Add class for 2d vectors about pyx HOT 4 CLOSED

For the time being, you could use numpy arrays as demonstrated in the
following example:

import numpy as np
from pyx import canvas, deco, path

c = canvas.canvas()
point_1 = np.array([1, 2])
vektor = np.array([4, -3])
point_2 = point_1 + vektor
c.stroke(path.line(*point_1, *point_2), [deco.earrow])
c.writePDFfile()

This approach works nicely in 3d as well which I use frequently together
with the projectors provided by graph.graphxyz to produce schematic 3d drawings.

from pyx.

Ah, I didn't know the tuple unpacking also works for numpy arrays. (And, to be honest, it didn't occur to me at all to use tuple unpacking.) I think there is still a case for a dedicated pyx vector class because lengths in pyx are not floats but instances of unit.length. For example, you may want to get a point on a path via the at method (which gives you a tuple of unit.length) and then do further calculations with it.

Would you still consider changing the path construction methods to accept vector arguments or do you think the tuple unpacking trick should be normal way to do things?

from pyx.

This is for the developers to decide. I am just a happy long-time user of PyX :-)

Concerning unpacking of tuples: I started using this only recently and found it quite useful with PyX at least in some situations.

from pyx.

This was an addition in Python 3.5, PEP 448. I remember quite well that I was very happy when learning about this feature. It is very handy for PyX. I also discussed it with Jörg and we kind of decided, that (a) it was good to not have used tuples in path elements constructors (and elsewhere) originally and (b) that due to this new feature we have even less reason to change anything in PyX. :-)

from pyx.