Path: utzoo!attcan!utgpu!jarvis.csri.toronto.edu!mailrus!tut.cis.ohio-state.edu!ucbvax!hplabs!hplabsz!mayer
From: mayer@hplabsz.HPL.HP.COM (Niels Mayer)
Newsgroups: comp.windows.x
Subject: Re: Motif Bugs - E-mail Address
Message-ID: <4255@hplabsz.HPL.HP.COM>
Date: 29 Oct 89 02:40:48 GMT
References: <8910131307.AA16724@osf.osf.org>
Reply-To: mayer@hplabs.hp.com (Niels Mayer)
Organization: Hewlett-Packard Labs, Software Technology Lab, Palo Alto, CA.
Lines: 21
Summary:
Expires:
Sender:
Followup-To:

> The MOTIF libraries have been passed with a Branch Flow Analysis
> Tool showing that the regression test suite is covering 85% of the code.
> We know the bugs are in those 15% left and there may be one per line
> in this code. However we are confident in the overall quality.

Does BFA make any sense at all for Xtoolkit based program? Much of the
program flow inside an Xt program comes from executing code indirectly, via
a function pointer whose value can seem pretty arbitrary from the point of
vue of an unintelligent analysis program. Can BFA really tell you anything
about programs that use such techniques to emulate "object oriented
programming"?  Seems like you'd need to do some pretty extensive dataflow
analyses, along with being able to predict the outcomes of all possible
input events sequences that could be generated when "1000 monkeys" get a
hold of your program and start poking at it's interface.

-------------------------------------------------------------------------------
	    Niels Mayer -- hplabs!mayer -- mayer@hplabs.hp.com
		  Human-Computer Interaction Department
		       Hewlett-Packard Laboratories
			      Palo Alto, CA.
				   *