Table of Contents

Struct b2Vec2

Namespace
Gym.Box2D
Assembly
Gym.Environments.dll

Box2D's 2-D column vector, float32, the type every position, velocity, force and vertex travels in, and the value pybox2d hands Python as b2Vec2. Arithmetic on it is float32, as in pybox2d, whose vector operators run in C++: v * 2.5 rounds 2.5 to float32 first and multiplies in float32.

public struct b2Vec2
Inherited Members
Extension Methods

Remarks

Footgun: reading x and y into Python-style double arithmetic is exact, but multiplying two components in C# (v.x * v.y) stays float32, where Gymnasium's Python computes a float64 product. Write (double)v.x * v.y wherever the Python operands are plain floats.

Reference: Box2D/Common/b2Math.h.

Constructors

b2Vec2(float, float)

A vector of the given coordinates.

public b2Vec2(float x, float y)

Parameters

x float

The x coordinate.

y float

The y coordinate.

Fields

x

The x coordinate.

public float x

Field Value

float

y

The y coordinate.

public float y

Field Value

float

Properties

this[int]

A coordinate by index, as pybox2d's getitem/setitem: 0 is x, 1 is y.

public float this[int i] { readonly get; set; }

Parameters

i int

0 or 1.

Property Value

float

The coordinate.

Exceptions

IndexError

Thrown for any other index, with pybox2d's message Index must be in (0,1).

length

The length (pybox2d's length property).

public readonly float length { get; }

Property Value

float

lengthSquared

The squared length (pybox2d's lengthSquared).

public readonly float lengthSquared { get; }

Property Value

float

skew

The skew vector (-y, x) (pybox2d's skew).

public readonly b2Vec2 skew { get; }

Property Value

b2Vec2

tuple

The coordinates as a tuple of Python floats (pybox2d's tuple); each float32 widens exactly.

public readonly (double x, double y) tuple { get; }

Property Value

(double From, double To)

valid

Whether both coordinates are finite (pybox2d's valid).

public readonly bool valid { get; }

Property Value

bool

Methods

Deconstruct(out float, out float)

Unpacks the coordinates, as Python's x, y = vec iterates a b2Vec2.

public readonly void Deconstruct(out float x, out float y)

Parameters

x float

Receives x.

y float

Receives y.

Equals(object?)

Equality with another boxed vector, by operator ==(b2Vec2, b2Vec2).

public override readonly bool Equals(object? obj)

Parameters

obj object

The other object.

Returns

bool

True for an equal b2Vec2.

GetHashCode()

A hash of both coordinates' bits.

public override readonly int GetHashCode()

Returns

int

The hash code.

IsValid()

Whether both coordinates are finite (not NaN or infinite).

public readonly bool IsValid()

Returns

bool

True for a finite vector.

Length()

The vector's length, sqrtf(x * x + y * y).

public readonly float Length()

Returns

float

The length.

LengthSquared()

The squared length, cheaper than Length() when only comparisons are needed.

public readonly float LengthSquared()

Returns

float

x * x + y * y.

Normalize()

Converts this vector into a unit vector in place; a vector shorter than FLT_EPSILON is left as it is.

public float Normalize()

Returns

float

The length before normalizing, or 0 for a vector too short to normalize.

Set(float, float)

Sets this vector to some specified coordinates.

public void Set(float x_, float y_)

Parameters

x_ float

The x coordinate.

y_ float

The y coordinate.

SetZero()

Sets this vector to all zeros.

public void SetZero()

Skew()

The skew vector, (-y, x), such that dot(skew_vec, other) == cross(vec, other).

public readonly b2Vec2 Skew()

Returns

b2Vec2

The skew vector.

ToString()

pybox2d's repr: b2Vec2(%g,%g).

public override readonly string ToString()

Returns

string

The text.

copy()

A copy (pybox2d's copy); C# structs copy on assignment, so this exists for readability.

public readonly b2Vec2 copy()

Returns

b2Vec2

The same coordinates.

cross(b2Vec2)

The 2-D cross product with another vector, in float32 (pybox2d's cross).

public readonly float cross(b2Vec2 other)

Parameters

other b2Vec2

The other vector.

Returns

float

x * other.y - y * other.x.

cross(double)

The cross product with a scalar, a vector (pybox2d's cross(float)).

public readonly b2Vec2 cross(double s)

Parameters

s double

The scalar, rounded to float32 as pybox2d's argument conversion does.

Returns

b2Vec2

(s * y, -s * x).

Exceptions

OverflowError

Thrown when s lies beyond the float32 range.

dot(b2Vec2)

The dot product with another vector, in float32 (pybox2d's dot).

public readonly float dot(b2Vec2 other)

Parameters

other b2Vec2

The other vector.

Returns

float

x * other.x + y * other.y.

Operators

operator +(b2Vec2, b2Vec2)

Adds two vectors coordinate by coordinate in float32.

public static b2Vec2 operator +(b2Vec2 a, b2Vec2 b)

Parameters

a b2Vec2

The first vector.

b b2Vec2

The second vector.

Returns

b2Vec2

(a.x + b.x, a.y + b.y).

operator /(b2Vec2, double)

Divides by a Python float, as pybox2d's truediv: the divisor rounds to float32 first.

public static b2Vec2 operator /(b2Vec2 a, double s)

Parameters

a b2Vec2

The vector.

s double

The divisor.

Returns

b2Vec2

(a.x / (float)s, a.y / (float)s).

Exceptions

OverflowError

Thrown when s lies beyond the float32 range.

operator /(b2Vec2, float)

Divides a vector by a scalar in float32 (pybox2d's truediv).

public static b2Vec2 operator /(b2Vec2 a, float s)

Parameters

a b2Vec2

The vector.

s float

The divisor.

Returns

b2Vec2

(a.x / s, a.y / s).

operator ==(b2Vec2, b2Vec2)

Exact coordinate equality, as C++'s operator == and pybox2d's eq (NaN never equals).

public static bool operator ==(b2Vec2 a, b2Vec2 b)

Parameters

a b2Vec2

The first vector.

b b2Vec2

The second vector.

Returns

bool

True when both coordinates compare equal.

implicit operator b2Vec2((double x, double y))

A Python sequence of two numbers becoming a vector, as pybox2d's typemap converts a tuple argument: each element is rounded to float32.

public static implicit operator b2Vec2((double x, double y) value)

Parameters

value (double From, double To)

The coordinates as Python floats.

Returns

b2Vec2

The vector.

Exceptions

TypeError

Thrown for an element beyond the float32 range (an infinity included), with pybox2d's message Converting from sequence to b2Vec2, expected int/float arguments index 0 (or index 1).

operator !=(b2Vec2, b2Vec2)

The negation of operator ==(b2Vec2, b2Vec2).

public static bool operator !=(b2Vec2 a, b2Vec2 b)

Parameters

a b2Vec2

The first vector.

b b2Vec2

The second vector.

Returns

bool

True when a coordinate differs (or either is NaN).

operator *(b2Vec2, double)

Scales by a Python float, as pybox2d's mul: the scalar rounds to float32, then float32 products.

public static b2Vec2 operator *(b2Vec2 a, double s)

Parameters

a b2Vec2

The vector.

s double

The scale.

Returns

b2Vec2

(a.x * (float)s, a.y * (float)s).

Exceptions

OverflowError

Thrown when s lies beyond the float32 range.

operator *(b2Vec2, float)

Scales a vector in float32 (pybox2d's mul, and C++'s *=).

public static b2Vec2 operator *(b2Vec2 a, float s)

Parameters

a b2Vec2

The vector.

s float

The scale.

Returns

b2Vec2

(a.x * s, a.y * s), the same bits as s * a.

operator *(double, b2Vec2)

Scales by a Python float from the left, as pybox2d's rmul (same bits as a * s).

public static b2Vec2 operator *(double s, b2Vec2 a)

Parameters

s double

The scale.

a b2Vec2

The vector.

Returns

b2Vec2

(a.x * (float)s, a.y * (float)s).

Exceptions

OverflowError

Thrown when s lies beyond the float32 range.

operator *(float, b2Vec2)

Scales a vector in float32.

public static b2Vec2 operator *(float s, b2Vec2 a)

Parameters

s float

The scale.

a b2Vec2

The vector.

Returns

b2Vec2

(s * a.x, s * a.y).

operator -(b2Vec2, b2Vec2)

Subtracts two vectors coordinate by coordinate in float32.

public static b2Vec2 operator -(b2Vec2 a, b2Vec2 b)

Parameters

a b2Vec2

The first vector.

b b2Vec2

The second vector.

Returns

b2Vec2

(a.x - b.x, a.y - b.y).

operator -(b2Vec2)

Negates both coordinates.

public static b2Vec2 operator -(b2Vec2 v)

Parameters

v b2Vec2

The vector.

Returns

b2Vec2

(-x, -y).