CODEWITHSUNDEEP
cstruct
user3542317
user3542317

Reputation: 313

What is each curly brace standing for when declaring/initializing a struct? (warning: missing braces around initializer [-Wmissing-braces])

I came across the following structure: 

static struct {
    unsigned char a[5];
} b[] = {       
        {1,1,1,1,1},
        {2,2,2,2,2}
};

However, if it is being compiled I get the following warning "warning: missing braces around initializer [-Wmissing-braces]".

If I alter it like so:

static struct {
    unsigned char a[5];
} b[] = {       
        {{1,1,1,1,1}},
        {{2,2,2,2,2}}
};

Then the warning is gone. What is each curly brace standing for after "b[] = " ? The innermost curly braces are obviously? standing for the initialization of char a[5]. But what are the other braces standing for? Obviously? one of the other curly braces has to account for the structure array b[] but which one? And why is there, as it seems, a third curly brace neccessary and what is it standing for? This confuses me.

Upvotes: 1

Views: 119

Answers (5)

pmg
pmg

Reputation: 108986

static struct {
    unsigned char a[5];
} b[] = {                 // array b
        {                 // struct
        {                 // array a
          1,1,1,1,1}},
        {                 // another struct
        {2,2,2,2,2}}
};

Upvotes: 0

babon
babon

Reputation: 3774

{1,1,1,1,1} - The five elements of an array.

{{1,1,1,1,1}} - That's how you initialize a struct with a field which is an array of five ints in this case.

{{{1,1,1,1,1}}, {{2,2,2,2,2}}} - This is an array of two elements. Where each element is an instance of a struct as mentioned above.

Upvotes: 1

KamilCuk
KamilCuk

Reputation: 142005

int a[2] = { 1, 2 };

Ok. Now this:

int a[2][2] = { { 1, 2 } , { 3, 4 } };

Right? a[0][0] = 1 and a[0][1] = 2 and a[1][0] = 3 and a[1][1] = 4. a[0] is memcmp equal to (int[]){ 1, 2 } and a[1] is memcmp equal to (int[]){ 3, 4 }.

Now this:

struct {
     int a[2];
} b = { { 1, 2 } };

So the first { } stand for structure initialization and the inner stand for b.a initialization. Now we make an array of structures:

struct b_s {
     int a[2];
} b[] = { 
      {
         { 1, 2 },
      },{ 
         { 3, 4, }
      },
};

So the first braces stand for array b[] initialization. The second one are for structure b[0] to initialize. The third one are to initialize the b[0].a array.

The b[0] is memcmp equal to &(struct b_s){ {1, 2 } }. The b[0].a is memcmp equal to (int[2]){ 1, 2 }. The b[0].a[0] is equal to 1.

Upvotes: 5

Eric Postpischil
Eric Postpischil

Reputation: 224546

Each brace corresponds to the start of a subobject (an object inside an aggregate object, such as an array or structure).

b[] is an array, so the first brace introduces elements of the array.

Each b is a structure, so the next brace introduces members of the structure.

The member a inside b is an array, so the next braces introduces elements of that array.

Upvotes: 1

Thomas Padron-McCarthy
Thomas Padron-McCarthy

Reputation: 27662

You have an array of structs containing arrays. Three pairs of braces.

Upvotes: 0

Related Questions