Thijs Koerselman

I came across this piece of code in jit.gl.gridshape:

jit_object_method(x,gensym("getcolor"),&ac,&av);

Is this somehow more efficient to get an objects attribute then doing

jit_attr_getfloat_array(x, gensym("color"), 4, color);

I think its obvious which one I'd prefer to use, but I'm just curious if

there is a good reason to use the first method. Or is it just "old" jitter

jit_object_method(x,gensym("getcolor"),&ac,&av);
if ((ac==4)&&av) {
   color[0] = jit_atom_getfloat(av);
   color[1] = jit_atom_getfloat(av+1);
   color[2] = jit_atom_getfloat(av+2);
   color[3] = jit_atom_getfloat(av+3);
   jit_freebytes(av,ac*sizeof(t_atom));
}

Is this somehow more efficient to get an objects attribute then doing
something like the following?

I think its obvious which one I'd prefer to use, but I'm just curious if
there is a good reason to use the first method. Or is it just "old" jitter
coding style?

2-ways-to-get-an-objects-attribute

You would use the first method if you don't know the size of the 

array, but the 2nd method is a convenience version of the 1st for 

situations in which you do. They are equivalently efficient.

Am 19.11.2007 um 16:34 schrieb Thijs Koerselman:

> I came across this piece of code in jit.gl.gridshape:

> jit_object_method(x,gensym("getcolor"),&ac,&av);

> Is this somehow more efficient to get an objects attribute then 

> jit_attr_getfloat_array(x, gensym("color"), 4, color);

> I think its obvious which one I'd prefer to use, but I'm just 

> curious if there is a good reason to use the first method. Or is it 

You would use the first method if you don't know the size of the  
array, but the 2nd method is a convenience version of the 1st for  
situations in which you do. They are equivalently efficient.

> Dear list,
>
> I came across this piece of code in jit.gl.gridshape:
>
> long ac=0;
> t_atom *av=NULL;
>
> jit_object_method(x,gensym("getcolor"),&ac,&av);
> if ((ac==4)&&av) {
>    color[0] = jit_atom_getfloat(av);
>    color[1] = jit_atom_getfloat(av+1);
>    color[2] = jit_atom_getfloat(av+2);
>    color[3] = jit_atom_getfloat(av+3);
>    jit_freebytes(av,ac*sizeof(t_atom));
> }
>
> Is this somehow more efficient to get an objects attribute then  
> doing something like the following?
>
> float color[4];
>
> jit_attr_getfloat_array(x, gensym("color"), 4, color);
>
> I think its obvious which one I'd prefer to use, but I'm just  
> curious if there is a good reason to use the first method. Or is it  
> just "old" jitter coding style?
>
> Thijs
>
>
>

On Nov 19, 2007 3:44 PM, Jeremy Bernstein wrote:

> You would use the first method if you don't know the size of the

> array, but the 2nd method is a convenience version of the 1st for

> situations in which you do. They are equivalently efficient.

Thanks for letting me know. I didn't think about the variable array size.

On Nov 19, 2007 3:44 PM, Jeremy Bernstein  wrote:

> You would use the first method if you don't know the size of the
> array, but the 2nd method is a convenience version of the 1st for
> situations in which you do. They are equivalently efficient.
>
>
Hi Jeremy,

2 ways to get an object's attribute