Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!ames!ncar!tank!mimsy!chris
From: chris@mimsy.UUCP (Chris Torek)
Newsgroups: comp.lang.misc
Subject: Re: "for" loops
Message-ID: <16022@mimsy.UUCP>
Date: 18 Feb 89 19:50:44 GMT
References: <15692@mimsy.UUCP> <8618@lanl.gov>
Organization: U of Maryland, Dept. of Computer Science, Coll. Pk., MD 20742
Lines: 24

In article <8618@lanl.gov> jlg@lanl.gov (Jim Giles) writes:
>Here's where we part company in a big way. I oppose ANY language construct
>that requires the programmer to know internals about how the construct
>is performed.

I will agree that `dark corners' or ambiguities are problematical (and
C is full of them, so that programmers must exercise restraint; alas,
many do not).  But the number of iterations of a loop had better not be
a `dark corner'!  If you cannot predict the trip count given the limits
and increments, you have no business writing that loop in the first
place!

>.... Many Fortran compilers actually implement DO loops with real limits
>in the above manner [avoiding accumulated roundoff errors].  The standard
>should really specify that this should be done so the user needn't worry
>about it.

And if the standard says it, or just as importantly if it does *not*,
programmers using the language should be aware of it!  If the standard
says that DO-loop trip counts depend on the phase of the moon, the
programmer should be prepared to look out the window as necessary.
-- 
In-Real-Life: Chris Torek, Univ of MD Comp Sci Dept (+1 301 454 7163)
Domain:	chris@mimsy.umd.edu	Path:	uunet!mimsy!chris