Re: No Go To's Forever!

[dvdeug@x8b4e53cd.dhcp.okstate.edu (David Starner)]

On Wed, 22 Mar 2000 00:00:10 +0100, Tim Gahnstr�m <md9tim@mdstud.chalmers.se> wrote:
>I notice that alot of people really hate gotos...
>But why is it sooo bad can someone tell me that???

This is some real code (from Axelrod's simulation of the
Prisoner's Dillema). This, mind you, is opening up the
file and basically randomly grabing a function - there
is much, much worse in that file alone.

    FUNCTION K58R(J,M,K,L,R, JA)
C BY GLEN ROWSAM
C TYPED BY JM
    k58r=ja    ! Added 7/27/93 to report own old value
    IF (M .GT. 1) GOTO 99
    KAM = 0
    NPHA = 0
99  IF (KAM .GT. 6) GOTO 87
    IF (NPHA .GE. 1) GOTO 89
    IF ((M / 18) * 18 .EQ. M .AND. KAM .GT. 2) KAM = KAM - 1
    IF ((M / 6) * 6 .NE. M) GOTO 88
    IF (K .LT. M) GOTO 10
    IF (K * 10 .LT. M * 15) GOTO 11
    IF (K .LT. M * 2) GOTO 12
    IF (K * 10 .LT. M * 25) GOTO 13
    GOTO 88
10  KAM = KAM + 2
11  KAM = KAM + 1
12  KAM = KAM + 1
13  KAM = KAM + 1
    NPHA = 2
    GOTO 87
89  NPHA = NPHA - 1
    IF (NPHA .EQ. 0) GOTO 87
88    K58R = 0
    GOTO 86
87    K58R = 1
86  RETURN
    END

Much of the blame for goto hating must go to FORTRAN <= 77, which
requires goto's for common stuff and provides only numeric labels.

>I assume this is a typical example of bad goto use

At first glance I don't think it works, and it would be
a little easier to read if you used control structures
instead of gotos, but it's not a typical example.

>Is goto statments particular slow?

No, but a modern compiler can frequently handle the other
control structures faster (more information to optimize with.)
Anyway, the arguments against gotos has always been that they're
hard to read, and that readibility should come before optimization.

-- 
David Starner - dstarner98@aasaa.ofe.org
Only a nerd would worry about wrong parentheses with
square brackets. But that's what mathematicians are.
   -- Dr. Burchard, math professor at OSU