[Skyy]

Sorry, I'm not sure if sourcemod has a function for this, and if it does if anyone can point me to the api related to it. Thanks!

[Fyren]

SM doesn't provide a function to tell you if three points are collinear, if that's what you're asking. If you need help implementing one, Google will definitely provide many samples in other languages.

[hmmmmm]

Don't think sourcemod has anything for this, but as Fyren suggested you can use some basic maths to implement it yourself. Off the top of my head you can use some slope formulas to see if slope of AC is same as AB, and if they are then they all lie on same line.

[Kryptanyte]

You could use something like this which I think is a bit simpler than what hmmmmm suggested.

PHP Code:

stock bool IsPointOnLine(float a[3], float b[3], float c[3])
{
    return (GetVectorDistance(a,c) + GetVectorDistance(b,c)) == GetVectorDistance(a,b);
} 


[klippy]

Originally Posted by Kryptanyte You could use something like this which I think is a bit simpler than what hmmmmm suggested. PHP Code: stock bool IsPointOnLine(float a[3], float b[3], float c[3]) {     return (GetVectorDistance(a,c) + GetVectorDistance(b,c)) == GetVectorDistance(a,b); }

Two floating point values that are a product of some arithmetic operation will almost never be exactly equal. You should check if those two are nearly equal.

[fakuivan]

I think this is the most efficient way to solve this

THIS WILL GIVE ERRONEOUS VALUES FOR COMBINATIONS LIKE {0.0, 1.0, 1.0} {0.0, 1.0, 2.0} {0.0, 1.0, 1.5}

PHP Code:

stock bool IsPointOnLine(const float a[3], const float b[3], const float c[3], float precision = 0.0001)
{
    float ab[3]; MakeVectorFromPoints(a, b, ab);
    float ac[3]; MakeVectorFromPoints(a, c, ac);
    
    // If a and c reside on the same point, lambda won't be a real number 
    // (no real number multiplied by zero will give you a non-zero result)
    if (FloatNearlyEqual(a[0], c[0], precision) && FloatNearlyEqual(a[1], c[1], precision) && FloatNearlyEqual(a[2], c[2], precision))
    {
        return true;
    }
    
    // If all lambdas are equal, there is a real lambda that turns ab into ac
    // assuming that lambda_ac is 1
    float lambda_x = ab[0] / ac[0];
    float lambda_y = ab[1] / ac[1];
    float lambda_z = ab[2] / ac[2];
    return FloatNearlyEqual(lambda_x, lambda_y, precision) && FloatNearlyEqual(lambda_y, lambda_z, precision);
}

stock bool FloatNearlyEqual(float x, float y , float precision = 0.0001)
{
    return FloatAbs( x - y ) < precision;
} 


[hmmmmm]

I'm not sure why you're using the FloatDiv and FloatCompare functions, but it makes your code much more confusing.

e.g. FloatDiv(x, y) == x / y
!FloatCompare(x, y) == x < y

[fakuivan]

Originally Posted by hmmmmm I'm not sure why you're using the FloatDiv and FloatCompare functions, but it makes your code much more confusing. e.g. FloatDiv(x, y) == x / y !FloatCompare(x, y) == x < y

Unless those are overloads (they are) I don't think those operators work the same for int and float.