Filling sourrounding cells in a 3D vector - vector

Question

I need to fill a 3d vector in a way that, once x,y,z are fixed, the code has to write numbers in all the nearest neighbours (i.e. all the cells with x+/-1, y+/-1, z+/-1). Of course I don't want the cells outside the boundaries to be filler. So far I've tried the following approach
void setSphericalSource(int x, int y, int z, float sigma, vector<vector<vector<float>>>& vec, float maxAct) {
float total_activity = 0.0;
vec[z][y][x] = maxAct ; //hotspot
total_activity += maxAct;
if (x > 0 && x < vec[0][0].size() && y > 0 && y < vec[0].size() && z > 0 && z < vec.size()) {
//cells with face in common
vec[z][y][x + 1] = maxAct/2;
vec[z][y][x - 1] = maxAct/2;
vec[z][y + 1][x] = maxAct/2;
vec[z][y - 1][x] = maxAct/2;
vec[z + 1][y][x] = maxAct/2;
vec[z - 1][y][x] = maxAct/2;
//cells with edge in common
vec[z][y + 1][x + 1] = maxAct/3;
vec[z][y + 1][x - 1] = maxAct/3;
vec[z][y - 1][x + 1] = maxAct/3;
vec[z][y - 1][x - 1] = maxAct/3;
vec[z + 1][y][x + 1] = maxAct/3;
vec[z + 1][y][x - 1] = maxAct/3;
vec[z - 1][y][x + 1] = maxAct/3;
vec[z - 1][y][x - 1] = maxAct/3;
vec[z + 1][y + 1][x] = maxAct/3;
vec[z + 1][y - 1][x] = maxAct/3;
vec[z - 1][y + 1][x] = maxAct/3;
vec[z - 1][y - 1][x] = maxAct/3;
//cells with a vertex in common
vec[z + 1][y + 1][x + 1] = maxAct/4;
vec[z + 1][y + 1][x - 1] = maxAct/4;
vec[z + 1][y - 1][x - 1] = maxAct/4;
vec[z - 1][y - 1][x - 1] = maxAct/4;
vec[z + 1][y - 1][x + 1] = maxAct/4;
vec[z - 1][y + 1][x + 1] = maxAct/4;
vec[z - 1][y - 1][x + 1] = maxAct/4;
vec[z - 1][y + 1][x - 1] = maxAct/4;
}
}
In practice I first check if x,y,z belong to the inner part of the 3d vector, then I go to the external cells and so on until a spherical volume around the chosen center is filled (skipping the cells that exceed the boundaries).
The problem rises when I want to set a spherical source with the center on a face, on an edge or on a vertex of the 3D grid. Is there a better way than repeating the code above for each face, each vertex and each corner?

Answer 1

Is there a better way of repeating the code above for each face, each vertex and each corner?
I take you actually meant "Is there a better way than repeating the code above for each face, each vertex and each corner?" - Here is another way; it's up to the reader to judge whether it's better:
for (int k = -1; k <= 1; ++k)
for (int j = -1; j <= 1; ++j)
for (int i = -1; i <= 1; ++i)
{
int d = abs(i)+abs(j)+abs(k)+1;
x += i, y += j, z += k;
if (0 <= x && x < vec[0][0].size()
&& 0 <= y && y < vec[0] .size()
&& 0 <= z && z < vec .size()
) vec[z][y][x] = maxAct/d;
x -= i, y -= j, z -= k;
}