Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!usc!zaphod.mps.ohio-state.edu!uakari.primate.wisc.edu!aplcen!haven!adm!cmcl2!lanl!jlg
From: jlg@lanl.gov (Jim Giles)
Newsgroups: comp.lang.misc
Subject: Re: Answers, Chapter 1: TeX (was C's sins... and others)
Message-ID: <3656@lanl.gov>
Date: 23 Oct 90 21:28:07 GMT
References: <66253@lanl.gov> <14269@goofy.megatest.UUCP>
Organization: Los Alamos Natl Lab, Los Alamos, N.M.
Lines: 32

In article <14269@goofy.megatest.UUCP>, djones@megatest.UUCP (Dave Jones) writes:
> From article <66253@lanl.gov>, by jlg@lanl.gov (Jim Giles):
> > ... There are _NO_ pointers in
> > the source code for TeX 'packed array tries'.  The code does everything
> > through the use of Pascal arrays.
> 
> For theoretical considerations, an index into an array is not much
> different from a pointer is it? Range-checking is the only significant
> difference I can think of, and that only makes incorrect programs
> fail nicer. (Pointers also have some measure of range-checking,
> "segmentation violation" for example. Never happened to *me* of
> course, but I've heard rumors.)

For thetheoretical considerations, pointers and one-dimensional arrays
are indeed very similar.  But the context of the discussion makes it
clear that the person who posted TeX as an example of pointer efficiency
was of the opinion that pointers and arrays differed importantly.  In
fact, in the same news article that he proposed TeX, he also gave
examples which attempted to prove the supposed superiority of pointers
to arrays.

Of course, pointers and one dimensional arrays - in spite of their
similarity - are indeed different.  For one thing: arrays are bounded.
For another: distinct arrays are generally not aliased.  These two
properties make debugging and maintaining array codes somewhat easier
than pointer code.  Similarly, these two differences (particularly the
no-alias information) is useful to the compiler in generating more
efficient code.  After all, all the common optimization techniques
that compilers use are inhibited to one extent or other by the
possible presence of aliasing.

J. Giles