.co.uk>
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 470

On 29 May 98 at 22:28, Paul Abrahams wrote:
> Given the variances among operating systems, perhaps there ought to be a
> convention that the first space-delimited item in each line produced by
> open(dirname) is the filename.

Unfortunately, space is a valid character in a filename in Win95/NT/Unix.  Of 
course, you could always return a record.

-- 
Iain Alexander               PGP 1024-bit key id B501A0AD
ia@stryx.demon.co.uk   I.Alexander@bsn0106.wins.icl.co.uk


From icon-group-sender  Tue Jun  2 17:20:07 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id RAA06770
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Tue, 2 Jun 1998 17:20:07 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA06570; Tue, 2 Jun 1998 17:20:00 -0700
Date: Tue, 02 Jun 98 17:46:37 -0400
Message-Id: <9806022146.AA0151@valinet.com>
From: Paul Abrahams <abrahams@acm.org>
To: gmt@baskerville.CS.Arizona.EDU
Cc: icon-group@baskerville.CS.Arizona.EDU
In-Reply-To: <9806021619.AA28016@hawk.CS.Arizona.EDU> (message from Gregg
	Townsend on Tue, 2 Jun 1998 09:19:59 -0700)
Subject: Re: Directory access facilities
Reply-To: abrahams@acm.org
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 1015

>>>>> On Tue, 2 Jun 1998 09:19:59 -0700, Gregg Townsend <gmt@baskerville.CS.Arizona.EDU> said:

|Gregg> I'd prefer a simple built-in function that would generate the
|Gregg> names of the files contained in a directory, e.g.  contents(s)
|Gregg> with other mechanisms (if any) used for finding attributes.

I wonder how useful it is to have names of subdirectories intermingled
with names of files.  Gordon Peterson's point about the importance of
traversing the directory tree is worth considering.  For instance,
opening a directory might yield a set (or sequence) containing two kinds
of objects, subdirectories and files.  This is an approach I hadn't
thought about previously.

There have been a lot of elegant approaches to this problem suggested.
My own bias is for whatever is least controversial, on the grounds that
that's the one most likely to be implemented.  Clearly the level of
aesthetic sensitivity in this community is very high, so the risk of
ending up with a kludge is small.

Paul Abrahams





From icon-group-sender  Wed Jun  3 08:17:33 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id IAA00838
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Wed, 3 Jun 1998 08:17:33 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA10605; Wed, 3 Jun 1998 08:17:26 -0700
From: gep2@computek.net
Date: Wed, 3 Jun 1998 00:21:44 -0500
Message-Id: <199806030521.AAA23309@axp.cmpu.net>
Mime-Version: 1.0
Content-Type: text/plain
Content-Transfer-Encoding: 7bit
Subject: Re:  Directory access facilities
To: icon-group@baskerville.CS.Arizona.EDU
X-Mailer: SPRY Mail Version: 04.00.06.17
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 404

> I'd prefer a simple built-in function that would generate the names
of the files contained in a directory, e.g.
	contents(s)
with other mechanisms (if any) used for finding attributes.

Personally, I think a bigger issue is how to handle recursion into lower-level 
subdirectories.

Gordon Peterson
http://www.computek.net/public/gep2/
Support the Anti-SPAM Amendment!  Join at http://www.cauce.org/



From icon-group-sender  Wed Jun  3 08:17:24 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id IAA00835
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Wed, 3 Jun 1998 08:17:23 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA07445; Wed, 3 Jun 1998 08:17:16 -0700
From: "Nevin :-] Liber" <nevin@pendragon-software.com>
To: <icon-group@optima.CS.Arizona.EDU>
Subject: Re: Directory access facilities
Date: Tue, 2 Jun 1998 21:00:33 -0500
Message-Id: <01bd8e93$64a4caa0$c37546cf@NEVIN>
Mime-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: 7bit
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
Content-Length: 1478

From: Gregg Townsend <gmt@baskerville.CS.Arizona.EDU>
To: icon-group@baskerville.CS.Arizona.EDU
<icon-group@baskerville.CS.Arizona.EDU>
Date: Tuesday, June 02, 1998 2:26 PM
Subject: Re: Directory access facilities


>I'd prefer a simple built-in function that would generate the names
>of the files contained in a directory, e.g.
> contents(s)
>with other mechanisms (if any) used for finding attributes.


I agree with you here, Gregg.  Probably need to, as a minimum, add a second
function to determine if a given string refers to a directory name.
Something like:

dir(sDirectory) - generator that returns filenames in directory
string(sDirectory), one name at a time.  No guarantee to the order in which
directory entries are returned.
isdir(sDirectory) - returns sDirectory if string(sDirectory) refers to a
directory; otherwise fails

One might be tempted to use initial failure from dirname() to determine if a
given directory is empty; however, this doesn't work in general, as on some
systems it might not be possible to differentiate between a non-existant
directory and an empty directory.

I would also vote against filtering out "." and ".."; dir() should return an
accurate representation of what is in the directory, and not require any
special casing (and it is trivial to filter this out at a higher level).
__
 Nevin ":-)" Liber  <mailto:nevin@pendragon-software.com>  (847) 816-9926
  Senior Software Engineer, Pendragon Software <http://www.webfayre.com/>



From icon-group-sender  Wed Jun  3 08:17:46 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id IAA00846
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Wed, 3 Jun 1998 08:17:46 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA09051; Wed, 3 Jun 1998 08:17:39 -0700
Message-Id: <s5751c70.061@housmtp.oceaneering.com>
X-Mailer: Novell GroupWise 4.1
Date: Wed, 03 Jun 1998 09:49:31 -0500
From: Charles Hethcoat <CHETHCOA@oss.oceaneering.com>
To: icon-group@optima.CS.Arizona.EDU
Subject: Retrieving directory contents -Reply
Mime-Version: 1.0
Content-Type: text/plain
Content-Disposition: inline
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 2707

"Richard A. O'Keefe" <ok@atlas.otago.ac.nz> said:

|Richard> A directory-with-file-attributes is NOT
a sequence of text
|Richard> lines; it is NOT a sequence, and its
elements are NOT lines of
|Richard> text.

>>> Paul Abrahams <abrahams@acm.org> 1998 Jun 01,
09:45pm >>> said:
>What I care about more
>than anything else is seeing *some*
>system-independent method of
>retrieving the names of the files in a given
>directory.

I think each of these gentlemen makes good points.
File and directory names are a problem, but a
bigger problem is the one of locating stuff (host
machines, servers, users, files, and devices like
printers) in general.

Where I work, we run PCs with Windows of various
flavors, Netware with NDS and bindery emulation,
and NT network with its means of naming things. 
We also have TCP/IP and its addressing methods. 
In addition, I run Linux. Accessing all this from
an Icon program isn't very easy.  I haven't
figured out a good and reasonably universal
answer, and it sounds like a lot of people have
the same problem.

Leaving aside a moment the question of what is a
good way to represent these directories in Icon, I
think the fundamentally correct answer is to find
a consistent, publically available programming
interface to all of these file and directory
systems that were designed by different parties at
different times for different reasons.

So I would ask the Icon group to think about how
all these system naming collisions and conflicts
might be reconciled in a sensible way--so that
programs that access them can work no matter what
system they run on, and in a vendor-independent
manner.  (Netware and Microsoft, maybe others, are
competing to define "the" correct way of doing it,
but who knows whose ox is being gored in the
process?)

I think Linux already has a good answer.  Its
/proc "virtual filesystem" is a clean and
undemanding way to put system-dependent
information up for programmers (meaning their
programs) to access.  The kernel's internal data
certainly isn't all string-based, but /proc makes
it look that way.  It is easy for a relatively
unsophisticated programmer to understand what is
being offered and to access it without having to
know how the internals work.  No language changes
are required to do this; just a one-time hack to
link this hypothetical /proc-like API into the
Icon interpreter and you have it.  Then all you
have to do in your Icon program is open the
appropriate /proc subdirectory and read it using
the existing facilities in the language.

The question is, can the Icon community come up
with this API once for every platform?  (Linux
source is available as a model.)

Charles Hethcoat
chethcoa@oss.oceaneering.com



From icon-group-sender  Wed Jun  3 12:46:07 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id MAA10570
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Wed, 3 Jun 1998 12:46:06 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA09162; Wed, 3 Jun 1998 12:45:59 -0700
From: gep2@computek.net
Date: Wed, 3 Jun 1998 12:45:30 -0500
Message-Id: <199806031745.MAA31997@axp.cmpu.net>
Mime-Version: 1.0
Content-Type: text/plain
Content-Transfer-Encoding: 7bit
Subject: Re: Directory access facilities
To: icon-group@optima.CS.Arizona.EDU
X-Mailer: SPRY Mail Version: 04.00.06.17
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 929

> One might be tempted to use initial failure from dirname() to determine if a
given directory is empty; however, this doesn't work in general, as on some
systems it might not be possible to differentiate between a non-existant
directory and an empty directory.

And how about directories which contain entries which you're not allowed to see?

> I would also vote against filtering out "." and ".."; dir() should return an
accurate representation of what is in the directory, 

Oh, jeez... I don't see ANY value in returning "." at least, whether it's 
"there" (for OS purposes) or not.  What's the point, after all???

> ...and not require any special casing (and it is trivial to filter this out at 
a higher level).

Trivial doesn't mean anything different, here, than "recurring and familiar 
nuisance".

Gordon Peterson
http://www.computek.net/public/gep2/
Support the Anti-SPAM Amendment!  Join at http://www.cauce.org/



From icon-group-sender  Thu Jun  4 08:06:16 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id IAA20231
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Thu, 4 Jun 1998 08:06:15 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA07072; Thu, 4 Jun 1998 08:06:08 -0700
Message-Id: <199806040951.LAA20259@capway.com>
From: "Vladimir Grodzenski" <grodzens@capway.com>
To: gep2@computek.net
Date: Thu, 4 Jun 1998 11:47:32 +0000
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7BIT
Subject: Re: Directory access facilities
Reply-To: <vladimir.grodzenski@capway.com>
Cc: icon-group@optima.CS.Arizona.EDU
Priority: urgent
In-Reply-To: <199806031745.MAA31997@axp.cmpu.net>
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 822

On  3 Jun 98 at 12:45, Gordon Peterson wrote:

> > I would also vote against filtering out "." and ".."; dir() should return an
> > accurate representation of what is in the directory, 
> 
> Oh, jeez... I don't see ANY value in returning "." at least, whether
> it's "there" (for OS purposes) or not.  What's the point, after
> all???

By the way, 
When you open(dirname) in Icon you'll get the list of files and 
directories, including "." and ".." 

The 'dirname' may be any valid "d:\\dir" string except the root!!! 
However open("d:\.") will provide the list in the root directory. All 
this is logical but far from being elegant :-)

Vladimir Grodzenski
=================================================
E-mail: vladimir.grodzenski@capway.com
CompuServe: 100700,526
=================================================


From icon-group-sender  Thu Jun  4 08:05:59 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id IAA20204
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Thu, 4 Jun 1998 08:05:59 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA10470; Thu, 4 Jun 1998 08:05:52 -0700
Message-Id: <35766713.6286@gte.net>
Date: Thu, 04 Jun 1998 04:21:23 -0500
From: MJE <evans@gte.net>
Reply-To: evans@gte.net
Organization: None
X-Mailer: Mozilla 3.01 (Win95; I)
Mime-Version: 1.0
To: icon-group@optima.CS.Arizona.EDU
Subject: Re: Directory access facilities
References: <01bd8e93$64a4caa0$c37546cf@NEVIN>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 223

I haven't followed this thread, but whatever you guys decide, make sure
that we can get a list of filenames that match some filtering criteria,
based on partial name, file type, modification date, or anything else.

Mark



From icon-group-sender  Thu Jun  4 08:05:51 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id IAA20190
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Thu, 4 Jun 1998 08:05:50 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA11122; Thu, 4 Jun 1998 08:05:42 -0700
Date: Thu, 4 Jun 1998 14:20:21 +1200 (NZST)
From: "Richard A. O'Keefe" <ok@atlas.otago.ac.nz>
Message-Id: <199806040220.OAA29849@atlas.otago.ac.nz>
To: icon-group@optima.CS.Arizona.EDU, nevin@pendragon-software.com
Subject: Re: Directory access facilities
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 5259

In reponse to messages like 

	>I'd prefer a simple built-in function that would generate the names
	>of the files contained in a directory, e.g.
	> contents(s)
	>with other mechanisms (if any) used for finding attributes.
	
and

	I agree with you here, Gregg.  Probably need to, as a minimum, add a second
	function to determine if a given string refers to a directory name.
	Something like:
	
I'd like to say that people seem to have missed the point of my reference
to Tcl/Tk.  The point is that there is an *existing* *multiplatform*
implementation of this idea that *works*.  We don't have to design.
We can steal.

The *last* thing we have any shadow of an excuse for doing is bodging
together yet another incompatible design by appealing to "preferences".

Of course Tcl/Tk isn't the only thing around that one might reasonably
copy.  There's the POSIX 'glob' function, for example, which is given
a pathname with wilcards and some options, and returns an array of
strings (in which any character whatsoever except \0 may occur).  One
could have a very nice interface _based_ on glob(3c) + stat(2); any
platform offering FindFirst/FindNext can emulate glob() easily enough.

Nevin Liber wrote:

	I would also vote against filtering out "." and ".."; dir() should return an
	accurate representation of what is in the directory, and not require any
	special casing (and it is trivial to filter this out at a higher level).

But where did you get the idea that the presence of "." and ".." was 'an
accurate representation of what is in the directory'?  There are POSIX
implementations that go to some trouble to fake these entries because the
_real_ directory format _doesn't_ have them.  Do you want to filter them
out when you know they've been faked and include them when you know they
haven't?  It's better if they are _always_ present or _always_ absent
whatever is really on the disc.  Note that the POSIX.1 standard (ok, so I
have an old copy) says very clearly and explicitly

	5.1.2.2
	43  The readdir() function shall not return directory entries
	    containing empty names.
===>	44  It is unspecified whether entries are returned for
	    dot or dot-dot.

	B5.1.2
	2956 The directory entries for dot and dot-dot are optional.
	     POSIX does not provide a
	2957 way to test a priori for their existence because
	     an application that is portable
	2958 must be written to look for (and usually ignore)
	     those entries.

There really are "POSIX" systems where those things AREN'T on the disk.
And of course there are non-POSIX systems that don't use "." or ".."
for anything particular.  Indeed, there's one file system I retain
considerable fondness for where the directory namespace and the file
namespace are disjoint, so that the same name _can_ be used for a directory
and a file at the same time, and often is.  E.g.
	(MAT212X)THESIS/PLOTTER/SOURCE ON USERPACK	=> Algol source form
	(MAT212X)THESIS/PLOTTER ON USERPACK     	=> executable code
where .../PLOTTER is a file name and .../PLOTTER/ is a directory name.
To the best of my knowledge there isn't any Icon implementation for that
machine, but there's a standard C compiler and a near-enough-to-POSIX
OS interface, so there _could_ be an Icon implementation.

This _does_ count as a criticism of 'imitate Tcl/Tk' and 'imitate POSIX'
because it strongly suggests that separate functions for finding
directories and finding files would be appropriate.  (That's the way I
did it for Quintus Prolog, with
	file_member_of_directory(Directory[, Pattern], File)
	directory_member_of_directory(Directory[, Pattern], SubDir)
	file_property(File, Property, Value)
	directory_property(Directory, Property, Value)
operations.  Mind you, that had to be portable across MacOS, MS-DOS,
Unix, VMS, VM/CMS, MVS, and Xerox Lisp (XNS) so it was a bit more
portable than Icon may need to be.)  As I've hinted above, we could
patch around this particular problem by insisting that directory names
always end with a (native) separator and file names not; this will work
straight out of the box on many UNIX systems where dirname/ is equivalent
to dirname/., but can reasonably be faked anywhere, so a single interface
will do.

One particular point about citing Tcl/Tk is the file license.terms in
the Tcl8.0 distribution, which says amongst other things

	The authors hereby grant permission to use, copy, modify,
	distribute, and license this software and its documentation
	for any purpose, provided that existing copyright notices
	are retained in all copies and that this notice is included
	verbatim in any distributions.  No written agreement, license,
	or royalty fee is required for any of the authorized uses.
	Modifications to this software may be copyrighted by their
	authors and need not follow the licensing terms described
	here, provided that the new terms are clearly indicated on
	the first pagev of each file where they apply.

What this means is that if the Icon interface is based on the Tcl/Tk
interface, two important benefits follows:

    - a multiplatform interface which is *known* to be usable will
      be available for very little design effort
    - a multiplatform *tested* implementation will be available *free*.

I must ask:  why reinvent what we can steal?



From icon-group-sender  Thu Jun  4 12:30:25 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id MAA06878
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Thu, 4 Jun 1998 12:30:25 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA12324; Thu, 4 Jun 1998 12:30:18 -0700
From: "Nevin :-] Liber" <nevin@pendragon-software.com>
To: <icon-group@optima.CS.Arizona.EDU>
Subject: Re: Directory access facilities
Date: Thu, 4 Jun 1998 12:47:30 -0500
Message-Id: <01bd8fe0$d860a2e0$c37546cf@NEVIN>
Mime-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: 7bit
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
Content-Length: 5285

-----Original Message-----
From: Richard A. O'Keefe <ok@atlas.otago.ac.nz>
To: icon-group@optima.CS.Arizona.EDU <icon-group@optima.CS.Arizona.EDU>;
nevin@pendragon-software.com <nevin@pendragon-software.com>
Date: Thursday, June 04, 1998 10:11 AM
Subject: Re: Directory access facilities


Let me preface this by saying that I know very little about Tcl/Tk;
basically, playing around with "tclsh" and checking out the man pages at
<http://cuiwww.unige.ch/eao/www/TclTkMan/Man.html>.

>I'd like to say that people seem to have missed the point of my reference
>to Tcl/Tk.  The point is that there is an *existing* *multiplatform*
>implementation of this idea that *works*.  We don't have to design.
>We can steal.

If it is the appropriate abstractions for the language at hand.  I don't
believe it is.

>The *last* thing we have any shadow of an excuse for doing is bodging
>together yet another incompatible design by appealing to "preferences".

My "preferences" come from using the language for the last 14 years, and I
suspect that Gregg Townsend has been using it even longer than I have.  I
think I have a pretty good sense of what fits in the language and what
doesn't.

It appears that Tcl/Tk is a language where everything is a string.  For
instance, "glob" returns a string of encoded filenames; one has to encode
some the filenames, because they may contain embedded seperators, embedded
encoding characters, etc.

In a shell this works, since a shell parses each and every command line, and
provides facilities for dividing a delimiter-seperated string into seperate
strings.

Icon does not do this, nor does it have builtin facilities for "unquoting"
an encoded filename.  In order to use this, there would have to be a
higher-level function that breaks up the string, gets rid of any quoting
characters that aren't part of the filename, and either returns a list of,
or more likely, generate a list of filenames.  Considering the lower level
system already keeps filenames as seperate entities, it seems that we are
doing a lot of work for nothing.

Now, one might argue that "glob" allows regular expressions for filtering.
However, regular expressions aren't used anywhere else in Icon; it would
look like a wart on the language.

What I am looking for is minimal functionality that will get me the
filenames broken up one filename at a time, since it is easy to build the
more complex stuff on top of that from within Icon.  Gregg's suggestion of a
generator that does just that meets that criterion nicely.

>But where did you get the idea that the presence of "." and ".." was 'an
>accurate representation of what is in the directory'?  There are POSIX
>implementations that go to some trouble to fake these entries because the
>_real_ directory format _doesn't_ have them.

And on "real" systems, "." and ".." don't necessarily do what you expect.
For instance, most shells interpret them differently.  Do a "cd
/foo/bar/../baz", and the shell would put you in "/foo/baz", yet if you
actually traversed the directories, you could be in "/baz", if "/foo/bar"
was a symbolic link to "/", for instance.  It's hard to encapsulate all of
this functionality without providing a full filesystem API.

Whatever the lower-level C functions return is what the Icon equivalent
should generate.


>To the best of my knowledge there isn't any Icon implementation for that
>machine, but there's a standard C compiler and a near-enough-to-POSIX
>OS interface, so there _could_ be an Icon implementation.

I don't believe POSIX is a requirement of Icon right now.


>As I've hinted above, we could
>patch around this particular problem by insisting that directory names
>always end with a (native) separator and file names not

But nothing else in Icon follows this convention.  Strings don't have
special embedded terminators; lists don't have special embedded terminators,
etc.  If you add a seperator or directory marker, you have to add a way of
quoting the seperator or directory marker (since a file name might
legitimately end with the directory marker) and functions to remove the
quotes.  This is much messier than having a generator suspend filenames one
at a time, and a seperate function to test if it is a directory.  Also, what
if you are creating a directory name somewhere else in your program?  You
have to have a function that turns a string into a directory name.

To use EE terms, I much prefer out-of-band signaling to in-band signaling.
Since we can get filenames already seperated (out-of-band signaling, since
something outside of the string itself indicates that this object represents
a directory), it is cleaner than having to parse a string (in-band
signaling, since whether or not it is a directory is encoded in the string)
when I just want to get the filename out of it.

Now, I do understand your point about systems where directories are in
seperate namespaces.  The cleanest solution would be to have two seperate
generators; one which returns regular files, and one which returns
directories.  I don't know if this is worth the effort, since so few systems
out there have seperate namespaces.
__
 Nevin ":-)" Liber  <mailto:nevin@pendragon-software.com>  (847) 816-9926
  Senior Software Engineer, Pendragon Software <http://www.webfayre.com/>



From icon-group-sender  Thu Jun  4 12:30:11 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id MAA06870
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Thu, 4 Jun 1998 12:30:10 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA10387; Thu, 4 Jun 1998 12:30:03 -0700
Date: Thu, 4 Jun 1998 11:38:20 -0400 (EDT)
From: "Christopher W. Milner" <cwm2n@cs.virginia.edu>
To: icon-group@optima.CS.Arizona.EDU
Subject: jcon question
In-Reply-To: <35766713.6286@gte.net>
Message-Id: <Pine.SOL.3.91.980604113239.5771C-100000@cobra.cs.virginia.edu>
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 498

I hope this is the correct forum for this type of question...

I am fairly innocent of the ways of Java. Is there a way to compile an 
icon program using jcont such that it will become an applet? Assuming 
this can be done, what sort of HTML incantation is required?

EXAMPLE:
go from the following icon program:

procedure main()
write("Hello, Madge")
end
 
to an applet and html file.

Many thanks,

Chris Milner
Graduate Research Assistant
Department of Computer Science
University of Virginia


From icon-group-sender  Thu Jun  4 12:30:17 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id MAA06873
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Thu, 4 Jun 1998 12:30:17 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA28252; Thu, 4 Jun 1998 12:30:10 -0700
From: gep2@computek.net
Date: Thu, 4 Jun 1998 10:54:24 -0500
Message-Id: <199806041554.KAA20858@axp.cmpu.net>
Mime-Version: 1.0
Content-Type: text/plain
Content-Transfer-Encoding: 7bit
Subject: Re: Directory access facilities
To: icon-group@optima.CS.Arizona.EDU
X-Mailer: SPRY Mail Version: 04.00.06.17
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 529

> I haven't followed this thread, but whatever you guys decide, make sure
that we can get a list of filenames that match some filtering criteria,
based on partial name, file type, modification date, or anything else.

I'll be happy if I can get a comprehensive list (INCLUDING subdirectories), I'll 
gladly do the other filtering (that's more likely to be application specific, 
anyhow) at the higher levels.

Gordon Peterson
http://www.computek.net/public/gep2/
Support the Anti-SPAM Amendment!  Join at http://www.cauce.org/



From icon-group-sender  Thu Jun  4 16:21:49 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id QAA18384
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Thu, 4 Jun 1998 16:21:49 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA10228; Thu, 4 Jun 1998 16:21:41 -0700
Message-Id: <3576F8AF.611F@gte.net>
Date: Thu, 04 Jun 1998 14:42:39 -0500
From: MJE <evans@gte.net>
Reply-To: evans@gte.net
Organization: None
X-Mailer: Mozilla 3.01 (Win95; I)
Mime-Version: 1.0
To: icon-group@optima.CS.Arizona.EDU
Subject: Re: Directory access facilities
References: <199806041554.KAA20858@axp.cmpu.net>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 872

gep2@computek.net wrote:
> 
> > I haven't followed this thread, but whatever you guys decide, make sure
> that we can get a list of filenames that match some filtering criteria,
> based on partial name, file type, modification date, or anything else.
> 
> I'll be happy if I can get a comprehensive list (INCLUDING subdirectories), I'll
> gladly do the other filtering (that's more likely to be application specific,
> anyhow) at the higher levels.
> 
> Gordon Peterson
> http://www.computek.net/public/gep2/
> Support the Anti-SPAM Amendment!  Join at http://www.cauce.org/


If you want to make work for yourself, OK, but I want calls to support
such filtering so that I don't have to write the code.  That's why I put
this note up.  I've done it just as you suggest, and I don't like it.

The filters chosen by the programmer are the application-specific part.

Mark



From icon-group-sender  Fri Jun  5 08:38:44 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id IAA18863
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Fri, 5 Jun 1998 08:38:43 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA12493; Fri, 5 Jun 1998 08:38:36 -0700
Date: Thu, 4 Jun 1998 17:33:41 -0700
From: Gregg Townsend <gmt>
Message-Id: <9806050033.AA02697@hawk.CS.Arizona.EDU>
To: cwm2n@cs.virginia.edu, icon-group
Subject: Re:  jcon question
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 698

    From icon-group-sender@baskerville.CS.Arizona.EDU Thu Jun  4 12:30:56 1998
    
    Is there a way to compile an 
    icon program using jcont such that it will become an applet?

No.  Jcon creates stand-alone Java programs and there is no
interface to the applet system or provision for creating applets.

   ---------------------------------------------------------------------------
   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 Jun  5 08:38:51 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id IAA18866
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Fri, 5 Jun 1998 08:38:50 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA12425; Fri, 5 Jun 1998 08:38:43 -0700
From: gep2@computek.net
Date: Thu, 4 Jun 1998 21:32:01 -0500
Message-Id: <199806050232.VAA10433@axp.cmpu.net>
Mime-Version: 1.0
Content-Type: text/plain
Content-Transfer-Encoding: 7bit
Subject: Re: Directory access facilities
To: icon-group@optima.CS.Arizona.EDU
X-Mailer: SPRY Mail Version: 04.00.06.17
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 1701

>>> I haven't followed this thread, but whatever you guys decide, make sure
> that we can get a list of filenames that match some filtering criteria,
> based on partial name, file type, modification date, or anything else.

>> I'll be happy if I can get a comprehensive list (INCLUDING subdirectories), 
I'll gladly do the other filtering (that's more likely to be application 
specific, anyhow) at the higher levels.

> If you want to make work for yourself, OK, but I want calls to support
such filtering so that I don't have to write the code.  

The problem with such things is that the "based on...or anything else" usually 
ends up making the filter parameterization so darned complicated that you end up 
always having to look it all up... there are so many possible tests (like "I 
want files with .dbf extensions, modified since April 30th 1997, with size 
greater than 5Mb but less than 10Mb... but only if the Accounting department has 
write permission for them...").

The thing I want to avoid is having to put OTHER tests in there just to exclude 
dumb stuff (like the basically useless "." subdirectory entry, and probably the 
".." one too) which you basically never need to see or use anyhow.  And I'd like 
to be able to avoid having to write (each time) all the recursion code to 
traverse the subdirectory tree... that's the kind of fiddly stuff that I'd like 
to see built in.

The kind of stuff that you're asking for can probably be built rather trivially 
at the user/application level, and without adding much to the code that they'd 
have to have anyhow.

Gordon Peterson
http://www.computek.net/public/gep2/
Support the Anti-SPAM Amendment!  Join at http://www.cauce.org/



From icon-group-sender  Fri Jun  5 12:55:16 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id MAA27539
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Fri, 5 Jun 1998 12:55:15 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA13632; Fri, 5 Jun 1998 12:55:08 -0700
Message-Id: <35782CAD.4F7E@gte.net>
Date: Fri, 05 Jun 1998 12:36:45 -0500
From: MJE <evans@gte.net>
Reply-To: evans@gte.net
Organization: None
X-Mailer: Mozilla 3.01 (Win95; I)
Mime-Version: 1.0
To: icon-group@optima.CS.Arizona.EDU
Subject: Re: Directory access facilities
References: <199806050232.VAA10433@axp.cmpu.net>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 261

Presumably one could ask for all files with no filters and get what you
want.  You and I have different opinions about what is trivial at the
application level.  I will put in my filters, and you can leave yours
out to do the filtering inside your app.

Mark



From icon-group-sender  Mon Jun  8 08:18:24 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id IAA19526
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Mon, 8 Jun 1998 08:18:22 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA13889; Mon, 8 Jun 1998 08:18:15 -0700
Message-Id: <199806071520.RAA01123@capway.com>
From: "Vladimir Grodzenski" <grodzens@capway.com>
To: evans@gte.net
Date: Sun, 7 Jun 1998 17:17:02 +0000
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7BIT
Subject: Re: Directory access facilities
Reply-To: <vladimir.grodzenski@capway.com>
Cc: icon-group@optima.CS.Arizona.EDU
Priority: urgent
In-Reply-To: <35782CAD.4F7E@gte.net>
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 635

On  5 Jun 98 at 12:36, MJE <evans@gte.net> wrote:

> Presumably one could ask for all files with no filters and get what
> you want.  You and I have different opinions about what is trivial
> at the application level.  I will put in my filters, and you can
> leave yours out to do the filtering inside your app.


procedure dirlist(path, recurse_flag, filter)
  return list of subdirectories
end

procedure filelist(-idem-)
  return list of files
end

Vladimir Grodzenski
=================================================
E-mail: vladimir.grodzenski@capway.com
CompuServe: 100700,526
=================================================


From icon-group-sender  Mon Jun  8 13:38:51 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id NAA14876
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Mon, 8 Jun 1998 13:38:51 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA16084; Mon, 8 Jun 1998 13:38:43 -0700
Date: Mon, 8 Jun 1998 08:34:15 -0700
From: Ralph Griswold <ralph>
Message-Id: <9806081534.AA20674@jupiter.CS.Arizona.EDU>
To: icon-group
Subject: Graphics programming book
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 348

The recently published book "Graphics Programming in Icon" is now
available from the Icon Project.

The price is $45, which includes parcel post shipping to addresses
in the United States, Canada, and Mexico.  Air parcel post shipping
to other countries is $18 additional.

For information about the book, see

	http://www.cs.arizona.edu/icon/gb/


From icon-group-sender  Tue Jun  9 08:29:15 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id IAA00493
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Tue, 9 Jun 1998 08:29:14 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA15308; Tue, 9 Jun 1998 08:29:06 -0700
Date: Tue, 9 Jun 1998 13:27:10 +1200 (NZST)
From: "Richard A. O'Keefe" <ok@atlas.otago.ac.nz>
Message-Id: <199806090127.NAA12029@atlas.otago.ac.nz>
To: evans@gte.net, vladimir.grodzenski@capway.com
Subject: Re: Directory access facilities
Cc: icon-group@optima.CS.Arizona.EDU
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 987

	procedure dirlist(path, recurse_flag, filter)
	  return list of subdirectories
	end
	
What's a recurse flag?  If it's just true/false, it's no use.  How do
you control whether mount points are crossed?  Whether symbolic links
(aliases in MacSpeak) are followed?  How cycles (possible in the
presence of aliases) are to be handled?  Is the list depth first,
breadth first, or alphabetic (ascii or by locale)?  What are the
arguments of the filter and what can it request?  What about directories
that are known to be directories but cannot be read?  What about
symbolic links whose target has disappeared?

	procedure filelist(-idem-)
	  return list of files
	end
	
Ditto.

People could at least take a serious look at ftw()/nftw().
Other people have had three-quarter baked ideas; no need
for us to half-bake our own.  nftw() exists because ftw(),
which is richer than this proposal, just plain wasn't good enough.

If people won't buy "copy tcl", will they buy "look hard at nftw()?"


From icon-group-sender  Tue Jun  9 17:37:20 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id RAA04371
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Tue, 9 Jun 1998 17:37:20 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA16062; Tue, 9 Jun 1998 17:37:12 -0700
Date: Tue, 9 Jun 1998 17:36:07 -0500 (CDT)
From: Chris Tenaglia <cdt@post.its.mcw.edu>
To: icon-group@optima.CS.Arizona.EDU
Subject: Icon for CGI
Message-Id: <Pine.SOL.3.96.980609172809.18527A-100000@post.its.mcw.edu>
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 991

Been trying to experiment with Icon V9 as a cgi engine.
wrote a little script that generates a value hmtl page
to stdout. It's called by an input form. I get the
following message from Netscape :
_______________________________________________________
Server Error

This server has encountered an internal error which prevents it from
fulfilling your request. The most likely cause is a misconfiguration.
Please ask the administrator to look for messages in the server's error
log.
_______________________________________________________

The chmod for executable and iconx and icont are 755.
icont works fine to compile, the telnet iconx runs
interactively. What am I missing to get the CGI to
run it? Thanks in advance

 
Chris Tenaglia   (system manager)     |  The future foretold,
Medical College of Wisconsin          |  The past explained,
8701 W. Watertown Plank Rd.           |  The present largely appologized for.
Milwaukee, WI 53226   (414)456-8765   |  Organon to the Doctor




From icon-group-sender  Wed Jun 10 08:14:31 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id IAA15048
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Wed, 10 Jun 1998 08:14:31 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA08142; Wed, 10 Jun 1998 08:14:23 -0700
From: gep2@computek.net
Date: Wed, 10 Jun 1998 09:01:52 -0500
Message-Id: <199806101401.JAA22815@axp.cmpu.net>
Mime-Version: 1.0
Content-Type: text/plain
Content-Transfer-Encoding: 7bit
Subject: Re: Directory access facilities
To: icon-group@optima.CS.Arizona.EDU
X-Mailer: SPRY Mail Version: 04.00.06.17
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 2143

> What's a recurse flag?  If it's just true/false, it's no use.  How do
you control whether mount points are crossed?  Whether symbolic links
(aliases in MacSpeak) are followed?  How cycles (possible in the
presence of aliases) are to be handled?  Is the list depth first,
breadth first, or alphabetic (ascii or by locale)?  What are the
arguments of the filter and what can it request?  What about directories
that are known to be directories but cannot be read?  What about
symbolic links whose target has disappeared?

> People could at least take a serious look at ftw()/nftw().
Other people have had three-quarter baked ideas; no need
for us to half-bake our own.  nftw() exists because ftw(),
which is richer than this proposal, just plain wasn't good enough.

...and this kind of stuff is an excellent example of why so many programmers 
just say "screw it!" and end up writing code that's appropriate to the OS they 
have to support... for 95% of us, that's Win95/NT/DOS.  

Frankly, if I have to go through all these dumb gyrations to support (in theory) 
some BS that I don't have to worry about anyhow, then it's easier to just solve 
the problem in some other way.

Example:  I don't have to worry about aliases and file structure closed loops.  
I don't care in the slightest if my program won't ever run on a Mac!  Sure, 
there are other markets (like Unix) but I'll gladly give those niche markets to 
other people, if I can have all the rest.  :-)

Hey, once you get the list, you can sort it (if you care) any way you like.  IF 
that matters to you.

As for the filter, the ultimate filter is to process the list elements using 
string scanning (I agree with the previous poster that it would be inappropriate 
to use a reg-ex style filter expression here).

What I'm concerned about is getting the basic data.  Once that's done, I'm 
confident I can do what I need to with it.  I don't so much like the system 
trying to second-guess me and trying to provide me (in advance) with every 
possible variant.

Gordon Peterson
http://www.computek.net/public/gep2/
Support the Anti-SPAM Amendment!  Join at http://www.cauce.org/



From icon-group-sender  Thu Jun 11 13:21:58 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id NAA13161
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Thu, 11 Jun 1998 13:21:54 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA18280; Thu, 11 Jun 1998 13:21:46 -0700
Message-Id: <3.0.5.32.19980611115520.00a2fc20@post.its.mcw.edu>
X-Sender: cdt@post.its.mcw.edu
X-Mailer: QUALCOMM Windows Eudora Light Version 3.0.5 (32)
Date: Thu, 11 Jun 1998 11:55:20
To: icon-group@optima.CS.Arizona.EDU
From: Chris Tenaglia <cdt@post.its.mcw.edu>
Subject: More CGI background
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 1523

I was having no progress. Here's a little more detail.
I found other resources on the web, even a teacher's
slide set. Seems easy in that form. Any programming
language should work, so I decided to try a simple
shell script like "testme" in the cgi directory.
----------------------------------------------------
#!/bin/csh
echo "Content type: text/html"
echo "<HTML>"
echo "<HEAD>"
echo "<TITLE>This is the title</TITLE>"
echo "</HEAD>"
echo "<BODY>"
echo "<H1>Beginning Page</H1>"
echo "<P>"
echo "Here is some free form text"
echo "</BODY>"
echo "</HTML>"
---------------------------------------------------

Which is called by this html page:
---------------------------------------------------
<HTML>
<HEAD>
<TITLE>This is the CGI testing Title</TITLE>
</HEAD>
<BODY>
<BODY BGCOLOR=#FFFFFF>
<Form action="http://www.mcw.edu/cgi-bin/bioet/testme" method=post>

<H1>Mondo Header</H1>
<P>
Here's a crazy TEST web page!
<P>
<select name="BINGO">
  <option selected>
  <option>Numero Uno
  <option>Numero Dos
  <option>Numero Tres
</select>
<P>
<input type=Submit value="DON'T PUSH THIS BUTTON">
</FORM>
</BODY>
</HTML>
---------------------------------------------------------
Still it sits for a few minutes and I get an internal server error message.
Put icont and iconx in the cgi-bin directory and did the patchstr.
Recompiled and the icode looks good and runs fine from telnet.
I have some other things I am trying. I hope to post the result
once I get past this hurdle. Thanks to all who responded so far.
Chris.




From icon-group-sender  Fri Jun 12 07:53:48 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id HAA25541
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Fri, 12 Jun 1998 07:53:48 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA16995; Fri, 12 Jun 1998 07:53:39 -0700
To: icon-group@optima.CS.Arizona.EDU
Date: Fri, 12 Jun 1998 04:21:46 GMT
From: mark@IcanFixYourEmail.com (Mark McWiggins)
Message-Id: <3580ac6c.321131@news.supernews.com>
Organization: The World's Usenet -- http://www.Supernews.com
Sender: icon-group-request@optima.CS.Arizona.EDU
Subject: NT "console application" version?
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 324

Hi --

Anybody have a binary for an NT "console application" version of Icon?
A client of mine is using an Icon-based app in  a Visual C++ "wizard",
and he doesn't want the extra windows popping up on every run.

Thanks in advance.

--
Mark McWiggins
mark@IcanFixYourEmail.com
+1 425 957 0848 (voice)
+1 425 957 7654 (fax)


From icon-group-sender  Fri Jun 12 12:59:26 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id MAA10316
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Fri, 12 Jun 1998 12:59:25 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA06248; Fri, 12 Jun 1998 12:59:17 -0700
Date: Fri, 12 Jun 1998 09:59:47 -0500
Message-Id: <199806121459.JAA18915@segfault.cs.utsa.edu>
From: Clinton Jeffery <jeffery@segfault.cs.utsa.edu>
To: mark@IcanFixYourEmail.com
Cc: icon-group@optima.CS.Arizona.EDU
In-Reply-To: <3580ac6c.321131@news.supernews.com> (mark@IcanFixYourEmail.com)
Subject: Re: NT "console application" version?
Reply-To: jeffery@cs.utsa.edu
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 610


>   Anybody have a binary for an NT "console application" version of Icon?

Windows Icon comes with a binary nticont.exe that produces console
applications based on nticonx.exe.  At the moment, you have to use
them from a command line (or a makefile or whatever).  I'll save this
note as a request to add support for nticont into the Wi program; it
will be trivial to do this in an ugly way, and more work (but easy)
to do it in a way that looks good.

Clint Jeffery, jeffery@cs.utsa.edu
Division of Computer Science, The University of Texas at San Antonio
Research http://www.cs.utsa.edu/research/plss.html


From icon-group-sender  Fri Jun 12 13:00:45 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id NAA10438
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Fri, 12 Jun 1998 13:00:44 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA09724; Fri, 12 Jun 1998 13:00:36 -0700
Message-Id: <3.0.5.32.19980612110154.00a1de00@post.its.mcw.edu>
X-Sender: cdt@post.its.mcw.edu
X-Mailer: QUALCOMM Windows Eudora Light Version 3.0.5 (32)
Date: Fri, 12 Jun 1998 11:01:54 -0500
To: icon-group@optima.CS.Arizona.EDU
From: Chris Tenaglia <cdt@post.its.mcw.edu>
Subject: cgi progress
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 1071

Many thanks again to those pointing out typos
and nuances. The next hurdle has to do with the
FORM

<Form action="http://www.mcw.edu/cgi-bin/bioet/iconx?docgi" method=post>

This works. icont actually generates a shell script
with a binary icode attachment. For some reason
the http server did not like that script. It looks like:
-------------------------------------------------------------------------
#!/bin/sh

case $# in
   0) exec "${ICONX-/db1/local/httpd/cgi-bin/bioet/iconx}" "$0";;
   *) exec "${ICONX-/db1/local/httpd/cgi-bin/bioet/iconx}" "$0" "$@";;
esac

[executable Icon binary follows]
...
-------------------------------------------------------------------------
But iconx can skip over the script and run the
icode. Nifty.

So next comes passing parameters. I now the form
should be using name/value pairs.

<Form method=post
action="http://www.mcw.edu/cgi-bin/bioet/iconx?docgi=1&value1=name1">

doesn't seem to work. Unfortunately they've removed the c compilers
so I can't generate a compiled executable via iconc. Any other tips?
Regards,
Chris.



From icon-group-sender  Mon Jun 15 08:24:23 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id IAA17281
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Mon, 15 Jun 1998 08:24:14 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA22228; Mon, 15 Jun 1998 08:24:05 -0700
Date: Sat, 13 Jun 1998 01:55:45 +0100
Message-Id: <199806130055.BAA12301@gairsay.aiai.ed.ac.uk>
From: Jeff Dalton <jeff@aiai.ed.ac.uk>
Subject: Re: More CGI background
To: Chris Tenaglia <cdt@post.its.mcw.edu>, icon-group@optima.CS.Arizona.EDU
In-Reply-To: Chris Tenaglia's message of Thu, 11 Jun 1998 11:55:20
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 502

> I was having no progress. Here's a little more detail.
> I found other resources on the web, even a teacher's
> slide set. Seems easy in that form. Any programming
> language should work, so I decided to try a simple
> shell script like "testme" in the cgi directory.
> ----------------------------------------------------
> #!/bin/csh
> echo "Content type: text/html"
> echo "<HTML>"

You need to echo a blank line after the headers, that is (in this
case), after the "Content type" line.

-- jeff


From icon-group-sender  Mon Jun 15 08:24:33 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id IAA17293
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Mon, 15 Jun 1998 08:24:32 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA18164; Mon, 15 Jun 1998 08:24:23 -0700
From: shamim@Synopsys.COM (Shamim Zvonko Mohamed)
Message-Id: <9806150114.AA18119@bermuda.synopsys.com>
Subject: Re: Directory access facilities
To: jeffery@cs.utsa.edu
Date: Sun, 14 Jun 1998 18:14:18 -0700 (PDT)
Cc: ok@atlas.otago.ac.nz, abrahams@acm.org, icon-group@optima.CS.Arizona.EDU
In-Reply-To: <199806020443.XAA08820@segfault.cs.utsa.edu> from "Clinton Jeffery" at Jun 1, 98 11:43:37 pm
X-Mailer: ELM [version 2.4 PL24]
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 1112

Clint writes:

> Anyone else have ideas (or votes) on how to do directory access
> facilities?

My admittedly biased and possibly Unix-centric view: we don't need any
elaborate solutions, I think the idea of each read() returning one entry
from the directory is the best one. This finesses all the issues of quoting
blanks, quotes, wildchars etc. etc. If you need more information about the
entry, you use the function that performs this query. In Unicon, you use
stat(), which returns a record. Here's the "ls" example:

   f := open(".") | stop( ... )
   while write( format( stat( read(f) ) ) )

(For clarity I left out some pretty features like sorting the names, and
that you'd want to use lstat instead of stat, etc. Look at the Unicon docs
for more info.)

Someone wanted a directory to be like a table. Yes, a directory is a
"lookup" type of object but not on all systems. We should stick to a clean
and simple interface that will work unambiguously without all sorts of
special cases. Others have already pointed out why a space-separated
string of all the entries is a bad idea.

-s

Shamim Mohamed




From icon-group-sender  Mon Jun 15 08:25:04 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id IAA17367
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Mon, 15 Jun 1998 08:25:04 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA20242; Mon, 15 Jun 1998 08:24:55 -0700
Message-Id: <199806151304.PAA12785@capway.com>
From: "Vladimir Grodzenski" <grodzens@capway.com>
To: mark@IcanFixYourEmail.com (Mark McWiggins)
Date: Mon, 15 Jun 1998 15:00:11 +0000
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7BIT
Subject: Re: NT "console application" version?
Reply-To: <vladimir.grodzenski@capway.com>
Cc: icon-group@optima.CS.Arizona.EDU
Priority: urgent
In-Reply-To: <3580ac6c.321131@news.supernews.com>
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 805

On 12 Jun 98 at 4:21, you wrote:

> Anybody have a binary for an NT "console application" version of
> Icon? A client of mine is using an Icon-based app in  a Visual C++
> "wizard", and he doesn't want the extra windows popping up on every
> run.

If your customer doesn't want to have windows popping up, tell him 
not to open windows in his program, that's it. The Windows Icon does 
not oblige you to open windows if you don't want to!  :-)))

OTOH, the console application means a console will be opened. Console 
is a window, isn't it?  In my opinion, you have to use the Windows 
Icon instead of "nt console" Icon.

Vladimir Grodzenski
=================================================
E-mail: vladimir.grodzenski@capway.com
CompuServe: 100700,526
=================================================


From icon-group-sender  Mon Jun 15 08:24:43 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id IAA17299
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Mon, 15 Jun 1998 08:24:43 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA13109; Mon, 15 Jun 1998 08:24:34 -0700
From: shamim@Synopsys.COM (Shamim Zvonko Mohamed)
Message-Id: <9806150137.AA18187@bermuda.synopsys.com>
Subject: Re: Directory access facilities
To: gep2@computek.net, icon-group@optima.CS.Arizona.EDU
Date: Sun, 14 Jun 1998 18:37:56 -0700 (PDT)
In-Reply-To: <199806031745.MAA31997@axp.cmpu.net> from "gep2@computek.net" at Jun 3, 98 12:45:30 pm
X-Mailer: ELM [version 2.4 PL24]
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 891

Gregg writes:
> I'd prefer a simple built-in function that would generate the names
> of the files contained in a directory

Gordon Peterson:
> Personally, I think a bigger issue is how to handle recursion into
> lower-level subdirectories.

Why? The obvious approach works:

procedure ls(s)
   local dir
   dir := open(s)
   while file := read(dir) do {
      if file == "." | ".." then next
      r := stat(file)
      if r.mode[1] == "d" then
         ls(s || delim || file)
   }
end

(I should point out that my view is Unix-centric so I think in terms of
entries like . and .. and queries called stat(); there are likely to be
similar things in other systems. Mutatis mutandis.)

Gordon Peterson:
> Oh, jeez... I don't see ANY value in returning "." at least, whether it's 
> "there" (for OS purposes) or not.  What's the point, after all???

Well, I might want to implement "ls"!

-s


From icon-group-sender  Mon Jun 15 16:24:49 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id QAA18510
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Mon, 15 Jun 1998 16:24:49 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA19502; Mon, 15 Jun 1998 16:24:40 -0700
From: gep2@computek.net
Date: Mon, 15 Jun 1998 14:52:04 -0500
Message-Id: <199806151952.OAA02701@axp.cmpu.net>
Mime-Version: 1.0
Content-Type: text/plain
Content-Transfer-Encoding: 7bit
Subject: Re: Directory access facilities
To: icon-group@optima.CS.Arizona.EDU
X-Mailer: SPRY Mail Version: 04.00.06.17
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 1993

>> Anyone else have ideas (or votes) on how to do directory access
> facilities?

> My admittedly biased and possibly Unix-centric view: we don't need any
elaborate solutions, I think the idea of each read() returning one entry
from the directory is the best one. This finesses all the issues of quoting
blanks, quotes, wildchars etc. etc. 

"Finesses" -> "Ignores".

> If you need more information about the
entry, you use the function that performs this query. In Unicon, you use
stat(), which returns a record. Here's the "ls" example:

>   f := open(".") | stop( ... )
   while write( format( stat( read(f) ) ) )

The problem (if you want something RELIABLE) with this kind of thing is that on 
multitasking systems (especially those crossing boundaries through a network) 
the entries can be changing (and entries even being added and deleted) WHILE you 
are retrieving your way through them.  It's much more robust to return a 
snapshot, which at least _was_ valid and consistent at the time the snapshot was 
taken.

It's entirely possible that (for example) the read(f) returns a temporary file 
name, which is no longer there by the time the stat() is issued for it.  If 
stat() failed (e.g.) in such a case, then you'd terminate your while loop 
prematurely.

Agreed that it's almost as flakey to try to load a snapshot into a table in 
memory... where that table could be of nearly arbitrary size... due to both the 
memory space available to hold it, and the (atomic) time it takes to generate 
such a potentially huge table.

> ...We should stick to a clean and simple interface that will work 
unambiguously without all sorts of special cases. 

Like the old saying, it should be "as simple as possible... BUT NO SIMPLER."

> Others have already pointed out why a space-separated string of all the 
entries is a bad idea.

Agreed, that's not a good solution.

Gordon Peterson
http://www.computek.net/public/gep2/
Support the Anti-SPAM Amendment!  Join at http://www.cauce.org/



From icon-group-sender  Tue Jun 16 08:31:21 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id IAA25119
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Tue, 16 Jun 1998 08:31:21 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA20768; Tue, 16 Jun 1998 08:31:12 -0700
From: "Nevin :-] Liber" <nevin@pendragon-software.com>
To: <icon-group@optima.CS.Arizona.EDU>
Subject: Re: Directory access facilities
Date: Mon, 15 Jun 1998 19:36:29 -0500
Message-Id: <01bd98be$cdb2d7a0$c37546cf@NEVIN>
Mime-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: 7bit
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
Content-Length: 2468

From: gep2@computek.net <gep2@computek.net>
To: icon-group@optima.CS.Arizona.EDU <icon-group@optima.CS.Arizona.EDU>
Date: Monday, June 15, 1998 6:26 PM
Subject: Re: Directory access facilities


>> My admittedly biased and possibly Unix-centric view: we don't need any
>elaborate solutions, I think the idea of each read() returning one entry
>from the directory is the best one. This finesses all the issues of quoting
>blanks, quotes, wildchars etc. etc.


>"Finesses" -> "Ignores".

But in this case, ignoring them is the Right Solution.


>The problem (if you want something RELIABLE) with this kind of thing is
that on
>multitasking systems (especially those crossing boundaries through a
network)
>the entries can be changing (and entries even being added and deleted)
WHILE you
>are retrieving your way through them.  It's much more robust to return a
>snapshot, which at least _was_ valid and consistent at the time the
snapshot was
>taken.

It's not ANY more robust, since the underlying C calls (which the Icon
runtime has to make in order to get the data) make no atomicity guarantees
(on Unix, MacOS, etc.) when querying all the entries in a directory.


>It's entirely possible that (for example) the read(f) returns a temporary
file
>name, which is no longer there by the time the stat() is issued for it.

And how is this magically fixed?  This happens with the C API, which means
that it will happen with the Icon API.  Pretending it doesn't won't make the
problem go away.

>If
>stat() failed (e.g.) in such a case, then you'd terminate your while loop
>prematurely.

Which makes the code incorrect even with your solution, since the window of
vunerability between the time you get the filename and the time you stat the
file is still there.  In either case, the code has to be modifed to
something like

every s := Files(sDirectory) do {
    write(format(stat(s)))
}

>Like the old saying, it should be "as simple as possible... BUT NO
SIMPLER."

And the solution where we have a function that generates directory entries
(and I don't believe we should overload read(), since this shares almost
nothing in common with read()) is, IMHO, that solution.  What you are trying
to graft onto it are atomicity guarantees, and that is impossible since you
need the underlying OS to make those guarantees.
__
 Nevin ":-)" Liber  <mailto:nevin@pendragon-software.com>  (847) 816-9926
  Senior Software Engineer, Pendragon Software <http://www.webfayre.com/>



From icon-group-sender  Tue Jun 16 12:49:59 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id MAA17945
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Tue, 16 Jun 1998 12:49:58 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA22238; Tue, 16 Jun 1998 12:49:49 -0700
Message-Id: <358690D9.58E@gte.net>
Date: Tue, 16 Jun 1998 10:35:53 -0500
From: MJE <evans@gte.net>
Reply-To: evans@gte.net
Organization: None
X-Mailer: Mozilla 3.01 (Win95; I)
Mime-Version: 1.0
To: icon-group@optima.CS.Arizona.EDU
Subject: Re: Directory access facilities
References: <01bd98be$cdb2d7a0$c37546cf@NEVIN>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 465

Could I propose that the authors of the more verbose commentaries in
this thread channel their energies into actually doing what they are
talking about?

I'm all for lively debate but at some juncture the gang should start
making demonstrations to prove points.  I mean, there are teenage
hackers out there doing useful shareware who don't sit around
contemplating how to do useful shareware.

Then we can all compare for ourselves and vote.

Cheers!

Mark Evans



From icon-group-sender  Tue Jun 16 16:22:31 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id QAA25846
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Tue, 16 Jun 1998 16:22:30 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA20066; Tue, 16 Jun 1998 16:22:21 -0700
Date: Tue, 16 Jun 98 13:00:25 PDT
Message-Id: <9806162000.AA20449@bermuda.synopsys.com>
From: Shamim Mohamed <shamim@Synopsys.COM>
To: evans@gte.net
Cc: icon-group@optima.CS.Arizona.EDU
Subject: Re: Directory access facilities
In-Reply-To: <358690D9.58E@gte.net>
References: <01bd98be$cdb2d7a0$c37546cf@NEVIN>
	<358690D9.58E@gte.net>
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 418

  > Could I propose that the authors of the more verbose commentaries in
  > this thread channel their energies into actually doing what they are
  > talking about?

Well, I'm not sure if my commentaries were among these verbose ones, but I
have already implemented my ideas! My apologies for yet another blatant
plug; they are part of Unicon:

   http://www.drones.com/unicon/

Any and all comments appreciated.

-s


From icon-group-sender  Wed Jun 17 09:13:55 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id JAA10731
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Wed, 17 Jun 1998 09:13:51 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA21188; Wed, 17 Jun 1998 09:13:41 -0700
From: gep2@computek.net
Date: Tue, 16 Jun 1998 20:34:21 -0500
Message-Id: <199806170134.UAA19159@axp.cmpu.net>
Mime-Version: 1.0
Content-Type: text/plain
Content-Transfer-Encoding: 7bit
Subject: Re: Directory access facilities
To: icon-group@optima.CS.Arizona.EDU
X-Mailer: SPRY Mail Version: 04.00.06.17
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 3521

>>> My admittedly biased and possibly Unix-centric view: we don't need any
>elaborate solutions, I think the idea of each read() returning one entry
>from the directory is the best one. This finesses all the issues of quoting
>blanks, quotes, wildchars etc. etc.


>> "Finesses" -> "Ignores".

> But in this case, ignoring them is the Right Solution.

I suppose you're entitled to your opinion, but I certainly don't agree.

>>The problem (if you want something RELIABLE) with this kind of thing is
that on multitasking systems (especially those crossing boundaries through a
network) the entries can be changing (and entries even being added and deleted)
WHILE you are retrieving your way through them.  It's much more robust to return 
a snapshot, which at least _was_ valid and consistent at the time the
snapshot was taken.

> It's not ANY more robust, since the underlying C calls (which the Icon
runtime has to make in order to get the data) make no atomicity guarantees
(on Unix, MacOS, etc.) when querying all the entries in a directory.

I don't argue that there might be bugs in the underlying calls, but the point is 
that you're using those current bugs as an excuse to propagate those BY DESIGN 
into the higher levels too.

It's entirely possible that one COULD fix the underlying problems... but it 
wouldn't help if the higher levels were still inherently unsound.

>> It's entirely possible that (for example) the read(f) returns a temporary
file name, which is no longer there by the time the stat() is issued for it.

> And how is this magically fixed?  

In this case, because you (presumably) wouldn't have to call STAT() at all.  
Even if it DID fail after a call, you would probably need to deal with that 
status differently than the "ambiguous failure" problem in the code's WHILE loop 
as was listed

> This happens with the C API, which means that it will happen with the Icon 
API.  Pretending it doesn't won't make the problem go away.

Again, it's one thing for there to be a bug in the underlying API.  It's another 
thing entirely for that inherently poor design to propagate those faults to 
higher level programs.

>>If stat() failed (e.g.) in such a case, then you'd terminate your while loop
>prematurely.

>Which makes the code incorrect even with your solution, since the window of
vunerability between the time you get the filename and the time you stat the
file is still there.  

If you get a table with filenames and directories, then you don't have to STAT 
yourself to find out whether you need to recurse or not.

> In either case, the code has to be modifed to something like

> every s := Files(sDirectory) do {
    write(format(stat(s)))
}

>> Like the old saying, it should be "as simple as possible... BUT NO
SIMPLER."

> And the solution where we have a function that generates directory entries
(and I don't believe we should overload read(), since this shares almost
nothing in common with read()) is, IMHO, that solution.  

Again, you're entitled to your opinion.  But I still don't agree with you.

> What you are trying to graft onto it are atomicity guarantees, and that is 
impossible since you need the underlying OS to make those guarantees.

It's not impossible that someday the underlying OS WILL make those guarantees.  
If the supporting calls are such that they are still ineffective, then fixing it 
in the OS doesn't solve anything.

Gordon Peterson
http://www.computek.net/public/gep2/
Support the Anti-SPAM Amendment!  Join at http://www.cauce.org/



From icon-group-sender  Wed Jun 17 13:57:47 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id NAA22965
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Wed, 17 Jun 1998 13:57:47 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA17602; Wed, 17 Jun 1998 13:57:38 -0700
From: WHITSTON@ltu.edu
Date: Wed, 17 Jun 1998 12:45:27 -0400 (EDT)
Subject: Icon and Linux
To: ICON-GROUP@optima.CS.Arizona.EDU
Message-Id: <01IYCDENCCIO8Y4XA8@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
Content-Length: 207

To the group:

In the July issue of the "Linux Journal" (Issue 51), there is an article on
about Icon -- written by Clinton Jeffery and Shamim Mohamed

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


From icon-group-sender  Thu Jun 18 12:58:11 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id MAA01774
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Thu, 18 Jun 1998 12:58:11 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA22775; Thu, 18 Jun 1998 12:58:01 -0700
Message-Id: <v03102801b1af164869df@[128.102.120.213]>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Date: Thu, 18 Jun 1998 11:28:04 -0800
To: icon-group@baskerville.CS.Arizona.EDU
From: Michael Shafto <mshafto@mail.arc.nasa.gov>
Subject: here's a funnny one!
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 2511

This is a kind of semi-meta-story about Icon:
It's not really about Icon, but it's a lot
funnier if you have written a bit of
Icon code.

I was at home playing with my laptop like
a cub bear with a box of candy.

I had 8 text files consisting of Icon
source code (about 10 pages of print-out
per file). I could have, as in years past,
copied these files onto a diskette, put the
diskette in my pocket, and trekked off to
work.

But no.

In the spirit of learning by doing, I dragged
the files into a 'Transfer' folder on my laptop,
and dragged them out of a similar folder on my
work computer.

I edited away for a while (mercifully, as
it turned out, altering only three of the
eight files), dragged 'em back into the
work 'Transfer' folder, and out again on
the laptop at home.

This is where the funny part happens: The
demons which had been lashed to the oars of the
file transfer process had surgically removed
every token preceded by a backslash (\). Let
that sink in for a minute while I provide a
few examples:

"\n"
"\B0"
and lots of other things
map to ""

\x[3]
maps to [3]

(\smfun)(ar)
maps to ()(ar)

\Ck_disable
maps to _disable [sic -- demons break for _]

if \altered then bazz(fazz)
maps to
if then bazz(fazz)

dyv <:= \ymin
dyv >:= \ymax
n := N+1

maps to

dyv <:=
dyv >:=
n := N+1


Funny, no?


Even funnier is the following conjecture:
Let wu (or Griswold's Fraction) be the probability
that an Icon procedure containing at least one such
alteration will compile with a "no errors" message.
Then wu is in the closed interval (0.67,0.75).


Telecommuters everywhere, beware!
I have no idea what "software product" is responsible
for this problem (certainly it has nothing to do with Icon),
but I'm a little suspicious of the RTF step which I know
is part of this transfer sequence.

Anyway, the experience is a bit like sitting on
a cactus, and I expect to be in recovery for a few more
days.

Mike


----- ----- ----- ----- ----- ----- ----- ----- ----- ----- -------
Mike Shafto (mshafto@mail.arc.nasa.gov)
Chief, Human-Automation Integration Research Branch
On detail (6/1/98 - 9/30/98) to the Computational Sciences Division
http://olias.arc.nasa.gov/
Mail Stop 262-4
NASA-Ames Research Center
Moffett Field,  CA       94035-1000
---------------------------------------------------------------
Those are my principles. If you don't like them I have others.
                                              -- Groucho Marx
---------------------------------------------------------------




From icon-group-sender  Thu Jun 18 16:57:11 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id QAA11670
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Thu, 18 Jun 1998 16:57:11 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA23991; Thu, 18 Jun 1998 16:57:01 -0700
Message-Id: <3.0.5.32.19980618154152.00a5c210@post.its.mcw.edu>
X-Sender: cdt@post.its.mcw.edu
X-Mailer: QUALCOMM Windows Eudora Light Version 3.0.5 (32)
Date: Thu, 18 Jun 1998 15:41:52 -0500
To: icon-group@optima.CS.Arizona.EDU
From: Chris Tenaglia <cdt@post.its.mcw.edu>
Subject: re: here's a funny one
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 661

Regarding the magic munge file transfer
mentioned by Mike Shafto, I have also
seen something like that. We have business
software that prints to local printers by
telneting to a dos session the output directly
to lpt1: So I think, cool, this is a nice little
file transfer program I'll make a handy
little programming environment by tweeking
some of the parameters and then up and down
loading the modules. I was disappointed that
that the transfer also did something with \
and { and } (curly braces). It would also
impact C programmers as well. So I'm just
sticking to a plain vanilla FTP for the time
being. When things get so smart, they're dumb.
Chris.




From icon-group-sender  Fri Jun 19 12:38:33 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id MAA22247
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Fri, 19 Jun 1998 12:38:33 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA28231; Fri, 19 Jun 1998 12:38:23 -0700
To: icon-group@optima.CS.Arizona.EDU
Date: 19 Jun 1998 13:30:56 GMT
From: cdt@post.its.mcw.edu (Chris Tenaglia)
Message-Id: <6mdp6g$k4g@wiscnews.wiscnet.net>
Organization: Medical College of Wisconsin - Milwaukee, WI
Sender: icon-group-request@optima.CS.Arizona.EDU
Subject: CGI progress
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 4413

Finally, some progress. Maybe its not the optimal approach,
but given this particular unix environment it can be made
to work well. Here's a summary ...

1. CGI program outputs Content-Type: text/html and a blank line.

2. The output is good HTML

3. The GET method is needed to accept data from a form.

4. There are useful environment variables.
   This is where the input data can be found.
   Icon can parse it quite easily.

5. icont and iconx installed in the CGI directory

6. I used a frontend to run the cgi program. This is because
   we did not have a C compiler to use iconc to make pure
   executables. So then the url looks like cgi-bin/iconx?program...
   and it seemed like parameters/variables beyond 'program'
   where not reachable.

###################################################################
#
# file : docgi.icn
# desc : this program is a cgi learning experiment
# use  : called by cgi in the web server
#
# update         by          what
# 08-jun-1998    tenaglia    initial write
#
global nvp   # name-value table
procedure main(param)
  query := getenv("QUERY_STRING") | "interactive"
  method:= getenv("REQUEST_METHOD") | "interactive"
  path  := getenv("PATH_INFO")      | "unknown"
  remote:= getenv("REMOTE_USER")    | "unknown"
  adrs  := getenv("REMOTE_ADDR")    | "interactive"

  log   := open("/db1/local/httpd/cgi-bin/bioet/docgi.log","a")
  write(log,adrs," Accessed DOCGI at ",&dateline," with ",query)
  close(log)

  fi := open("/db1/local/httpd/cgi-bin/bioet/docgi.cnt")
  bump := read(fi) + 1
  close(fi)
  fo := open("/db1/local/httpd/cgi-bin/bioet/docgi.cnt","w")
  write(fo,bump)
  close(fo)

  nvp := table("n/a")
  parts := parse(query,'&')
  every part := !parts do
    {
    pair := parse(part,'=')
    ident := unurl(pair[1])
    value := unurl(pair[2])
    nvp[ident] := value
    }
  write("Content-type: text/html")
  write("")
  write("<HTML>")
  write("<HEAD>")
  write("<TITLE>This is the CGI Generated Title</TITLE>")
  write("</HEAD>")
  write("<BODY>")
  write("<BODY BGCOLOR=#FFFFFF>")
  write("<H1>Big Header</H1>")
  write("<P><P><P>")
  write("Here's a crazy web page!<P>")
  write("The parameter was :",param[1],"<P>")
  write("The query string var is :",query,"<P>")
  write("<TABLE border=5>")
  write("<TH>Name Parts</TH><TH>Value Parts</TH>")
  write("<TR><TD>NAME</TD><TD>VALUE</TD></TR>")
  every xxx := key(nvp) do
    {
    write("<TR><TD>",xxx,"</TD><TD>",nvp[xxx],"</TD></TR>")
    }
  write("</TABLE>")
  write("<P>")
  write("<P>")
  write("<TABLE BORDER=10>")
  write("<TH>CGI Parameters</TH><TH>CGI Values</TH>")
  write("<TR><TD>Method</TD><TD>",method,"</TD></TR>")
  write("<TR><TD>Path</TD><TD>",path,"</TD></TR>")
  write("<TR><TD>Username</TD><TD>",remote,"</TD</TR>")
  write("<TR><TD>Remote address</TD><TD>",adrs,"</TD></TR>")
  write("<TR><TD>Page counter</TD><TD>",bump,"</TD></TR>")
  write("</TABLE>")
  write("</BODY>")
  write("</HTML>")
  end

#
# parse a string into a list with respect to a delimiter
#
procedure parse(line,delims)
  static chars
  chars  := &cset -- delims
  tokens := []
  line ? while tab(upto(chars)) do put(tokens,tab(many(chars)))
  return tokens
  end

#
# This procedure maps the wierd characters out of cgi query strings
#
procedure unurl(text)
  static  hex
  initial {
          hex := table("")
          hex["0"] := 0 ; hex["1"] := 1 ; hex["2"] := 2 ; hex["3"] := 3
          hex["4"] := 4 ; hex["5"] := 5 ; hex["6"] := 6 ; hex["7"] := 7
          hex["8"] := 8 ; hex["9"] := 9 ; hex["a"] := 10; hex["A"] := 10
          hex["b"] := 11; hex["B"] := 11; hex["c"] := 12; hex["C"] := 12
          hex["d"] := 13; hex["D"] := 13; hex["e"] := 14; hex["e"] := 14
          hex["f"] := 15; hex["F"] := 15
          }
  work  := map(text,"+"," ")
  cache := []
  every i := 1 to *work do if work[i] == "%" then put(cache,i)
  while i := pop(cache) do
   {
   nyb1 := work[i+1]
   nyb2 := work[i+2]
   if (nyb1 == "") | (nyb2 == "") then next
   raw  := hex[nyb1] * 16 + hex[nyb2]
   byte := char(raw)
   work[i+:3] := byte
   }
  return work
  end

###################################################################

--
Chris Tenaglia   (system manager)     |  The future foretold,
Medical College of Wisconsin          |  The past explained,
8701 W. Watertown Plank Rd.           |  The present largely appologized for.
Milwaukee, WI 53226   (414)456-8765   |  Organon to the Doctor



From icon-group-sender  Fri Jun 19 12:38:53 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id MAA22260
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Fri, 19 Jun 1998 12:38:53 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA13686; Fri, 19 Jun 1998 12:38:43 -0700
From: pygmy@eskimo.com (Frank Sergeant)
To: icon-group@baskerville.CS.Arizona.EDU
Subject: using Icon for database application
Date: Fri, 19 Jun 1998 12:11:28 -0500
Reply-To: frank.sergeant@pobox.com
Message-Id: <Avpi1Yv1ukQG084yn@eskimo.com>
Lines: 186
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 8752

     I was in the process of attempting to pick my new
language for a particular application when the July '98
_Linux Journal_ arrived containing the Jeffery/Mohamed
article "A Glimpse of Icon".  I enjoyed the article and
that led to reading the Linux Gazette article and various
Icon Newsletters, archived mailing list, etc.  I'm now
reading the 1983 edition of _The Icon Programming Language_
(Icon v. 5).

     I'd like to discuss some of my questions and concerns
related to Icon and its use in my application in hopes of
hearing your comments and suggestions.  Please feel free to
email me or to post to the list as you prefer.

     I maintain a medical accounts receivable application
that is installed in 12 or 13 small medical offices.  Two of
us are involved: Hal for sales (if you can call it that) and
me for the programming and technical support.  I took it over
nearly 4 years ago and have been improving it and cleaning
it up since then.  It is written in Clipper (the dBASE or
xBASE language) and consists of around 150,000 lines of
source.  It runs under DOS (or a DOS box in Windows).  More
work is needed to clean it up further to make it easier to
maintain.  Our typical customer has two or three PCs and
runs under a "file server" model with either DOS/Lantastic
or with W95 for the networking.  Thus, all workstations
access the database files (*.DBF) and index files.  (Any
workstation crashing puts the data files on the server at
risk.)  I am interested in moving more toward a
"client/server" model where all access to the data files is
handled by a process on the server ("put all your eggs in
one basket then _watch_ the basket").  I am not interested
in buying or having our customers buy a commercial SQL
database system.  I plan to handle the programming of the
data access myself.  One obvious decision to make is whether
to do the cleaning up in Clipper before I convert or to
convert and clean up as I go.

     Here are the reasons I am considering changing
languages (to the degree I know my own thinking):

     1. Protect the data better by restricting its access
        to a single program on the server.

     2. Reduce network traffic (by not dragging the database
        and index files over the ethernet) to speed up
        the application.  This is not a problem on an
        ethernet, as it is fast enough, but opens the door
        to making the application usable even over a modem.

     3. By going to a 32-bit Windows program we get out of
        the 16-bit Windows step child environment.  My
        feeling is that 16-bit applications are more likely
        to suffer from Windows bugs than 32-bit "native"
        applications.

     4. Offering a "Windows" version of the program might
        occasionally attract a customer who would not
        consider a plain DOS application -- and might help
        keep current customers.

     5. Open the door to working in a mixed operating system
        environment.  In particular, for even better
        reliability, customers could use a Linux box as the
        server.

     6. Our history of sales has been dismal.  Still, we
        have our core of loyal customers that use it year
        after year and cheerfully pay our small ($300/year)
        annual maintenance fee.  I am in this for the long
        run.  If we never make another sale, I intend to
        support these customers forever.  However, I need to
        do this as efficiently as possible, as I need to
        take on other work to keep eating.  So, platform and
        vendor independence seem like worthy goals.  This
        seems to rule out VO (CA-Visual Objects, supposedly
        the Windows-based successor to Clipper) and Delphi
        which run only on Windows.  Perhaps I should
        consider Java, but I think I am suffering from hype
        poisoning.  Python and Perl sound good to me
        primarily because of the delightfully literateness
        of the Pythona Perl books' authors.  I'm agreeable
        to the idea that Perl's syntax is too muddy.  Perl
        still looks good.  (Can anyone comment on the
        various tradeoffs between these languages and Icon?)


Icon questions (or comments)

     1. How do you think all this fits with Icon?

     2. I gather Windows Icon can make a normal looking
        Windows application (rather than a Motif-ish
        application) and that I probably could figure out
        how to do this by reviewing the source code for Wi.
        Does that sound reasonable?  Would _Graphics
        Programming in Icon_ help?  (I'm not interested in
        drawing pictures, etc, just in the UI part.)

     3. How fast is an Icon application?  For example, I ran
        the Python Megawidgets demo program that puts up a
        screen full of entry fields (Pentium 90 with 48
        MB RAM) and the fields rippled slowly onto the screen
        rather than just popping up instantly.  I'm afraid that
        would be too slow.  Many of our customers' machines
        are not even up to a P90/48MB, e.g. '486 8MB.  (Some
        even use slow '386s and one uses one or two
        '286s, but I would expect they would be willing to
        upgrade.)  My point is the application must be fast
        enough on moderate hardware as I can't expect all
        customer machines to be the latest and fastest.

     4. How big would the .EXE file be?  From a newsletter,
        I understand there is the 200K or 300K overhead for
        the Icon interpreter.  How does .EXE file size grow
        after this?  (I imagine this is hard to answer, but
        perhaps you can give me some idea.)  Currently our
        .EXE is about 850K, so I can live with that size,
        but would prefer smaller.

     5. Is a sockets interface only available for Unicon or
        is it also available for Wicon?  I would hope the
        client programs on the workstations could connect
        via sockets to the server running either on a
        Windows 95 or Linux machine.  The client machines
        would almost certainly be W95 machines.  If sockets
        are not built in to Icon, how hard is it to make
        calls to the Win32 system?  How about to routines
        produced by MSVC or Borland C?  Perhaps I could
        write the socket access routines in C and call them
        from Icon?

     6. So far I haven't been able to track down Idol.  A
        url or two for Idol on the web site seemed to give
        an "invalid" message.  Since I had also been
        considering Python for this application, it seems I
        should consider using Idol if I use Icon, right?

     7. I currently make extensive use of Clipper's "ragged
        arrays".  These are essentially lists where the
        elements can be of any type including other lists.
        It looks like Icon's lists would just drop in and
        work the same.

     8. The current app is about 150,000 lines (over 100
        source code files).  How does Icon handle
        applications of this size?  I think the app is currently
        more complex than is required.  All we do, really,
        is keep track of, display, modify various lists --
        lists of doctors, patients, insurance companies,
        charges, paygments, and such.  (How hard can it be?)

     9. The data, ignoring index files, probably runs 2 to
        10 MB in most offices.  At the larger sizes, much of
        the data is relatively inactive and could be rolled
        out to archives.  On a modern machine, all the data
        would fit in RAM.  I am not talking about an
        application with gigabytes of data and 200
        workstations and multiple servers.  We currently use
        dBASE style .DBF files.  These have fixed width
        fields.  As an experiment, I converted some of our
        data files to variable width, delimited files

           Smith, George|1234 W. 17th Street|etc|etc|etc

        and was surprised to find it reduced file size by
        about two-thirds.  I'm not sure this is typical, but
        it looks like going to variable width fields would
        save a lot of space, making the data fit even more
        easily into RAM, reducing the time of reading from
        disk, etc.  The cost would be the slower parsing of
        records to locate the individual fields.  Any
        thoughts about the tradeoffs here?

    10. I have been very impressed with Bertrand Meyer's
        _Object Oriented Software Construction_.  Can anyone
        comment on the differences between Eiffel and Icon,
        that is, why I might prefer Icon over Eiffel, etc?


     All comments, suggestions, corrections gratefully
received.


  -- Frank
  frank.sergeant@pobox.com


From icon-group-sender  Mon Jun 22 08:02:49 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id IAA28044
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Mon, 22 Jun 1998 08:02:49 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA25812; Mon, 22 Jun 1998 08:02:38 -0700
To: icon-group@optima.CS.Arizona.EDU
Date: 19 Jun 1998 14:28:57 -0500
From: moss@tim.risc.sps.mot.com (Matthew Moss)
Message-Id: <6mee5pINNhc0@tim.risc.sps.mot.com>
Organization: Motorola, Inc. -- Austin,TX
Sender: icon-group-request@optima.CS.Arizona.EDU
Subject: Wanted: small meta-data language
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 1638


I'm hoping someone can help identify an existing language that fills
my needs.  I've posted to the .awk and .icon groups as well, since my
needs are similar (and maybe a match) to those languages, but I am 
unsure of how well they would work on binary data chunks.

Anyway, I'm looking for an existing "meta-data" language.  My purpose
is to port such a beast to the PalmOS.  Such a language would most 
likely be interpreted, not compiled, would provide some methods for
pattern matching in both text and binary, and would operate primarily
on binary data (with text objects within the binary stream).

The idea is to easily allow 3rd-party applications a simple and common
method to accessing other databases.  (IE, I might create a scheduling
app that reads through the built-in datebook database to see if there
are any conflicts).  Such a language would define the structure of the
database and allow in-place modification and reading of data.  

For example, assuming that the address book data structure was defined,
I could write C-code somewhat like this:

    if (!strcmp ("Acme Company",  GET_STRING(ref, "company")))
	/* do action */

where GET_STRING is an entry point into the gawk/icon/sed/other meta-data
language, ref is a pointer to the data record, and "company" is the field
name within the meta-data language definition for the address book 
database.

The whole idea is to provide one library with multiple meta-data
definitions for each library.  I don't know if such a thing/language exists,
so I'm hoping someone out there can tell me "yes" or "no" before I write
the whole thing myself.

Thanks in advance....


From icon-group-sender  Mon Jun 22 08:03:22 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id IAA28059
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Mon, 22 Jun 1998 08:03:21 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA26604; Mon, 22 Jun 1998 08:03:11 -0700
From: gep2@computek.net
Date: Fri, 19 Jun 1998 20:37:26 -0500
Message-Id: <199806200137.UAA25365@axp.cmpu.net>
Mime-Version: 1.0
Content-Type: text/plain
Content-Transfer-Encoding: 7bit
Subject: using Icon for database application
To: icon-group@baskerville.CS.Arizona.EDU
X-Mailer: SPRY Mail Version: 04.00.06.17
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 10567

> ...I maintain a medical accounts receivable application
that is installed in 12 or 13 small medical offices.  Two of
us are involved: Hal for sales (if you can call it that) and
me for the programming and technical support.  I took it over
nearly 4 years ago and have been improving it and cleaning
it up since then.  It is written in Clipper (the dBASE or
xBASE language) and consists of around 150,000 lines of
source.  It runs under DOS (or a DOS box in Windows).  

That's a fairly huge Clipper application... I'd think that if it were written 
really utilizing the tools present to best advantage (native report writers etc 
etc... I use FoxPro for several of my clients) that it would be able to be a 
whole lot smaller than that.

>... (Any workstation crashing puts the data files on the server at
risk.)  

I've written a program (in S*BOL) which reads through xBASE .DBF files and 
"repairs" this kind of "not a database file" problems.  It fixes many kinds of 
scrogged fields in the header, and particularly fixes the length irregularities 
which usually create this symptom.

> I am interested in moving more toward a
"client/server" model where all access to the data files is
handled by a process on the server ("put all your eggs in
one basket then _watch_ the basket").  

Personally, I don't much like that approach (no surprise, as I was the guy who 
invented the first commercially successful LAN software in 1976).  It's not only 
just putting all your eggs in one basket (from a security standpoint) but it's 
also a reversion to what I call "mainframe think".  It puts a limitation of how 
much processor power you can cluster around that shared database... your 
potential user/application fanout is greatly reduced, IMHO.

> I am not interested in buying or having our customers buy a commercial SQL
database system.  

I agree with you there... I've (for example) yet to see a SINGLE Oracle-type 
application which I think has responsiveness, cost, and efficiency which are all 
respectable.

> I plan to handle the programming of the
data access myself.  One obvious decision to make is whether
to do the cleaning up in Clipper before I convert or to
convert and clean up as I go.

My personal feeling is that I'd leave the code pretty much as is, and probably 
leave it pretty much in xBASE.  For all the wonderful things in Icon, Icon (at 
least I've not seen them) doesn't have the kind of sweeping database-style 
features that xBASE-type systems have.  (And here I'm talking about things like 
the wonderful report writers and such which are so cool in FoxPro and the like). 
 Screen-and-field-oriented data entry features (with verification features, 
automatic relationships, etc etc) really are a dandy thing to have.  Icon 
doesn't even have native ISAM file support, and that's just about a minimum 
pre-requisite for any serious business use (IMHO).

>     1. Protect the data better by restricting its access
        to a single program on the server.

I believe that there are FoxPro (xBASE) flavor "server" engines available... 
which would satisfy this goal (FWIW) without requiring a complete application 
rewrite.

>     2. Reduce network traffic (by not dragging the database
        and index files over the ethernet) to speed up
        the application.  This is not a problem on an
        ethernet, as it is fast enough, but opens the door
        to making the application usable even over a modem.

There are other ways to do that, whether by thin-client approaches (ICK) or by 
things like ReachOut, PC-Anywhere, etc etc.  

>     3. By going to a 32-bit Windows program we get out of
        the 16-bit Windows step child environment.  My
        feeling is that 16-bit applications are more likely
        to suffer from Windows bugs than 32-bit "native"
        applications.

Perhaps, although they also tend to be simpler and more mature, too... so I 
don't think it's all that certain that the applications are really more 
reliable.  In all the time I've been working with FoxPro (for DOS) I have seen 
few (if any) legitimate BUGS in the FoxPro package itself.  It seems to overall 
be very stable.

>     4. Offering a "Windows" version of the program might
        occasionally attract a customer who would not
        consider a plain DOS application -- and might help
        keep current customers.

I haven't found this to be an issue (with my current customers, anyhow) and they 
seem to greatly appreciate the better responsiveness of the text-mode 
applications.

>     5. Open the door to working in a mixed operating system
        environment.  In particular, for even better
        reliability, customers could use a Linux box as the
        server.

I think that, from a reliability and support standpoint, it's in most cases just 
plain dumb to go to a mixed-OS environment.  You're talking about a minor 
advantage at best, and usually introducing an awful lot of potential for 
incompatibilities.  Especially at smaller clients, I think that maintaining a 
Linux system is so alien to everything else they know that it's at minimum 
something to do only with the greatest of trepidation.

>     6. Our history of sales has been dismal.  Still, we
        have our core of loyal customers that use it year
        after year and cheerfully pay our small ($300/year)
        annual maintenance fee.  I am in this for the long
        run.  If we never make another sale, I intend to
        support these customers forever.  However, I need to
        do this as efficiently as possible, as I need to
        take on other work to keep eating.  

Personally, especially in that case, I wouldn't want to do a rewrite of the 
package just to change languages!!!

>       ...So, platform and
        vendor independence seem like worthy goals.  

What a strange conclusion to come to, based on those concerns!  Given that 
probably NONE of your current clients are running Linux, and that none of them 
likely have any desire to run Linux, I'm wondering how on earth you think it's a 
great idea to lob that grenade their direction.

>       ...This
        seems to rule out VO (CA-Visual Objects, supposedly
        the Windows-based successor to Clipper) and Delphi
        which run only on Windows.  

I think those also run on NT, which (if you're going to not use W9x) is probably 
a better alternative than going to Linux.  My perception is that in five or ten 
years, the corporate environment is going to be a lot more NT-like than it will 
be Unix-like.

If and when we get ready to migrate most of my current clients' systems from 
FoxPro or xBASE to Win-native stuff, it will more likely be to FoxPro for 
Windows.

>       ....Perhaps I should
        consider Java, but I think I am suffering from hype
        poisoning.  

Not just that, but I don't want to be in the middle of a pissing match between 
Bill Gates and Sun.  (If it came to that, Bill will win it).

>       ....Python and Perl sound good to me
        primarily because of the delightfully literateness
        of the Pythona Perl books' authors.  

My reaction to Perl is that I can't imagine for the life of me why anybody 
familiar with either S*BOL or Icon would use it in preference to the other two.

>       ...I'm agreeable
        to the idea that Perl's syntax is too muddy.  Perl
        still looks good.  (Can anyone comment on the
        various tradeoffs between these languages and Icon?)

You know, for "general purpose programming" I still really like S*BOL and Icon 
(hell, even C/C++ for stuff that's that intensive).  But business programming is 
NOT really "general purpose programming"... it's much more file-and-database 
intensive, has more demanding requirements on data entry and verification, 
involves a lot more report generation, and tends to be less algorithm-intensive. 
 
>     8. The current app is about 150,000 lines (over 100
        source code files).  How does Icon handle
        applications of this size?  

It probably would handle it fine, although (depending on how the application is 
written... the fact that it's 150K lines is perhaps a red flag) my suspicion is 
that the app in Icon would be substantially LARGER in terms of lines of code.

>       ...I think the app is currently
        more complex than is required.  All we do, really,
        is keep track of, display, modify various lists --
        lists of doctors, patients, insurance companies,
        charges, paygments, and such.  (How hard can it be?)

Don't underestimate the value of all the behind-the-scenes features that 
xBASE-style languages provide to simplify the programmer's life!!

>       9. ...We currently use
        dBASE style .DBF files.  These have fixed width
        fields.  As an experiment, I converted some of our
        data files to variable width, delimited files

>           Smith, George|1234 W. 17th Street|etc|etc|etc

>        and was surprised to find it reduced file size by
        about two-thirds.  I'm not sure this is typical, but
        it looks like going to variable width fields would
        save a lot of space, making the data fit even more
        easily into RAM, reducing the time of reading from
        disk, etc.  The cost would be the slower parsing of
        records to locate the individual fields.  Any
        thoughts about the tradeoffs here?

I think that, with 32Mb of RAM costing maybe $50 or thereabouts and 6Gb hard 
drives going for about $200-250, any discussion of saving bytes by increasing 
programmer effort is probably pretty silly.  If you're really worried about 
stuff like this, let me suggest that (instead of application-specific forms of 
compression like this) you'd be *far* better off to use Stacker or even 
Microsoft's own disk compression software.  It's also a big PITA (doing it 
yourself) when it comes time to update records, since compressed records usually 
are nontrivial to update in place.

>     All comments, suggestions, corrections gratefully
received.

I hope this helps you!

I'm NOT trying to turn you away from Icon or S*BOL... they're both wonderful 
languages... but personally, when I'm presented with a problem like you're 
dealing with, I tend to favor FoxPro (essentially, like Clipper) to solve it... 
and adding some modules in S*BOL or whatever to attack the problems more 
amenable to those languages when appropriate.

The right tool for the right job.

Gordon Peterson
http://www.computek.net/public/gep2/
Support the Anti-SPAM Amendment!  Join at http://www.cauce.org/



From icon-group-sender  Mon Jun 22 08:03:36 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id IAA28069
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Mon, 22 Jun 1998 08:03:35 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA28396; Mon, 22 Jun 1998 08:03:25 -0700
To: icon-group@optima.CS.Arizona.EDU
Date: Sat, 20 Jun 1998 10:27:31 +0100
From: Ian Stirling <0000358B8083.NO_UCE@mauve.demon.co.uk>
Message-Id: <0000358B8083.NO_UCE@mauve.demon.co.uk>
Organization: None.
Sender: icon-group-request@optima.CS.Arizona.EDU
References: <6mee5pINNhc0@tim.risc.sps.mot.com>, <6mf0fo$a2l$1@supernews.com>
Subject: Re: Wanted: small meta-data language
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 1080

In comp.lang.awk Scott <none@nospam.com> wrote:
: Have you considered something like JavaScript (ECMA Script).  The language
: itself is pretty easy, and it shouldn't be that hard to implement from
: scratch if needed be.  That shouldn't be necessary though as both Netscape
: and Microsoft have released source for it (I don't know if MS is still doing
: that though).

<snip>

: Matthew Moss wrote in message <6mee5pINNhc0@tim.risc.sps.mot.com>...
:>
:>I'm hoping someone can help identify an existing language that fills
:>my needs.  I've posted to the .awk and .icon groups as well, since my
:>needs are similar (and maybe a match) to those languages, but I am
:>unsure of how well they would work on binary data chunks.

I assume you've looked at the palmpilot software archive, Lots of languages
have been ported.


-- 
See http://www.mauve.demon.co.uk/    |Linux PDA, cheap electronics/PC bits sale.
See_header,_for_UCE_policy___________|_____________________________Ian_Stirling.
If it can't be expressed in figures, it is not science, it is opinion.
Robert A Heinlein.



From icon-group-sender  Mon Jun 22 12:50:51 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id MAA11483
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Mon, 22 Jun 1998 12:50:51 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA27571; Mon, 22 Jun 1998 12:50:41 -0700
Message-Id: <3.0.5.32.19980622102318.00a34530@post.its.mcw.edu>
X-Sender: cdt@post.its.mcw.edu
X-Mailer: QUALCOMM Windows Eudora Light Version 3.0.5 (32)
Date: Mon, 22 Jun 1998 10:23:18
To: icon-group@optima.CS.Arizona.EDU
From: Chris Tenaglia <cdt@post.its.mcw.edu>
Subject: RE: Wanted : Small meta data language
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 1808

I designed a similar database package years ago. I think what you
mean is an INI file design. I called my package Attercop, wrote it
around 1983, and ported to Icon in 1989. The database is stored
in a flat file with \xFE as field delimiters. I thought about having an
\xFF delimiter for a field with "many" potential elements. 
Typically, data is in a .dat file. In this example, the metafile is called
userbase.atcp. So typing "attercop userbase" opens this metafile
which tells attercop where the data is, how it's layed out, and also
contains some display, sort, and print preferences. attercop then
displays a menu for performing various maintenance and reporting
on the data. But with a known interface other programs (even CGI)
could make use of it. A unix compressed version of attercop is
downloadable from http://www.execpc.com/~cdt if one wants to
look more closely at it.
____________________________________________________________________________
___________
data in /export/home/cdt/srch/userbase.dat
header is General User Accounts
fields are UserName UID_____ GID_____ Phone FullName Dept Email_
__ HomeDir_ Profile_
lengths are 10 6 6 14 30 35 35 21 8
look up 1
sort on 5
print using 1 57 mishp5/200_dplx n/a
title begin
MCWSUN USER ACCOUNTS REPORT                             &PAGE
                     FOR &DATE
USERNAME    UID     GID     PHONE NUMBER    FULLNAME
LAST
 NAME             DEPARTMENT NAME                      EMAIL ADDRESS
            HOMEDIR                PROFILE
==========  ======  ======  ==============  ==============================
====
================  ===================================
=========================
==========  =====================  =======
title end
____________________________________________________________________________
____________




From icon-group-sender  Mon Jun 22 16:44:43 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id QAA22859
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Mon, 22 Jun 1998 16:44:43 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA28717; Mon, 22 Jun 1998 16:44:32 -0700
From: pygmy@eskimo.com (Frank Sergeant)
To: icon-group@baskerville.CS.Arizona.EDU
Subject: Re: using Icon for database application
Date: Mon, 22 Jun 1998 15:22:26 -0500
Reply-To: frank.sergeant@pobox.com
Message-Id: <C0rj1Yv1ucpV084yn@eskimo.com>
Lines: 18
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 623

I recently wrote:

>                     Python and Perl sound good to me
>         primarily because of the delightfully literateness
>         of the Pythona Perl books' authors.  I'm agreeable
>         to the idea that Perl's syntax is too muddy.  Perl
                                                        ^^^^
>         still looks good.  (Can anyone comment on the
>         various tradeoffs between these languages and Icon?)

     Oops, that was a typo.  I meant to say that, while I'm
tentatively ready to reject Perl because of its syntax, _Python_
still looks good.


  -- Frank
  frank.sergeant@pobox.com



From icon-group-sender  Mon Jun 22 16:44:52 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id QAA22862
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Mon, 22 Jun 1998 16:44:51 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA23562; Mon, 22 Jun 1998 16:44:41 -0700
Date: Mon, 22 Jun 1998 16:39:26 -0500
Message-Id: <199806222139.QAA25152@segfault.cs.utsa.edu>
From: Clinton Jeffery <jeffery@segfault.cs.utsa.edu>
To: frank.sergeant@pobox.com
Cc: icon-group@baskerville.CS.Arizona.EDU
In-Reply-To: <Avpi1Yv1ukQG084yn@eskimo.com> (pygmy@eskimo.com)
Subject: Re: using Icon for database application
Reply-To: jeffery@cs.utsa.edu
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 7013

[Frank Sergeant offered a discussion of a large Clipper application that
he is maintaining, followed by a number of Icon questions.]

Hi Frank,

I'm glad you liked the Linux Journal article on Icon.  While I disagree with
Gordon Peterson's answers to your questions on many or most points, he did
say several things that are very true; you will have to figure out for
yourself which of his comments are relevant for your application.  I suspect
Gordon has more experience than I with xBase languages (although I did
program-for-hire in dBase III and Foxbase).  I've done a lot more Icon
programming than Gordon, who frequently posts about Snobol in the Icon
group, bless his heart!

>   Icon questions (or comments)
>	1. How do you think all this fits with Icon?

Icon isn't a special-purpose database language.  Its built-in types are
designed for flexible data manipulation in main memory.  If that's what
you want to do, you are all set; if not, you should look for a database
language, or think about connecting Icon to a third-party database engine.
There's more on memory requirements in your question 9 below.

>	2. I gather Windows Icon can make a normal looking
>	   Windows application (rather than a Motif-ish
>	   application) and that I probably could figure out
>	   how to do this by reviewing the source code for Wi.
>	   Does that sound reasonable?  Would _Graphics
>	   Programming in Icon_ help?

Well, you shouldn't have to learn it by studying source code.  Both the
on-line reference and an appendix in "Graphics Programming in Icon" discuss
these functions, which are quite simple.

>	3. How fast is an Icon application?

In your hypothetical client/server architecture, the database engine
performance will be your bottleneck.  Icon's graphics are plenty fast enough
for 486 clients, but Icon would not be suitable for the server software
unless either (a) enough main memory was available to fit all the indices
and preferably the data as well, or (b) you extended Icon to access a
third-party database, such as Postgres or mSQL.

>	4. How big would the .EXE file be?
	   How does .EXE file size grow after [the 200/300K interpreter]?

It grows at a moderate rate :-) depending on how effectively you are using
the built-in features.  The virtual machine is by nature higher-level and
potentially more compact than many languages, but if you write many thousands
of lines of Pascal-like (or xBase-like) code, it will be huge.

One wart I know of at present: if you use tons of different record types you
get penalized (this is true, for example, of the Motif-like portable widgets
in the Icon Program Library).  I thought a student had fixed this for me,
but his "field table compression" has some bugs that need fixing before we
turn it on and leave it on by default.

>	   Currently our .EXE is about 850K, so I can live with that size,
>	   but would prefer smaller.

I think the Icon program would be larger, unless your rewrite substantially
reduces the number of lines of code to exploit Icon's built-in features.  If
you just translated xBase line-by-line into Icon the result will be huge.
Clipper is (was?) a fine commercial-quality product.

>	5. Is a sockets interface only available for Unicon or
>	   is it also available for Wicon?

Ah, its time for some vaporware.  Sockets are only available on Unix at the
present time.  I've merged Unicon into my main source tree so that I can
start porting it to Windows, but I have to wrangle with some other problems
in Icon 9.3.2 before I take that on.

>	   If sockets are not built in to Icon, how hard is it to make
>	   calls to the Win32 system?  How about to routines
>	   produced by MSVC or Borland C?  Perhaps I could
>	   write the socket access routines in C and call them
>	   from Icon?

I keep getting nibbles, but so far no one has volunteered to implement this
kind of access.  You can get the Windows Icon source and make such
extensions yourself, and if you build something really useful I'd certainly
like to incorporate it into the main public domain Windows Icon
distribution.

>	6. So far I haven't been able to track down Idol.  A url or two for
>          Idol on the web site seemed to give an "invalid" message.
>	   it seems I should consider using Idol if I use Icon, right?

My apologies; there was a reorganization that left a bad symbolic link.
The main Idol web site is now www.cs.utsa.edu/research/idol/ and I am
in the process of packaging a release of Windows Idol.  You should consider
using Idol if you want to write object-oriented applications or have a large
complex application with many user-defined types.

>	7. I currently make extensive use of Clipper's "ragged arrays".
>          These are essentially lists where the elements can be of any
>          type including other lists.  It looks like Icon's lists would
>          just drop in and work the same.

Yes, or better.

>	8. The current app is about 150,000 lines (over 100 source code
>          files).  How does Icon handle applications of this size?

Icon is generally tuned for speed, rather than to minimize space
requirements.  I've heard reports of Unix Icon programs in the million line
range, but haven't seen a Windows Icon program in the 100K range yet.  My
impression is that Windows manages virtual memory much worse than UNIX or
Linux.  On old 486's with 8MB of memory you may run into terrible problems
unless you split the application up into a thin "client" program that can
access a server with more muscle.

>           I think the app is currently more complex than is required.
>           All we do, really, is keep track of, display, modify various
>           lists --  lists of doctors, patients, insurance companies,
>	    charges, paygments, and such.  (How hard can it be?)

This is generally vastly easier in Icon than in an xBase language.

>	9. The data, ignoring index files, probably runs 2 to
>	   10 MB in most offices.  At the larger sizes, much of
>	   the data is relatively inactive and could be rolled
>	   out to archives.  On a modern machine, all the data
>	   would fit in RAM.

I heartily recommend memory-based databases!  Memory is cheap.
The server should have lots of memory.

>       10. I have been very impressed with Bertrand Meyer's
>	   _Object Oriented Software Construction_.  Can anyone
>	   comment on the differences between Eiffel and Icon,
>	   that is, why I might prefer Icon over Eiffel, etc?

Icon is dynamically typed, offers a lot more flexibility, and has a much
higher-level set of built-in data structures and operators.  Eiffel is a
very good object-oriented systems language geared more for multi-programmer
efforts that might otherwise be written in C++ or Ada.  You write a lot more
lines of code in Eiffel to do the same work.  Eiffel may run quite a bit
faster; I am not sure how much.

Cheers,

Clint Jeffery, jeffery@cs.utsa.edu
Division of Computer Science, The University of Texas at San Antonio
Research http://www.cs.utsa.edu/research/plss.html


From icon-group-sender  Wed Jun 24 08:08:45 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id IAA01699
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Wed, 24 Jun 1998 08:08:44 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA29429; Wed, 24 Jun 1998 08:08:34 -0700
From: pygmy@eskimo.com (Frank Sergeant)
To: icon-group@baskerville.CS.Arizona.EDU
Subject: Re: using Icon for database application
Date: Tue, 23 Jun 1998 16:33:31 -0500
Reply-To: frank.sergeant@pobox.com
Message-Id: <r8Bk1Yv1usle084yn@eskimo.com>
In-Reply-To: <199806200137.UAA25365@axp.cmpu.net>
Lines: 233
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 10427

Gordon, I very much appreciate your comments and suggestions.
I don't think I can agree completely yet, but you've given
me food for thought.

gep2@computek.net (Gordon Peterson) wrote:
> > xBASE language) and consists of around 150,000 lines of
> > source.  It runs under DOS (or a DOS box in Windows).
> 
> That's a fairly huge Clipper application... I'd think that if it
> were written really utilizing the tools present to best advantage
...
> that it would be able to be a whole lot smaller than that.

     I agree that it is large and that it could/should be
smaller.  I attribute this to my predecessors on the project.
There has been a good bit of redundancy in the system.  For
example, when I took it over I found 6 monthly statement
routines: 1 set of 3 for each of 3 possible forms for
printing a single statement and a similar set of 3 routines
for printing all statements.  Obviously, this multiplies the
opportunities for errors and increases the difficulty of
wrapping one's mind around the system.  All reports were
written individually, duplicating the line counting and
report heading logic each time.  Etc.  I have gradually been
factoring out the commonalities but a lot of work remains.
I understand how such (unfactored) code comes about: there
is no money to be made yet there is an obligation to do
something (or at least appear to do something) to solve a
current emergency, and so the most hurried quickest immediate
fix is applied without regard to the long range maintenance
problems it creates, cutting and pasting rather than
analyzing and simplifying.


> >... (Any workstation crashing puts the data files on the server at
> risk.)  
> 
> I've written a program (in S*BOL) which reads through xBASE
> .DBF files and "repairs" this kind of "not a database file"
> problems.  It fixes many kinds of scrogged fields in the
> header, and particularly fixes the length irregularities
> which usually create this symptom.

     I'm going to quote two more parts of your message out of
order so they will be near the above quote.

> reliable.  In all the time I've been working with FoxPro
> (for DOS) I have seen few (if any) legitimate BUGS in the
> FoxPro package itself.  It seems to overall be very stable.

> I think that, from a reliability and support standpoint,
> it's in most cases just plain dumb to go to a mixed-OS
> environment.

     I'm not sure it is a bug in the FoxPro package (or in
Clipper), but _how come you have those S*BOL repair
programs_ if FoxPro and the environment (and the OS) are so
damn stable?  I think there are several factors.  One is
whether the hardware, especially the RAM, is fully reliable
or partially (intermittently) defective.  The best method I
know of to determine this is to run what I call the Linux
stress test (compile the Linux kernel).  I've written up a
little article about this ("Stress Testing Hardware") and
put it on my web site.  So, being able to run (and later
rerun) this stress test to verify the hardware is a
beneficial side effect of having Linux available.  Also, I
am convinced Linux is a more reliable operating system than
W95 (and perhaps WNT).

     Another factor in data integrity is what I was
referring to when I called it "putting all my eggs in
one basket", but that isn't really the best analogy.
The eggs were already in one basket, that is, the
application's data resided on a single machine, the
file server.  The problem is that all the workstations
independently access that data.  If any one of those
machines fails, the data is at risk.  Putting a UPS
on the server isn't enough, etc.  But, if I move to
"client/server" then it seems I could arrange things
so nothing the clients do can endanger the data on
the server.  Therefore, I remove several potential
points of failure.

     Another factor _may_ be bugs in W95.  It was
interesting to read over the bug history list for
Kermit.  Item after item referred to bugs they had
located in W95 that were not present either in OS/2
or in WNT.  With regard to file server record locking
and such, there is the W95 "vredir" problem and several
"fixes" from Microsoft and several Knowledge Base articles
about it.  The Knowledge Base articles are gibberish as
far as I am concerned.

     So, it may be dumb to go to a mixed OS environment.
But, it may be dumber to use Windows by itself.  Anyway,
I'm not saying you are wrong about this.  I've had my own
strong reservations about recommending Linux for my
customers due to the support problem.  I have figured that
if W95 screws up, they might perhaps blame Microsoft, but
if Linux screws up they will blame me.  Nevertheless, I
have talked customers through (or have gone to the office
to do it myself) various reconfigurations of W95 and its
networking.  I am beginning to think Linux cannot cause
me any more trouble than W95 is causing me.  Plus, Linux
should be much easier to administer remotely, and that
could be worth a lot.  So, I am not saying I will demand
customers use a Linux server, just that I see a possible
advantage to keeping my options open.  (Although it was
W3.11, not W95, I've had a customer phone because our
app would no longer run.  "What do you mean", I asked.
It turned out the shortcut icon on the desktop had
disappeared.  Well, I talked them through setting up
the shortcut again.  Can you believe it?  The shortcut
disappeared and _they_ didn't remove it (so they said, but
I suspect they did somehow by accident).  I know our
program didn't remove it.  Would Linux really cause me more
trouble than that?)

> Icon doesn't even have native ISAM file support, and that's
> just about a minimum pre-requisite for any serious business
> use (IMHO).

     Hmmm.  So it looks like I am back to the question of
whether all of the data would fit in RAM.  I think it would.
I'll try to run some tests.


> I believe that there are FoxPro (xBASE) flavor "server"
> engines available... which would satisfy this goal (FWIW)
> without requiring a complete application rewrite.

     This is a good point.  There are two products in
particular that I have heard good things about.  One is
Advantage Database Server.  It just costs money and requires
a dedicated NT (or Novell) server which costs money.  I
think we are talking about some thousands of dollars per
customer (that I think they do not wish to pay).  But,
Advantage gets rave reviews.  The other is RASQL/B which
provides an interface from Clipper to Btrieve.  Again,
substantial money for each customer.  Both of these seem
like overkill, remembering that our typical office has
two or three computers.

> >     2. Reduce network traffic (by not dragging the database
> 
> There are other ways to do that, whether by thin-client
> approaches (ICK) or by things like ReachOut, PC-Anywhere, etc etc.

     Well, these solutions seem to require dedicating a
machine at a central office for use by remote office (and
possibly vice versa), since Windows isn't multi-user.  I
could see it for occasional use.  We do have one customer
using PC-Anywhere sometimes to access the office machine
from home.

> I think that, with 32Mb of RAM costing maybe $50 or
> thereabouts and 6Gb hard drives going for about $200-250,
> any discussion of saving bytes by increasing programmer
> effort is probably pretty silly.

     I generally agree with this.  My point was not so
much to save bytes on disk, but to allow the entire
database to reside in RAM.  I am not saying that is a
general solution for every business application, just
that it might well fit our application considering the
size of our databases.  Then, bang!, with it all in RAM,
certain complexities seem to disappear.  Well, I need
to do some testing.  (And, although 32MB of RAM might
seem cheap at $50, there is also the problem of how to
get it put into the machine.  Will the customer do that
himself?  I don't want to have to do it for him, but I
might have to.  So, really, the cost is greater than just
the $50.)

> If you're really worried about stuff like this, let me
> suggest that (instead of application-specific forms of
> compression like this) you'd be *far* better off to use
> Stacker or even Microsoft's own disk compression software.

     It will be a cold day in hell before I use Microsoft's
disk compression.  (Do I feel a chill in the room?).  Maybe
I'm living too far in the past, but disk compression has
caused too many problems.  Is it really reliable now?  I'm
still very suspicious of all the disk compression products.
Certainly (until convinced otherwise), I would buy larger
hard disks first and recommend the same to my customers.

> It's also a big PITA (doing it yourself) when it comes time
> to update records, since compressed records usually are
> nontrivial to update in place.

     Well, I've got to think all this through.  But, if
the data would indeed fit in RAM, wouldn't Icon take care
of the updating in place automatically?  I'd have the
responsibility of initially loading the data from disk
at the start of the day (or the start of the year or the
start of the hour depending on which operating system was
in use) and writing it back out later.  I envision writing
out log files as I go so the database could be reconstructed
in case of a power failure, etc.

> I hope this helps you!

     Yes.  At least it makes me think and reconsider my
opinions and prejudices.


> The right tool for the right job.

     It is hard to argue with that.  On the other hand,
it is not impossible.  There are costs to sticking entirely
with Clipper.  There are costs to change.  Which will be
best considering the short term and the long term?

> The right tool for the right job.

     It seems possible that there could be a language,
possibly Icon or Python, or possibly a combination of two
languages, that might generally be good enough for most
programming tasks even if it wasn't the absolute best
for any single one of those tasks, and not _too_ bad
for all the other tasks so that I would be better off
doing _everything_ in that one language (or two) when
you consider the efficiencies of working with just one
or two languages instead of 30 (fill in the blank) perfect
languages each perfect for its own task.  This is just
a conjecture, but it seems likely to me to be true.


  -- Frank  (in sunny and hot San Marcos, Texas)

  frank.sergeant@pobox.com
  http://www.eskimo.com/~pygmy   (for the Stress Testing article)


From icon-group-sender  Wed Jun 24 08:08:34 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id IAA01691
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Wed, 24 Jun 1998 08:08:34 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA13548; Wed, 24 Jun 1998 08:08:23 -0700
From: pygmy@eskimo.com (Frank Sergeant)
To: icon-group@baskerville.CS.Arizona.EDU
Subject: Re: using Icon for database application
Date: Tue, 23 Jun 1998 20:03:06 -0500
Reply-To: frank.sergeant@pobox.com
Message-Id: <KBFk1Yv1uYeV084yn@eskimo.com>
In-Reply-To: <199806222139.QAA25152@segfault.cs.utsa.edu>
Lines: 179
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 7986

Clint Jeffery, jeffery@cs.utsa.edu wrote:

> While I disagree with
> Gordon Peterson's answers to your questions on many or most points,

     I'm glad to hear that.  Even if Gordon is correct in all his
points, I'm going to have to work toward them slowly.  He does
bring up (albeit politely) the possibility, worth considering,
that I am out of my mind.

> I've done a lot more Icon
> programming than Gordon, who frequently posts about Snobol in the Icon
> group, bless his heart!

     All I know (?) about Snobol was that I'd heard over the
years that it was "good for string processing".  Now I gather it
is the father, or favorite uncle, of the Icon language.

> Icon isn't a special-purpose database language.  Its built-in types are
> designed for flexible data manipulation in main memory.  If that's what
> you want to do, you are all set;

     When you put it that way, I think it is clear that that _is_
what I want to do.  (Also, it should be perfectly suitable for
sequetial file processing, right?  That should work very well for
producing reports.)  There remains the question as to whether
the data will actually fit in RAM so that I _can_ do what I want
to do.  I'm inclined to think it will fit, but I need to do some
tests.

> if not, you should look for a database language, or think about
> connecting Icon to a third-party database engine. There's more on
> memory requirements in your question 9 below.

     I am now thinking about using the variable length,
delimited format with all the data in RAM.  If that doesn't
work out, I gather from a later comment you made that I
could write extension routines in C as modifications to the
Icon source.  I think I could do that to give Icon
direct/random access to database files.  I am set up here
with MSVC/C++, BC/C++, Watcom, and several other C compilers
(which strikes me as very strange for someone who definitely
is not a fan of C -- oh well).  Of course, I suspect there
are details to study in order to coordinate properly with
Icon, so I'd rather not rush into this.  However, I'm
familiar with the C and Win32 file access routines.  Another
possibility would seem to be doing the same thing to get
access to Win32 socket routines and use them to communicate
with a (possibly bare-bones) database server, written in
anything, possibly C.  So, I am encouraged to think the
straight forward, all Icon, data in RAM approach might work
fine but that if it doesn't, I have not necessarily painted
myself into a corner.

> >	2. I gather Windows Icon can make a normal looking
> >	   Windows application (rather than a Motif-ish
> >	   application) and that I probably could figure out
> >	   how to do this by reviewing the source code for Wi.

> Well, you shouldn't have to learn it by studying source code.

     You understood I meant to use the source for Wi as
an example of how to put up the various menus, entry fields,
and such, right?  That is, the menus and such in the Wi
application look "normal" for Windows, whereas those in VIB
do not.  Since Wi is written in Icon, I figured it must be
possible to get that "normal" Windows look and that Wi
would show how.

> Both the on-line reference and an appendix in "Graphics
> Programming in Icon" discuss these functions, which are quite
> simple.

     I have the book on order.  I'm very much looking
forward to it.  At the moment, Python and Icon are largely
tied in my mind.  I would be delighted to hear more comments
that might help me understand their relative tradeoffs.  One
thing that worries me about Python is that its preferred GUI
seems to be Tk.  I used Tk a little (with Tcl) several years
ago, prior to it getting a "native look and feel" and wasn't
crazy about it.  Also, a demo of a extension to Tk, PMW
(Python Mega Widgets), seemed way too slow.  So, that Icon
might have an easy to use, fast GUI makes me think I will
like it better than Python with Tk.

> Icon's graphics are plenty fast enough for 486 clients, but
> Icon would not be suitable for the server software unless
> either (a) enough main memory was available to fit all the
> indices and preferably the data as well, or

     How does keeping just the indices in RAM solve the
problem if Icon does not have direct access to the data
file?

> One wart I know of at present: if you use tons of different
> record types you get penalized (this is true, for example,

     I guess I would have a record type for each database
plus other record types for anything I could treat as an
object.  On the bright side, you might well have this fixed
before I'm ready to put the app in production.

> I think the Icon program would be larger, unless your
> rewrite substantially reduces the number of lines of code to
> exploit Icon's built-in features.  If you just translated
> xBase line-by-line into Icon the result will be huge.
> Clipper is (was?) a fine commercial-quality product.

     I think I could knock perhaps a third off the source
code side by rewriting it properly in Clipper.  I would hope
to take advantage of Icon's features to make it even
smaller.  All this is guess work until I try it out.

> Ah, its time for some vaporware.  Sockets are only
> available on Unix at the present time.
 < calling Win32 routines from Icon >
> I keep getting nibbles, but so far no one has volunteered to implement this
> kind of access.  You can get the Windows Icon source and make such
> extensions yourself, and if you build something really useful I'd certainly
> like to incorporate it into the main public domain Windows Icon
> distribution.

     Ok.  I find this very encouraging.  As slowly as I am
moving, you may well have it all ready before I need it.
Or, if I do need it sooner, I might be able to write
specific calls to the socket routines.  I'll be glad to send
you anything I do in this area, but I'm a long way from it
at the moment.

     Another idea that occurred to me, if only as a proof of
concept, if sockets aren't available, would be to have the
client and server communicate via files on a RAM disk on the
server.  Another possibility might be to have the server
generate HTML and let the clients be any old web browsers.
(I have some user interface concerns here, though.)

> The main Idol web site is now www.cs.utsa.edu/research/idol/ and I am
> in the process of packaging a release of Windows Idol.  You should consider
> using Idol if you want to write object-oriented applications or have a large
> complex application with many user-defined types.

     Thanks.  I'll take a look when it is ready.

> Icon is generally tuned for speed, rather than to minimize space
> requirements.  I've heard reports of Unix Icon programs in the million line
> range, but haven't seen a Windows Icon program in the 100K range yet.

     If things go well, my rewrite would be under 100K, so
that sounds ok.

> My impression is that Windows manages virtual memory much
> worse than UNIX or Linux.

     I don't doubt it.  I'm rather annoyed with Windows at
the moment, so I'm ready to believe any bad thing about it.
Although, W95 almost never crashes on me any more now that
I fixed my hardware.  Prior to that, it would crash perhaps
one or more times a day.  But, at that same time, with the
defective hardware, I would usually run Linux for more hours
per day on the same machine and Linux would essentially
_never_ crash on me.  So, my conclusion is that with
equivalent hardware Linux is the better operating system
and goes out of its way to take care of you and your data
while Windows does not.

> On old 486's with 8MB of memory you may run into terrible
> problems unless you split the application up into a thin
> "client" program that can access a server with more muscle.

     I think I can do that.  It would be easier to say a
customer must have _one_ fast machine as a server, than to
say _every_ machine in the office must be a fast machine.

     Thanks for your other comments as well.


  -- Frank
  frank.sergeant@pobox.com



From icon-group-sender  Thu Jun 25 08:29:42 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id IAA12971
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Thu, 25 Jun 1998 08:29:42 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA31216; Thu, 25 Jun 1998 08:29:31 -0700
From: pygmy@eskimo.com (Frank Sergeant)
To: icon-group@baskerville.CS.Arizona.EDU
Subject: Re: using Icon for database application
Date: Wed, 24 Jun 1998 21:42:41 -0500
Reply-To: frank.sergeant@pobox.com
Message-Id: <hkbk1Yv1uAmb084yn@eskimo.com>
In-Reply-To: <KBFk1Yv1uYeV084yn@eskimo.com>
Lines: 28
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 933

I wrote:

> > Icon's graphics are plenty fast enough for 486 clients, but
> > Icon would not be suitable for the server software unless
> > either (a) enough main memory was available to fit all the
> > indices and preferably the data as well, or
> 
>      How does keeping just the indices in RAM solve the
> problem if Icon does not have direct access to the data
> file?

     What led me to the above comment was

     "Input and output in Icon are sequential and line oriented.
      There are neither random access facilities nor format
      conversions."

on page 88 of _The Icon Programming Language_ (1st edition?
anyway dated 1983).  I realize I'm reading an out of date book
which describes version 5, but I saw a comment or two on the list
that I misunderstood as confirming the above quote.  Now I realize
Icon has seek() which solves my concerns about random access to
files.


  -- Frank
  frank.sergeant@pobox.com



From icon-group-sender  Thu Jun 25 12:22:16 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id MAA06255
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Thu, 25 Jun 1998 12:22:16 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA30605; Thu, 25 Jun 1998 12:22:05 -0700
Message-Id: <35927650.3744@gte.net>
Date: Thu, 25 Jun 1998 11:09:52 -0500
From: MJE <evans@gte.net>
Reply-To: evans@gte.net
Organization: None
X-Mailer: Mozilla 3.01 (Win95; I)
Mime-Version: 1.0
To: icon-group@baskerville.CS.Arizona.EDU
Subject: Icon Variants
References: <199806222139.QAA25152@segfault.cs.utsa.edu>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 535

Dear Group,

Would someone who is knowledgeable please provide a list of all current
variants on the Icon language, including (1) web URL's, (2) platform
support data, (3) how many people are working on each and within what
time frame (i.e. current alpha/beta/release status), and (4) likelihood
of future incorporation into the "official" Icon.

I have seen blurbs about Godiva, Jcon, Unicon, and on and on, but I
don't have a clue whether these projects are simply tangents or
fundamental to the future of Icon itself.

Mark Evans



From icon-group-sender  Fri Jun 26 15:40:05 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id PAA16803
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Fri, 26 Jun 1998 15:40:04 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA30768; Fri, 26 Jun 1998 15:39:53 -0700
Sender: rwg@fns.com
Message-Id: <3593FF9A.4A241736@fns.com>
Date: Fri, 26 Jun 1998 16:07:54 -0400
From: Robbie Gilbert <rwg@fns.com>
Organization: Fujitsu Network Communications
X-Mailer: Mozilla 4.05 [en] (X11; I; SunOS 5.5.1 sun4u)
Mime-Version: 1.0
To: icon-group@optima.CS.Arizona.EDU
Subject: Forcing Generator to Return a List of Values
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 1016

Hi,

I have a newbie question:

I would like to know if there is a function or procedure
that will repeatedly evaluate an expression which is a
generator and return ALL the results in a list.  I could
write such a procedure very easily (maybe only 2
lines?) using the every-do control structure but if there is
one already in the program library, I would prefer to use
the standard.

I found evallist but that seems to construct an entire
program and exec it which seems expensive for my purposes.
I also checked list construction and type conversion to see
if there was a way to "force" a list context around an
expression evaluation but couldn't find any.  I want to
assign the list of values and since assignment only takes
the first value, things like "| expr" only seem to produce
one value.  I need some intervening procedure or function to
"collect" all the values generated.  The list constructor
doesn't do the job since it only evaluates its second
argument, the expr, once.

Robbie Gilbert
rwg@fns.com




From icon-group-sender  Fri Jun 26 15:39:37 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id PAA16608
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Fri, 26 Jun 1998 15:39:36 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA18130; Fri, 26 Jun 1998 15:39:25 -0700
Message-Id: <s5938445.031@housmtp.oceaneering.com>
X-Mailer: Novell GroupWise 4.1
Date: Fri, 26 Jun 1998 16:19:56 -0500
From: Charles Hethcoat <CHETHCOA@oss.oceaneering.com>
To: icon-group@optima.CS.Arizona.EDU
Subject: TeX editor (Newsletter #47)
Mime-Version: 1.0
Content-Type: text/plain
Content-Disposition: inline
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 621

Hi,

I was reviewing some old Icon newsletters and came across the item in
Newsletter 47 about some student projects in graphics programming.  There
is mention that an unnamed student wrote a TeX editor.  I found no further
references to this program.

I could use a graphical TeX editor.  Is this one available?  I would be
very interested in taking a look at the source code and trying to run it.
If it's incomplete that doesn't bother me as long as the basic design is
good.

Thanks.
		
Charles Hethcoat
Oceaneering Space Systems, Inc.
16665 Space Center Blvd., Houston, Texas 77058 USA
chethcoa@oss.oceaneering.com



From icon-group-sender  Mon Jun 29 08:41:51 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id IAA06745
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Mon, 29 Jun 1998 08:41:51 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA29145; Mon, 29 Jun 1998 08:41:39 -0700
Date: Sun, 28 Jun 1998 22:37:17 -0500
Message-Id: <199806290337.WAA12295@segfault.cs.utsa.edu>
From: Clinton Jeffery <jeffery@segfault.cs.utsa.edu>
To: rwg@fns.com
Cc: icon-group@optima.CS.Arizona.EDU
In-Reply-To: <35970836.34465163@fns.com> (message from Robbie Gilbert on Sun,
	28 Jun 1998 23:21:26 -0400)
Subject: Re: Forcing Generator to Return a List of Values
Reply-To: jeffery@segfault.cs.utsa.edu
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 1299

[Robbie Gilbert provides a helper procedure and uses co-expression invocation
p{} to solve his problem of turning a generator's result sequence into a list.]

Hi Robbie,

You've got some coding problems to fix, but when you fix them the general
approach you describe will work and is attractively concise.  However, it is
not cheap.  If you need to do this 10 times in your program, you have a
great solution.  If you need to do it many thousands of times, you had best
not create a new co-expression every time you capture a generator's results.
The reason is that each co-expression requires substantial memory for its
own stack, and the memory for co-expressions is not allocated and reclaimed
in the same, efficient way that most Icon values are.

This is just my impression.  I am a big co-expression advocate and would
welcome corrections and anecdotes from persons who have happily used
thousands of co-expressions in a single program!

Remember, you are only going to save one line of code over:
	L := []
	every put(L, generator_expression)

Perhaps if you do an experiment and collect some timings, that will influence
you!

Cheers,

Clint Jeffery, jeffery@cs.utsa.edu
Division of Computer Science, The University of Texas at San Antonio
Research http://www.cs.utsa.edu/research/plss.html


From icon-group-sender  Mon Jun 29 08:42:08 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id IAA06781
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Mon, 29 Jun 1998 08:42:06 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA30182; Mon, 29 Jun 1998 08:41:55 -0700
Message-Id: <199806291302.PAA27566@capway.com>
From: "Vladimir Grodzenski" <grodzens@capway.com>
To: icon-group@optima.CS.Arizona.EDU
Date: Mon, 29 Jun 1998 15:02:12 +0000
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7BIT
Subject: Bug in FillRectangle function ?
Reply-To: <vladimir.grodzenski@capway.com>
Cc: jeffery@cs.utsa.edu
Priority: urgent
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 1282

(I am using the last Windows Icon downloaded from -UTSA- FTP site)

Here is just a very quick note on one example from Graphics 
book which uses "drawop=reverse" attribute (i.e. drawing the same 
figure twice in "reverse" mode must erase the figure).

Page 88.  (or on the CD, Examples Chapter 4, glide.icn), the 
following example does not work:

WAttrib("drawop=reverse")
every x := 1 to 100 do {
   FillRectangle(x, 100, 10, 20)
   WDelay(1)
   FillRectangle(x, 100, 10, 20)
   }

Here the rectangle is simply not filled!

The effect may be obtained with the different code:

every x := 1 to 100 do {
   WAttrib("drawop=copy")
   FillRectangle(x, 100, 10, 20)
   WDelay(1)
   WAttrib("drawop=reverse")
   FillRectangle(x, 100, 10, 20)
   }

I also tried the DrawCircle(), DrawRectangle() and even 
FillCircle() instead of FillRectangle(). All of these functions 
worked fine in the way described in the Graphics book.

Well, assuming that FillRectangle is just buggy and all Draw***() 
functions work with "drawop=reverse" attribute, I am surprised that 
DrawString() does not. Why doesn't it?


Vladimir Grodzenski
=================================================
E-mail: vladimir.grodzenski@capway.com
CompuServe: 100700,526
=================================================


From icon-group-sender  Mon Jun 29 08:41:34 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id IAA06696
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Mon, 29 Jun 1998 08:41:33 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA21982; Mon, 29 Jun 1998 08:41:22 -0700
Sender: rwg@fns.com
Message-Id: <35970836.34465163@fns.com>
Date: Sun, 28 Jun 1998 23:21:26 -0400
From: Robbie Gilbert <rwg@fns.com>
Organization: Fujitsu Network Communications
X-Mailer: Mozilla 4.05 [en] (X11; I; SunOS 5.5.1 sun4u)
Mime-Version: 1.0
To: jeffery@segfault.cs.utsa.edu
Cc: icon-group@optima.CS.Arizona.EDU
Subject: Re: Forcing Generator to Return a List of Values
References: <199806270618.BAA10281@segfault.cs.utsa.edu>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 1928

Clinton Jeffery wrote:

> Hi,
>
> I am sure you got deluged with answers, so I will be brief.  Yes, it is
> useful occasionally to collect all the results from a generator and put them
> in a list, and no, there isn't a standard library function.  In fact, if you
> think about it, a library procedure would not help you.  You need to write
> the code yourself, either in your own helper procedure, or directly in the
> place where you've got the generator and need the list.  Fortunately, as you
> say, its only a couple of lines.
>
> Clint Jeffery

Clint,

Thanks for the response.  Actually yours was the only one I've gotten so far.
I think a procedure can help; how about:

     procedure collectAll (expressions)
       local results
       results := []

       every C := !expressions do
         every result := |@C do
           results |||:= [result]

       return results
     end

This procedure is invoked using braces as explained on page 116 of the The Icon
Programming Language (3rd ed.), in the section on user-defined control
structures:

     # args is some list
     foo := collectAll { ! args }

This procedure takes a comma-separated list of co-expressions (in this example,
there is only one, !args), grabs all the results and returns them in a list.
(This may need some refinement to work properly for all kinds of expressions,
I just finished writing it so I haven't tested it thoroughly yet.  Also the
list concatenation may not be very efficient--there may be a better way but I'm
new to this language.)

Hopefully I can pass any valid expression as an arg and get a list of the
results back.  This is what I was trying to accomplish and seems to be fairly
general.  This seems like a natural, common task and given Icon's strengths in
this type of thing I thought there would already be something like this in the
program library--didn't want to reinvent the wheel.

Thanks,

Robbie
rwg@fns.com




From icon-group-sender  Mon Jun 29 08:41:59 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id IAA06764
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Mon, 29 Jun 1998 08:41:59 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA00698; Mon, 29 Jun 1998 08:41:47 -0700
Sender: rwg@fns.com
Message-Id: <35970F72.40125E54@fns.com>
Date: Sun, 28 Jun 1998 23:52:18 -0400
From: Robbie Gilbert <rwg@fns.com>
Organization: Fujitsu Network Communications
X-Mailer: Mozilla 4.05 [en] (X11; I; SunOS 5.5.1 sun4u)
Mime-Version: 1.0
To: jeffery@segfault.cs.utsa.edu
Cc: Icon Mailing List <icon-group@optima.CS.Arizona.EDU>
Subject: Re: Forcing Generator to Return a List of Values
References: <199806290337.WAA12295@segfault.cs.utsa.edu>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 2116

Clint,

Thanks for the input!  Yes, I've already discovered some coding problems and am
working on fixing them.  This is only for a small (as you say, around 10) number
of iterations, so it shouldn't be too expensive in that sense, but being a newbie
I didn't realize co-expressions were memory-intensive.

I realized I was only saving one line of code but I hate to use a line of code for
something as trivial as initialization. Collecting all the values and assigning
them seems such a natural operation, especially if this is an operation that can
be reused.  As you say, though, the performance hit may not justify the
succintness of the expression.  I will do some timings and see.

Thanks again,

Robbie
rwg@fns.com

Clinton Jeffery wrote:

> [Robbie Gilbert provides a helper procedure and uses co-expression invocation
> p{} to solve his problem of turning a generator's result sequence into a list.]
>
> Hi Robbie,
>
> You've got some coding problems to fix, but when you fix them the general
> approach you describe will work and is attractively concise.  However, it is
> not cheap.  If you need to do this 10 times in your program, you have a
> great solution.  If you need to do it many thousands of times, you had best
> not create a new co-expression every time you capture a generator's results.
> The reason is that each co-expression requires substantial memory for its
> own stack, and the memory for co-expressions is not allocated and reclaimed
> in the same, efficient way that most Icon values are.
>
> This is just my impression.  I am a big co-expression advocate and would
> welcome corrections and anecdotes from persons who have happily used
> thousands of co-expressions in a single program!
>
> Remember, you are only going to save one line of code over:
>         L := []
>         every put(L, generator_expression)
>
> Perhaps if you do an experiment and collect some timings, that will influence
> you!
>
> Cheers,
>
> Clint Jeffery, jeffery@cs.utsa.edu
> Division of Computer Science, The University of Texas at San Antonio
> Research http://www.cs.utsa.edu/research/plss.html





From icon-group-sender  Mon Jun 29 08:41:14 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id IAA06667
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Mon, 29 Jun 1998 08:41:14 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA31475; Mon, 29 Jun 1998 08:41:02 -0700
X-Sender: nevin@pop.interaccess.com (Unverified)
Message-Id: <v04003a04b1ba60b6229a@[204.148.87.119]>
In-Reply-To: <01bda15c$16b0f180$c37546cf@NEVIN>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Date: Sat, 27 Jun 1998 03:51:31 -0500
To: icon-group@optima.CS.Arizona.EDU
From: "Nevin :-] Liber" <nevin@pendragon-software.com>
Subject: 
 Re:
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 943

Robbie Gilbert <mailto:rwg@fns.com> wrote:

>I would like to know if there is a function or procedure
>that will repeatedly evaluate an expression which is a
>generator and return ALL the results in a list.  I could
>write such a procedure very easily (maybe only 2
>lines?) using the every-do control structure but if there is
>one already in the program library, I would prefer to use
>the standard.

I don't know if there is a standard one in the IPL, but as you pointed out,
it is fairly trivial to write:

	procedure genlist(pGenerator, LArguments[])

		local   L

		L := list()
		every put(L, pGenerator ! LArguments)

		return L

	end

--
 Nevin ":-)" Liber  <mailto:nevin@pendragon-software.com>  (847) 816-9926
  Senior Software Engineer, Pendragon Software <http://www.webfayre.com/>

	Pendragon Software Corporation
	1580 South Milwaukee Avenue, Suite 515
	Libertyville, IL 60048-3776
	voice:	(847) 816-9660
	fax:	(847) 816-9710




From icon-group-sender  Mon Jun 29 08:41:21 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id IAA06676
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Mon, 29 Jun 1998 08:41:21 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA01032; Mon, 29 Jun 1998 08:41:09 -0700
Message-Id: <35970078.4332DA12@ix.netcom.com>
Date: Sun, 28 Jun 1998 22:48:24 -0400
From: Phillip Lee Thomas <teruthom@ix.netcom.com>
Reply-To: thomaspl@acm.org
X-Mailer: Mozilla 4.05 [en] (Win95; U)
Mime-Version: 1.0
To: icon-group@optima.CS.Arizona.EDU
Subject: Retrieving lists from tables
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 575

I'm working on a project where it would be most convenient to store
information in a table where the index is a list and the value is a
record; however, I find it not obvious how to retrieve the data as lists
are retrieved by their ID number, not their value. Any suggestions?

record ref(part1,part2)
# Initially...
        t := table()
        L := ["house","horse"]
        t[L] := ref(1,2)
# Later...
       L2 := ["house","horse"]
Now I want to find out if a list with the content of L2 is in table t,
and if so to augment the value of its part1.

-- Phillip Thomas





From icon-group-sender  Mon Jun 29 16:24:34 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id QAA25642
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Mon, 29 Jun 1998 16:24:34 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA01832; Mon, 29 Jun 1998 16:24:22 -0700
From: Guy Laden <guy@math.tau.ac.il>
Message-Id: <199806291912.WAA25423@orion.math.tau.ac.il>
Subject: Re: Retrieving lists from tables
To: thomaspl@acm.org
Date: Mon, 29 Jun 1998 22:12:36 +0300 (GMT+0300)
Cc: icon-group@optima.CS.Arizona.EDU
In-Reply-To: <35970078.4332DA12@ix.netcom.com> from "Phillip Lee Thomas" at Jun 28, 98 10:48:24 pm
X-Mailer: ELM [version 2.4 PL25]
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 943

> I'm working on a project where it would be most convenient to store
> information in a table where the index is a list and the value is a
> record; however, I find it not obvious how to retrieve the data as lists
> are retrieved by their ID number, not their value. Any suggestions?
>
> record ref(part1,part2)
> # Initially...
>         t := table()
>         L := ["house","horse"]
>         t[L] := ref(1,2)
> # Later...
>        L2 := ["house","horse"]
> Now I want to find out if a list with the content of L2 is in table t,
> and if so to augment the value of its part1.

I'd be happy to hear of solutions from experienced Icon programmers. My 
solution in a similar case was to create a string out of the list and to
use it as key. I didn't have to handle arbitrary strings in my application
so encoding the list as a string was simple. Perhaps the codeobj module
from the IPL can be used if more generality is needed.

Regards,
Guy


From icon-group-sender  Mon Jun 29 16:24:27 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id QAA25635
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Mon, 29 Jun 1998 16:24:26 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA01169; Mon, 29 Jun 1998 16:24:14 -0700
Message-Id: <3597BA27.532@gte.net>
Date: Mon, 29 Jun 1998 11:00:39 -0500
From: MJE <evans@gte.net>
Reply-To: evans@gte.net
Organization: None
X-Mailer: Mozilla 3.01 (Win95; I)
Mime-Version: 1.0
To: icon-group@optima.CS.Arizona.EDU
Subject: Re: Retrieving lists from tables
References: <35970078.4332DA12@ix.netcom.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 876

Would a two-way table help?  You would have to maintain a second table
indexed by record that stores the list keys.  Every time you do an
action on the first table, you have to do it also on the second one to
keep them synchronized.

Mark



Phillip Lee Thomas wrote:
> 
> I'm working on a project where it would be most convenient to store
> information in a table where the index is a list and the value is a
> record; however, I find it not obvious how to retrieve the data as lists
> are retrieved by their ID number, not their value. Any suggestions?
> 
> record ref(part1,part2)
> # Initially...
>         t := table()
>         L := ["house","horse"]
>         t[L] := ref(1,2)
> # Later...
>        L2 := ["house","horse"]
> Now I want to find out if a list with the content of L2 is in table t,
> and if so to augment the value of its part1.
> 
> -- Phillip Thomas



From icon-group-sender  Mon Jun 29 16:25:14 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id QAA25709
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Mon, 29 Jun 1998 16:25:14 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA29304; Mon, 29 Jun 1998 16:25:02 -0700
Date: Mon, 29 Jun 1998 14:06:04 -0700
From: Gregg Townsend <gmt>
Message-Id: <9806292106.AA22085@hawk.CS.Arizona.EDU>
To: icon-group
Subject: Re:  Icon Variants
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 1612

[ In response to Mark Evans' query about the various versions of Icon: ]

We (Ralph Griswold and Gregg Townsend) maintain the "master source copy"
of Icon at the University of Arizona.  The source code is distributed as
part of packages configured for particular platforms.  Periodic updates
are available by subscription.

We also produce and distribute the source and binary packages for Unix
and provide web/FTP space for other platforms.  The Icon web site is at
	http://www.cs.arizona.edu/icon/
and it includes a list of versions available for various platforms.

We regard Icon as a mature language and we have no plans for major
changes or additions.

Clint Jeffery of UT San Antonio maintains the Windows version of Icon.
This is also a released product, but work continues on a more active
basis.  See
	http://www.cs.utsa.edu/research/icon/
Clint also distributes Idol, an object-oriented variant of Icon.

Unicon, from Shamim Mohamed, adds many POSIX-inspired built-in functions
to Unix Icon.  It is available through his web page at
	http://www.drones.com/unicon/

Todd Proebsting and Gregg Townsend are actively working on Jcon, a new
Java-based implementation of Icon, and intend to release a new and faster
version later this year.  We don't know where this will ultimately lead.
See
	http://www.cs.arizona.edu/icon/jcon/

These are the versions of Icon that have been in the news lately.
There may be other things going on that we're not aware of.

   ---------------------------------------------------------------------------
   Gregg Townsend  
   Ralph Griswold
   The University of Arizona



From icon-group-sender  Tue Jun 30 08:00:15 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id IAA28489
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Tue, 30 Jun 1998 08:00:15 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA02125; Tue, 30 Jun 1998 08:00:03 -0700
Message-Id: <35983EEB.F39CA405@ix.netcom.com>
Date: Mon, 29 Jun 1998 21:27:08 -0400
From: Phillip Lee Thomas <teruthom@ix.netcom.com>
Reply-To: thomaspl@acm.org
X-Mailer: Mozilla 4.05 [en] (Win95; U)
Mime-Version: 1.0
To: icon-group@optima.CS.Arizona.EDU
Subject: Re: Retrieving lists from tables
References: <199806300034.TAA13464@segfault.cs.utsa.edu>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 560

Thanks to Mark Evans, R. Clayton, Shamim Mohamed, Guy Laden, and Clint Jeffery
for your very acute replies.

It looks like double-entry bookkeeping has found its place in the solution to
this problem. In the particular instance there appears to be little to help
avoiding the maintenance of both string and list representations during the
entire life of the processing.

The IPL has a list function l_find that does match one list as a substring of
another but that doesn't help much in table lookups.

OK. Time to ponder your collective insights.

Phillip




From icon-group-sender  Tue Jun 30 08:00:26 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id IAA28495
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Tue, 30 Jun 1998 08:00:25 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA01984; Tue, 30 Jun 1998 08:00:13 -0700
Message-Id: <35985E94.B8AC0267@ix.netcom.com>
Date: Mon, 29 Jun 1998 23:42:12 -0400
From: Phillip Lee Thomas <teruthom@ix.netcom.com>
Reply-To: thomaspl@acm.org
X-Mailer: Mozilla 4.05 [en] (Win95; U)
Mime-Version: 1.0
To: icon-group@optima.CS.Arizona.EDU
Subject: Re: Retrieving lists from tables
References: <35970078.4332DA12@ix.netcom.com> <3597BA27.532@gte.net>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 645

After pondering the solutions sent in, the following appears to involve the
least processing without making the code absolutely undebugable:

record ref(L,part1,part2)  # keep list as first element of record
.....
         t := table()
         hashval := hash(L)
         if \t[hashval] then
           t[hashval].part1 +:= 1
         else
           t[hashval] := ref(L,1,1)
....

# later retrieval of the list for further processing can be done through
#    converting the table to a sorted list or whatever is appropriate.

procedure hash(L)
    # return some string unique to each list L
end

Thanks again to all who responded.

Phillip




From icon-group-sender  Tue Jun 30 08:00:45 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id IAA28503
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Tue, 30 Jun 1998 08:00:45 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA00768; Tue, 30 Jun 1998 08:00:32 -0700
Date: Tue, 30 Jun 1998 08:23:23 -0500 (CDT)
From: Chris Tenaglia <cdt@post.its.mcw.edu>
To: icon-group@optima.CS.Arizona.EDU
Subject: sample CGI
Message-Id: <Pine.SOL.3.96.980630081906.29720A-100000@post.its.mcw.edu>
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 2869

This CGI stuff is a snap once you get the hang of it. Here's a
sample. This is a CGI to return the time remainting till the
turn of the century. It also records visits in a logfile.
I was able to run it at http://www.mcw.edu/cgi-bin/bioet/cdown
where it will be for a few days. As usual the file permissions
are the most sensitive issue. Here's the source;
_______________________________________________________________

#
# file : cdown.icn
# desc : This program returns countdown information till
#        the turn of the century
# use  : to be called by a cgi script in a web page : .../cgi-bin/cdown
# note : 730 days from 1/1/1998 to 1/1/2000
#
# update          by          what
# 29-jun-1998     tenaglia    initial write
#
procedure main()
  write("content-type: text/html")
  write()
  write("<HTML>")
  write("<HEAD>")
  write("<TITLE>")
  write("Century Countdown")
  write("</TITLE>")
  write("</HEAD>")
  write("<BODY BGCOLOR=#ffffff>")
  write("<CENTER><FONT COLOR=#c00000>")
  write("<H1>Century CountDown !</H1>")
  write("<P>")

  write("<TABLE border=10>")
  write("<TH><H3><FONT COLOR=#800080>")
  write("Today is ",&dateline,"</TH>")

#
# gather the time fields from the current moment
#
  date := &date
  time := &clock
  delta:= [0,31,59,90,120,151,181,212,243,273,304,334,365]

  year := parse(date,'/')[1]
  month:= parse(date,'/')[2]
  day  := parse(date,'/')[3]

  hour := parse(time,':')[1]
  min  := parse(time,':')[2]
  sec  := parse(time,':')[3]

#
# calculate the time left until 1/1/2000
#
  diff    := (year - 1998) * 365 + delta[month] + day
  remains := 730 - diff
  part1   :=  24 - hour
  part2   :=  60 - min
  part3   :=  60 - sec

  write("<TR><TD><FONT COLOR=#800080><CENTER>There are ",remains," days,</CENTER></TD></TR>")
  write("<TR><TD><FONT COLOR=#800080><CENTER>And ",part1," hours,</CENTER></TD></TR>",
        "<TR><TD><FONT COLOR=#800080><CENTER>",part2," minutes,<CENTER></TD></TR>",
        "<TR><TD><FONT COLOR=#800080><CENTER>",part3," seconds until 2000 A.D.</CENTER></TD></TR>")
  write("</FONT></TABLE>")
  write("<P></FONT>")
  write("</BODY>")
  write("</HTML>")

#
# record visits
#
  computer:= getenv("REMOTE_ADDR")
  logfile := open("cdown.log","a")
  write(logfile,"CDOWN accessed ",&dateline," from ",computer)
  close(logfile)

  end

#
# parse a string into a list with respect to a delimiter
#
procedure parse(line,delims)
  static chars
  chars  := &cset -- delims
  tokens := []
  line ? while tab(upto(chars)) do put(tokens,tab(many(chars)))
  return tokens
  end

_______________________________________________________________
Chris Tenaglia   (system manager)     |  The future foretold,
Medical College of Wisconsin          |  The past explained,
8701 W. Watertown Plank Rd.           |  The present largely appologized for.
Milwaukee, WI 53226   (414)456-8765   |  Organon to the Doctor




From icon-group-sender  Tue Jun 30 16:30:43 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id QAA28910
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Tue, 30 Jun 1998 16:30:37 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA01012; Tue, 30 Jun 1998 16:30:25 -0700
Date: Tue, 30 Jun 1998 13:51:21 -0700
From: Ralph Griswold <ralph>
Message-Id: <9806302051.AA16354@jupiter.CS.Arizona.EDU>
To: vladimir.grodzenski@capway.com
Subject: Re:  Bug in FillRectangle function ?
Cc: icon-group
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 148

drawop=reverse does not work properly in Windows Icon.  You'll find this
and a few other things on pages 474-475 of the graphics programming
book.


From icon-group-sender  Mon Jul  6 08:16:31 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id IAA17672
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Mon, 6 Jul 1998 08:16:30 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA04087; Mon, 6 Jul 1998 08:16:17 -0700
Message-Id: <199807040137.UAA13830@newmonics.com>
X-Sender: whm@newmonics.com (Unverified)
X-Mailer: QUALCOMM Windows Eudora Pro Version 4.0
Date: Fri, 03 Jul 1998 18:43:06 -0700
To: icon-group@optima.CS.Arizona.EDU
From: "William H. Mitchell" <whm@newmonics.com>
Subject: Icon routine to convert UNIX internal time to ASCII?
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 204

Has anybody got an Icon routine to convert a UNIX internal time, measured
as the number of seconds since midnight GMT, 1/1/1970, into ASCII?  I
couldn't find anything in the IPL that seemed applicable.



From icon-group-sender  Tue Jul  7 17:00:37 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id RAA03697
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Tue, 7 Jul 1998 17:00:36 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA08522; Tue, 7 Jul 1998 17:00:23 -0700
Message-Id: <35A2750A.E12A0673@aasaa.ofe.org>
Date: Tue, 07 Jul 1998 13:20:42 -0600
From: David Starner <dstarner98@aasaa.ofe.org>
X-Mailer: Mozilla 4.05 [en] (Win95; I)
Mime-Version: 1.0
To: icon-group@optima.CS.Arizona.EDU
Subject: Stripping excessive spaces
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 549

I've just started programming in Icon, and I came across the problem of
removing all the starting and ending spaces from a string. This was my
solution, but I think it's terribly inefficent. Can some one help me
with a better way to do this?

procedure clip(s1)
	if type(s1) == "null" then fail
	s1 ?:= {
		tab(many(' ')) & tab(0)
		}
	s1 := reverse(s1)
	s1 ?:= {
		tab(many(' ')) & tab(0)
		}
	s1 := reverse(s1)
	return s1
end
-- 
David Starner - dvdeug@hotmail.com, dstarner98@aasaa.ofe.org
GURPS: http://www.geocities.com/TimesSquare/Ring/7895/


From icon-group-sender  Wed Jul  8 14:01:48 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id OAA02640
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Wed, 8 Jul 1998 14:01:47 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA10420; Wed, 8 Jul 1998 14:01:34 -0700
From: "Nevin :-] Liber" <nevin@pendragon-software.com>
To: <icon-group@optima.CS.Arizona.EDU>
Cc: "David Starner" <dstarner98@aasaa.ofe.org>
Subject: Re: Stripping excessive spaces
Date: Tue, 7 Jul 1998 19:55:41 -0500
Message-Id: <01bdaa0b$2191b000$c37546cf@NEVIN>
Mime-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: 7bit
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
Content-Length: 2099


>I've just started programming in Icon, and I came across the problem of
>removing all the starting and ending spaces from a string. This was my
>solution, but I think it's terribly inefficent. Can some one help me
>with a better way to do this?

The built-in function trim() does the bulk of what you want to do.  Check
out the documentation at
<http://www.cs.arizona.edu/icon/refernce/funcsz.htm#trim>.

If you are looking to express the algorithm as a one-liner, the following
should suffice:


    procedure clip(s, c)

        /c := ' '

        s := string(s) | runerr(103, s)
        c := cset(c) | runerr(104, c)

        return reverse(trim(reverse(trim(s, c)), c))

    end

If the "inefficiency" of reversing a string twice bothers you, you might
wish to write it as:

    procedure clip(s, c)

        /c := ' '

        s := string(s) | runerr(103, s)
        c := cset(c) | runerr(104, c)

        trim(s, c) ? {
            tab(many(c))
            return tab(0)
        }

    end

These are both a little more general than what you asked for, as you can
clip any arbitrary cset (which defaults to space) instead of just spaces.
Also, this behaves a little differently than your algorithm does, as it
aborts with a runtime error instead of just failing if s is not convertible
to a string (and c is not null and not convertible to a cset).

If you wanted to generalize this a little more, say to use different csets
for the left and right sides (defaulting the left clip to space if none is
provided, and defaulting the right clip to be the same as the left clip if
non is provided), you might write it as:

    procedure clip(s, cLeft, cRight)

        /cLeft := ' '
        /cRight := cLeft

        s := string(s) | runerr(103, s)
        cLeft := cset(cLeft) | runerr(104, cLeft)
        cRight := cset(cRight) | runerr(104, cRight)

        trim(s, cRight) {
            tab(many(cLeft))
            return tab(0)
        }
__
 Nevin ":-)" Liber  <mailto:nevin@pendragon-software.com>  (847) 816-9926
  Senior Software Engineer, Pendragon Software <http://www.webfayre.com/>



From icon-group-sender  Wed Jul  8 14:02:10 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id OAA02671
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Wed, 8 Jul 1998 14:02:09 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA10436; Wed, 8 Jul 1998 14:01:56 -0700
Date: Tue, 7 Jul 1998 21:42:48 -0500 (CDT)
From: Chris Tenaglia <cdt@post.its.mcw.edu>
To: David Starner <dstarner98@aasaa.ofe.org>
Cc: icon-group@optima.CS.Arizona.EDU
Subject: Re: Stripping excessive spaces
In-Reply-To: <35A2750A.E12A0673@aasaa.ofe.org>
Message-Id: <Pine.SOL.3.96.980707214057.20216A-100000@post.its.mcw.edu>
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 1117

try

new := reverse(trim(reverse(trim(str))))

Chris Tenaglia   (system manager)     |  The future foretold,
Medical College of Wisconsin          |  The past explained,
8701 W. Watertown Plank Rd.           |  The present largely appologized for.
Milwaukee, WI 53226   (414)456-8765   |  Organon to the Doctor


On Tue, 7 Jul 1998, David Starner wrote:

> Date: Tue, 07 Jul 1998 13:20:42 -0600
> From: David Starner <dstarner98@aasaa.ofe.org>
> To: icon-group@optima.CS.Arizona.EDU
> Subject: Stripping excessive spaces
> 
> I've just started programming in Icon, and I came across the problem of
> removing all the starting and ending spaces from a string. This was my
> solution, but I think it's terribly inefficent. Can some one help me
> with a better way to do this?
> 
> procedure clip(s1)
> 	if type(s1) == "null" then fail
> 	s1 ?:= {
> 		tab(many(' ')) & tab(0)
> 		}
> 	s1 := reverse(s1)
> 	s1 ?:= {
> 		tab(many(' ')) & tab(0)
> 		}
> 	s1 := reverse(s1)
> 	return s1
> end
> -- 
> David Starner - dvdeug@hotmail.com, dstarner98@aasaa.ofe.org
> GURPS: http://www.geocities.com/TimesSquare/Ring/7895/
> 



From icon-group-sender  Wed Jul  8 14:02:01 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id OAA02663
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Wed, 8 Jul 1998 14:02:01 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA09408; Wed, 8 Jul 1998 14:01:47 -0700
Date: Tue, 7 Jul 1998 18:22:02 -0700
From: Ralph Griswold <ralph>
Message-Id: <9807080122.AA26095@jupiter.CS.Arizona.EDU>
To: icon-group
Subject: Forwarded information request
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 1878

>From shawakfa@charlie.cns.iit.edu Sun Jul  5 22:54:08 1998
Received: from baskerville.CS.Arizona.EDU by jupiter.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA22602; Sun, 5 Jul 1998 22:54:05 -0700
Received: from optima.cs.arizona.edu (optima.CS.Arizona.EDU [192.12.69.5])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with ESMTP id WAA22879
	for <icon-project@baskerville.cs.arizona.edu>; Sun, 5 Jul 1998 22:46:11 -0700 (MST)
Received: from charlie.cns.iit.edu (root@charlie.cns.iit.edu [198.87.195.70])
	by optima.cs.arizona.edu (8.9.0.Beta5/8.9.0.Beta5) with SMTP id WAA19228
	for <icon-project@cs.arizona.edu>; Sun, 5 Jul 1998 22:46:09 -0700 (MST)
Received: from charlie.cns.iit.edu (as5121.dialup.iit.edu [198.87.167.62]) by charlie.cns.iit.edu (950413.SGI.8.6.12/950213.SGI.AUTOCF) via ESMTP id AAA23478 for <icon-project@cs.arizona.edu>; Mon, 6 Jul 1998 00:46:05 -0500
Message-Id: <35A065D7.3C850679@charlie.cns.iit.edu>
Date: Mon, 06 Jul 1998 00:51:19 -0500
From: Emad Shawakfa <shawakfa@charlie.cns.iit.edu>
Reply-To: shawakfa@charlie.cns.iit.edu
Organization: Illinois Institute of Technology
X-Mailer: Mozilla 4.05 [en] (Win95; I)
Mime-Version: 1.0
To: icon-project@optima.CS.Arizona.EDU
Subject: Calling Icon programs from other languages.
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Status: RO

Hello,

    I am currently a PhD student at IL Institute of Technology where I
am doing some natural language processing. Icon is a very good
programming language to deal with text and do parsing for any text inpu.
Is it possible for me to call a parser written in icon from other
programming languages such as VC++ 5.0? if not, are there any
translators that can translate from/to Icon source code?

Thanks,

- Emad Shawakfa
   PhD Student,
   Computer Science Dept,
   IL Institute of Technology
  shawakfa@charlie.cns.iit.edu




From icon-group-sender  Wed Jul  8 14:02:22 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id OAA02674
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Wed, 8 Jul 1998 14:02:16 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA28049; Wed, 8 Jul 1998 14:02:03 -0700
Message-Id: <199807081244.OAA00040@capway.com>
From: "Vladimir Grodzenski" <grodzens@capway.com>
To: David Starner <dstarner98@aasaa.ofe.org>
Date: Wed, 8 Jul 1998 14:44:08 +0000
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7BIT
Subject: Re: Stripping excessive spaces
Reply-To: <vladimir.grodzenski@capway.com>
Cc: icon-group@optima.CS.Arizona.EDU
Priority: urgent
In-Reply-To: <35A2750A.E12A0673@aasaa.ofe.org>
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 945

I'd write your procedure in this way:

procedure clip(s)
   s ? {
     tab(many(' '))
     return trim(tab(0))
    }
end


On  7 Jul 98 at 13:20, David Starner wrote:
> I've just started programming in Icon, and I came across the problem
> of removing all the starting and ending spaces from a string. This
> was my solution, but I think it's terribly inefficent. Can some one
> help me with a better way to do this?
> 
> procedure clip(s1)
>  if type(s1) == "null" then fail
>  s1 ?:= {
>   tab(many(' ')) & tab(0)
>   }
>  s1 := reverse(s1)
>  s1 ?:= {
>   tab(many(' ')) & tab(0)
>   }
>  s1 := reverse(s1)
>  return s1
> end
> -- 
> David Starner - dvdeug@hotmail.com, dstarner98@aasaa.ofe.org
> GURPS: http://www.geocities.com/TimesSquare/Ring/7895/
> 
> 

Vladimir Grodzenski
=================================================
E-mail: vladimir.grodzenski@capway.com
CompuServe: 100700,526
=================================================


From icon-group-sender  Wed Jul  8 14:01:45 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id OAA02622
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Wed, 8 Jul 1998 14:01:39 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA05744; Wed, 8 Jul 1998 14:01:25 -0700
Date: Tue, 7 Jul 1998 17:14:37 -0700
From: kwalker@sfo.harbinger.com (Ken Walker)
Message-Id: <199807080014.RAA02444@varda.premenos.com>
To: icon-group@optima.CS.Arizona.EDU, dstarner98@aasaa.ofe.org
Subject: Re: Stripping excessive spaces
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Md5: IBoK7ybSEBHiGKAtRZMTtw==
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 536

> Date: Tue, 07 Jul 1998 13:20:42 -0600
> From: David Starner <dstarner98@aasaa.ofe.org>
> 
> I've just started programming in Icon, and I came across the problem of
> removing all the starting and ending spaces from a string. This was my
> solution, but I think it's terribly inefficent. Can some one help me
> with a better way to do this?

Your code to remove leading spaces looks fine to me, but I would
use trim() to remove trailing spaces.


Ken Walker, kenneth.walker@sfo.harbinger.com
Harbinger Coporation, Concord, Ca. 94520



From icon-group-sender  Tue Jul 14 08:17:12 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id IAA10004
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Tue, 14 Jul 1998 08:17:08 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA26913; Tue, 14 Jul 1998 08:16:54 -0700
To: icon-group@optima.CS.Arizona.EDU
Date: Tue, 14 Jul 1998 10:15:40 +0200
From: Anders Holtsberg <andersh@maths.lth.se>
Message-Id: <35AB13AC.145D@maths.lth.se>
Organization: Lund University
Sender: icon-group-request@optima.CS.Arizona.EDU
Subject: Icon language: Sockets
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 515

Is networking supported in Icon? I tried to look in the book and its
index and in the FAQ but I can't find it. I found file and pipe support 
but not socket support. What I want to do is a small multiplayer game 
program.

-- Andy

== Anders Holtsberg =========================== andersh@maths.lth.se ==
   Department of Mathematical Statistics       Phone +46 46 222 4953
   Lund University                               Fax +46 46 222 4623
== Box 118, S-221 00 LUND, Sweden == http://www.maths.lth.se/matstat ==


From icon-group-sender  Tue Jul 14 16:29:09 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id QAA08073
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Tue, 14 Jul 1998 16:29:07 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA10864; Tue, 14 Jul 1998 16:28:53 -0700
From: "Mike Levine" <mike@aftergov.u-net.com>
To: <icon-group@optima.CS.Arizona.EDU>
Subject: RE: Icon language: Sockets
Date: Tue, 14 Jul 1998 20:24:54 +0100
Message-Id: <000001bdaf5d$14513e90$08c566c3@aftergov>
Mime-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: 7bit
X-Priority: 3 (Normal)
X-Msmail-Priority: Normal
X-Mailer: Microsoft Outlook 8.5, Build 4.71.2173.0
Importance: Normal
In-Reply-To: <35AB13AC.145D@maths.lth.se>
X-Mimeole: Produced By Microsoft MimeOLE V4.72.2106.4
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 1253

Good question. I am very interested in this too.

Games, especially Multi-user (MU*) types are used in many
fields of
Psychology and Sociology and are written in C, C++ to parse
multi-user
input (100+ simultaneous socket connections). The language
efficiency
of 3GL/4GL programming language text processing has been the
subject of debate for some time and I would like to conduct
research of ICON and traditional languages within this niche.

Kind Regards
Mike Levine




> -----Original Message-----
> From: icon-group-request@CS.Arizona.EDU
> [mailto:icon-group-request@CS.Arizona.EDU]On Behalf
> Of Anders Holtsberg
> Sent: Tuesday, July 14, 1998 9:16 AM
> To: icon-group@optima.CS.Arizona.EDU
> Subject: Icon language: Sockets
>
>
> Is networking supported in Icon? I tried to look in
> the book and its
> index and in the FAQ but I can't find it. I found
> file and pipe support
> but not socket support. What I want to do is a
> small multiplayer game
> program.
>
> -- Andy
>
> == Anders Holtsberg ===========================
> andersh@maths.lth.se ==
>    Department of Mathematical Statistics
> Phone +46 46 222 4953
>    Lund University
> Fax +46 46 222 4623
> == Box 118, S-221 00 LUND, Sweden ==
> http://www.maths.lth.se/matstat ==
>
>
>



From icon-group-sender  Tue Jul 14 16:29:20 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id QAA08109
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Tue, 14 Jul 1998 16:29:19 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA13928; Tue, 14 Jul 1998 16:29:05 -0700
Date: Tue, 14 Jul 1998 15:26:25 -0500
Message-Id: <199807142026.PAA25970@segfault.cs.utsa.edu>
From: Clinton Jeffery <jeffery@segfault.cs.utsa.edu>
To: andersh@maths.lth.se
Cc: icon-group@optima.CS.Arizona.EDU
In-Reply-To: <35AB13AC.145D@maths.lth.se> (message from Anders Holtsberg on
	Tue, 14 Jul 1998 10:15:40 +0200)
Subject: Re: Icon language: Sockets
Reply-To: jeffery@segfault.cs.utsa.edu
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 1021


>   Is networking supported in Icon? I tried to look in the book and its
>   index and in the FAQ but I can't find it. I found file and pipe support 
>   but not socket support. What I want to do is a small multiplayer game 
>   program.

On common UNIX platforms, the answer is yes; on other platforms the answer
is so far, no.

On UNIX, your options are:

	1. use loadfunc() to call C functions in the Icon Program Library,
	   developed by Gregg Townsend, or
	2. use Unicon, the variant of Icon with high level POSIX extensions
	   developed by Shamim Mohamed, available at www.drones.com/unicon

You are right, this sounds like a good question for the FAQ.  :-)

As a silly aside, it is possible to develop multiplayer X Window games
using the graphics facilities of "100% Pure Icon", just open windows on
each player's display!  X Windows is great that way.

Clint Jeffery, jeffery@cs.utsa.edu
Division of Computer Science, The University of Texas at San Antonio
Research http://www.cs.utsa.edu/research/plss.html


From icon-group-sender  Tue Jul 14 16:29:31 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id QAA08113
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Tue, 14 Jul 1998 16:29:31 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA14528; Tue, 14 Jul 1998 16:29:16 -0700
To: icon-group@optima.CS.Arizona.EDU
Date: 14 Jul 1998 20:07:39 GMT
From: espie@liafa.jussieu.fr (Marc Espie)
Message-Id: <6ogdqb$js3$1@vishnu.jussieu.fr>
Organization: None
Sender: icon-group-request@optima.CS.Arizona.EDU
References: <35AB13AC.145D@maths.lth.se>
Subject: Re: Icon language: Sockets
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 681

In article <35AB13AC.145D@maths.lth.se>,
Anders Holtsberg  <andersh@maths.lth.se> wrote:
>Is networking supported in Icon? I tried to look in the book and its
>index and in the FAQ but I can't find it. I found file and pipe support 
>but not socket support. What I want to do is a small multiplayer game 
>program.

I don't have the url handy, but if you look more closely at the icon
web pages, you'll find a project which extends Icon to support common
Unix system calls, including socket support.
-- 
	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  Thu Jul 16 08:48:22 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id IAA26365
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Thu, 16 Jul 1998 08:48:21 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA11710; Thu, 16 Jul 1998 08:48:06 -0700
From: Otfried Geffert <og@stardivision.de>
Date: Thu, 16 Jul 1998 09:10:27 GMT
Message-Id: <19980716.9102789@og-1042.stardiv.de>
Subject: No graphics books
To: icon-group@optima.CS.Arizona.EDU
X-Mailer: Mozilla/3.0 (compatible; StarOffice/4.0; WinNT/Win95)
X-Priority: 3 (Normal)
Mime-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 722

	Unpublished in Europe

Once upon a time=20
there was a programmer in Germany,
who used Icon=20
for everyday work.
He heard about graphics=20
and about a book to come.

     He waponed himself=20
     with patience,
     because writing a book
     was a hard and a time=20
     consuming work.

January of '98
he ordered the book.
A couple of month later
the author sent a mail
that the book was published.

     And the retailer for Europe
     sent a mail to the bookstore
     'We are sorry the book is sold out'.

And the programmer sang:
(apologies to the Beatles)

     Whisper words of wisdom
     Let it be
     Write in C

And hoped
that this was not the end of the story.

Angrily=20
from Germany, Otfried






From icon-group-sender  Thu Jul 16 12:22:56 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id MAA05954
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Thu, 16 Jul 1998 12:22:55 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA05494; Thu, 16 Jul 1998 12:22:40 -0700
Date: Thu, 16 Jul 1998 13:13:06 -0500
Message-Id: <199807161813.NAA27758@segfault.cs.utsa.edu>
From: Clinton Jeffery <jeffery@segfault.cs.utsa.edu>
To: og@stardiv.de
Cc: icon-group@optima.CS.Arizona.EDU
In-Reply-To: <19980716.9102789@og-1042.stardiv.de> (message from Otfried
	Geffert on Thu, 16 Jul 1998 09:10:27 GMT)
Subject: Re: Yes graphics books
Reply-To: jeffery@segfault.cs.utsa.edu
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 566


Otfried,

The book is not sold out.  The book is brand new and has not even appeared
in American bookstores yet.  It is available right now from the Icon
Project, and it can be ordered on-line from www.peer-to-peer.com.  It can
also be ordered from www.amazon.com, but it only just appeared there and
they don't have it in stock yet.

My uneducated guess is that it should start appearing in stores in the next
2-3 months.  If you don't want to wait for the retail book distribution
channel, I would order it from Icon Project!

Cheers,
Clint, jeffery@cs.utsa.edu


From icon-group-sender  Fri Jul 17 08:42:11 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id IAA07264
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Fri, 17 Jul 1998 08:42:10 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA10953; Fri, 17 Jul 1998 08:41:55 -0700
Message-Id: <199807171230.OAA02421@capway.com>
From: "Vladimir Grodzenski" <grodzens@capway.com>
To: Otfried Geffert <og@stardivision.de>
Date: Fri, 17 Jul 1998 14:30:06 +0000
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7BIT
Subject: Re: No graphics books
Reply-To: <vladimir.grodzenski@capway.com>
Cc: icon-group@optima.CS.Arizona.EDU
Priority: urgent
In-Reply-To: <19980716.9102789@og-1042.stardiv.de>
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 450

On 16 Jul 98 at 9:10, Otfried Geffert wrote:

> January of '98
> he ordered the book.

I ordered the book directly online from the Editor Peer-to-Peer 
about 5 weeks ago, an I got it three weeks ago.

    http://www.peer-to-peer.com/catalog/language/icong.html






Vladimir Grodzenski
=================================================
E-mail: vladimir.grodzenski@capway.com
CompuServe: 100700,526
=================================================


From icon-group-sender  Fri Jul 17 08:42:00 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id IAA07261
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Fri, 17 Jul 1998 08:41:59 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA13986; Fri, 17 Jul 1998 08:41:44 -0700
Message-Id: <199807170754.IAA01247@holyrood.ed.ac.uk>
Date: 17 Jul 98  08:54:49 BST
From: Roger Hare <rjhare@holyrood.ed.ac.uk>
Subject: Graphics book
To: icon-group@optima.CS.Arizona.EDU
Reply-To: rjhare@ed.ac.uk
Organisation:  Edinburgh Parallel Computing Centre
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 569

> I would order it from Icon Project!

Seconded!

My local bookshop here in Edinburgh told me that it wasn't available
yet. These tales are put out by bookshops when they don't actually
know the answer to your query, I am sure.

I ordered mine from the Icon Project - it arrived in about 14 days
(less time than I would have expected from the bookshop), and the
extra postage costs just about equalled the markup which would have
been added on by the bookshop, so it cost the same (within 13 pence
if I remember rightly - about 8 cents or 30 pfennig (?)).

Roger Hare


From icon-group-sender  Fri Jul 17 16:50:18 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id QAA11906
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Fri, 17 Jul 1998 16:50:14 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA18627; Fri, 17 Jul 1998 16:49:58 -0700
From: pygmy@eskimo.com (Frank Sergeant)
To: icon-group@baskerville.CS.Arizona.EDU
Subject: Re: using Icon for database application
Date: Fri, 17 Jul 1998 14:25:32 -0500
Reply-To: frank.sergeant@pobox.com
Message-Id: <sU6r1Yv1u83e084yn@eskimo.com>
In-Reply-To: <hkbk1Yv1uAmb084yn@eskimo.com>
Lines: 120
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 4593

My fooling around with Icon continues as I try to decide on
the future direction of the database application (currently
in Clipper) that I discussed earlier.  Here is an update
along with some additional questions.

I have had the _Graphics Programming in Icon_ book for
awhile and am enjoying it.

I suppose I still do not fully appreciate some of the
special/unique features of Icon.  I guess that will come
with time.  Meanwhile, it looks like all the data processing
can be done comfortably in Icon by using its simple features
that I already understand.  One appeal to me of Icon was that
much of what I do in Clipper seems to map conveniently to
Icon.  For example, the arrays/lists in both are 1-based and
can hold elements of mixed types, missing parameters default
to Nil/&null, etc.

That leaves two areas of concern: the GUI (under Windows) and
TCP/IP socket connections (to communicate between the workstations
and the server, both running Icon).

GUI:

I tried the TextDialog() function and at first thought
it might be a show stopper.  The appearance looked a little
too Unixy and not enough like native Windows.  By trying
various color combinations and font sizes, however, I got
something that I might be able to live with.  I put up a
simple example with 4 text entry fields and two buttons.
Unfortunately, the tab key only moves among the 4 text
fields and never gets to the buttons.  Is there a way
to add the buttons to the tab list?  Also, the editing
in the text fields was not at all what a Windows user
would expect (home, end, high-lighted replacement, etc
did not seem to work at all).

I need to try it some more, but the first example
I tried with VIB with 3 entry fields and a few buttons
just didn't work right at all.  I could click in one
field with the mouse and see the I-bar cursor appear
but when I started typing, the text appeared in a
different entry field.  Again, the editing didn't work
and so forth.

Anyway, for my database project, I might not want
to build the screens with VIB.  I think I would prefer
to build them procedurally.  The main reason is that
the Clipper code is filled with "@ SAY GET" statements
such as

     @ 10, 20 SAY "First Name" GET FNAME PICTURE "@!"

and I might rather mechanically convert all of them
to put up a "reasonable" entry form (and later make
it prettier).  So, I might not need VIB after all.
If you think I'm overlooking some strong points of
VIB, please enlighten me.

Then I stumbled upon a mention of some native
Windows features in Technical Report CS-97-9 "Version
9 of Icon for Microsoft Windows" on the UTSA site.
I experimented a little with WinMenuBar(), WinScrollBar(),
and WinButton() which all seem to work perfectly and give
the conventional Windows sculpted appearance.

I even understood how to attach events to the
menu items and buttons.

I tried WinEditRegion() but didn't fully understand
how to work with it.  The editing part did seem to work
properly with home, end, high-lighted replacement, etc.
However, I didn't see how to get out of it or how to make
it into a single line edit field (or even a fixed multi
line edit field), that is, I couldn't get rid of the
scroll bars and couldn't "escape" from the edit region
or see exactly how to tie it into the rest of the program.
I haven't tried looking for the source for this.  It might
explain everything.

Are there other native Windows widgets?  Especially,
a single line edit field?

Anyway, I am now so close to seeing how I would do
everything GUI-related that my application would need,
that I feel sure the remaining pieces can either be found
or worked around.


SOCKETS:

I gathered from some other postings that the Unix
version has a socket interface and also a general mechanism
for calling arbitrary external (e.g. C) routines but that
the Windows version does not.  (?)

I tried looking briefly at the Icon source to see
whether I might see how to add the socket routines and such.
I decided now is not the time for me to try to do this.
It reminded me of the joke I saw posted to comp.lang.python
that went something like this:

     You have a problem.  You say to yourself, "I know!
     I'll use Regular Expressions."  Now you have two
     problems.

Have I overlooked a way already built into Windows Icon
to call an external routine?  If so, that might solve the
socket question.  If not, I probably can wait on that and
perhaps used a shared directory and let the processes
communicate via files for the present.


All comments and suggestions and criticisms welcomed.


  -- Frank
  frank.sergeant@pobox.com




From icon-group-sender  Mon Jul 20 08:13:34 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id IAA20103
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Mon, 20 Jul 1998 08:13:31 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA19698; Mon, 20 Jul 1998 08:13:15 -0700
Date: Sun, 19 Jul 1998 21:16:59 -0500
Message-Id: <199807200216.VAA31067@segfault.cs.utsa.edu>
From: Clinton Jeffery <jeffery@segfault.cs.utsa.edu>
To: frank.sergeant@pobox.com
Cc: icon-group@baskerville.CS.Arizona.EDU
In-Reply-To: <sU6r1Yv1u83e084yn@eskimo.com> (pygmy@eskimo.com)
Subject: Re: using Icon for database application
Reply-To: jeffery@cs.utsa.edu
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 3705


> Frank Sergeant wrote:
>   I tried the TextDialog()...
>   Unfortunately, the tab key only moves among the 4 text
>   fields and never gets to the buttons.  Is there a way
>   to add the buttons to the tab list?

This is a good question!  TextDialog's author(s) can answer it better
than I could.

>   the editing in the text fields was not at all what a Windows user
>   would expect (home, end, high-lighted replacement, etc
>   did not seem to work at all).

Well, it is not native to Windows and was not written what that in mind.
It would be possible to enhance the text vidgets to support more special
keys, but no one is working on it that I know of.

>   Anyway, for my database project, I might not want to build the screens
>   with VIB.  I think I would prefer to build them procedurally.

You could certainly write a program to translate clipper codes into IPL
calls to create the appropriate vidgets directly.  We have also had
success here in dynamically constructing VIB-compatible lists of strings
in order to produce interfaces that are customized at runtime.  To do this
we started with a VIB interface and then added parameters to the VIB-generated
code that are inserted into the list of interface component strings.

>   I tried WinEditRegion() but didn't fully understand...
>   I didn't see how to get out of it or how to make
>   it into a single line edit field (or even a fixed multi
>   line edit field)

The facility is at present very simple and very general purpose.  It was not
intended to replace single-line text input vidgets.  To use WinEditRegion
you define a rectangle area on the screen in which the user can edit
multiple lines of text.  The program views the result as one big string.
You can't "get out of it"; if it is part of a temporary interaction you can
put it in a dialog, in its own temporary window, or resize it to be
0 pixels in size.

>   I couldn't see exactly how to tie it into the rest of the program.

There is, deviously, no event produced during editing and no changes to the
program's control flow as a result of editing; the user just changes the
contents of the region when they feel like it.  You can check a flag to tell
if a region is modified, and you read the contents of the region (a string)
via a single call to WinEditRegion().  A typical application might do this,
for example, in response to a Save... menu option.  The program wi.icn is
an example.

>   Are there other native Windows widgets?  Especially,
>   a single line edit field?

One could be added.  This is the first request I've had for it.
All the native facilities presently supported are described in the
Windows Icon technical report.  This fall I should get an opportunity
to look at what else might be added from Windows' core API.

>   I gathered from some other postings that the Unix version has a socket
>   interface and also a general mechanism for calling arbitrary external
>   (e.g. C) routines but that the Windows version does not.  (?)

This is correct.

>   Have I overlooked a way already built into Windows Icon
>   to call an external routine?

No, there is no external interface at present.  I would love it if someone
would implement loadfunc() for Windows.  I've had 2-3 rumors of projects
that might include such an effort, but nothing concrete.  My own plans
include porting most or all of Unicon to Windows (Unicon's easy network
facilities should be reasonably easy to port), but I have no immediate
plans to work on an external interface until a student volunteers to help
do it.

Clint Jeffery, jeffery@cs.utsa.edu
Division of Computer Science, The University of Texas at San Antonio
Research http://www.cs.utsa.edu/research/plss.html


From icon-group-sender  Mon Jul 20 08:13:41 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id IAA20109
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Mon, 20 Jul 1998 08:13:40 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA21274; Mon, 20 Jul 1998 08:13:24 -0700
Message-Id: <BC1970247DD6D111949900A0C9AA8B52A04D20@il0015exch001u.ih.lucent.com>
From: "Hall, Joe T (Joe)" <jthall@lucent.com>
To: icon-group@optima.CS.Arizona.EDU
Subject: New graphics book
Date: Mon, 20 Jul 1998 09:19:44 -0500
Mime-Version: 1.0
X-Mailer: Internet Mail Service (5.5.1960.3)
Content-Type: text/plain
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 487

I think that Chapter 2 of the new graphics books is the very best
and most concise description of Icon I've every read.

	
_______________________________________________________________
	
Lucent Technologies  O
	
Bell Labs Innovations

		Joe T. Hall
		Distinguished Member of Technical Staff	     Ph:
630-713-1237
		Flexent Architecture & Estimation                     Fax:
630-713-3888
		Email:  jthall@lucent.com	
	
_______________________________________________________________




From icon-group-sender  Mon Jul 20 12:38:50 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id MAA12033
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Mon, 20 Jul 1998 12:38:49 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA07615; Mon, 20 Jul 1998 12:38:34 -0700
From: Kostas Oikonomou <ko@mara.ho.att.com>
Date: Mon, 20 Jul 1998 15:35:07 -0400 (EDT)
To: icon-group@optima.CS.Arizona.EDU
Subject: Reading numbers from a file
X-Mailer: VM 6.43 under 20.4 "Emerald" XEmacs  Lucid
Message-Id: <13747.39630.559540.280143@mara>
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 484


Hello,

This is probably a dumb question, but there doesn't seem to be an easy way to
read a numbers from a file with Icon.  I mean a file consisting of lines, each
line being a list of numbers (integers or reals), separated by blanks.

I finally wrote my own routine, but I don't want to reinvent thw wheel...

(I looked in the procs/ library, and found pack/unpack in binary.icn.  But I
don't see how to get unpack to do what I want.)

Thanks for your help.

				Kostas Oikonomou


From icon-group-sender  Mon Jul 20 12:38:33 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id MAA12025
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Mon, 20 Jul 1998 12:38:32 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA18623; Mon, 20 Jul 1998 12:38:17 -0700
Date: Mon, 20 Jul 1998 08:39:20 -0700
From: Ralph Griswold <ralph@optima.CS.Arizona.EDU>
Message-Id: <9807201539.AA00844@jupiter.CS.Arizona.EDU>
To: icon-group@optima.CS.Arizona.EDU, jthall@lucent.com
Subject: Re:  New graphics book
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 215

Thanks for your nice remark about chapter 2 of the Icon graphics
programming book (wrote that chapter myself).

Would you consider adding your thoughts to be "reader's comments"
for the book on amazon.com?

	Ralph


From icon-group-sender  Wed Jul 22 09:36:58 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id JAA02246
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Wed, 22 Jul 1998 09:36:57 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA18600; Wed, 22 Jul 1998 09:36:41 -0700
From: dae_alt@juno.com
To: icon-group@optima.CS.Arizona.EDU
Date: Tue, 21 Jul 1998 17:41:48 PDT
Subject: Prior Messages
Message-Id: <19980721.174149.3439.0.dae_alt@juno.com>
X-Mailer: Juno 1.49
X-Juno-Line-Breaks: 0-8
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 409

Hello:

Is there a place on the web to look at
previous messages from icon-group?

As a newbie to Icon, most likely my
question have been asked and answered.

Thanks. 

_____________________________________________________________________
You don't need to buy Internet access to use free Internet e-mail.
Get completely free e-mail from Juno at http://www.juno.com
Or call Juno at (800) 654-JUNO [654-5866]


From icon-group-sender  Wed Jul 22 13:51:31 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id NAA20629
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Wed, 22 Jul 1998 13:51:30 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA20771; Wed, 22 Jul 1998 13:51:14 -0700
Date: Wed, 22 Jul 1998 10:15:36 -0700
From: Ralph Griswold <ralph@optima.CS.Arizona.EDU>
Message-Id: <9807221715.AA04959@jupiter.CS.Arizona.EDU>
To: dae_alt@juno.com, icon-group@optima.CS.Arizona.EDU
Subject: Re:  Prior Messages
Cc: slm@optima.CS.Arizona.EDU
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 208

You can get prior messages to icon-group from either our ftp or web sites.

For the web, connect to

	http://www.cs.arizona.edu/icon/ftp/newsgrp

You'll see message by year and for recent years, subsection.


From icon-group-sender  Wed Jul 22 13:51:48 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id NAA20647
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Wed, 22 Jul 1998 13:51:48 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA23494; Wed, 22 Jul 1998 13:51:32 -0700
From: pygmy@eskimo.com (Frank Sergeant)
To: icon-group@baskerville.CS.Arizona.EDU
Subject: Re: using Icon for database application
Date: Wed, 22 Jul 1998 15:03:02 -0500
Reply-To: frank.sergeant@pobox.com
Message-Id: <2Vkt1Yv1uQST084yn@eskimo.com>
Lines: 126
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 5611

Oh, I've been moving slowly.  I think I got too excited
seeing the native W95 menus and buttons working and
unjustifiably thought everything else would work out
also.  I overlooked the text entry fields, the
picklists, and so forth.  TextDialog() as well as VIB
seem useless to me at the moment.  Perhaps I could dive
into the Icon source code and modify TextDialog() to
work the way I want.  Ditto for implementing native
text entry fields etc based upon the current interface
to native buttons.  However, from the brief glance I
gave the code a few weeks ago I think I just do not
want to take on what may be such a large and possibly
fruitless task.  I believe Clint has confirmed that
winicon does not have a built-in method of calling
external C routines, so that makes adding other native
W95 widgets (and sockets) more difficult.

I have been almost ready to give up Icon for this
project (but glad to keep it available for plain text
processing).

However, I see two possibilities that remain for using
winicon.  The first is to write my own widgets using
the built-in Icon graphics routines.  That is, instead
of trying to call native W95 widgets, I just "draw"
them on the screen myself, position the cursor
explicitly, catch each keystroke, etc. etc.  This could
well be easier than trying to figure out the interface
to Windows and the Icon source code.  This might be a
reasonable approach.  It still leaves the question of
socket access.  I think I could write a tiny program in
C that handles the socket connection and communicates
with the Icon process via a shared disk file.  This
would seem to be an acceptable work around until
winicon gets socket support.

The other possibility is to decide to let web browsers
handle the user interface.  Then, I'm back to being
able to use Icon in purely text mode.  It can generate
the HTML to send via the web server to the browser.  My
book on JavaScript arrived yesterday.  Using an older
MSIE browser to view some of the book's examples makes
me think this just might work.  JavaScript seems to let
me catch the keystroke events (I really want my users
to be able to do with keystrokes everything they can
now do with keystrokes!) and generate my own 'submit'
requests, rather than requiring the user to click on
'submit'.  Anyway, the examples didn't look too bad.
I've downloaded a newer Netscape which I'll try to
install later today and then take another look at the
JavaScript examples.  The buttons, picklists, text
entry fields, and so forth looked pretty good under
MSIE.  The fonts were ugly, but perhaps they could be
changed.  The browser's own tool bar and menu bar still
took up far too much screen real estate but I think
(from the JavaScript book) that I can create a
full-screen window.

In my depression (following my earlier elation) over
Icon, I've reconsidered VO (Computer Assoicates' Visual
Objects), Delphi, Python with Tk, Python with a COM
interface to Delphi, Object REXX, a raw C engine that
reads the widget information from a text file (which
Icon could then write), etc.

Object REXX is definitely out.  The examples I tried
were too slow even on a P166 with 48MB RAM and a copy
of the Object REXX run-time would need to be purchased
for each machine in each customer's office.  To hell
with that.  I probably could learn VO or Delphi
eventually.  VO looks better but I found periodic
slow-downs when browsing over the network that seem
unacceptable (and unexplainable).  The Python COM
Delphi might produce very pretty screens but is a whole
'nuther learning experience I'd rather avoid.  I may
decide that Python with Tk is fast enough after all.
At least Python has sockets built in (as I understand
it).  I don't really want to build my own GUI engine in
raw C.  The graphapp C GUI library looks good so that's
still a possibility.  JPython (Python written in Java,
and so having access to the Java GUI widgets) is
another possibility.

I guess I'll try to push further with JavaScript/Icon
and possibly the Python/Tk and see what happens.

> I suppose what I'm finding odd for me is that I usually
> use icon because of the way it handles text, not
> because of its graphics capabilities.

So, we have a terrible inversion.  (Well, perhaps only
I have it).  Would should be the difficult/important
part of the application (the actual data processing) is
very easy.  Would should be the trivial part (put up the
GUI) is a nightmare of endless "gotchas".

> You might also find the zero notation useful (end of
> record, line, whatever) if you have data kept in a list
> of varying length.

     Yes, I love that in Icon.  Python is zero-based
and so cannot use 0 as the end of list indicator.
Other than that, Python's array/list/dictionary
handling looks very much like Icon's (to my untrained
eye).  Of course, since my Clipper code is already
one-based, it would be less error prone to move to
one-based Icon rather than to zero-based Python.

> Anyway, I was glad to see your post to the icon-group.
> I think we should see some work soon on the problem out
> of Arizona and Texas. My understanding is that the
> Arizona group did most of the work on vib.

Disclaimer: This post is meant mostly to express my
current state of mind (it is possible this state of
mind was affected somewhat by working until very late
last night and then awakening to customer problems
early this morning).  No offense is meant to the
hard working, good natured, helpful Icon and Python
people.  All suggestions, comments, criticisms
cheerfully received either by posting or direct email.


  -- Frank
  frank.sergeant@pobox.com



From icon-group-sender  Wed Jul 22 17:37:52 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id RAA27739
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Wed, 22 Jul 1998 17:37:51 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA04329; Wed, 22 Jul 1998 17:37:35 -0700
Message-Id: <3.0.5.32.19980722163020.00a5fbc0@post.its.mcw.edu>
X-Sender: cdt@post.its.mcw.edu
X-Mailer: QUALCOMM Windows Eudora Light Version 3.0.5 (32)
Date: Wed, 22 Jul 1998 16:30:20
To: Kostas Oikonomou <ko@mara.ho.att.com>, icon-group@optima.CS.Arizona.EDU
From: Chris Tenaglia <cdt@post.its.mcw.edu>
Subject: Re: Reading numbers from a file
In-Reply-To: <13747.39630.559540.280143@mara>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 1212

try a variation of this:
------------------------------
#
# read space delimited lines of numbers
#
procedure main()
  while line := read() do
     {
      numbers := parse(line,' ')                              #split the
line into numbers
      every value := !numbers do process(value)   # process each number
      }
  end

#
# parse a string into a list with respect to a delimiter
#
procedure parse(line,delims)
  static chars
  chars  := &cset -- delims
  tokens := []
  line ? while tab(upto(chars)) do put(tokens,tab(many(chars)))
  return tokens
  end

#
# whatever
#
procedure process(n)
   ..........
  end

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

At 03:35 PM 7/20/98 -0400, Kostas Oikonomou wrote:
>
>Hello,
>
>This is probably a dumb question, but there doesn't seem to be an easy way to
>read a numbers from a file with Icon.  I mean a file consisting of lines,
each
>line being a list of numbers (integers or reals), separated by blanks.
>
>I finally wrote my own routine, but I don't want to reinvent thw wheel...
>
>(I looked in the procs/ library, and found pack/unpack in binary.icn.  But I
>don't see how to get unpack to do what I want.)
>
>Thanks for your help.
>
>				Kostas Oikonomou
>
>



From icon-group-sender  Tue Jul 28 16:33:27 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.8/8.8.7) with SMTP id QAA08147
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Tue, 28 Jul 1998 16:33:26 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA28615; Tue, 28 Jul 1998 16:33:09 -0700
From: dae_alt@juno.com
To: icon-group@optima.CS.Arizona.EDU
Date: Tue, 28 Jul 1998 14:32:18 PDT
Subject: Xcode puzzlement
Message-Id: <19980728.143221.2807.0.dae_alt@juno.com>
X-Mailer: Juno 1.49
X-Juno-Line-Breaks: 0-2,4-65
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 1483

To Icon user group:

I cannot figure this out.
Program 1 writes to disk and reads from disk the file 'xstore', which it
uses in xdecode().
Program 2 reads the same file as Program 1, but xdecode() fails.
(Run Program1 first, to create the disk file 'xstore')
Help!

#---------------------------------------------
#program1.icn
link xcode
record testit( a,b,c)
procedure main()
#write it
  f := open("xstore","w")
  x := testit()
  x.a := "doug"
  x.b := 34
  x.c := "mardi gras"
  xencode(x,f)
  close(f)
#read it
  g := open("xstore")
  y := xdecode(g)
  \y | stop("y is null")
  close(g)
  write(y.a) 
  write(y.b)
  write(y.c)
end
#---------------------------------------------

#program2.icn
link xcode
record testit( a,b,c)
procedure main()
#write it
#  f := open("xstore","w")
# x := testit()
#  x.a := "doug"
#  x.b := 34
#  x.c := "mardi gras"
#  xencode(x,f)
#  close(f)
#read it
  g := open("xstore")
  y := xdecode(g)
  \y | stop("y is null")
  close(g)
  write(y.a)
  write(y.b)
  write(y.c)
end
#-----------------------------------

I have no idea why xdecode() fails in program2.
The code is identical to that in program1




 - Doug Edmunds - 
edmunds@ethergate.com
--- END OF REAL MESSAGE ---
(Don't bother to read past here)

_____________________________________________________________________
You don't need to buy Internet access to use free Internet e-mail.
Get completely free e-mail from Juno at http://www.juno.com
Or call Juno at (800) 654-JUNO [654-5866]


From icon-group-sender  Wed Jul 29 16:48:56 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.7/8.8.7) with SMTP id QAA22260
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Wed, 29 Jul 1998 16:48:55 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA22653; Wed, 29 Jul 1998 16:48:38 -0700
Date: Wed, 29 Jul 1998 15:34:00 -0700
From: Sandy Miller <slm>
Message-Id: <9807292234.AA25219@kingfisher.CS.Arizona.EDU>
To: icon-group@optima.CS.Arizona.EDU
Subject: moderator on vacation
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Md5: NrzGn1zYZvFn8i+R6M6kpQ==
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 247

I will be on vacation from Thursday, July 29 until Monday, August 10.
There will be nobody moderating the incoming icon-group and forwarding
it to the mailing list until I return.  I hope this doesn't cause any
major problems for anyone.

-sandy


From icon-group-sender  Mon Aug 10 09:00:05 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.7/8.8.7) with SMTP id JAA06286
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Mon, 10 Aug 1998 09:00:05 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA01042; Mon, 10 Aug 1998 08:59:44 -0700
To: icon-group@optima.CS.Arizona.EDU
Date: 7 Aug 1998 05:30:55 GMT
From: jeffery@cs.utsa.edu (Clinton Jeffery)
Message-Id: <6qe3ef$kr8$1@ringer.cs.utsa.edu>
Organization: The University of Texas at San Antonio
Sender: icon-group-request@optima.CS.Arizona.EDU
References: <35C9CEBC.4C1B@wi.net>
Reply-To: jeffery@cs.utsa.edu
Subject: Re: Learning Icon
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 553

J W Tatay (jwtatay@wi.net) wrote:
: If anyone has links to pages about learning icon please reply.

See John Shipman's tutorial (www.nmt.edu/tcc/help/lang/icon/) and
Dave Hanson's introduction (www.cs.arizona.edu/icon/intro.htm)

Note that these webpages are also provided with Windows Icon in Windows Help
format; Windows Icon is at
	ftp://ringer.cs.utsa.edu/pub/icon/nt/binaries/winicon.zip

--
Clint Jeffery, jeffery@cs.utsa.edu
Division of Computer Science, The University of Texas at San Antonio
Research http://www.cs.utsa.edu/research/plss.html


From icon-group-sender  Mon Aug 10 09:00:16 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.7/8.8.7) with SMTP id JAA06289
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Mon, 10 Aug 1998 09:00:16 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA01050; Mon, 10 Aug 1998 08:59:55 -0700
To: icon-group@optima.CS.Arizona.EDU
Date: Mon, 10 Aug 1998 11:34:28 -0400
From: Steve Holden <sholden@bellatlantic.net>
Message-Id: <35CF1303.E048B09E@bellatlantic.net>
Organization: Bell Atlantic Internet Solutions
Sender: icon-group-request@optima.CS.Arizona.EDU
Reply-To: sholden@bellatlantic.net
Subject: Help wanted with SPARC linux implementation
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 628

If there's anyone involved with the Icon language
still reading this list (;-) I'd be glad of some help with
a port to SPARC linux.  I've got the basic complilation
done, and can "make Samples", but "make Tests" shows
up some pretty serious errors with the floating-point
arithmetic -- not an area I would have expected trouble.

It's probably to do with the dumb way I've patched the
Intel Linux implementation.  I am NOT a SPARC assembly
language programmer...

For the record, I'm running on a SPARCstation LX under
Red Hat Linux 4.2 (with 5.1 on the way soon, if that is
likely to help).

regards
Steve Holden
Holden Web




From icon-group-sender  Mon Aug 10 08:59:53 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.7/8.8.7) with SMTP id IAA06204
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Mon, 10 Aug 1998 08:59:52 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA01020; Mon, 10 Aug 1998 08:59:31 -0700
To: icon-group@optima.CS.Arizona.EDU
Date: Thu, 06 Aug 1998 10:41:48 -0500
From: J W Tatay <jwtatay@wi.net>
Message-Id: <35C9CEBC.4C1B@wi.net>
Organization: Inc.Net http://www.inc.net
Sender: icon-group-request@optima.CS.Arizona.EDU
Subject: Learning Icon
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 122

Hello.

If anyone has links to pages about learning icon please reply.

Thank You.  

August 8 1998 CE  10 37 00 CDT EST


From icon-group-sender  Wed Aug 12 12:27:29 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.8.7/8.8.7) with SMTP id MAA18912
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Wed, 12 Aug 1998 12:27:29 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA03728; Wed, 12 Aug 1998 12:27:08 -0700
To: icon-group@optima.CS.Arizona.EDU
Date: Wed, 12 Aug 1998 08:26:19 -0400
From: Steve Holden <sholden@bellatlantic.net>
Message-Id: <35D189EA.10AA320A@bellatlantic.net>
Organization: Bell Atlantic Internet Solutions
Sender: icon-group-request@optima.CS.Arizona.EDU
Reply-To: sholden@bellatlantic.net
Subject: SPARC linux arithmetic error
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 1297

This is a multi-part message in MIME format.
--------------AD5DCC7A639DF857AED33B1B
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Would someone with access to a sun_* implementation let
me know if they see an error running a standard test program.

I am comparing the output of the "numeric" test and see
problems with the modulo operator:

     Running numeric
     Checking numeric
     59,60c59,60
     < 36 % -7 ----> 6
     < -36 % -7 ----> -6
     ---
     > 36 % -7 ----> 1
     > -36 % -7 ----> -1

Any suggestions gratefully received. I would be especially
interested in hearing from other SPARC Linux users - not
a major market for Red Hat just yet, I suspect :-)

regards
 Steve


--------------AD5DCC7A639DF857AED33B1B
Content-Type: text/x-vcard; charset=us-ascii; name="vcard.vcf"
Content-Transfer-Encoding: 7bit
Content-Description: Card for Steve Holden
Content-Disposition: attachment; filename="vcard.vcf"

begin:          vcard
fn:             Steve Holden
n:              Holden;Steve
org:            Holden Web
adr:            10908 Justin Knoll Road;;;Oakton;VA;22124-1006;USA
email;internet: sholden@bellatlantic.net
x-mozilla-cpt:  ;0
x-mozilla-html: FALSE
version:        2.1
end:            vcard


--------------AD5DCC7A639DF857AED33B1B--



From icon-group-sender  Thu Aug 20 08:32:54 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) with SMTP id IAA07359
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Thu, 20 Aug 1998 08:32:54 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA08139; Thu, 20 Aug 1998 08:32:31 -0700
Message-Id: <199808200556.FAA26562@out1.ibm.net>
X-Sender: usinet.laturk@pop5.ibm.net
X-Mailer: QUALCOMM Windows Eudora Pro Version 4.0
Date: Thu, 20 Aug 1998 00:50:38 +0700
To: icon-group@optima.CS.Arizona.EDU
From: "Dr. Louis A. Turk" <laturk@ibm.net>
Subject: Graphics Book and getch() and Word Processor
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 1461

The day before yesterday my copy of Graphics Programming In Icon arrived.
The reason I ordered this book is because I have a command line Icon
concordance program that writes to screen lines of text containing a chosen
word; this program needed to use a different font so as to be able to put
more characters on a line.  I am happy to report that converting my program
to Icon windows so it could do that only took a few hours including the
time spent in reading the book, once I had it.  It is a very interesting,
easy to understand book.

The CD ROM that came with the book is a great source of tools and help.  I
was able to install MS Windows Icon in just a few minutes.  Wi is easy to
use, and on line help (language reference, etc.) is immediately available.  

I did experience one problem: getch() did not compile under Icon windows.
Is there a reason for this?  I changed getch() to read() for a quick fix,
but that requires a carriage return.  

Also, I am wondering if anyone has written a word processor in Icon?  Icon
seems to be perfect for such a project.  I have a few specialized functions
I would like to add to a word processor for my own personal use, but I am
afraid that I don't have the knowledge or time to create a word processor
from scratch.  Or would even that be fast and easy in Icon, with perhaps a
little help from the experts on this list?  Any advice, directions, or
comments would be appreciated.  

Thanks,
Louis Turk








From icon-group-sender  Fri Aug 21 16:24:42 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) with SMTP id QAA09140
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Fri, 21 Aug 1998 16:24:42 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA12580; Fri, 21 Aug 1998 16:24:19 -0700
Date: Fri, 21 Aug 1998 10:01:16 -0500 (CDT)
From: Ray Pereda <rpereda@ringer.cs.utsa.edu>
X-Sender: rpereda@pandora
To: "Dr. Louis A. Turk" <laturk@ibm.net>
Cc: icon-group@optima.CS.Arizona.EDU
Subject: Re: Graphics Book and getch() and Word Processor
In-Reply-To: <199808200556.FAA26562@out1.ibm.net>
Message-Id: <Pine.GSO.3.95.980821095934.9992A-100000@pandora>
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 674

On Thu, 20 Aug 1998, Dr. Louis A. Turk wrote:

> Also, I am wondering if anyone has written a word processor in Icon?  Icon
> seems to be perfect for such a project.  I have a few specialized functions
> I would like to add to a word processor for my own personal use, but I am
> afraid that I don't have the knowledge or time to create a word processor
> from scratch.  Or would even that be fast and easy in Icon, with perhaps a
> little help from the experts on this list?  Any advice, directions, or
> comments would be appreciated.  

I remember seeing a word processor written by Bob Alexander.  I can't
find it right now though.  Check out the Icon web page.

-ray



From icon-group-sender  Fri Aug 21 16:24:54 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) with SMTP id QAA09171
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Fri, 21 Aug 1998 16:24:53 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA12654; Fri, 21 Aug 1998 16:24:31 -0700
To: icon-group@optima.CS.Arizona.EDU
Date: Fri, 21 Aug 1998 16:17:32 BST
From: Bob Ardler <ardler@argonet.co.uk>
Message-Id: <na.f4257c4879.a600a0ardler@argonet.co.uk>
Organization: ArgoNet, but does not reflect its views
Sender: icon-group-request@optima.CS.Arizona.EDU
Reply-To: Bob Ardler <ardler@argonet.co.uk>
Subject: Acorn Icon: anyone alive out there?
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 543

Acorn-specific documentation?

Giovanni Lo Conti has done a port to ACORN-ARCHIMEDES of Icon v9.3 which
works well on the v8 programs I've tried.

It comes with the usual U. of Arizona general documents but I can't find
anything Acorn-specific. Anyone know?

Anyone tried Wimp programming in Icon?

 


-- 
 __  __  __  __  __  __ ___   _____________________________________________
|__| R   G   O   N   E   |   /
|  |        Bob          |  /...Internet access for all Acorn RISC machines
___________________________/ ardler@argonet.co.uk




From icon-group-sender  Mon Aug 24 09:08:44 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) with SMTP id JAA03330
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Mon, 24 Aug 1998 09:08:44 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA14216; Mon, 24 Aug 1998 09:08:20 -0700
Message-Id: <2.2.32.19980822053809.003390d8@pop5.ibm.net>
X-Sender: usinet.laturk@pop5.ibm.net (Unverified)
X-Mailer: Windows Eudora Pro Version 2.2 (32)
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Date: Sat, 22 Aug 1998 00:38:09 -0500
To: icon-group@optima.CS.Arizona.EDU
From: "Dr. Louis A. Turk" <laturk@ibm.net>
Subject: Why doen't this work?
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 1859


Can anybody tell me why this code only removes CR/LF's every other paragraph
that contains them? Why
does it skip a paragraph?

Louis

Obviously, there will be more to this program, once I get past this problem.

############################################################################
#############
#
#  HTML TO Nota Bene 4.5 FILTER
#  Ver. 1.0 Aug.
#  Programmer: Louis A. Turk
#
#  USE: Coverts HTML to Note Bene using two passes.  FIRST PASS:
#	1. Removes the CR/LF's between <P> and </P> 
#	2. Removes the CR/LF's between <UL> and </UL> and also removes right
indention.
#	SECOND PASS:
#	3. Replaces all HTML code with Nota Bene code.
#
############################################################################
##############


link graphics

procedure main(arg)

	WOpen("size=1005,850")


	infile    := arg[1]
	outfile   := arg[2]
	tempfile  := "temp3.txt"

        in  := open(infile,"r") | stop("Can't open file: ",in)
   	out := open(outfile,"w") | stop("Can't open file: ",out)
	tmp := open(tempfile,"c") | stop("Can't open file: ",tmp)

        #### FIRST PASS: REMOVE EXCESS CR/LF's
######################################

	while line := read(in) do {
		if find(line,"<P>") then {                            # Beginning of
defective code
			WWrites(line," ")
			writes(tmp,line," ")
			until find(line := read(in),"</P>") do {
				WWrites(line," ")
				writes(tmp,line," ")
			}
			WWrite(line)
			write(tmp,line)
		}
		else if find(line,"<UL>") then {
			WWrites(line," ")
			writes(tmp,line," ")
			until find(line := read(in),"</UL>") do {
				WWrites(line," ")
				writes(tmp,line," ")
			}
			WWrite(line)
			write(tmp,line)
		}                                                     # End of defective
code		
		else {
			WWrite(line)
			write(tmp,line)
		}

	}
	##### SECOND PASS: #######################################
Event()
end




From icon-group-sender  Mon Aug 24 09:08:17 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) with SMTP id JAA03321
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Mon, 24 Aug 1998 09:08:17 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA13229; Mon, 24 Aug 1998 09:07:54 -0700
Message-Id: <35DE42C1.CFE616E2@ix.netcom.com>
Date: Sat, 22 Aug 1998 00:02:09 -0400
From: Phillip Lee Thomas <teruthom@ix.netcom.com>
Reply-To: thomaspl@acm.org
X-Mailer: Mozilla 4.05 [en] (Win95; U)
Mime-Version: 1.0
To: icon-group@optima.CS.Arizona.EDU
Subject: Re: Graphics Book and getch() and Word Processor
References: <Pine.GSO.3.95.980821095934.9992A-100000@pandora>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Content-Transfer-Encoding: 7bit
Status: RO
Content-Length: 538



Ray Pereda wrote:

> On Thu, 20 Aug 1998, Dr. Louis A. Turk wrote:
>
> > Also, I am wondering if anyone has written a word processor in Icon?  Icon
> > seems to be perfect for such a project.  I have a few specialized functions

Other than the windows version in the Icon Programming Library, I wrote an
experimental word processor. The code is in the one of the last two ICEBOL
conference volumes. It was designed for use with structured data (e.g., poems)
and was tuned to the natural language which, in my case was Malay.

Phillip



From icon-group-sender  Mon Aug 24 09:09:08 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) with SMTP id JAA03355
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Mon, 24 Aug 1998 09:09:07 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA14276; Mon, 24 Aug 1998 09:08:44 -0700
Message-Id: <35DFEDF3.1E73@cauchy.demon.co.uk>
Date: Sun, 23 Aug 1998 11:24:51 +0100
From: Steve Harborne <stephen@cauchy.demon.co.uk>
Organization: Science is only a theory
X-Mailer: Mozilla 3.01Gold (Win95; I)
Mime-Version: 1.0
To: icon-group@optima.CS.Arizona.EDU
Subject: Release Notes for Icon
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Content-Transfer-Encoding: 7bit
Status: RO
Content-Length: 517

I am a large user of Icon and have a large number of C-extensions
and I have being converting from one version of the language to
another. I rely very much on the low-level functionality you provide.

I wondered if there were any detailed release notes that quantified
all the low-level changes to the language at each release. Most
of the changes have no effect on user functionality but do have an
effect at the developer level.


Regards

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


From icon-group-sender  Tue Aug 25 08:21:48 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) with SMTP id IAA16983
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Tue, 25 Aug 1998 08:21:48 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA15430; Tue, 25 Aug 1998 08:21:24 -0700
Message-Id: <35E20EF3.6DAEA82A@ix.netcom.com>
Date: Mon, 24 Aug 1998 21:10:12 -0400
From: Phillip Lee Thomas <teruthom@ix.netcom.com>
Reply-To: thomaspl@acm.org
X-Mailer: Mozilla 4.05 [en] (Win95; U)
Mime-Version: 1.0
To: "Dr. Louis A. Turk" <laturk@ibm.net>, icon-group@optima.CS.Arizona.EDU
Subject: Re: Why doen't this work?
References: <2.2.32.19980822053809.003390d8@pop5.ibm.net>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Content-Transfer-Encoding: 7bit
Status: RO
Content-Length: 3736

I'd have to see how this works against a piece of text, but you have a problem in the first
few lines if a single line has the shape:   <P>some text</P> as you write this line out and
then read in further lines until you find one ending in </P>. Looking at this rather
briefly, I would think you would produce a number of lines glued together as "<P>some
text</P> some more text and some more and <P> something that meets condition 2</P>".

Secondly, HTML glues the whole thing together for parsing purposes so it is quite possible
that in a single input line you could get several "<P>...</P>" sequences.

I suggest that you read the whole document in with a single reads() if this doesn't blow
your memory. Then use map(line,"\n"," ") to convert new lines to spaces, and finally do a
string scan along the line, chopping off pieces as you go:
    megaline ? {
        while line  :=tab(find("<P>")) do  {
             line ||:=  tab(find("</P>") +4)
             write(out, line)
             }  # while in <P>...</P>
      }  # while scan

You'd have to fiddle a bit with this but it's close to being right.

Sound possible?

Phillip Thomas

Dr. Louis A. Turk wrote:

> Can anybody tell me why this code only removes CR/LF's every other paragraph
> that contains them? Why
> does it skip a paragraph?
>
> Louis
>
> Obviously, there will be more to this program, once I get past this problem.
>
> ############################################################################
> #############
> #
> #  HTML TO Nota Bene 4.5 FILTER
> #  Ver. 1.0 Aug.
> #  Programmer: Louis A. Turk
> #
> #  USE: Coverts HTML to Note Bene using two passes.  FIRST PASS:
> #       1. Removes the CR/LF's between <P> and </P>
> #       2. Removes the CR/LF's between <UL> and </UL> and also removes right
> indention.
> #       SECOND PASS:
> #       3. Replaces all HTML code with Nota Bene code.
> #
> ############################################################################
> ##############
>
> link graphics
>
> procedure main(arg)
>
>         WOpen("size=1005,850")
>
>         infile    := arg[1]
>         outfile   := arg[2]
>         tempfile  := "temp3.txt"
>
>         in  := open(infile,"r") | stop("Can't open file: ",in)
>         out := open(outfile,"w") | stop("Can't open file: ",out)
>         tmp := open(tempfile,"c") | stop("Can't open file: ",tmp)
>
>         #### FIRST PASS: REMOVE EXCESS CR/LF's
> ######################################
>
>         while line := read(in) do {
>                 if find(line,"<P>") then {                            # Beginning of
> defective code
>                         WWrites(line," ")
>                         writes(tmp,line," ")
>                         until find(line := read(in),"</P>") do {
>                                 WWrites(line," ")
>                                 writes(tmp,line," ")
>                         }
>                         WWrite(line)
>                         write(tmp,line)
>                 }
>                 else if find(line,"<UL>") then {
>                         WWrites(line," ")
>                         writes(tmp,line," ")
>                         until find(line := read(in),"</UL>") do {
>                                 WWrites(line," ")
>                                 writes(tmp,line," ")
>                         }
>                         WWrite(line)
>                         write(tmp,line)
>                 }                                                     # End of defective
> code
>                 else {
>                         WWrite(line)
>                         write(tmp,line)
>                 }
>
>         }
>         ##### SECOND PASS: #######################################
> Event()
> end





From icon-group-sender  Tue Aug 25 16:22:31 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) with SMTP id QAA16662
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Tue, 25 Aug 1998 16:22:31 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA16002; Tue, 25 Aug 1998 16:22:07 -0700
Message-Id: <35E322BE.6889D8C@charlie.cns.iit.edu>
Date: Tue, 25 Aug 1998 15:46:54 -0500
From: "Thomas W. Christopher" <tc@charlie.cns.iit.edu>
Organization: Illinois Institute of Technology
X-Mailer: Mozilla 4.05 [en] (WinNT; I)
Mime-Version: 1.0
To: icon-group@optima.CS.Arizona.EDU
Subject: Icon Programming Language Hanbook available on line
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Content-Transfer-Encoding: 7bit
Status: RO
Content-Length: 641

I have placed my Icon Programming Language Handbook on-line in .PDF
format for personal use. (That is, you may run of single copies for
individual use;
no mass production without permission please.)

You can get it from my company web site:
    http://www.toc-press.com/freesoftware.htm

This is a Beta edition. Please report all errors to me. (Make it easy on
me:
include the page number where you found the bug.) Thanks for your help.

--
Thomas W. Christopher -- tc@charlie.cns.iit.edu , tc@toc-press.com
Principal, Tools of Computing LLC. -- http://www.toc-press.com
Associate Prof., Illinois Inst. of Tech. -- http://www.iit.edu/~tc




From icon-group-sender  Fri Aug 28 13:11:48 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) with SMTP id NAA02870
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Fri, 28 Aug 1998 13:11:47 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA19641; Fri, 28 Aug 1998 13:11:22 -0700
To: icon-group@optima.CS.Arizona.EDU
Date: Fri, 28 Aug 1998 16:59:49 BST
From: Bob Ardler <ardler@argonet.co.uk>
Message-Id: <na.ecc435487c.a600a0ardler@argonet.co.uk>
Organization: ArgoNet, but does not reflect its views
Sender: icon-group-request@optima.CS.Arizona.EDU
Reply-To: Bob Ardler <ardler@argonet.co.uk>
Subject: A hood fan is guard to mind.
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 2161

As long as nobody reads this newsgroup, I might as well ask
somnolent iconians whether the following program, for generating
polyspoonerisms, can be shortened by icon clever stuff. The nested
"every"s seem reminiscent of lesser languages:-

procedure main()
  L := ["g","m","h","f"]
  every h:=1 to 4 do {
    every i:=1 to 4 do {
      every j:=1 to 4 do {
        every k:=1 to 4 do {
          write(fit(h,i,j,k))
  }}}}
end

global L,h,i,j,k

procedure fit(a,b,c,d)
  if a=b|a=c|a=d|b=c|b=d|c=d then fail
  return "A "||L[a]||"ood "||L[b]||"an is "||L[c]||"ard to "||L[d]||"ind."
end

Of the generated spoonerisms, only the following, with suitably amended
spelling, appear to be viable:-

A good man is hard to find.
A mood fan is guard to hind.
A good man is fard to hind.
A hood man is guard to find.
A good fan is marred to hind.
A hood fan is guard to mind.
A good fan is hard to mind.
A food man is guard to hind.

Malheureusement, ca ne marche point en francais:-

Un bon homme est difficile a trouver.
Un bon homme est tifficile a drouver.
Un bon domme est hifficile a trouver.
Un bon domme est tifficile a hrouver.
Un bon tomme est hifficile a drouver.
Un bon tomme est difficile a hrouver.
Un hon bomme est difficile a trouver.
Un hon bomme est tifficile a drouver.
Un hon domme est bifficile a trouver.
Un hon domme est tifficile a brouver.
Un hon tomme est bifficile a drouver.
Un hon tomme est difficile a brouver.
Un don bomme est hifficile a trouver.
Un don bomme est tifficile a hrouver.
Un don homme est bifficile a trouver.
Un don homme est tifficile a brouver.
Un don tomme est bifficile a hrouver.
Un don tomme est hifficile a brouver.
Un ton bomme est hifficile a drouver.
Un ton bomme est difficile a hrouver.
Un ton homme est bifficile a drouver.
Un ton homme est difficile a brouver.
Un ton domme est bifficile a hrouver.
Un ton domme est hifficile a brouver.

Dormez bien, mes amis,
Bob.




-- 
 __  __  __  __  __  __ ___   _____________________________________________
|__| R   G   O   N   E   |   /
|  |        Bob          |  /...Internet access for all Acorn RISC machines
___________________________/ ardler@argonet.co.uk




From icon-group-sender  Fri Aug 28 13:10:42 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) with SMTP id NAA02753
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Fri, 28 Aug 1998 13:10:42 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA19576; Fri, 28 Aug 1998 13:10:17 -0700
To: icon-group@optima.CS.Arizona.EDU
Date: Fri, 28 Aug 1998 16:13:32 GMT
From: gcavazos@yahoo.com
Message-Id: <6s6kvc$lrd$1@nnrp1.dejanews.com>
Organization: Deja News - The Leader in Internet Discussion
Sender: icon-group-request@optima.CS.Arizona.EDU
Subject: Club created for Icon
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 406

I have created a club at Yahoo, so people can go leave posts, chat with
persons interested in Icon, also, you can post links, etc.
The address is:

http://clubs.yahoo.com/clubs/iconprogramminglanguage


I hope everyone enjoys this forum...

Regards
Gustavo...

-----== Posted via Deja News, The Leader in Internet Discussion ==-----
http://www.dejanews.com/rg_mkgrp.xp   Create Your Own Free Member Forum


From icon-group-sender  Fri Aug 28 13:11:54 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) with SMTP id NAA02894
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Fri, 28 Aug 1998 13:11:54 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA19675; Fri, 28 Aug 1998 13:11:30 -0700
Date: Fri, 28 Aug 1998 10:40:49 -0700
From: Gregg Townsend <gmt>
Message-Id: <9808281740.AA29523@hawk.CS.Arizona.EDU>
To: icon-group
Subject: Announcing Version 2 of Jcon
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 810

We are pleased to announce the availability of Version 2 of Jcon.

Jcon is a new Java-based implementation of the Icon programming language.
The Jcon translator, written in Icon, generates Java class files that
execute in conjunction with a run-time system written in Java. 

Version 2 of Jcon is an essentially complete implementation of Icon,
including graphics and large integers. A few minor features are missing
-- mostly things like chdir() that cannot be done in Java. 

The performance of Version 2 is considerably improved over Version 1,
although it still isn't as fast as the standard Icon interpreter. 

For more information, or to download Jcon, go to our web page:
        http://www.cs.arizona.edu/icon/jcon/

-- Gregg Townsend, The University of Arizona
   Todd Proebsting, Microsoft Research


From icon-group-sender  Fri Aug 28 13:12:14 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) with SMTP id NAA02925
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Fri, 28 Aug 1998 13:12:14 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA19771; Fri, 28 Aug 1998 13:11:49 -0700
To: icon-group@optima.CS.Arizona.EDU
Date: 28 Aug 1998 13:58:54 -0400
From: clayton@jazz.cc.gatech.edu (R. Clayton)
Message-Id: <ml3iujdyom9.fsf@jazz.cc.gatech.edu>
Organization: College of Computing, Georgia Tech
Sender: icon-group-request@optima.CS.Arizona.EDU
Subject: A prooner spogram.
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 1313

  ja cat t.icn
  procedure main()

    every p := perms(set(["g","m","h","f"])) do
       write("A ", p[1], "ood ", p[2], "an is ", p[3], "ard to ", p[4], "ind.")

    end # main


  procedure perms(letters)

    if *letters = 0 then
      suspend ""
    else
      every l := !letters do
	every suspend l || perms(delete(copy(letters), l))

    end # perms

  ja icont t.icn
  Translating:
  t.icn:
    main
    perms
  No errors
  Linking:

  ja ./t
  A hood man is gard to find.
  A hood man is fard to gind.
  A hood gan is mard to find.
  A hood gan is fard to mind.
  A hood fan is mard to gind.
  A hood fan is gard to mind.
  A mood han is gard to find.
  A mood han is fard to gind.
  A mood gan is hard to find.
  A mood gan is fard to hind.
  A mood fan is hard to gind.
  A mood fan is gard to hind.
  A good han is mard to find.
  A good han is fard to mind.
  A good man is hard to find.
  A good man is fard to hind.
  A good fan is hard to mind.
  A good fan is mard to hind.
  A food han is mard to gind.
  A food han is gard to mind.
  A food man is hard to gind.
  A food man is gard to hind.
  A food gan is hard to mind.
  A food gan is mard to hind.

  ja 

Why I wrote this program in five minutes instead of one minute:  I deleted l
from letters itself instead of from a copy of letters.


From icon-group-sender  Mon Aug 31 09:29:15 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) with SMTP id JAA00967
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Mon, 31 Aug 1998 09:29:15 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA21360; Mon, 31 Aug 1998 09:28:50 -0700
Message-Id: <35E814C9.A0E6FEBC@charlie.cns.iit.edu>
Date: Sat, 29 Aug 1998 09:48:41 -0500
From: "Thomas W. Christopher" <tc@charlie.cns.iit.edu>
Organization: Illinois Institute of Technology
X-Mailer: Mozilla 4.06 [en] (WinNT; I)
Mime-Version: 1.0
To: "icon-group@CS.Arizona.EDU" <icon-group@optima.CS.Arizona.EDU>
Subject: Spoonerisms
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Content-Transfer-Encoding: 7bit
Status: RO
Content-Length: 614

Bob Ardler,

Your spoonerism program can be obfuscated a bit as follows:
procedure main()
  L := ["g","m","h","f"]
  every h:=1 to 4 & i:=1 to 4 & j:=1 to 4 & k:=1 to 4 do {
          write(fit(h,i,j,k))
  }
end

global L

procedure fit(a,b,c,d)
  if a=(b|c|d)|b=(c|d)|c=d then fail
  return "A "||L[a]||"ood "||L[b]||"an is "||L[c]||"ard to
"||L[d]||"ind."
end

Although Clayton's solution is the best so far.
--
Thomas W. Christopher -- tc@charlie.cns.iit.edu , tc@toc-press.com
Principal, Tools of Computing LLC. -- http://www.toc-press.com
Associate Prof., Illinois Inst. of Tech. -- http://www.iit.edu/~tc




From icon-group-sender  Mon Aug 31 09:28:25 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) with SMTP id JAA00752
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Mon, 31 Aug 1998 09:28:25 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA21311; Mon, 31 Aug 1998 09:28:00 -0700
To: icon-group@optima.CS.Arizona.EDU
Date: Fri, 28 Aug 1998 21:36:12 -0400
From: Jules Gilbert <jules@aasp.net>
Message-Id: <35E75B0B.142E9658@aasp.net>
Organization: Synbolic Mechanics
Sender: icon-group-request@optima.CS.Arizona.EDU
Subject: Icon, OS/2, and interfacing C code
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 693

Hello:

I use Icon under OS/2 and I am interested in interfacing some Icon code
to a C program.

I don't have the (obsolete) CSET/2 compiler, nor the Visual Age stuff.

Instead, I use EMX/GCC and various other commercial products.

So, is this possible?  Will my application live?  Hello?

Icon is a really great language, too.   Is their anyone interested in
building an Icon to MAINTAINABLE C source code converter with me.

I once built (and still have) a Snobol to C source code converter,
designed to translate the Snobol source generated via the Rebus product
into C source, again MAINTAINABLE C.

Thank you,
Please respond to both this group and me.  I am jules@aasp.net

Thank you!




From icon-group-sender  Mon Aug 31 09:29:09 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) with SMTP id JAA00909
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Mon, 31 Aug 1998 09:29:08 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA21342; Mon, 31 Aug 1998 09:28:43 -0700
From: gep2@computek.net
Date: Sat, 29 Aug 1998 01:50:56 -0500
Message-Id: <199808290650.BAA28892@axp.cmpu.net>
Mime-Version: 1.0
Content-Type: text/plain
Content-Transfer-Encoding: 7bit
Subject: Re: A hood fan is guard to mind.
To: icon-group@optima.CS.Arizona.EDU
In-Reply-To: <na.ecc435487c.a600a0ardler@argonet.co.uk>
X-Mailer: SPRY Mail Version: 04.00.06.17
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Content-Transfer-Encoding: 7bit
Status: RO
Content-Length: 2742

> can [this program] be shortened by icon clever stuff. The nested
"every"s seem reminiscent of lesser languages:-

Indeed!

> procedure main()
  L := ["g","m","h","f"]
  every h:=1 to 4 do {
    every i:=1 to 4 do {
      every j:=1 to 4 do {
        every k:=1 to 4 do {
          write(fit(h,i,j,k))
  }}}}
end
global L,h,i,j,k
procedure fit(a,b,c,d)
  if a=b|a=c|a=d|b=c|b=d|c=d then fail
  return "A "||L[a]||"ood "||L[b]||"an is "||L[c]||"ard to "||L[d]||"ind."
end

How about something more like:

procedure main()
  L := ['g','m','h','f']
  every write(fit(!L,!L,!L,!L))
end
global L
procedure fit(a,b,c,d)
  if *(a ++ b ++ c ++ d) < 4 then fail
  return "A "|| a ||"ood "|| b ||"an is "|| c ||"ard to "|| d ||"ind."
end

Note that L is now a list of character sets rather than character strings!  This 
should reduce the number of conversions necessary.  It also reduces the number 
of character set to string conversions required until the final return statement 
where everything's concatenated into a final string to return.

The question is whether the character set union (and size) operation takes less 
time (or more) than the equals tests used in the original program.  (It's 
certainly more Icon-like to use the character set union and size).  Also, why 
convert all the time between subscript numbers and the corresponding characters, 
as the original did?  Icon iterates just great across strings.  

In general, character sets are one of the most sadly neglected features of 
Icon... a wonderful feature that deserves to be used more.

You might also want to reduce the list, calling and parameter passing overhead 
and move the whole thing into the main procedure, more like:

procedure main()
  L = 'gmhf'
  every *((a := !L) ++ (b := !L) ++ (c := !L) ++ (d := !L)) = 4 do
   write("A "|| a ||"ood "|| b ||"an is "|| c ||"ard to "|| d ||"ind.")
end

You might even want to take advantage of this kind of structure to use four 
separate character set constants in the character set concatenation, to 
eliminate in advance those words which you know won't be legal words... such as:

procedure main()
  every *((a := 'gmhf') ++ (b := 'mf') ++ (c := 'gh') ++ (d := 'mhf')) = 4 do
   write("A "|| a ||"ood "|| b ||"an is "|| c ||"ard to "|| d ||"ind.")
end

It's probably possible to use character set differences to extend this principle 
and use the difference set for subsequent generations (if you use the same set 
for all four iterations), which would eliminate the need for the test for 4 
different characters in the resulting set.  But that might go beyond elegance 
and just get obscure.  :-)

Gordon Peterson
http://www.computek.net/public/gep2/
Support the Anti-SPAM Amendment!  Join at http://www.cauce.org/



From icon-group-sender  Mon Aug 31 09:27:50 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) with SMTP id JAA00692
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Mon, 31 Aug 1998 09:27:49 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA21268; Mon, 31 Aug 1998 09:27:24 -0700
Date: Fri, 28 Aug 1998 20:54:19 -0400
From: Jules Gilbert <jules@aasp.net>
Subject: V9 and OS/2
To: icon-group@arizona.edu
Message-Id: <35E7513A.87455022@aasp.net>
Organization: Synbolic Mechanics
Mime-Version: 1.0
X-Mailer: Mozilla 4.04b1 [en] (OS/2; I)
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Content-Transfer-Encoding: 7bit
Status: RO
Content-Length: 357

I have an application for which I would like very much to use Icon.

My environment is OS/2 and v9 of Icon.

1)  I want to tie a fairly simple C program to Icon.

2)  If possible, I would like to generate .gif files, also.

By the way, many people who use OS/2 are sad to see your organization
stop supporting OS/2 based Icon.

Sincerely,
Jules Gilbert





From icon-group-sender  Mon Aug 31 09:29:37 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) with SMTP id JAA00986
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Mon, 31 Aug 1998 09:29:36 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA21386; Mon, 31 Aug 1998 09:29:11 -0700
To: icon-group@optima.CS.Arizona.EDU
Date: Sat, 29 Aug 1998 22:59:13 BST
From: Bob Ardler <ardler@argonet.co.uk>
Message-Id: <na.4a68f3487d.a600a0ardler@argonet.co.uk>
Organization: ArgoNet, but does not reflect its views
Sender: icon-group-request@optima.CS.Arizona.EDU
References: <ml3iujdyom9.fsf@jazz.cc.gatech.edu>
Reply-To: Bob Ardler <ardler@argonet.co.uk>
Subject: Re: A prooner spogram.
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 714

 R. Clayton:

>   procedure perms(letters)
>     if *letters = 0 then
>       suspend ""
>     else
>       every l := !letters do
>  every suspend l || perms(delete(copy(letters), l))
>     end # perms

Help! suspend, copy, pling, zero set size, implicit function value.

It will lake me mery vuch tonger than your five minutes writing time to
prasp your grogram.

However, it's just the sort of thiconic inking I beed to be nattered with.

Thany manks.

Bob Ardler.






-- 
 __  __  __  __  __  __ ___   _____________________________________________
|__| R   G   O   N   E   |   /
|  |        Bob          |  /...Internet access for all Acorn RISC machines
___________________________/ ardler@argonet.co.uk




From icon-group-sender  Mon Aug 31 09:29:53 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) with SMTP id JAA01014
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Mon, 31 Aug 1998 09:29:52 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA21433; Mon, 31 Aug 1998 09:29:27 -0700
To: icon-group@optima.CS.Arizona.EDU
Date: Sat, 29 Aug 1998 22:43:11 BST
From: Bob Ardler <ardler@argonet.co.uk>
Message-Id: <na.ac677b487d.a600a0ardler@argonet.co.uk>
Organization: ArgoNet, but does not reflect its views
Sender: icon-group-request@optima.CS.Arizona.EDU
References: <na.ecc435487c.a600a0ardler@argonet.co.uk>, <6s6r1e$qa7$1@shell14.ba.best.com>
Reply-To: Bob Ardler <ardler@argonet.co.uk>
Subject: Re: A hood fan is guard to mind.
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 710



Grateful for Hamim Mohamed's program and mailing-list recommendation: have
found what may be the mailing list and have (I hope) subscribed to it.

His: if *cset(a||b||c||d) = 4   is very pleasant.

The: every write(fit(!chars, !chars, !chars, !chars)) is exactly the sort of
thing (proper Icon as opposed to Icon being a better Algol) that I can only
do by trial and accident and was hoping (but not expecting) to see.

Instructive. Cheers indeed!

Bob Ardler.
  




-- 
 __  __  __  __  __  __ ___   _____________________________________________
|__| R   G   O   N   E   |   /
|  |        Bob          |  /...Internet access for all Acorn RISC machines
___________________________/ ardler@argonet.co.uk




From icon-group-sender  Mon Aug 31 09:30:05 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) with SMTP id JAA01077
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Mon, 31 Aug 1998 09:30:03 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA21426; Mon, 31 Aug 1998 09:29:37 -0700
Date: Sun, 30 Aug 1998 14:54:38 -0300 (ADT)
From: Larry Bezeau <bezeau@unb.ca>
X-Sender: bezeau@sol.sun.csd.unb.ca
Reply-To: Larry Bezeau <bezeau@unb.ca>
To: icon-group@optima.CS.Arizona.EDU
Subject: Integers in Posix-compliant icon
Message-Id: <Pine.SOL.3.96.980830144657.1473A-100000@sol.sun.csd.unb.ca>
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 682

        In writing a procedure to change a date in most any form
to a standard form, I encountered some strange behavior by 
Posix compliant icon on a Unix platform.  This is illustrated by
the following short program.

cat integers.icn
procedure main(A)
  write()
  every (arg := !A) do  {
    if integer(arg) then write(" ",arg," is the integer ",integer(arg),".")
                    else write(" ",arg," is not an integer.")
    write()
  }
end
./integers 3rd 23rd

 3rd is not an integer.

 23rd is the integer 13.

        In my opinion, neither of these is an integer.  Is
this a bug or am I missing something?
                                      Larry   Bezeau@UNB.Ca





From icon-group-sender  Mon Aug 31 09:30:21 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) with SMTP id JAA01105
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Mon, 31 Aug 1998 09:30:21 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA21456; Mon, 31 Aug 1998 09:29:56 -0700
Message-Id: <35E9AA94.7A5F5147@charlie.cns.iit.edu>
Date: Sun, 30 Aug 1998 14:40:04 -0500
From: "Thomas W. Christopher" <tc@charlie.cns.iit.edu>
Organization: Illinois Institute of Technology
X-Mailer: Mozilla 4.06 [en] (WinNT; I)
Mime-Version: 1.0
To: "icon-group@CS.Arizona.EDU" <icon-group@optima.CS.Arizona.EDU>
Subject: LL(1) and LL(k) parser generators and educational compiler available
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Content-Transfer-Encoding: 7bit
Status: RO
Content-Length: 866

I have placed on the web a couple of parser generators and an example
compiler
for your use. All are written in Icon. They are available at:
    http://www.toc-press.com/freesoftware.htm

The parser generators are:

TCLL1 - an LL(1) parser generator. It includes panic-mode error repair.

TCLLk - an LL(k) parser generator (technically, "strong LL(k)"). I have
been using it
in classes and it appears to work, but complete tests are not complete
yet. It appears
to be comparable to LALR(1).

EULER - an EULER compiler written in Icon using TCLL1.

Illinois Institute of Technology uses these for their undergraduate
Programming
Languages and Translators course.

--
Thomas W. Christopher -- tc@charlie.cns.iit.edu , tc@toc-press.com
Principal, Tools of Computing LLC. -- http://www.toc-press.com
Associate Prof., Illinois Inst. of Tech. -- http://www.iit.edu/~tc




From icon-group-sender  Mon Aug 31 09:30:28 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) with SMTP id JAA01114
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Mon, 31 Aug 1998 09:30:28 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA21473; Mon, 31 Aug 1998 09:30:03 -0700
Date: Mon, 31 Aug 1998 09:05:25 -0700
From: kwalker@sfo.harbinger.com (Ken Walker)
Message-Id: <199808311605.JAA20565@varda.premenos.com>
To: icon-group@optima.CS.Arizona.EDU
Subject: Re: A hood fan is guard to mind.
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Md5: +9a2kKmLIC6wRuNurqTDbA==
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Content-Transfer-Encoding: 7bit
Status: RO
Content-Length: 1542

How about a version of polyspoonerisms that uses recursive generators.
It allows the sentence to be arbitrarily long by changing only data
and not the number of nested "every" expressions.

procedure main()
   local fillers, firstChar, remainingChars

   #
   # The sentence is broken into three categories:
   #   parts not participating in the substitution
   #   the initial characters of the words involved in substitution
   #   the tail of the words involved in substitution
   #
   fillers :=    ["A ",    " ",  " is ", " to ",    "."]
   firstChar :=     ["g",    "m",   "h",     "f"]
   remainingChars := ["ood",  "an",  "ard",    "ind"]

   polyspoonerism(fillers, firstChar, remainingChars, "")
end

procedure polyspoonerism(
  fillers,        # words before each participating word
  firstChar,      # first characters to use in substitutions
  remainingChars, # words (w/out 1st char) that are participating
  sentence)       # sentence so far

  local indx

  if *remainingChars = 0 then {
     write(sentence, fillers[1] | "")
     return
  }

  #
  # Use each of the remaining first characters
  #
  every indx := 1 to *firstChar  do {
     #
     # Add the next set of parts to the sentence then call
     # the function recursively.
     #
     polyspoonerism(fillers[2:0],
        firstChar[1:indx] ||| firstChar[indx+1:0],
        remainingChars[2:0],
        sentence || fillers[1] || firstChar[indx] || remainingChars[1])
  }
end


Ken Walker, kenneth.walker@sfo.harbinger.com
Harbinger Coporation, Concord, Ca. 94520



From icon-group-sender  Mon Aug 31 09:30:37 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) with SMTP id JAA01124
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Mon, 31 Aug 1998 09:30:36 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA20694; Mon, 31 Aug 1998 09:30:10 -0700
Date: Mon, 31 Aug 1998 09:11:59 -0700
From: kwalker@sfo.harbinger.com (Ken Walker)
Message-Id: <199808311611.JAA20585@varda.premenos.com>
To: icon-group@optima.CS.Arizona.EDU
Subject: Re: A hood fan is guard to mind.
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Md5: 0iWoS3eDDqPo4J69Q7wyKA==
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Content-Transfer-Encoding: 7bit
Status: RO
Content-Length: 1524

I apologize. The version of the program I sent out did not have
a procedure that generated results, so it was not a recursive generator.
Here is a version that is.

procedure main()
   local fillers, firstChar, remainingChars

   #
   # The sentence is broken into three categories:
   #   parts not participating in the substitution
   #   the initial characters of the words involved in substitution
   #   the tail of the words involved in substitution
   #
   fillers :=    ["A ",    " ",  " is ", " to ",    "."]
   firstChar :=     ["g",    "m",   "h",     "f"]
   remainingChars := ["ood",  "an",  "ard",    "ind"]

   every write(polyspoonerism(fillers, firstChar, remainingChars, ""))
end

procedure polyspoonerism(
  fillers,        # words before each participating word
  firstChar,      # first characters to use in substitutions
  remainingChars, # words (w/out 1st char) that are participating
  sentence)       # sentence so far

  local indx

  if *remainingChars = 0 then
     return sentence || (fillers[1] | "")

  #
  # Use each of the remaining first characters
  #
  every indx := 1 to *firstChar  do {
     #
     # Add the next set of parts to the sentence then call
     # the function recursively.
     #
     suspend polyspoonerism(fillers[2:0],
        firstChar[1:indx] ||| firstChar[indx+1:0],
        remainingChars[2:0],
        sentence || fillers[1] || firstChar[indx] || remainingChars[1])
  }
end


Ken Walker, kenneth.walker@sfo.harbinger.com
Harbinger Coporation, Concord, Ca. 94520



From icon-group-sender  Mon Aug 31 09:30:47 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) with SMTP id JAA01134
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Mon, 31 Aug 1998 09:30:45 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA20152; Mon, 31 Aug 1998 09:30:17 -0700
Date: Mon, 31 Aug 1998 09:23:41 -0700
From: kwalker@sfo.harbinger.com (Ken Walker)
Message-Id: <199808311623.JAA20603@varda.premenos.com>
To: icon-group@optima.CS.Arizona.EDU
Subject: Re: A hood fan is guard to mind.
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Md5: 5qDOZPLcfV6WSo1n3qDQDQ==
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Content-Transfer-Encoding: 7bit
Status: RO
Content-Length: 1450

And here's a solution that does not involve passing "sentence" down
the call chain.

procedure main()
   local fillers, firstChar, remainingChars

   #
   # The sentence is broken into three categories:
   #   parts not participating in the substitution
   #   the initial characters of the words involved in substitution
   #   the tail of the words involved in substitution
   #
   fillers :=    ["A ",    " ",  " is ", " to ",    "."]
   firstChar :=     ["g",    "m",   "h",     "f"]
   remainingChars := ["ood",  "an",  "ard",    "ind"]

   every write(polyspoonerism(fillers, firstChar, remainingChars))
end

procedure polyspoonerism(
  fillers,        # words before each participating word
  firstChar,      # first characters to use in substitutions
  remainingChars) # words (w/out 1st char) that are participating

  local indx

  if *remainingChars = 0 then
     return fillers[1] | ""

  #
  # Use each of the remaining first characters
  #
  every indx := 1 to *firstChar  do {
     #
     # Construct the first part of the sentence then make a
     # recursive call to construct the rest.
     #
     suspend fillers[1] || firstChar[indx] || remainingChars[1] ||
        polyspoonerism(fillers[2:0], firstChar[1:indx] ||| firstChar[indx+1:0],
        remainingChars[2:0])
  }
end

This is finally starting to look reasonably elegant and Iconish.

Ken Walker, kenneth.walker@sfo.harbinger.com
Harbinger Coporation, Concord, Ca. 94520



From icon-group-sender  Mon Aug 31 12:22:25 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) with SMTP id MAA08634
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Mon, 31 Aug 1998 12:22:24 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA21839; Mon, 31 Aug 1998 12:22:00 -0700
Message-Id: <4FD6422BE942D111908D00805F3158DF0757B531@RED-MSG-52>
From: Todd Proebsting <toddpro@microsoft.com>
To: icon-group@optima.CS.Arizona.EDU
Subject: RE: A hood fan is guard to mind.
Date: Mon, 31 Aug 1998 10:19:34 -0700
X-Mailer: Internet Mail Service (5.5.2232.9)
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 219

Somebody might find the following interesting.  On a slow day.

procedure main()
   local L, t
   L := "gmhf"
   every t := !L || !L || !L || !L & *cset(t)=4 do
      write(map("A good man is hard to find", L, t))
end


From icon-group-sender  Mon Aug 31 12:22:17 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) with SMTP id MAA08631
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Mon, 31 Aug 1998 12:22:16 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA21841; Mon, 31 Aug 1998 12:21:52 -0700
Date: Mon, 31 Aug 1998 09:54:27 -0700
From: swampler@noao.edu (Steve Wampler)
Subject: Re: Integers in Posix-compliant icon
To: bezeau@unb.ca
Cc: icon-group@optima.CS.Arizona.EDU
Message-Id: <swampler-9807311654.AA001317106@orpheus.gemini.edu>
In-Reply-To: <Pine.SOL.3.96.980830144657.1473A-100000@sol.sun.csd.unb.ca>
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 706


Larry - Icon accepts integers expressed in 'radix' notation.  For example,
   integers represented in binary are 2r1010, 2r111, etc.  Icon handles
   any base up to base 36, so 23rd is the base 23 number 'd'.  3rd is
   not an integer because 'd' is not a legal digit in base 3.  Note
   that hexidecimal numbers are represented as 16rFF, 16r1af9, etc.

   In general the digits allowed in integers in base N are the first
   N characters of the string "0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZ"
   (ok, so upper- or lowercase letters work...).



--
Steve Wampler - swampler@gemini.edu [Gemini 8m Telescopes Project (under AURA)]
The gods that smiled at your birth are now laughing openly. (Fortune Cookie)


From icon-group-sender  Tue Sep  1 08:26:55 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) with SMTP id IAA24338
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Tue, 1 Sep 1998 08:26:55 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA22303; Tue, 1 Sep 1998 08:26:30 -0700
To: icon-group@optima.CS.Arizona.EDU
Date: 1 Sep 1998 00:45:00 GMT
From: sobotka@axess.com (Henry Sobotka)
Message-Id: <6sfg2c$m9e$1@news2.tor.accglobal.net>
Organization: ICAN.Net Customer
Sender: icon-group-request@optima.CS.Arizona.EDU
References: <35E75B0B.142E9658@aasp.net>
Subject: Re: Icon, OS/2, and interfacing C code
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 996

In article <35E75B0B.142E9658@aasp.net>, jules@aasp.net says...
>
>Hello:
>
>I use Icon under OS/2 and I am interested in interfacing some Icon code
>to a C program.
>
>I don't have the (obsolete) CSET/2 compiler, nor the Visual Age stuff.
>
>Instead, I use EMX/GCC and various other commercial products.
>
>So, is this possible?  Will my application live?  Hello?

Yes. I built an older Icon 9 with emx+gcc last fall, then this spring merged 
the changes with the (then?) latest sources. Unfortunately, the emx+gcc 
binaries create a very stiff (reboot) zombie on exit which I've wanted to get 
rid of before making the update available. If you're interested in building 
it, I'd be happy to provide you with a copy of my code; another pair of eyes 
might be able to spot the bug more easily.

Also, now that I'm far more at ease using the VisualAge compiler than back 
then, I want to try an icc build in the hoping of producing at least one clean 
set of binaries for OS/2 Icon users.

Henr



From icon-group-sender  Tue Sep  1 08:27:05 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) with SMTP id IAA24346
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Tue, 1 Sep 1998 08:27:05 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA22934; Tue, 1 Sep 1998 08:26:39 -0700
Date: Tue, 1 Sep 1998 09:54:19 -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: Integers in Posix-compliant Icon
Message-Id: <Pine.SOL.3.96.980901095141.25092A-100000@sol.sun.csd.unb.ca>
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 590

        Many thanks to Steve Wampler and R. Clayton for reminding
me of the radix notion for Icon integers and for explaining why
"3rd" is not an integer while "23rd" is.  I have since re-read the
relevant sections in Griswold and Griswold.  In retrospect, I
should have suspected something like this since any attempt to
convert normal integers followed by the other two ordinal suffixes
(in English) to integers always failed.  In my program, I now 
search for and remove all ordinal suffixes very early in the
analysis process.
                                   Larry   Bezeau@UNB.Ca



From icon-group-sender  Tue Sep  1 12:40:11 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) with SMTP id MAA08404
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Tue, 1 Sep 1998 12:40:04 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA22417; Tue, 1 Sep 1998 12:39:39 -0700
From: abs@reba.csd.sgi.com (Alan Saichek)
Message-Id: <199809011555.PAA07853@reba.csd.sgi.com>
Subject: Jcon port to Windows: MKS Korn shell in Msoft NT Add-On Pack
To: icon-group@baskerville.CS.Arizona.EDU
Date: Tue, 1 Sep 1998 08:55:33 -0700 (PDT)
X-Mailer: ELM [version 2.4 PL23]
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
X-Mime-Autoconverted: from 8bit to quoted-printable by rock.csd.sgi.com id IAA06035
Content-Transfer-Encoding: 8bit
X-Mime-Autoconverted: from quoted-printable to 8bit by baskerville.CS.Arizona.EDU id IAA25888
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Content-Transfer-Encoding: 8bit
Status: RO
Content-Length: 1780



	On Jcon portability, an immediate (but costly?) solution for 
	Windows is available:

	http://www.cs.arizona.edu/icon/jcon/doc/intro.htm

	    mentions:

	"The jcont script that directs compilation and linking is a 
	400-line Korn shell script. While this works well for Unix, 
	it is the single largest impediment to a non-Unix port. We 
	believe that writing a replacement program would be the 
	easiest route to a Windows or Macintosh port."

	On the Windows front, a solution is available now from MKS for 
	all Windows platforms, and a Microsoft solution for Windows NT
	is/will be available, per the May Microsoft press release:

	http://www.microsoft.com/presspass/press/1998/may98/ntunixpr.htm

	"LAS VEGAS - May 6, 1998 Microsoft Corp. today announced 
	development of the Microsoft. Windows NT. Services for 
	UNIX Add-On Pack ... is scheduled to deliver a beta version 
	of the Windows NT Services for UNIX Add-On Pack this summer 
	for both Intel and Alpha platforms. 

	Common scripting across platforms. Windows NT Services for UNIX 
	will also allow users to run many existing UNIX scripts on Windows 
	NT-based systems. As part of these scripting capabilities, 
	Microsoft has licensed from Mortice Kern Systems Inc. (MKS) more 
	than 25 UNIX scripting commands and its leading KornShell, which
        gives users the ability to automate common processes and 
	administrative tasks across both Windows NT and UNIX platforms. 
	MKS is a leading provider of Windows NT scripting and migration 
	tools and software configuration management (SCM) products."
	
	http://www.mks.com

	Enjoy, Alan
-- 
Alan B. Saichek                                 	abs@csd.sgi.com
Member Technical Staff - Engineering			(650) 933-2507
Worldwide Customer Service, Service Tools R&D


From icon-group-sender  Wed Sep  2 13:20:43 1998
Return-Path: <icon-group-sender>
Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
	by baskerville.CS.Arizona.EDU (8.9.1a/8.9.1) with SMTP id NAA10101
	for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Wed, 2 Sep 1998 13:20:43 -0700 (MST)
Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
	id AA24624; Wed, 2 Sep 1998 13:20:17 -0700
To: icon-group@optima.CS.Arizona.EDU
Date: Wed, 02 Sep 1998 13:58:50 GMT
From: cwills@bix.com
Message-Id: <6sjius$i4v$1@nnrp1.dejanews.com>
Organization: Deja News - The Leader in Internet Discussion
Sender: icon-group-request@optima.CS.Arizona.EDU
References: <35E75B0B.142E9658@aasp.net>, <6sfg2c$m9e$1@news2.tor.accglobal.net>
Subject: Re: Icon, OS/2, and interfacing C code
Errors-To: icon-group-errors@optima.CS.Arizona.EDU
Status: RO
Content-Length: 717

In article <6sfg2c$m9e$1@news2.tor.accglobal.net>,
  sobotka@axess.com (Henry Sobotka) wrote:
>
> Also, now that I'm far more at ease using the VisualAge compiler than back
> then, I want to try an icc build in the hoping of producing at least one
clean
> set of binaries for OS/2 Icon users.
>
> Henr
>

I'm just about to send in a 9.3.1 version of Icon for OS/2.
I have the compiles done, and just need to "package" it up.  I've included
external functions, and dynamic loading as well.  Note that this is not
the graphical version, just the line mode.

Cheyenne Wills

-----== Posted via Deja News, The Leader in Internet Discussion ==-----
http://www.dejanews.com/rg_mkgrp.xp   Create Your Own Free Member Forum