NetNews Offline 2/NetNews Offline Volume 2.iso/news/comp/lang/c++-part1/9220

Text File | 1996-08-05 | 3.5 KB | 72 lines

Newsgroups: comp.lang.c++
Path: netcom.com!marnold
From: [e-mail] (Matt Arnold)
Subject: Re: Is there a standard for * and & placement style?
Message-ID: <[e-mail]>
Organization: NETCOM On-line Communication Services (408 261-4700 guest)
References: <[e-mail]> <[e-mail]> <4gckd5$[e-mail]> <[e-mail]> <4gk1b8$[e-mail]> <[e-mail]> <4gumu0$[e-mail]>
Date: Thu, 29 Feb 1996 08:15:35 GMT
Sender: [e-mail]

[e-mail] (David White) writes:

>[e-mail] (Matt Arnold) writes:


>>...is a "special case" for the "char *"-style, simply highlights that the
>>style, IMO, is conceptually inconsistent.  Certainly, the function itself
>>doens't have pointer-ness (although, as for as the language it concerned, it
>>resovles to an expression that does) and so it doesn't seem quite right to
>>associate the * with the function name.  It seems inconsistent that you
>>would make this concession, but turn around and insist the "char *p" makes
>>sense for the declaration of p.

>It wasn't a concession. I would use 'char *foo()'. I was just saying that
>I can understand why some people prefer 'char* foo()' (and 'char* p').

I see.

>I don't want to drag this on, because it doesn't bother me if others use
>the 'char* p' form, but I don't think my point came across too well.
>My main point concerned you and another poster suggesting that the form:
>    char* p, q;
>should declare both 'p' and 'q' to be char*. I

I didn't suggest that.  I pointed out that I personally never use that
form because I find it visually inconsistent and potentially bug-prone
(for me anyway).

I suppose it follows that I never use mutliple code statements on a
single line, delimited by commas, either.  Yuck!

> was just trying to demonstrate
>that, without making special rules for '*' (and '&'), this doesn't fit in
>with the language. As far as the compiler is concerned, *, &, [],
>() are operators that apply to an operand. In its parsing, it treats
>them that way whether it is an expression or a declaration. It is only
>because '*' happens to be a pre-operand operator that you are able
>to place the '*' with the 'char'. That's why I prefer the 'char *p'
>(and 'char *foo()') form; I just follow the same rules as the compiler
>for everything. I don't see any fundamental difference between
>'array of char' and 'pointer to char'; arrays and pointers are both
>types derived from some other, but you can't write 'char[10] p', simply
>because thej[10] is on the wrong side of 'p'.

Well, as I said before, this is one of those preference things, just like
where curly braces should go.

I still think it makes more sense to think of declaration statements as
types and instances of those types.  In a declaration, I'd rather see *
as an operator on the type rather than on the instance.  Because the
"loose" multiple-instance per type declaration rules interfere with this
view, I never use that form.

Regards,
-------------------------------------------------------------------------
Matt Arnold                       |        | ||| | |||| |  | | || ||
[e-mail]                |        | ||| | |||| |  | | || ||
Boston, MA                        |      0 | ||| | |||| |  | | || ||
617.389.7384 (h) 617.576.2760 (w) |        | ||| | |||| |  | | || ||
C++, MIDI, Win32/95 developer     |        | ||| 4 3 1   0 8 3 || ||
-------------------------------------------------------------------------