Structures, Pointer and gets

[Joe Wright]

Re: Structures, Pointer and gets

Thomas Stegen wrote:[color=blue]
>
> Joe Wright wrote:
>[color=green]
> > Thomas Stegen wrote:
> >[color=darkred]
> >>Joe Wright wrote:
> >>
> >>
> >>>Yes indeed. I am of the old school that differentiates between values
> >>>and objects. I read pointer as an object which may hold a value which is
> >>>the address of another object. Others call an address value a pointer. I
> >>>understand but I disagree.
> >>
> >>You disagree with a standard definition? Interesting... :)
> >>[/color]
> >
> > Why interesting?[/color]
>
> Because the most you can say is that the definition is flawed.
> A definition is, by definition, correct.
>[color=green]
> > Does the Standard confuse value and object? Maybe.[/color]
>
> It does not. Objects can have type pointer to T. Values can have type
> pointer to T. Sheldon quoted an appropriate from the standard with which
> you disagree (You are saying that the address of an object is not a
> pointer.)
>[/color]
I agree that both objects and values can have type 'pointer to T'.
Because a value has pointer type does not make it a pointer. From K&R2
the first sentence of Chapter 5: reads "A pointer is a variable that
contains the address of a variable." I know that K&R is not the Standard
but if Kernighan and Ritchie agree on something about C, it's good
enough for me.
[color=blue][color=green]
> > Let's examine two objects..
> >
> > char arr[10];
> > char *arp;[/color]
>
> As your example demonstrates something different from what you were
> saying I snipped most of it.
>
> You are saying (at least said) that arp is a pointer, whereas &arp is
> not. Note that when arr is evaluated it is converted to a value which
> is the address of the first element in the array. This value is not
> stored in any object and so is by your disagreement with the standard
> not a pointer. You just have to look at clause 6.3.2.1#2 to see that
> this value is indeed a pointer.
>[/color]
I'll 'say' it again for the listening impaired. A value is not a pointer
even if it has pointer type. The pointer is the object that holds the
value.

It's not just about pointers. Consider..

int a; /* a is an object of type int */

a = 5; /* 5 is a literal value of type int */

Although (a == 5) is true, 5 is not an object.

I've had this discussion before. It is clear as crystal to me. That
others can't see it as I do truly baffles me.
--
Joe Wright http://www.jw-wright.com
"Everything should be made as simple as possible, but not simpler."
--- Albert Einstein ---

[Thomas Stegen]

Re: Structures, Pointer and gets

Joe Wright wrote:
[color=blue]
> I've had this discussion before. It is clear as crystal to me.[/color]

It is clear as crystal to me what you mean. It is still wrong.

--
Thomas.

[Joe Wright]

Re: Structures, Pointer and gets

Thomas Stegen wrote:[color=blue]
>
> Joe Wright wrote:
>[color=green]
> > I've had this discussion before. It is clear as crystal to me.[/color]
>
> It is clear as crystal to me what you mean. It is still wrong.
>[/color]
Pray tell. What is 'wrong' about anything I've said here?
--
Joe Wright http://www.jw-wright.com
"Everything should be made as simple as possible, but not simpler."
--- Albert Einstein ---

[Thomas Stegen]

Re: Structures, Pointer and gets

Joe Wright wrote:
[color=blue]
> Thomas Stegen wrote:[/color]
[color=blue][color=green]
>>It is clear as crystal to me what you mean. It is still wrong.
>>[/color]
>
> Pray tell. What is 'wrong' about anything I've said here?[/color]

"Others call an address value a pointer. I understand but I disagree."

I hope I never see you talking about a function returning pointers.
Oh and did you know that the unary & operator returns a pointer?
The standard is very clear on the matter.

If this is just your opinion please say so, because I am not
interested in discussing your opinion proper, which is what we
appear to be doing.

--
Thomas.

[Thomas Stegen]

Re: Structures, Pointer and gets

Joe Wright wrote:
[color=blue]
> Thomas Stegen wrote:
>[color=green]
>>Joe Wright wrote:
>>
>>[color=darkred]
>>>I've had this discussion before. It is clear as crystal to me.[/color]
>>
>>It is clear as crystal to me what you mean. It is still wrong.
>>[/color]
>
> Pray tell. What is 'wrong' about anything I've said here?[/color]

Ok, I've had a look at the standard and it does not define the
term "pointer" at all. So saying you disagree that the unary &
operator returns a pointer has no foundation anywhere. You even
disagree with K&R which states at least once that function
such and such returns a pointer to whatever. The standard also
does this. So the consensus between the standard and K&R is that
functions do return pointers. And the 2 other books I checked
also does this.

Also consider how awkward it is if only objects of type pointer
to T can be called pointer. You cannot say that a function takes
a pointer as an argument for example. It seems that pointer
should be used to mean anything (object or value) that has pointer
to T and if you want to make a distinction you use object of type
pointer to T or value of type pointer to T.

It certainly seems like the consensus is against you when the two
most authorative sources available on C programming is against you.

--
Thomas.

[Joe Wright]

Re: Structures, Pointer and gets

Thomas Stegen wrote:[color=blue]
>
> Joe Wright wrote:
>[color=green]
> > Thomas Stegen wrote:[/color]
>[color=green][color=darkred]
> >>It is clear as crystal to me what you mean. It is still wrong.
> >>[/color]
> >
> > Pray tell. What is 'wrong' about anything I've said here?[/color]
>
> "Others call an address value a pointer. I understand but I disagree."
>
> I hope I never see you talking about a function returning pointers.
> Oh and did you know that the unary & operator returns a pointer?
> The standard is very clear on the matter.
>[/color]
Really? Functions 'return' values. Operators 'yield' values. Values
which are addresses of objects have type 'pointer to T' but they are
still values and not pointers. This because pointers are objects!
[color=blue]
> If this is just your opinion please say so, because I am not
> interested in discussing your opinion proper, which is what we
> appear to be doing.
>[/color]
Are you becoming bored with me Thomas?

Can you give me your definition of 'pointer' which is not ambiguous? Is
it a value or an object? Is it both? Is it neither? Straighten me out
here.
--
Joe Wright http://www.jw-wright.com
"Everything should be made as simple as possible, but not simpler."
--- Albert Einstein ---

[Thomas Stegen]

Re: Structures, Pointer and gets

Joe Wright wrote:
[color=blue]
> Are you becoming bored with me Thomas?[/color]

[snip]

Sorry about that. I didn't mean to send that article and when
I did I tried to cancel it, but I was too late apparently. Apologies
for that. Everything I said in that article is hereby declared null
and void. The other article I sent a bit later is better argued and
more decent in all respects.

--
Thomas.

[Joe Wright]

Re: Structures, Pointer and gets

Thomas Stegen wrote:[color=blue]
>
> Joe Wright wrote:
>[color=green]
> > Thomas Stegen wrote:
> >[color=darkred]
> >>Joe Wright wrote:
> >>
> >>
> >>>I've had this discussion before. It is clear as crystal to me.
> >>
> >>It is clear as crystal to me what you mean. It is still wrong.
> >>[/color]
> >
> > Pray tell. What is 'wrong' about anything I've said here?[/color]
>
> Ok, I've had a look at the standard and it does not define the
> term "pointer" at all. So saying you disagree that the unary &
> operator returns a pointer has no foundation anywhere. You even
> disagree with K&R which states at least once that function
> such and such returns a pointer to whatever. The standard also
> does this. So the consensus between the standard and K&R is that
> functions do return pointers. And the 2 other books I checked
> also does this.
>
> Also consider how awkward it is if only objects of type pointer
> to T can be called pointer. You cannot say that a function takes
> a pointer as an argument for example. It seems that pointer
> should be used to mean anything (object or value) that has pointer
> to T and if you want to make a distinction you use object of type
> pointer to T or value of type pointer to T.
>
> It certainly seems like the consensus is against you when the two
> most authorative sources available on C programming is against you.
>[/color]
Only two? Anyway Thomas you have been civil and a credit to our
community. I concede. I thank you for your interest in my proposition
and note (sadly) that no one else here seems interested at all. :-)

Via con Dios.
--
Joe Wright http://www.jw-wright.com
"Everything should be made as simple as possible, but not simpler."
--- Albert Einstein ---

[Irrwahn Grausewitz]

Re: Structures, Pointer and gets

Joe Wright <joewwright@ear thlink.net> wrote:

[Are addresses pointers?]
[color=blue]
> and note (sadly) that no one else here seems interested at all. :-)[/color]

Hmmmmmm...... ;-)

How about:

In C, variables designate objects, and objects contain values.

Thus we can say:

Pointer variables designate objects that hold pointer values.

Expressions that evaluate to the address of an object yield such
a pointer value that can be stored in objects designated by a
pointer variable, a la:

int i = 42, *p = &i;

where p is a pointer variable and &i is a pointer value.

IIRC that was the consensus in one of the last discussions about
this matter. A google groups search will easily bring up some
threads where this has been discussed.

Regards
--
Irrwahn
(irrwahn33@free net.de)

[Richard Heathfield]

Re: Structures, Pointer and gets

Joe Wright wrote:
[color=blue]
> Thomas Stegen wrote:[color=green]
>>
>> It certainly seems like the consensus is against you when the two
>> most authorative sources available on C programming is against you.
>>[/color]
> Only two? Anyway Thomas you have been civil and a credit to our
> community. I concede. I thank you for your interest in my proposition
> and note (sadly) that no one else here seems interested at all. :-)[/color]

Oh, we're interested all right. It's just that we already had this
discussion quite recently, so we're just watching the fireworks. :-)

--
Richard Heathfield : binary@eton.pow ernet.co.uk
"Usenet is a strange place." - Dennis M Ritchie, 29 July 1999.
C FAQ: http://www.eskimo.com/~scs/C-faq/top.html
K&R answers, C books, etc: http://users.powernet.co.uk/eton

[Joe Wright]

Re: Structures, Pointer and gets

Richard Heathfield wrote:[color=blue]
>
> Joe Wright wrote:
>[color=green]
> > Thomas Stegen wrote:[color=darkred]
> >>
> >> It certainly seems like the consensus is against you when the two
> >> most authorative sources available on C programming is against you.
> >>[/color]
> > Only two? Anyway Thomas you have been civil and a credit to our
> > community. I concede. I thank you for your interest in my proposition
> > and note (sadly) that no one else here seems interested at all. :-)[/color]
>
> Oh, we're interested all right. It's just that we already had this
> discussion quite recently, so we're just watching the fireworks. :-)
>[/color]
Voyeurs? I brought this up here several years ago and Kaz shot it down.
I bring it up about once a year now because it really bugs me. Thanks
for listening.
--
Joe Wright http://www.jw-wright.com
"Everything should be made as simple as possible, but not simpler."
--- Albert Einstein ---

[Joona I Palaste]

[OT] Structures, Pointer and gets

Joe Wright <joewwright@ear thlink.net> scribbled the following:[color=blue]
> Via con Dios.[/color]

"Road with God"? Don't you mean "Vaya con Dios", "Walk with God"?

--
/-- Joona Palaste (palaste@cc.hel sinki.fi) ------------- Finland --------\
\-- http://www.helsinki.fi/~palaste --------------------- rules! --------/
"Roses are red, violets are blue, I'm a schitzophrenic and so am I."
- Bob Wiley

[Joe Wright]

Re: [OT] Structures, Pointer and gets

Joona I Palaste wrote:[color=blue]
>
> Joe Wright <joewwright@ear thlink.net> scribbled the following:[color=green]
> > Via con Dios.[/color]
>
> "Road with God"? Don't you mean "Vaya con Dios", "Walk with God"?
>[/color]
Yes, of course.
--
Joe Wright http://www.jw-wright.com
"Everything should be made as simple as possible, but not simpler."
--- Albert Einstein ---