Differece between two ways of using scanf - c

Question

What is the difference between these two segments of code:
1
scanf("%d%d", p1, p2);
2
scanf("%d", p1);
scanf("%d", p2);

Answer 1

Since you're not checking the return values, there is no difference in the behavior. If you were checking the return values, the second option potentially gives you a little more detail (two return values to check) for a little more work. If the input is a single number followed by an EOF, the second will return 1,EOF. If the input is a single number followed by a non-number, it will return 1,0. The first option will return 1 in either of the above cases, so you can't tell the difference without another call (if you care).

Answer 2

I am posting this because it will helpful for you to use scanf in different way::
if you input like::
below all input taking in different variable and array
8 5
2 3 1 2 3 2 3 3
0 2
0 1
6 7
3 5
0 7
then how to use scanf
if you want to take input from console as "space separated input" ex:: 1 2 3 4 5 5 6
then you can use scanf() as below
for(i=0; i < no_you_want;i++)
{
// single space before %d in below scanf function
scanf(" %d",&a[i]);
}

Answer 3

Firstly you need to add '&' in your inputs ;-) Like Chris mentioned there is no difference in working unless you handle the return value. I tried to compare the assembly code, of-course there are couple of extra instructions as below.
scanf("%d%d", &p1, &p2);
main:
.LFB0:
.cfi_startproc
pushq %rbp
.cfi_def_cfa_offset 16
.cfi_offset 6, -16
movq %rsp, %rbp
.cfi_def_cfa_register 6
subq $16, %rsp
movl $.LC0, %eax
leaq -4(%rbp), %rdx
leaq -8(%rbp), %rcx
movq %rcx, %rsi
movq %rax, %rdi
movl $0, %eax
call __isoc99_scanf
movl $0, %eax
leave
.cfi_def_cfa 7, 8
ret
.cfi_endproc
scanf("%d", &p1);
scanf("%d", &p2);
main:
.LFB0:
.cfi_startproc
pushq %rbp
.cfi_def_cfa_offset 16
.cfi_offset 6, -16
movq %rsp, %rbp
.cfi_def_cfa_register 6
subq $16, %rsp
movl $.LC0, %eax
leaq -8(%rbp), %rdx
movq %rdx, %rsi
movq %rax, %rdi
movl $0, %eax
call __isoc99_scanf
movl $.LC0, %eax
leaq -4(%rbp), %rdx
movq %rdx, %rsi
movq %rax, %rdi
movl $0, %eax
call __isoc99_scanf
movl $0, %eax
leave
.cfi_def_cfa 7, 8
ret
.cfi_endproc

Answer 4

An expected difference between scanf("%d%d", p1, p2); and scanf("%d", p1);
scanf("%d", p2); occurs when the input text is "-+2\n".
With scanf("%d%d", p1, p2);, scanf() read "-+" and seeing that it is not a valid int sequence, ungets the + for subsequent IO. *p1 and *p2 remain unchanged.
With scanf("%d", p1);, scanf() read "-+" and seeing that it is not a valid int sequence, (*p1 remains unchanged) ungets the + for subsequent IO which is scanf("%d", p2); which scans "+3\n", save 3 into *p2 and ungets the \n for subsequent IO.
C11dr §7.21.6.2 9 footnote about fscanf(), which applies to scanf(): "fscanf pushes back at most one input character onto the input stream. Therefore, some sequences that are acceptable to strtod, strtol, etc., are unacceptable to fscanf.
But I can not make it differ on my machine.
My cygwin Kepler gcc unexpectedly pushes back both "-+" causing both codes to operate the same. Your mileage may vary. My gcc appears to not comply with the C spec.