Newsgroups: comp.lang.c++,comp.lang.smalltalk,comp.object
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news2.near.net!satisfied.elf.com!news.mathworks.com!uunet!allegra!alice!ark
From: ark@research.att.com (Andrew Koenig)
Subject: Re: C++ Productivity
Message-ID: <D3y9My.5CH@research.att.com>
Organization: AT&T Bell Laboratories, Murray Hill NJ
References: <3hgmu5$qh9@panix.com> <3hmo3n$jid@uuneo.neosoft.com> <3hmr7j$k9@panix.com>
Date: Mon, 13 Feb 1995 17:52:10 GMT
Lines: 12
Xref: glinda.oz.cs.cmu.edu comp.lang.c++:112526 comp.lang.smalltalk:20793 comp.object:26657

In article <3hmr7j$k9@panix.com> dsiegel@panix.com (David Siegel) writes:

> If an automated conversion tool inserted "explicit" before every declaration
> of a single argument constructor, would that not guarantee identical behavior
> to the old standard (warts and all)?  New code, though, wouldn't be crippled
> by default.

Yes indeed.  Unfortunately, I don't know how to construct such
a tool that will always work in the presence of the C preprocessor.
-- 
				--Andrew Koenig
				  ark@research.att.com
