Reset a pointer

[steve1_rm]

Hello,

I have a structure and a pointer to point to this structure on the heap.

I am incrementing the pointer to add new data into the structure. After I want to display all the records in this structure. However, once the pointer has been incremented, is there a way that I can reset it to point to the first record in the structure?

My code is below:

#include <stdlib.h>
#include <stdio.h>
#include <string.h>

typedef struct
{
	char name[20];
	char city[20];
	char state[10];
}record;

typedef record *pointerRecord;

int main(int argc, char** argv)
{
	int i = 0;
	//Allocate meory for 2 records
	pointerRecord ptrRecord = (pointerRecord) malloc(sizeof(record[2]));

	//Fill struct with data
	strncpy(ptrRecord->name, "steve", sizeof(ptrRecord->name));
	strncpy(ptrRecord->city, "Reading", sizeof(ptrRecord->city));
	strncpy(ptrRecord->state, "Bracknell", sizeof(ptrRecord->state));

	ptrRecord++;
	strncpy(ptrRecord->name, "Robert", sizeof(ptrRecord->name));
	strncpy(ptrRecord->city, "Tibury", sizeof(ptrRecord->city));
	strncpy(ptrRecord->state, "Berks", sizeof(ptrRecord->state));

	//Display contents of the structure.
	//Reset pointer HERE to point to the first structure.	
	while(ptrRecord != 0)
	{
		printf("Name: %s", ptrRecord->name);
		printf("City: %s", ptrRecord->city);
		printf("State: %s", ptrRecord->state);
		//Increment pointer HERE
	}

	free(ptrRecord);

	getchar();

	return 0;
}

[brewbuck]

Just store a copy of the original pointer in another variable for use at the end.

[steve1_rm]

Thanks brewbuck that worked. Solution below.

However, I have noticed that when I try to get to the end of the structure when I am printing it will display a lot of rubbish.

while(ptrRecord != 0)

I was thinking the above line would get to the last record and break out of the while loop. However, it doesn't break and keeps on looping. Is there any other condition to set this loop too.

Thanks,

#include <stdlib.h>
#include <stdio.h>
#include <string.h>

typedef struct
{
	char name[20];
	char city[20];
	char state[10];
}record;

typedef record *pointerRecord;

int main(int argc, char** argv)
{
	int i = 0;
	//Allocate meory for 2 records
	pointerRecord ptrRecord = (pointerRecord) malloc(sizeof(record[2]));
	pointerRecord resetPointer;
	resetPointer = ptrRecord; //Will point to the orginal address

	//Fill struct with data
	strncpy(ptrRecord->name, "steve", sizeof(ptrRecord->name));
	strncpy(ptrRecord->city, "Reading", sizeof(ptrRecord->city));
	strncpy(ptrRecord->state, "Bracknell", sizeof(ptrRecord->state));

	ptrRecord++;
	strncpy(ptrRecord->name, "Robert", sizeof(ptrRecord->name));
	strncpy(ptrRecord->city, "Tibury", sizeof(ptrRecord->city));
	strncpy(ptrRecord->state, "Berks", sizeof(ptrRecord->state));

	//Display contents of the structure.
	//Reset pointer HERE to point to the first structure.	
	ptrRecord = resetPointer;
	while(ptrRecord != 0)
	{
		printf("Name: &#37;s\n", ptrRecord->name);
		printf("City: %s\n", ptrRecord->city);
		printf("State: %s\n", ptrRecord->state);
		//Increment pointer HERE
		ptrRecord++;
	}

	free(ptrRecord);

	getchar();

	return 0;
}

[vart]

you know the count of mebers in the array
use for loop with index
for(i=0;i<array_size;i++)

[Elysia]

typedef record *pointerRecord;

Pointer typedefs obfuscates the code. Better not to use them. Don't be lazy and type record* instead.

[rdrast]

Er... Instead of:

	
         //Allocate meory for 2 records
	pointerRecord ptrRecord = (pointerRecord) malloc(sizeof(record[2]));

maybe

	//Allocate meory for 2 records
	pointerRecord ptrRecord = (pointerRecord) malloc(2 * sizeof(record));

[steve1_rm]

Hello,

Just wondering what is the difference between writing

malloc(sizeof(record[2]));

and

malloc(2 * sizeof(rocord));

I have never seen the second method been used before. And advantages?

Thanks,

Steve

[Elysia]

1) Allocates 1 x size of element 3 in variable record (bad).
2) Allocates 2 x size of record.

[robwhit]

no, record is a type. record[2] is type array of records with 2 elements. They are both functionally identical. If you don't have any optimizing at all in a very non optimizing compiler, the first one might avoid a multiplication, but don't quote me on that.

If 2 * sizeof(record) overflows the size of size_t, you'd be more likely to receive a compile-time error message, I think. But don't quote me on that either.