Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!sun-barr!olivea!uunet!spool.mu.edu!sdd.hp.com!hplabs!hplabsb!seroussi
From: seroussi@hplabsb.HP.COM (Gadiel Seroussi)
Newsgroups: comp.sys.handhelds
Subject: Re: HP48 compact arrays
Message-ID: <5968@hplabsb.HP.COM>
Date: 8 Feb 91 18:51:40 GMT
References: <5966@hplabsb.HP.COM> <2918@charon.cwi.nl>
Organization: Hewlett Packard Laboratories, Palo Alto
Lines: 21

In article <2918@charon.cwi.nl>, jurjen@cwi.nl (Jurjen NE Bos) writes:
> How do we know for sure that the HP48 ROM does not make any assumptions about
> the comtents of a Linked Array?  The name suggests that the contents may ahve
> a structure that contains addresses of other objects............

We don't know for sure. The question is, who looks at the linked array? These 
programs do not use any built-in function or ROM object to access the 
linked array. The only access is thru the ML code, which actually ignores the 
object header (the latter is only looked at during argument checking). 
I agree that there is a remote possibility that the garbage collector could 
object, but that seems unlikely, and after very extensive use of these objects, 
I have yet to encounter that problem.

> Until that moment, I don't consider it a safe idea to use Linked Arrays as
> objects in a program.

That is, of course, very understandable (remember all disclaimers), and  
a matter of personal preference. I think that, among all the "officially unsafe"
things we do with our calculators, this one is probably on the mild side.

Gadiel