[Oberon] Re (2): GOTO

[peter at easthope.ca]

From:	Chris Burrows <cfbsoftware at gmail.com>
Date:	Mon, 13 Jan 2025 10:56:42 +1030
> We've been too busy having a lot of fun working on Oberon software and
> releasing new versions of Astrobe to spend much time chatting here  ;-)

Good to hear.

> https://xkcd.com/292/
> 
> But for a more serious observation (from 20 years ago!) refer to Section
> 4.2. *The GO TO statement* in *Good Ideas, Through the Looking Glass*,
> Niklaus Wirth, 2005:
> 
> https://people.inf.ethz.ch/wirth/Articles/GoodIdeas_origFig.pdf

Probably nobody here argues with prohibition of GOTO.  I wondered 
whether someone might comment on techniques for the difficulties 
alluded to. 
https://lists.dyne.org/lurker/message/20241102.144628.9f84c05a.en.html

(1) "... recursively wander around the incomplete LR automaton ..."     
I know about recursion.  Automaton (abstract machine, 
https://en.wikipedia.org/wiki/Abstract_machine) was covered briefly in 
CS 220.  LR = left-to-right.  I don't catch meaning beyond that.

(2) "... two independent sets of nested searches ..." I guess that 
means one loop inside another.  Two loops not really independent.   
I imagine a flag or return value which the inner search sets or returns.
After completion of the inner search, the outer loop branches 
according to the flag or return value.  No particular difficulty?

(3) "... Modula 3's exceptions ..." I haven't encountered Modula 3.  I 
guess these exceptions don't exist in the other Oberon languages.  A 
flag is declared where needed?

Regards,                              ... P.

-- 
VoIP:   +1 604 670 0140
work: https://en.wikibooks.org/wiki/User:PeterEasthope