V G wrote:

> On Sat, Oct 6, 2012 at 6:12 AM, V G <x.solarwind.x@gmail.com> wrote:
>=20
>> On Sat, Oct 6, 2012 at 2:16 AM, Tamas Rudnai <tamas.rudnai@gmail.com>wro=
te:
>>
>>> I am not arguing here, just saying that the type is the pointer or '*' =
in
>>> C. The mark 'int' is needed by the pointer arithmetic and to casting th=
e
>>> memory selected by the pointer. That's why it makes perfect sense to me=
 to
>>> use the version B 'int *a, *b, *c;' notation.
>>>
>>> By the way, we always can create our own type, so maybe what OP is look=
ing
>>> for is something like:
>>>
>>>     typedef int* pInt;
>>>     pInt d,e,f;
>>>
>>>
>> I'm not really "looking for" a new type. I'm just the type of guy who
>> needs to settle on a standard before I write 100 000 lines of code based=
 on
>> it. It helps me sleep at night knowing that my asterisks are aligned
>> correctly, you know?
>>
>> In either case, I'm sold on the type A (asterisk near the type) style,
>> since boost (http://boost.org) with 18 million+ lines of code is using
>> it. So I'm not the only one who likes it. As I said, it helps me sleep a=
t
>> night.
>>
>=20
> Edit: My word count tool says ~2 million lines of code in the boost sourc=
e
> tree. I misinterpreted the 18 million from this site:
> http://www.ohloh.net/p/boost

FWIW, Boost is a large library with a number of different styles. The
style is mostly consistent within a given library, but changes between
libraries. For example, a part of it uses the older "const int", a part
of it uses the newer "int const". I've never paid specific attention to
the pointer asterisk and reference ampersand placement, but if I had to
bet, I'd bet that they vary between libraries.

Also FWIW, I prefer form B, but there are people on my team who prefer
form A. (I write C++ for a living.)

Another consideration is the dereferencing usage. When you have a
pointer to int and you refer to its int value, you generally write
something like

  int a;
  int *p;
  *p =3D a;

The notion that "*p" is an int is part of pretty much every style, even
the ones that write "int* p;". I've yet to see a style that would write
this like

  int a;
  int* p;
  * p =3D a;

But then, this isn't a binding argument. Form A is quite viable and
often used, and so is form B.

Gerhard
--=20
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist
.