pass by value

[mdh]

May I just confirm my understanding that:

If a pointer is passed as an argument to a function, manipulation of
the function pointer ( eg pointer ++ till some condition....) will not
effect the value of the calling pointer?

So;

myf(p) /* p is ptr to char) */


void myf( char *q){

while (...)

*q++;

}

leaves p unchanged?

 

[osmium]

May I just confirm my understanding that:

If a pointer is passed as an argument to a function, manipulation of
the function pointer ( eg pointer ++ till some condition....) will not
effect the value of the calling pointer?

So;

myf(p) /* p is ptr to char) */


void myf( char *q){

while (...)

*q++;

}

leaves p unchanged?

Yes, that's right. One nitpick, I wouldn't use "function pointer" in this
discussion, it usually means something else.

 

[mdh]

One nitpick, I wouldn't use "function pointer" in this
discussion, it usually means something else.

Good point...thank you

 

[Richard Heathfield]

mdh said:

May I just confirm my understanding that:

If a pointer is passed as an argument to a function, manipulation of
the function pointer ( eg pointer ++ till some condition....) will not
effect the value of the calling pointer?

So;

myf(p) /* p is ptr to char) */


void myf( char *q){

while (...)

*q++;

}

leaves p unchanged?

It does indeed leave p unchanged. It doesn't change anything in the
thing pointed to, either, though. Were you aware of that? It is
possible that *q++; doesn't do quite what you think it does.

 

[Richard]

mdh said:


It does indeed leave p unchanged. It doesn't change anything in the
thing pointed to, either, though. Were you aware of that? It is
possible that *q++; doesn't do quite what you think it does.

In this sample code that is misleading.

the "..." could have been anything manipulating exactly what p was
pointing to.

 

[Richard Heathfield]

Richard said:

In this sample code that is misleading.

No, it isn't.

the "..." could have been anything manipulating exactly what p was
pointing to.

Indeed, but we can only go on the code that is posted. My point refers
specifically to the statement: *q++;

 

[mdh]

Richard said:

Nope...but then I was just "making up" some code to be sure I
understood "pass by value" and in the past, I have had my knuckles
rapped for doing this!!!
As I look at the hasty code I wrote, I think what this would do would
simply point at subsequent values of an array and not change a
thing!!! The wonder of terse code

 

[mdh]

to, will not alter p.

Neither Richard(I believe) nor I were talking about altering p.