Path: utzoo!attcan!utgpu!news-server.csri.toronto.edu!mailrus!cs.utexas.edu!sdd.hp.com!zaphod.mps.ohio-state.edu!rpi!sci.ccny.cuny.edu!phri!cmcl2!lanl!beta.lanl.gov!hinker
From: hinker@acl.lanl.gov (Paul J. Hinker)
Newsgroups: comp.sys.amiga
Subject: Re: FORTRAN recommendation?
Message-ID: <HINKER.90Jun4091248@nohope.acl.lanl.gov>
Date: 4 Jun 90 15:12:48 GMT
References: <Jun.2.10.31.07.1990.15594@pilot.njin.net> <30484@cup.portal.com>
Sender: news@lanl.gov
Reply-To: hinker@acl.lanl.gov
Distribution: usa
Organization: Advanced Computing Lab, LANL, NM
Lines: 6
In-reply-to: dedina@cup.portal.com's message of 3 Jun 90 07:08:48 GMT


   Close, Michael, but no banana.  FORTRAN 66 requires that all DO loops
execute at least once.  FORTRAN 77 will allow the body of a DO to be skipped
if the loop bounds are such that the start is larger than the end.  Most
FORTRAN 77 compilers have a command line flag that will follow the FORTRAN 66
way of handling DO loops.  I can't comment on Amiga FORTRAN compilers.