Path: utzoo!censor!geac!torsqnt!news-server.csri.toronto.edu!cs.utexas.edu!sdd.hp.com!wuarchive!uunet!microsoft!jimad
From: jimad@microsoft.UUCP (Jim ADCOCK)
Newsgroups: comp.lang.c++
Subject: Re: Incompatible changes in C++
Message-ID: <59766@microsoft.UUCP>
Date: 12 Dec 90 20:16:58 GMT
References: <9011190326.AA13113@mole.ai.mit.edu> <11642@alice.att.com> <59305@microsoft.UUCP> <192@nazgul.UUCP>
Reply-To: jimad@microsoft.UUCP (Jim ADCOCK)
Organization: Microsoft Corp., Redmond WA
Lines: 27

In article <192@nazgul.UUCP> bright@nazgul.UUCP (Walter Bright) writes:
|In article <59305@microsoft.UUCP> jimad@microsoft.UUCP (Jim ADCOCK) writes:
|/*I'd* rather see C++ cleaned up a little bit, make some of the rules a
|/little simpler rather stretching the rules are far as possible in order
|/to encompass as many old C and C++ programs as possible.
|
|Could you please post some rules that you feel should be simplified? I'm
|interested as I have my own agenda :-)

....

Here's my list off the top of my head [not particularly ordered]

1) Issues of const, volatile and aliasing need to be seriously addressed.

2) Coercion rules ought to be more C++-consistent.  Arrays come
first to mind. [Practically, I think the historical implied coercion
of an array to a pointer to its first element has to stay, but to me
that doesn't imply exact array types as formal arguments ought to be
mis-represented as being pointer declarations [perhaps they should be
mis-represented as references to the exact type instead? 1/2 :-]]

3) The rules for argument matching takes three full pages in the working
paper.  I am not optimistic that any two vendors are going to implement
these the same, nor that any programmer can remember even the smallest
part of these rules.  Can't they be simplified, so that they fit on
maybe *two* pages, for example?