NetNews Usenet Archive 1993 #3/NN_1993_3.iso/spool/comp/lang/cplus/20080

Internet Message Format | 1993-01-28 | 3.1 KB

Path: sparky!uunet!crdgw1!rpi!usc!howland.reston.ans.net!spool.mu.edu!agate!doc.ic.ac.uk!pipex!demon!trmphrst.demon.co.uk!nikki
From: [e-mail] (Nikki Locke)
Newsgroups: comp.lang.c++
Subject: Re: Help w/ default arguments and more
Message-ID: <[e-mail]>
Date: 27 Jan 93 11:32:09 GMT
References: <[e-mail]>
Sender: [e-mail]
Reply-To: [e-mail]
Organization: Trumphurst Ltd.
Lines: 76
X-Mailer: cppnews $Revision: 1.31 $

In article <[e-mail]> [e-mail] (Suman Kumar Inala) writes:
> I would like to be able to do something like this:
>
> class X {
>     int x;
> public:
>     X(void) { x = 3; }
>     int func(int arg = this->x) { return arg; }
> };
>
> In other words, I would like the default argument to
> one of a classes methods to be one of the members of
> the class.  This is of course possible if I were
> to make x static, but this is not acceptable.  BC complains
> that this may only be used in member functions; simply
> expressing "int arg = x" gives the legitimate complaint
> that x is not associated with an object.  The
> following is a workaround, but I don't like it much:
>
> class X {
>     int x;
> public:
>     X(void) { x = 3; }
>     int func(void) { return x; }
>     int func(int arg) { return arg; }
> };
>
> Is there a better solution?

I don't think there is. If arg never takes a particular value (e.g. -1),
then you can do ...
      int func(int arg = -1) { return (arg == -1) ? x : arg; }

This is often most useful for pointer arguments, which you default to
NULL.

> Also, I know I'm missing
> something simple here, but the following doesn't work:
>
>
> #define SWAP_INT(a, b) { a ^= b; b ^= a; a ^= b; }
> inline void swap(int a, int b) {  a ^= b; b ^= a; a ^= b; }
>
> The SWAP_INT macro works, but not my inline function.  Why?

Because you are passing the arguments by value. An inline function has
exactly the same effect as a non-inline function (indeed, the compiler may
arbitrarily decide not to inline it). You need ...

  inline void swap(int &a, int &b) {  a ^= b; b ^= a; a ^= b; }

> Also, I declared a class Map which contains a 2D array.  I
> would have liked to overload the [] operator to take two
> arguments, so that I could treat it in a natural way,
> indexing the elements of the map using array notation.
> (Is this perhaps bad practise?)
>
> Map map(80, 20); // Dimensions of map are 80 columns, 20 rows
>
> cout << map[3][4];
>
> But the [] operator may only be declared to take one argument.
> It would be nice if the following notation could be used were
> it possible for [] to take more than one argument, though it
> is not traditional C notation:
>
> cout << map[3, 4];

Surely it would be better to be able to use map[3][4], just like a "real"
array. There are ways of doing this - usually using a helper class which
emulates a single dimension of the array. I believe there is an example
in the FAQ for this group.

--
Nikki Locke,Trumphurst Ltd.(PC and Unix consultancy) [e-mail]
trmphrst.demon.co.uk is NOT affiliated with ANY other sites at demon.co.uk.