Path: csiph.com!v102.xanadu-bbs.net!xanadu-bbs.net!feeder.erje.net!eu.feeder.erje.net!newsfeed.fsmpi.rwth-aachen.de!eternal-september.org!feeder.eternal-september.org!news.eternal-september.org!.POSTED!not-for-mail
From: Martin Gregorie <martin@address-in-sig.invalid>
Newsgroups: comp.lang.java.programmer
Subject: Re: Java blunders
Date: Sat, 26 Jul 2014 21:04:10 +0000 (UTC)
Organization: A noiseless patient Spider
Lines: 55
Message-ID: <lr154a$rlo$2@dont-email.me>
References: <jvbts91r2f9onj1e3dbr5lvqp70hmofkum@4ax.com> <53cef75d$0$291$14726298@news.sunsite.dk> <7s6vs9p395bretqrtdvh9i6til83vfk52p@4ax.com> <53d04b5f$0$291$14726298@news.sunsite.dk> <sip0t9dqg0c77e7u4o7vo2brs8chsdqok7@4ax.com> <53d19752$0$295$14726298@news.sunsite.dk> <l2i7t9pahlpdmhlur2585hc21ilq8cg6c4@4ax.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
Injection-Date: Sat, 26 Jul 2014 21:04:10 +0000 (UTC)
Injection-Info: mx05.eternal-september.org; posting-host="e5160e2be48f1ea550cf61512446f525"; logging-data="28344"; mail-complaints-to="abuse@eternal-september.org";	posting-account="U2FsdGVkX18+g2kwy/5mr5FKUupPwDPb6HJ5LrhLgvQ="
User-Agent: Pan/0.139 (Sexual Chocolate; GIT bf56508 git://git.gnome.org/pan2)
Cancel-Lock: sha1:18BXN9WWccZv+Xr56OP1BDloeLs=
Xref: csiph.com comp.lang.java.programmer:29353

On Sat, 26 Jul 2014 08:31:10 -0700, Roedy Green wrote:

> On Thu, 24 Jul 2014 19:31:29 -0400, Arne Vajhøj <arne@vajhoej.dk> wrote,
> quoted or indirectly quoted someone who said :
> 
>>A new concatenation operator would not solve that problem in all cases.
> 
> Assume there were two distinct + operators.
>
That's just operator overloading. Algol 68 explicitly allowed that. The 
following was how the standard operator for adding integers was defined 
in the Algol 68R standard prelude:

op + = (int a,b) int: ( a + b );

But its easy to overload. If we assume that the string mode was declared 
as:

mode string = [1: 0 flex]]char;

then we can overload + to concatenate two strings:

op + = (string a,b) string: 
                    begin string c; 
                          for i from lwb a to upb a do c[i] := a[i] od;
                          int n := upb c;
                          for i from lwb b to upb b do c[i+n] := b[i] od;
                          c;
                    end

The syntax may not be quite right (its been 37 years and I don't have a 
compiler to check it with), but in Algol 68R you could (and I did) 
overload any operator by declaring extensions in just that way. 
  
> If you used the new operators to express your intent, an expression that
> is legal now, but gives the wrong result would become an illegal
> expression.
>
It can be very elegant. There was am Algol 68R extension for defining and 
plotting 2D diagrams It used two base modes: a sketch (which was a 
plottable object but yet scaled or coloured, etc. and a picture, which 
was scaled and ready for output. There were also a number of predefined 
sketch extensions, e.g. circle, rectangle, text and a few more overloads 
for + so you could write something like this:

picture p = circle(3.0) + rectangle(3.5, 4.5) + text("Title");
print(p);

Easy to read and write or what?


-- 
martin@   | Martin Gregorie
gregorie. | Essex, UK
org       |