Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.2 9/18/84; site unc.unc.UUCP
Path: utzoo!linus!decvax!mcnc!unc!rentsch
From: rentsch@unc.UUCP (Tim Rentsch)
Newsgroups: net.lang.st80
Subject: Re: BS II user exchange?
Message-ID: <975@unc.unc.UUCP>
Date: Tue, 11-Feb-86 18:31:44 EST
Article-I.D.: unc.975
Posted: Tue Feb 11 18:31:44 1986
Date-Received: Thu, 13-Feb-86 19:13:31 EST
References: <1@randvax.UUCP>
Reply-To: rentsch@unc.UUCP (Tim Rentsch)
Distribution: net
Organization: CS Dept, U. of N. Carolina, Chapel Hill
Lines: 34
Summary: 


First, I would like to applaud Mike's initiative for BSII and
Smalltalk information exchange.  Let's build a real user community
out here.

Second, here is a horror story of my own.  This is not limited to
sun2's, or even suns;  this is a problem of the BSII interpreter in
general.  The problem?

	The message "at:put:" to subclasses of Array (that only have
	indexable instance variables) gets intercepted by the 
	interpreter!  For example, the method

		at: i put: v

		| |
		self halt.
		^ super at: i put: v

	in a subclass of Array will *not* be executed if the message
	is sent directly.  (Single stepping the debugger through the
	code will get to the method.)  The interpreter clearly states
	that it checks for this message "for speed".

I don't have a fix for the problem (probably taking out the special
check would work nicely).  I haven't minded because once the problem
is known the work arounds are easy.  Finding it, though, is another
thing altogether.  (Remember the announcement said to expect bugs!)

Lastly, I think that information exchange should not be limited to
the interpreter itself.  Let's exchange some Smalltalk goodies and
hints about using Smalltalk as a programming language/environment.

Next: a Smalltalk goodie to do filename expansion