tuplevector: Treat tuples (and NamedTuples) as vectors
Tuplevector is a simple library of pure functions that allow for element-wise arithmetic operations on numeric tuples that are the same length.
Could you have done this yourself? Yes, probably. But now you don't have to.
Why tuplevector? What's the advantage?
The big advantage is that you can do vector arithmetic on your own, user-defined NamedTuple and the result will be given back to you in your same user-defined NamedTuple! Tuplevector preserves the type of your first input NamedTuple.
Example:
from collections import namedtuple
from typing import NamedTuple
import tuplevector as vec
Vector = namedtuple("Vector", ["i", "j", "k"])
Point = NamedTuple("Point", [("x", float),
("y", float),
("z", float)])
V1 = Vector(3, 4, 5)
P1 = Point(6, 7, 8)
T1 = (1, 2, 3)
print(vec.add(V1, P1))
>>> Vector(i=9, j=11, k=13)
print(vec.add(P1, T1))
>>> Point(x=4, y=6, z=8)Wouldn't I just use NumPy for vector math?
Sure, you could. But:
- Do you really need to import all of NumPy to add some vectors?
- You would have to give up your elegant namedtuple data type
- Some platforms cannot run NumPy (e.g. MicroPython)
Available vector functions
vec.add(t1: tuple, t2: tuple)
vec.subtract(t1: tuple, t2: tuple)
vec.multiply(t1: tuple, t2: tuple)
vec.divide(t1: tuple, t2: tuple, ignore_zeros:bool = False)
vec.angle(t1: tuple, t2: tuple)
vec.dot(t1: tuple, t2: tuple)
vec.cross(t1: tuple, t2: tuple)
vec.mean(t: tuple)
vec.magnitude(t: tuple)
vec.normalize(t: tuple)