1
\$\begingroup\$

A small script to calculate total units of distance between two vector entities. Especially useful in games and grid snaps, inventory snaps etc..

typedef struct COORD
{
    int X, Y;
} coord;

int get_dist (coord entity1, coord entity2)
{
    int minuendX    = max(entity1.X, entity2.X);
    int minuendY    = max(entity2.Y, entity2.Y);
    int subtrahendX = min(entity1.X, entity2.X);
    int subtrahendY = min(entity2.Y, entity2.Y);

    return (minuendX - subtrahendX) + (minuendY - subtrahendY);
}
\$\endgroup\$
3
  • \$\begingroup\$ What do you mean by "radial distance"? \$\endgroup\$ Oct 23, 2015 at 15:53
  • \$\begingroup\$ Well regularly it refers to the euclidean vector but in this case.. a radius. \$\endgroup\$
    – Malina
    Oct 23, 2015 at 15:58
  • 1
    \$\begingroup\$ Does this code give the intended results? \$\endgroup\$ Oct 23, 2015 at 16:02

1 Answer 1

6
\$\begingroup\$

There is a typo in the y components, both times only entity2 is used. I believe that is a copy-paste error.

Your distance is better known as the manhattan distance. I suggest naming the method after it to avoid surprises.

An other method of getting it uses the absolute value of the differences:

int get_manhattan_dist (coord entity1, coord entity2)
{
    return abs(entity1.X - entity2.X) + abs(entity1.Y - entity2.Y);
}

Of course if you have the manhattan distance you can easily provide the euclidean distance

double get_euclid_dist (coord entity1, coord entity2)
{
    double diffX    = entity1.X - entity2.X;
    double diffY    = entity1.Y - entity2.Y;

    return hypot(diffX, diffY);
}
\$\endgroup\$
4
  • \$\begingroup\$ Oh you are right about the typos. I re-created my code in here. I am testing on the correct version though. Thank you, didn't know of its name. \$\endgroup\$
    – Malina
    Oct 23, 2015 at 16:05
  • \$\begingroup\$ Note: double diffX = entity1.X; diffX -= entity2.X; costs 1 more int/double conversion, but prevents int overflow. \$\endgroup\$ Oct 28, 2015 at 21:52
  • \$\begingroup\$ @chux not really more int->double conversion because hypot takes 2 doubles and the params get converted before the function gets called \$\endgroup\$ Oct 28, 2015 at 22:03
  • \$\begingroup\$ double diffX = entity1.X - entity2.X; employs 1 int/double conversion. double diffX = entity1.X; diffX -= entity2.X; uses 2 int/double conversions. The call of hypot() does not affect that count. \$\endgroup\$ Oct 28, 2015 at 22:06

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.