pymel.util.arrays.VectorN¶

digraph inheritancec3f42b6b11 { rankdir=TB; ranksep=0.15; nodesep=0.15; size="8.0, 12.0"; "Array" [fontname="Vera Sans, DejaVu Sans, Liberation Sans, Arial, Helvetica, sans",URL="pymel.util.arrays.Array.html#pymel.util.arrays.Array",style="setlinewidth(0.5)",tooltip="A generic n-dimensional array class using nested lists for storage.",height=0.25,shape=box,fontsize=8]; "VectorN" [fontname="Vera Sans, DejaVu Sans, Liberation Sans, Arial, Helvetica, sans",URL="#pymel.util.arrays.VectorN",style="setlinewidth(0.5)",tooltip="A generic size VectorN class derived from Array, basically a 1 dimensional Array.",height=0.25,shape=box,fontsize=8]; "Array" -> "VectorN" [arrowsize=0.5,style="setlinewidth(0.5)"]; }

class VectorN(...)¶

A generic size VectorN class derived from Array, basically a 1 dimensional Array.

Most methods and behavior are herited from Array, with the limitation that a MatrixN must have exactly 2 dimensions.

>>> V = VectorN()
>>> V
VectorN([])
>>> V = VectorN([0, 1, 2])
>>> V
VectorN([0, 1, 2])
>>> V = VectorN(0, 1, 2)
>>> V
VectorN([0, 1, 2])
>>> M = MatrixN([[0], [1], [2]])
>>> print M.formated()
[[0],
 [1],
 [2]]
>>> V = VectorN(M.col[0])
>>> V
VectorN([0, 1, 2])

The VectorN class has a constant ndim of 1

>>> VectorN.ndim
1
>>> V.ndim
1
>>> VectorN.ndim = 2
Traceback (most recent call last):
    ...
AttributeError: attribute ndim is a read only class attribute and cannot be modified on class VectorN
>>> V.ndim = 2
Traceback (most recent call last):
    ...
AttributeError: 'VectorN' object attribute 'ndim' is read-only

It’s protected against initialization or resizing to a shape that wouldn’t be a VectorN anymore

>>> V = VectorN([[0, 1], [2, 3]])
Traceback (most recent call last):
    ...
TypeError: cannot initialize a VectorN of shape (4,) from [[0, 1], [2, 3]] of shape (2, 2),
as it would truncate data or reduce the number of dimensions

>>> V.resize((2, 2))
Traceback (most recent call last):
    ...
TypeError: new shape (2, 2) is not compatible with class VectorN

Other Array types can be cast to VectorN, but truncating data or reducing dimensions is not allowed to avoid silent loss of data in a conversion, use an explicit resize / trim / sub-array extraction

>>> A = Array(range(4), shape=(4,))
>>> V = VectorN(A)
>>> V
VectorN([0, 1, 2, 3])

>>> A = Array(range(4), shape=(2, 2))
>>> V = VectorN(A)
Traceback (most recent call last):
    ...
TypeError: cannot cast a Array of shape (2, 2) to a VectorN of shape (4,),
as it would truncate data or reduce the number of dimensions

As for Array, __init__ is a shallow copy, note that as VectorN don’t have sub-arrays, shallow and deep copy amounts to the same thing.

>>> A = Array(range(4), shape=(4,))
>>> V = VectorN(A)
>>> V == A
False
>>> V is A
False
>>> V.isEquivalent(A)
True
>>> V[0] == A[0]
True
>>> V[0] is A[0]
True

__imul__(other)¶

a.__imul__(b) <==> a *= b

In place multiplication of VectorN a and b, see __mul__, result must fit a’s type

__ixor__(other)¶

a.__xor__(b) <==> a^=b

Inplace cross product or transformation by inverse transpose MatrixN of b is v is a MatrixN

__mul__(other)¶

a.__mul__(b) <==> a*b

If b is a VectorN, __mul__ is mapped to the dot product of the two vectors a and b, If b is a MatrixN, __mul__ is mapped to VectorN a by MatrixN b multiplication (post multiplication or transformation of a by b), otherwise, returns the result of the element wise multiplication of a and b if b is convertible to Array, multiplies every component of a by b if b is a single numeric value

__rmul__(other)¶

a.__rmul__(b) <==> b*a

If b is a VectorN, __rmul__ is mapped to the dot product of the two vectors a and b, If b is a MatrixN, __rmul__ is mapped to MatrixN b by VectorN a multiplication, otherwise, returns the result of the element wise multiplication of b and a if b is convertible to Array, multiplies every component of a by b if b is a single numeric value

__xor__(other)¶

a.__xor__(b) <==> a^b

Defines the cross product operator between two vectors, if b is a MatrixN, a^b is equivalent to transforming a by the inverse transpose MatrixN of b, often used to transform normals

angle(other, third=None)¶

u.angle(v) <==> angle(u, v) –> float

Returns the angle of rotation between u and v, 3 dimensional Vectors representing 3D vectors.

Note : this angle is not signed, use axis to know the direction of the rotation

Alternatively can use the form a.angle(b, c), where a, b, c are 4 dimensional Vectors representing 3D points, it is then equivalent to angle(b-a, c-a)

axis(v[, normalize=False]) <==> axis(u, v[, normalize=False])¶

Returns the axis of rotation from u to v as the vector n = u ^ v, u and v being 3 dimensional Vectors representing 3D vectors. If the normalize keyword argument is set to True, n is also normalized.

Alternatively can use the form a.axis(b, c), where a, b, c are 4 dimensional Vectors representing 3D points, it is then equivalent to axis(b-a, c-a).

cotan(v) <==> cotan(u, v)¶

Returns the cotangent of the u, v angle, u and v should be 3 dimensional Vectors representing 3D vectors.

Alternatively can use the form a.cotan(b, c), where a, b, c are 4 dimensional Vectors representing 3D points, it is then equivalent to cotan(b-a, c-a)

cross(v) <==> cross(u, v)¶

cross product of u and v, u and v should be 3 dimensional vectors.

dot(v) <==> dot(u, v)¶

dot product of u and v, u and v should be Vectors of identical size.

isParallel(other, tol=9.313225746154785e-10)¶

u.isParallel(v[, tol]) –> bool

Returns True if both arguments considered as VectorN are parallel within the specified tolerance

length()¶

u.length() –> float

Returns the length of u, ie sqrt(u.dot(u))

ndim = 1¶

normal()¶

u.normal() –> VectorN

Returns a normalized copy of self. Overriden to be consistant with Maya API and MEL unit command, does not raise an exception if self if of zero length, instead returns a copy of self

outer(v) <==> outer(u, v)¶

Outer product of vectors u and v

projectionOnto(other)¶

Returns the projection of this vector onto other vector.

shape¶

v.shape – tuple of one int

Shape of the VectorN, as Vectors are one-dimensional Arrays: v.shape = (v.size,).

It can be queried, or set to change the VectorN’s shape similarly to the resize method, as the only way to change a VectorN’s shape is to resize it.

>>> V = VectorN(1, 2, 3)
>>> V
VectorN([1, 2, 3])
>>> V.shape=(4)
>>> V
VectorN([1, 2, 3, 0])
>>> V.shape=(2, 2)
Traceback (most recent call last):
    ...
TypeError: new shape (2, 2) is not compatible with class VectorN

Related : see Array.resize method.

size¶

Number of components in the VectorN

sqlength()¶

u.sqlength() –> float

Returns the square length of u, ie u.dot(u).

transformAsNormal(other)¶

u.transformAsNormal(m) –> VectorN

Equivalent to transforming u by the inverse transpose MatrixN of m, used to transform normals.

unit()¶

u.normal() –> VectorN

Returns a normalized copy of self. Overriden to be consistant with Maya API and MEL unit command, does not raise an exception if self if of zero length, instead returns a copy of self