Path: utzoo!utgpu!jarvis.csri.toronto.edu!rutgers!usc!cs.utexas.edu!yale!cs.yale.edu!news
From: news@cs.yale.edu (Usenet News)
Newsgroups: comp.lang.eiffel
Subject: Re: Questions on Eiffel inheritance and assertions
Summary: ALSO info on Meyer's book
Message-ID: <12654@cs.yale.edu>
Date: 24 Jan 90 07:50:35 GMT
References: <120002@gore.com>
Reply-To: jellinghaus-robert@yale.UUCP
Organization: Yale University Computer Science Dept, New Haven CT  06520-2158
Lines: 47

The questions you posed are specifically answered in each of two different
sources I have:  Bertrand Meyer's book _Object-Oriented Software Construction_
(New York, Prentice-Hall, 1988), ISBN 0-13-629049-3 in hardback (which is the
only available U.S. version), and the _Eiffel: The Language_ manual, 
TR-EI-17/RM from ISE.  The first will be referred to as [Meyer] and the second
as [Ref].

In article <120002@gore.com> jacob@gore.com (Jacob Gore) writes:
>1.  Are non-exported ("secret", right?) routines inheritable?

Yes.  All features defined in a class are inherited by any descendants of
that class, subject to the normal redefinition and renaming rules.  [Meyer,
section 1.5.3] [Ref, section 7.6].  Export clauses have nothing to do with
inheritance.

>2.  Is there a short way to include the precondition in the postcondition?
>For example, is there a shorthand for:
>
>	require
>		"a long list of assertions"
>	do
>		"routine body"
>	ensure
>		"same list of assertions as in the require clause"
>		"some new assertions"
>	end

What you are looking for is some set of assertions that is true for the
class whenever a routine is not being executed, right?  "A long list of
assertions" is true before every routine starts and after every routine
ends?  You want to use the "invariant" clause [Meyer, sections 7.4.1 and
7.4.2] [Ref, section 12.5].

>(I don't have an Eiffel compiler to try these; my computer is a NeXT.)

Procure the Eiffel 2.2 reference manual described above, and you'll have 
enough information to _write_ an Eiffel compiler.  Post to comp.lang.eiffel
for an Eiffel grammar (posted recently by ISE).  Many people will be
grateful for a non-ISE Eiffel implementation.

>Jacob Gore		Jacob@Gore.Com			boulder!gore!jacob


Rob Jellinghaus                | "Next time you see a lie being spread or a
jellinghaus-robert@CS.Yale.EDU |  bad decision being made out of sheer ignor-
ROBERTJ@{yalecs,yalevm}.BITNET |  ance, pause, and think of hypertext."
{everyone}!decvax!yale!robertj |     -- K. Eric Drexler, _Engines of Creation_