Path: utzoo!censor!geac!torsqnt!news-server.csri.toronto.edu!cs.utexas.edu!wuarchive!uunet!mcsun!ukc!icdoc!cc.ic.ac.uk!umapd51
From: umapd51@cc.ic.ac.uk (W.A.C. Mier-Jedrzejowicz)
Newsgroups: comp.sys.handhelds
Subject: Re: BREAK and RETURN on the 48sx: where are they?
Summary: Not built in, but can be simulated.
Keywords: Structured programming
Message-ID: <1990Dec19.041347.26105@cc.ic.ac.uk>
Date: 19 Dec 90 04:13:47 GMT
References: <8601@dog.ee.lbl.gov>
Sender: Wlodek Mier-Jedrzejowicz
Organization: Imperial College Computer Centre
Lines: 31
Nntp-Posting-Host: suni2cc


The author asks - where are BREAK and RETURN on the HP48 - is he missing
something very basic? Well, yes Matthew, you are missing something very
basic - the designers of the RPL language believe deeply in structured
programming where blocks of code have only one entry point and one exit
point. This IS the basic point - so there is no method designed to let
you break out of a loop or return before the end of a subprogram.
Providing such commands would be contrary to the nature of the language.

This is not to say there is anything intrinsically wrong with BREAK or
RETURN - and being a typical HP hacker, I wondered how these could be
achieved - in fact I am writing a series of articles on that very topic
right now in DATAFILE (the journal of the British club for users of HP
handhelds - available to members regularly 8 times a year). Bill Wickes
himself provides some very welcome clues in his book HP41/HP48
Transitions. One example is that an error in an IFERR causes the rest of
the IFERR to be skipped - and you can use DOERR to force an error where
you wish - this combination lets you BREAK out of a block of code if you
want to do that. Other tricks are available too, and are mentioned in
passing in Bill's book. You can try causing an error in a subprogram to
leave that subprogram prematurely too.

Alternatively, you can treat RPL as a language which forces you to learn
structured programming with only one entry point and one exit point per
block of code - it is quite interesting to learn how this can be done if
you are coming from an unstructured language, such as FORTRAN.

Best Wishes,

Wlodek Mier-Jedrzejowicz, Space & Atmospheric Physics, Imperial College,
London.