Path: utzoo!utgpu!jarvis.csri.toronto.edu!helios.physics.utoronto.ca!ists!yunexus!maccs!howard
From: howard@maccs.dcss.mcmaster.ca (Howard Betel)
Newsgroups: comp.lang.eiffel
Subject: Coping with failure
Message-ID: <25EB6B0D.2249@maccs.dcss.mcmaster.ca>
Date: 28 Feb 90 06:09:16 GMT
Reply-To: howard@maccs.dcss.mcmaster.ca (Howard Betel)
Organization: McMaster University, Hamilton, Ontario
Lines: 34

I'm an undergrad who is taking a 4'th year MIS course.  Our text for the      
course is OBJECT-ORIENTED SOFTWARE CONSTRUCTION by Betrand Meyer.  The 
way the course is working is that each person gives an approximately
1.5 hour seminar on part of a chapter.  Mine was 7.9 to the end of the
chapter.  I had a question from the class that I wasn't sure of the
answer, so here goes.

Meyer states on page 168 (7.10.3)
   "Another way to express the same remark is to say that an operation
    may only succeed or fail; there is no middle ground.  (Trivial as
    this comment may seem, it is not satisfied by the exception  
    mechanisms of such languages as ADA, CLU or PL/I, which make
    it possible to exit from a routine without achieving the routine's
    intent, and without signalling failure to the caller."

Now I really do understand the importance of this (and the logic) however
since the assertion defaults are:
   "NO_ASSERTION_CHECK (Y):
    PRECONDITIONS (Y): ALL
    ALL_ASSERTIONS (Y):"

doesn't this allow failures?  If a function is executing and the     
postconditions are not met, it is possible for a function to complete
with a failure and the client not know that this occured.  Not only
does this seem possible, it seems probable considering that by default
postconditions are not checked.

Any answers?


-- 
Howard Betel                                    Howard@maccs.dcss.McMaster.CA
Dept of Computer Science                     ...!unet!utai!utgpu!maccs!howard
McMaster University