Performance issues

[nammenam]

Hey, great macro!
it works but it is very laggy for me, I gained some performance by converting to an approximate BSpline. I was thinking that others may also benefit from this. Here is the code I changed

    def run(self):
        # generate coordinates
        # thetaList = [j / self.steps * 2 * pi for j in range(self.steps + 1)]
        thetaList = [rescale_func(j / self.steps, 0.8) for j in range(self.steps + 1)]
        points = []
        # thetaList = [rescale_func(theta, 0.8) for theta in thetaList]
        xList = [epi.x(theta * 2 * pi) for theta in thetaList]
        yList = [epi.y(theta * 2 * pi) for theta in thetaList]
        # add lines to the sketch
        for j in range(self.steps):
            x1, x2 = xList[j:j+2]
            y1, y2 = yList[j:j+2]
            # self.sketch.addGeometry(Part.LineSegment(App.Vector(x1, y1, 0),
            #                             App.Vector(x2, y2, 0)), False)

            points.append(App.Vector(x1,y1,0))
        points.append(points[0]) # to close the loop
        bs = Part.BSplineCurve()
        bs.approximate(points)
        self.sketch.addGeometry(bs)
        # self.sketch.addConstraint(Sketcher.Constraint("Coincident", 0, 1, self.steps - 1, 2))
        self.doc.recompute()

[Widdi97]

Good point. I will add it as a kwarg. I also tried this once, but went back to using line segments because my CAM software (EstlCAM) had trouble reading DXF sketches with splines.