Path: utzoo!attcan!uunet!lll-winken!csd4.milw.wisc.edu!mailrus!tut.cis.ohio-state.edu!ucbvax!ucsd!ucsbcsl!eiffel!bertrand
From: bertrand@eiffel.UUCP (Bertrand Meyer)
Newsgroups: comp.lang.eiffel
Subject: Re: Multi-branch instructions and unique values (Version 2.2 preview)
Message-ID: <130@eiffel.UUCP>
Date: 23 Mar 89 03:18:14 GMT
References: <122@eiffel.UUCP> <4377@enea.se>
Organization: Interactive Software Engineering, Santa Barbara CA
Lines: 36

From <4377@enea.se>, sommar@enea.se (Erland Sommarskog):

> I object to [...] the comparison to Ada and Pascal.
> [... Discussion of Pascal and Ada mechanism ..]
    
    I checked my original message and I don't think anything it said
contradicted Mr. Sommarskog's discussion of the Pascal and
Ada mechanisms. My only precise reference to these languages in
the description of the multi-branch choice <122@eiffel.UUCP> was:

>> The test instruction (as its Pascal  and  Ada  counterparts)  is
>> potentially  dangerous  because it reflects the programmer's
>> expectation that the value of e will be one  of  the  stated test values.

    I think this statement is correct; in particular, the Pascal and Ada
techniques described by Mr. Sommarskog in his message do not invalidate it.

    I also considered in detail Mr. Sommarskog's arguments in favor of an
optional ``else'' clause and I think he is right. The mechanism will
be emended accordingly. An exception will still be raised if no ``else''
clause is present and the value of the discriminant matches none of the
values listed in the ``when'' clause. 

    There will also be one minor other change: the keyword for the
multi-branch instruction will be something else than ``test''.
It occurred to me that at least one out of every three new users
is bound to write a first trial class called TEST, implying that
his first encounter with the Eiffel tools would now be some message
of the form ``test is a reserved word and cannot be used as a class
name''. Better choose a less obvious keyword.

    I am grateful to Mr. Sommarskog for his observations.
-- 

-- Bertrand Meyer
bertrand@eiffel.com