Path: utzoo!attcan!utgpu!jarvis.csri.toronto.edu!cs.utexas.edu!usc!merlin.usc.edu!nunki.usc.edu!jeenglis
From: jeenglis@nunki.usc.edu (Joe English)
Newsgroups: comp.lang.c++
Subject: Re: Extension to switch statement.
Message-ID: <6844@merlin.usc.edu>
Date: 4 Dec 89 22:19:36 GMT
References: <492@contex.UUCP>
Sender: news@merlin.usc.edu
Reply-To: jeenglis@nunki.usc.edu (Joe English)
Organization: University of Southern California, Los Angeles, CA
Lines: 19

ed@contex.UUCP (Edward Bagdonas) writes:
>	I am curious if any discussion or thought has been given to 
>	extending the "switch" statment in C++ to handle non constant 
>	expressions.

That got hashed out in comp.lang.c++ a while ago;
I think the main reason for not allowing non-constant
and non-integer (class) expressions in case labels is
that it's impossible to verify at compile time that
each of the cases will have distinct values.  That
would significantly change the semantics of switch().

(Changing the semantics of switch() has been discussed
as well, but most of the alternate proposals so far
would violate the tenets of structured programming.)

--Joe English

  jeenglis@nunki.usc.edu