NetNews Usenet Archive 1993 #1/NN_1993_1.iso/spool/comp/lang/scheme/2878

Text File | 1993-01-06 | 2.7 KB | 56 lines

Newsgroups: comp.lang.scheme
Path: sparky!uunet!zaphod.mps.ohio-state.edu!rpi!ghost.dsi.unimi.it!univ-lyon1.fr!chx400!bernina!neptune!nugget.inf.ethz.ch!marti
From: [e-mail] (Robert Marti)
Subject: Re: applying or
Message-ID: <[e-mail]>
Sender: [e-mail] (Mr News)
Nntp-Posting-Host: nugget.inf.ethz.ch
Organization: Dept. Informatik, Swiss Federal Institute of Technology (ETH), Zurich, CH
References: <[e-mail]> <[e-mail]> <[e-mail]> <[e-mail]>
Date: Wed, 6 Jan 1993 21:54:11 GMT
Lines: 43

In article <[e-mail]>,
[e-mail] (Doug Quale) writes in response
to an earlier article by me:
|> >Pascal as described in the 1st ed. of Jensen & Wirth did indeed specify
|> >that both the left _and_ right Boolean sub-expression of AND and OR must
|> >_always_ be evaluated.
|>
|> Could you provide the page number that says this?  This directly conflicts
|> with my recollection that the User Manual and Report is completely silent
|> on this issue.  This is one of the Report's several bugs, but it was
|> pretty good for its time.
|>
|> In a quick scan of the second edition (the one I had handy, but I can
|> get the first edition from the library) I couldn't find anything.  My
|> understanding is that short-circuit evaluation is compatible with the
|> requirements of the User Manual and Report, but is not required.  Many
|> early Pascal compilers didn't do short-circuit evaluation, and ultimately
|> many used the CDC implementation from ETH as the final arbiter on the
|> ambiguities in the definition.  (This was clearly less than ideal.)

Ahem.  I think you're pretty much correct here.  I couldn't find anything
in the 2nd Edition of J&W either, although I didn't search extensively.
(Interestingly enough, I had absolutely no problem to find the relevant
stuff in the "Report on the Programming Language Modula-2" contained in
N. Wirth, "Programming in Modula-2", 3rd, Corrected Ed., Springer, 1985,
where it says on p. 153:  "p OR q means 'if p then TRUE otherwise q' and
p AND q means 'if p then q otherwise FALSE'".)

However, Wirth himself assumed that Pascal implementations would always
evaluate both Boolean sub-expressons as can be seen e.g. in his Pascal
code in the 1st edition of "Algorithms and Data Structures".
Example:
  if p <> nil then
    if p^.key = x then
      begin
        (* do something here *)
      end

--
Robert Marti                    |  Phone:    [phone]
Informationssysteme             |  FAX:      [phone]
ETH-Zentrum                     |  E-Mail:   [e-mail]
CH-8092 Zurich, Switzerland     |