Parallelogram

public final @NonNull ImmutableBox computeBoundingBox()

Returns the minimum bounding box containing the Parallelogram.

Performance-sensitive code should use the computeBoundingBox overload that takes a pre-allocated MutableBox, so that instance can be reused across multiple calls.

public final @NonNull MutableBox computeBoundingBox(@NonNull MutableBox outBox)

Returns the minimum bounding box containing the Parallelogram.

public final @NonNull List<@NonNull ImmutableVec> computeCorners()

Returns a list containing the 4 corners of the Parallelogram.

Corners are numbered 0, 1, 2, 3. In a Y-up coordinate system, the corners of the base rectangle are, in order: bottom-left, bottom-right, top-right, top-left. In a Y-down coordinate system, they are: top-left, top-right, bottom-right, bottom-left. The corners keep their numbering through any shear and/or rotation applied to the base rectangle. Numerically, the corners are equivalent to: C - u - v C + u - v C + u + v C - u + v Where C = center, and u and v are the semiAxes.

Performance-sensitive code should use the computeCorners overload that takes pre-allocated MutableVecs, so that instances can be reused across multiple calls.

public final void computeCorners(
    @NonNull MutableVec outCorner1,
    @NonNull MutableVec outCorner2,
    @NonNull MutableVec outCorner3,
    @NonNull MutableVec outCorner4
)

Populates the 4 output points with the corners of the Parallelogram.

For explanation of order, please see computeCorners above.

public final @NonNull List<@NonNull ImmutableVec> computeSemiAxes()

Returns the semi axes of this Parallelogram. These are equal to:

- (.5 * w * cos(θ), .5 * w * sin(θ))
- (.5 * h * (s * cos(θ) - sin(θ)), .5 * h * (s * sin(θ) + cos(θ)))

respectively, where w = width, h = height, θ = rotation, and s = shearFactor

The semi-axes of a parallelogram are two vectors. Each one points from the center to the midpoint of an edge. The first semi-axis points from the center to the midpoint of the edge between corners 1 and 2, and the second semi-axis points from the center to the midpoint of the edge between corners 2 and 3. In a Y-up coordinate system, on the base rectangle, these two edges are the right and top, respectively. In a Y-down coordinate system, on the base rectangle, they are the right and bottom, respectively.

Performance-sensitive code should use the computeSemiAxes overload that takes a pre-allocated MutableVecs, so that instances can be reused across multiple calls.

public final void computeSemiAxes(@NonNull MutableVec outAxis1, @NonNull MutableVec outAxis2)

Fills the MutableVecs with the semi axes of this Parallelogram. For definition please see computeSemiAxes above.

public final float computeSignedArea()

Returns the signed area of the Parallelogram. If either the width or the height is zero, this will be equal to zero; if the width is non-zero, then this will have the same sign as the height.

public final boolean contains(@NonNull ImmutableVec point)

Returns whether the given point is contained within the Box. Points that lie exactly on the Box's boundary are considered to be contained.

public abstract float getHeight()

A Parallelogram may have a positive or negative height; a positive height indicates that the angle from the first semi-axis to the second will also be positive.

public abstract float getShearFactor()

A Parallelogram] may have a positive or negative shear factor; a positive shear factor indicates a smaller absolute angle between the semi-axes (the shear factor is, in fact, the cotangent of that angle).

public abstract @FloatRange(from = 0.0) float getWidth()

A Parallelogram may not have a negative width. If an operation on a parallelogram would result in a negative width, it is instead normalized, by negating both the width and the height, adding π to the angle of rotation, and normalizing rotation to the range [0, 2π).

public final boolean isAlmostEqual(
    @NonNull Parallelogram other,
    @FloatRange(from = 0.0) float tolerance
)

Compares this Parallelogram with other, and returns true if both center points are considered almost equal with the given tolerance, and the difference between width and other.width is less than tolerance, and likewise for height, rotation, and shearFactor.