Path: utzoo!attcan!uunet!decwrl!bacchus.pa.dec.com!news.crl.dec.com!decvax.dec.com!hollie.rdg.dec.com!ryn.esg.dec.com!shlump.nac.dec.com!tkou02.enet.dec.com!diamond
From: diamond@tkou02.enet.dec.com (diamond@tkovoa)
Newsgroups: comp.std.c++
Subject: Re: A Proposal to Extend C++ with Variable-length Arrays
Message-ID: <1990Oct29.004609.19264@tkou02.enet.dec.com>
Date: 29 Oct 90 00:46:09 GMT
References: <11578@spool.cs.wisc.edu>
Reply-To: diamond@tkou02.enet.dec.com (diamond@tkovoa)
Organization: Digital Equipment Corporation Japan , Tokyo
Lines: 22

In article <11578@spool.cs.wisc.edu> bothner@sevenlayer.cs.wisc.edu (Per Bothner) writes:

[perfectly reasonable suggestion for variable-sized array syntax and
semantics, with perfectly reasonable justification]

>PRIOR ART:
>The GNU compilers gcc and g++ allow variable-sized auto (stack-based)
>arrays. Otherwise, I know of no prior C++ art.

Is the "Not Invented Here" disease really running so rampant in the C++
camp now?  Yes, (dare I say "obviously") variable-sized auto arrays are
a good idea, as has been proved in dozens of other languages.  PL/I had
it 26 years ago and there were surely others before that.

If "Not Invented Here" were always so rampant, C would not have enumerated
types (copied from a language which also sadly lacked variable-sized arrays
in its early years), and it wouldn't use "*" as a multiply operator.

-- 
Norman Diamond, Nihon DEC    diamond@tkov50.enet.dec.com
                                    (tkou02 is scheduled for demolition)
We steer like a sports car:  I use opinions; the company uses the rack.