How to convert and adapt a script to Cython?

I have this simple script:

def snap_vert(og, vv, verts, dist):
    lvl = None
    n0,n1,n2 = 0,1,2
    for v in verts:
        vd1 = og[n1]-v[n1]
        vd2 = og[n2]-v[n2]
        dx = abs(vv[n1]*vd2-vv[n2]*vd1)
        if dx < dist:
            vd0 = og[n0]-v[n0]
            dy = abs(vv[n2]*vd0-vv[n0]*vd2)
            if dy < dist:
                dz = abs(vv[n0]*vd1-vv[n1]*vd0)
                if dz < dist:
                    dist = max([dx,dy,dz])
                    lcox, lcoy, lcoz = v
                    lvl = vert
                    tmp = n1
                    n1 = n2
                    n2 = tmp
                tmp = n1
                n1 = n0
                n0 = tmp
    return lvl

This is a pure python code where:
vv is a Vector
og is a Vector
v is a vector list
dist is a float

(vector in this case can be an array)

But I want to optimize and test on a windows 64 bits.

For this I need to convert this .py code for .pyd with Cython.
But I’m having a lot of difficulties with compilation using python 3.4. I can not :no:
If anyone can help me doing it or, indicating a simple way to do this, I would be grateful.