Need Explanation on Lvalue Assignment for the following snippet

[Pranav]

#include<stdio.h>
#include<conio.h>
int main()
{
const int i=6;
int k = 9;
int a[i], *c = a, *v = a ;
a[2] = 9;

*c++ = *v++; // ( 1 ) This line does not generate any exception /
error
k++ = k++; // ( 2 ) But this line is generating
k++ = 6; // ( 3 ) This Also Generating

getch();
}

Now I know that you cannot assign a value to an expression; But why
does the above (1) expression not generating a exception/error/
warning? and (2) & (3) are generating as they are supposed to. Please
explain this.
Is this due to operator precedence in the case(1), ie the expression
access the value and then assign them and then increment the pointers
and move forward. Correct if I am wrong..,

(I am using DevC++ Compiler for testing)

[pete]

Pranav wrote:

#include<stdio.h>
#include<conio.h>
int main()
{
const int i=6;
int k = 9;
int a[i], *c = a, *v = a ;
a[2] = 9;

*c++ = *v++; // ( 1 ) This line does not generate any exception /

You're assigning a value to (*c) there.
There nothing wrong with that.

error
k++ = k++; // ( 2 ) But this line is generating
k++ = 6; // ( 3 ) This Also Generating

Though (k++) has the same value
as (k) did before the increment,
(k++) is the result of an arithmetic operation.

A simpler example of the same problem, would be:

(k + 1) = 6;

(&k) is the address of (k), but (k + 1) doesn't have an address,
so there is no place to put the value of (6) into.

--
pete

[Richard Heathfield]

pete said:

Pranav wrote:

<snip>

>const int i=6;
int k = 9;
int a[i], *c = a, *v = a ;
a[2] = 9;

*c++ = *v++; // ( 1 ) This line does not generate any exception /

You're assigning a value to (*c) there.
There nothing wrong with that.

Look again! :-)

<snip>

--
Richard Heathfield <http://www.cpax.org.uk>
Email: -http://www. +rjh@
Google users: <http://www.cpax.org.uk/prg/writings/googly.php>
"Usenet is a strange place" - dmr 29 July 1999

[pete]

Richard Heathfield wrote:

pete said:

>Pranav wrote:

<snip>

>>const int i=6;
int k = 9;
int a[i], *c = a, *v = a ;
a[2] = 9;

*c++ = *v++; // ( 1 ) This line does not generate any exception /

You're assigning a value to (*c) there.
There nothing wrong with that.

Look again! :-)

I was just addressing the lvalue issue from the subject line.
The code has other problems too.
I read you other post in this thread.
I think you got them all.

--
pete

[vippstar]

On Sep 4, 11:19 am, pete <[e-mail]:

Richard Heathfield wrote:

pete said:

Pranav wrote:

<snip>

>const int i=6;
int k = 9;
int a[i], *c = a, *v = a ;
a[2] = 9;

>*c++ = *v++; // ( 1 ) This line does not generate any exception /
You're assigning a value to (*c) there.
There nothing wrong with that.

Look again! :-)

I was just addressing the lvalue issue from the subject line.
The code has other problems too.
I read you other post in this thread.
I think you got them all.

He did not mention <conio.hand getchr() not being defined by the
standard.

[pete]

vi******@gmail.com wrote:

On Sep 4, 11:19 am, pete <[e-mail]:

>Richard Heathfield wrote:

>>pete said:
Pranav wrote:
<snip>
const int i=6;
int k = 9;
int a[i], *c = a, *v = a ;
a[2] = 9;
*c++ = *v++; // ( 1 ) This line does not generate any exception /
You're assigning a value to (*c) there.
There nothing wrong with that.
Look again! :-)

I was just addressing the lvalue issue from the subject line.
The code has other problems too.
I read you other post in this thread.
I think you got them all.

He did not mention <conio.hand getchr() not being defined by the
standard.

Even though getchr() isn't defined by the standard,
I think you mean getch().

--
pete

[Nick Keighley]

On 4 Sep, 09:11, Richard Heathfield <[e-mail]:

Pranav said:

#include<stdio.h>
#include<conio.h>

non-standard header

int main()

int main(void)
is better style

{
const int i=6;
int k = 9;
int a[i], *c = a, *v = a ;

"a" appears to be a C99 VLA (older versions of C
don't support variable length arrays)

a[2] = 9;

*c++ = *v++; // ( 1 ) This line does not generate any exception /
error

Nevertheless, the behaviour of the statement, and therefore the program, is
undefined, because *v++ reads a[0], the value of which is indeterminate.

as c and v point to the same thing wouldn't there be UB even if
a[0] was initialised?

#include <stdio.h>

int main (void)
{
int a [6] = {0};
int *c = a;
int *v = a;

*c++ = *v++; /* UB I believe */
return 0;
}

<snip>

--
Nick Keighley

[vippstar]

On Sep 4, 11:33 am, pete <[e-mail]:

[e-mail] wrote:

He did not mention <conio.hand getchr() not being defined by the
standard.

Even though getchr() isn't defined by the standard,
I think you mean getch().

That's right.

[Willem]

Nick Keighley wrote:
) as c and v point to the same thing wouldn't there be UB even if
) a[0] was initialised?
)
) #include <stdio.h>
)
) int main (void)
) {
) int a [6] = {0};
) int *c = a;
) int *v = a;
)
) *c++ = *v++; /* UB I believe */
) return 0;
) }

Nope, that's not UB.
SaSW, Willem
--
Disclaimer: I am in no way responsible for any of the statements
made in the above text. For all I know I might be
drugged or something..
No I'm not paranoid. You all think I'm paranoid, don't you !
#EOT

[Richard Heathfield]

vi******@gmail.com said:

On Sep 4, 11:19 am, pete <[e-mail]:

<snip>

>I think you got them all.

He did not mention <conio.hand getchr() not being defined by the
standard.

That's true - I didn't. Feel free to do so yourself.
--
Richard Heathfield <http://www.cpax.org.uk>
Email: -http://www. +rjh@
Google users: <http://www.cpax.org.uk/prg/writings/googly.php>
"Usenet is a strange place" - dmr 29 July 1999

[Richard Heathfield]

Nick Keighley said:

On 4 Sep, 09:11, Richard Heathfield <[e-mail]:

>Pranav said:

#include<stdio.h>
#include<conio.h>

non-standard header

<sighYes, I know.

>

int main()

int main(void)
is better style

True enough.

>

{
const int i=6;
int k = 9;
int a[i], *c = a, *v = a ;

"a" appears to be a C99 VLA (older versions of C
don't support variable length arrays)

Yes, but it's still legal C, if he's using a C99 compiler, which is not
beyond the bounds of possibility. Another indication that he might be is
that he's using BCPL-style comments.

a[2] = 9;

*c++ = *v++; // ( 1 ) This line does not generate any exception /
error

Nevertheless, the behaviour of the statement, and therefore the program,
is undefined, because *v++ reads a[0], the value of which is
indeterminate.

as c and v point to the same thing wouldn't there be UB even if
a[0] was initialised?

No: ignoring the ++s for a while, it would simply be the equivalent of a[0]
= a[0]; which is perfectly legal. The ++s affect the pointers (which are
different objects), not the object to which they both point.

>
#include <stdio.h>

Unnecessary header (since we're being picky). :-)

int main (void)
{
int a [6] = {0};
int *c = a;
int *v = a;

*c++ = *v++; /* UB I believe */

You believe wrong, for the reason stated.

<snip>

--
Richard Heathfield <http://www.cpax.org.uk>
Email: -http://www. +rjh@
Google users: <http://www.cpax.org.uk/prg/writings/googly.php>
"Usenet is a strange place" - dmr 29 July 1999

[vippstar]

On Sep 4, 2:13 pm, Richard Heathfield <[e-mail]:

[e-mail] said:

On Sep 4, 11:19 am, pete <[e-mail]:

<snip>

I think you got them all.

He did not mention <conio.hand getchr() not being defined by the
standard.

That's true - I didn't. Feel free to do so yourself.

I don't understand your reply, I did so. If my reply came out as
offensive, it was not my intention. I was not trying to correct you, I
had the genuine intentions of replying to the OP with a full post, but
after seeing yours, the only thing to add was that conio/getch is not
standard.

[Pranav]

Thank You All For quick response and in depth explanation.

Pranav

[Antoninus Twink]

On 4 Sep 2008 at 11:39, vi******@gmail.com wrote:

On Sep 4, 2:13 pm, Richard Heathfield <[e-mail]:

>[e-mail] said:

He did not mention <conio.hand getchr() not being defined by the
standard.

That's true - I didn't. Feel free to do so yourself.

I had the genuine intentions of replying to the OP with a full post,
but after seeing yours, the only thing to add was that conio/getch is
not standard.

It's crystal clear that your intention was to be a prissy little
dickhead as usual. If the OP is writing a console program for Windows,
WHY IN THE HELL should he listen to your whining about non-"Standard"
functions THAT HE NEEDS TO USE?

[Richard]

Antoninus Twink <no****@nospam.invalidwrites:

On 4 Sep 2008 at 11:39, vi******@gmail.com wrote:

>On Sep 4, 2:13 pm, Richard Heathfield <[e-mail]:

>>[e-mail] said:
He did not mention <conio.hand getchr() not being defined by the
standard.

That's true - I didn't. Feel free to do so yourself.

I had the genuine intentions of replying to the OP with a full post,
but after seeing yours, the only thing to add was that conio/getch is
not standard.

It's crystal clear that your intention was to be a prissy little
dickhead as usual. If the OP is writing a console program for Windows,
WHY IN THE HELL should he listen to your whining about non-"Standard"
functions THAT HE NEEDS TO USE?

It was blatantly obvious at an early stage that vippstar is only
interested in currying favour with the clique at each and every
opportunity.

[Richard]

vi******@gmail.com writes:

On Sep 4, 2:13 pm, Richard Heathfield <[e-mail]:

>[e-mail] said:

On Sep 4, 11:19 am, pete <[e-mail]:

<snip>

>I think you got them all.

He did not mention <conio.hand getchr() not being defined by the
standard.

That's true - I didn't. Feel free to do so yourself.

I don't understand your reply, I did so. If my reply came out as
offensive, it was not my intention. I was not trying to correct you, I
had the genuine intentions of replying to the OP with a full post, but
after seeing yours, the only thing to add was that conio/getch is not
standard.

Try reading the thread before offering advice. It saves a lot of clc
clique in fighting and tedious repeated answers.