home *** CD-ROM | disk | FTP | other *** search
- Xref: sparky comp.edu:1332 comp.lang.fortran:3158
- Path: sparky!uunet!gatech!rpi!uwm.edu!ogicse!mintaka.lcs.mit.edu!ai-lab!life.ai.mit.edu!burley
- From: burley@geech.gnu.ai.mit.edu (Craig Burley)
- Newsgroups: comp.edu,comp.lang.fortran
- Subject: Re: Small Language Wanted
- Message-ID: <BURLEY.92Aug25164419@geech.gnu.ai.mit.edu>
- Date: 25 Aug 92 20:44:19 GMT
- Article-I.D.: geech.BURLEY.92Aug25164419
- References: <1992Aug25.095623.18808@news.Hawaii.Edu>
- Sender: news@ai.mit.edu
- Followup-To: comp.edu
- Organization: Free Software Foundation 545 Tech Square Cambridge, MA 02139
- Lines: 14
- In-reply-to: tholen@galileo.ifa.hawaii.edu's message of 25 Aug 92 09:56:23 GMT
-
- Presumably people complaining about Fortran's "computed GOTOs" really mean
- _assigned_ GOTOs, right? I mean, computed GOTOs are fairly easy to figure
- out; they're just a SELECT CASE with automatic "GOTO label" filled in for
- each case from 1 to N. Assigned GOTOs, on the other hand, as in "GOTO I",
- are really bizarre and hard to read and maintain, and often are used to
- implement "inline subroutines" and the like.
-
- Given SELECT CASE in F90, one doesn't really need computed GOTOs, but they
- aren't really bad for readability. But assigned GOTOs are really a pain
- for everone, programmer and implementer both!
- --
-
- James Craig Burley, Software Craftsperson burley@gnu.ai.mit.edu
- Member of the League for Programming Freedom (LPF)
-