home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
ftp.cs.arizona.edu
/
ftp.cs.arizona.edu.tar
/
ftp.cs.arizona.edu
/
icon
/
newsgrp
/
group95a.txt
/
000005_icon-group-sender _Sun Jan 8 17:04:03 1995.msg
< prev
next >
Wrap
Internet Message Format
|
1995-02-09
|
2KB
Received: by cheltenham.cs.arizona.edu; Sun, 8 Jan 1995 16:04:13 MST
Date: Sun, 8 Jan 1995 17:04:03 -0600
From: jeffery@runner.jpl.utsa.edu (Clinton L. Jeffery)
Message-Id: <9501082304.AA14576@runner.utsa.edu>
To: blake@edge.ercnet.com
Cc: icon-group@cs.arizona.edu
In-Reply-To: <3eo07m$k8h@edge.ercnet.com> (message from Blake McBride on 8 Jan 1995 06:20:38 GMT)
Subject: Re: Some Questions
Content-Length: 1052
Errors-To: icon-group-errors@cs.arizona.edu
3. Are there many people making real use of IDOL?
No. Idol is used by only a tiny segment of the Icon community.
Most Icon programmers either don't care for object-oriented concepts,
or never write large programs where objects would be useful. Icon's
built-in repertoire of data types and operations are so sophisticated
that using objects for smaller programs does not help very much.
Will it work in conjunction with the ICON->C compiler?
Yes. The current version of idol generates Icon code as its output,
which may then be used as input to iconc.
How effecient is it?
Idol's objects are just records, and member function invocations are just
field accesses -- which require constant time (and the constant is small).
Idol programs use a bit more space than Icon programs. Idol's efficiency
is one of its strong points. Its weak points (so far) stem from its
preprocessor-based implementation, which is not as convenient or as
efficient as would be an implementation that directly modifies Icon's
translator and run-time system.