ofn-bend-path

[Ofnuts]

Looking into it. The strangeness is that the 3rd degree polynomial is actually a 1st degree one in some curve elements.

Btw, how did you generate that sine wave?

---

OK, figured out the cause, the individual Bézier curves in your top curve have the X coordinates of the tangent be exactly 1/3 from the extremities(*):

Code:

PointsX: [0.0, 12.5, 25.0, 37.5], coeffsX: [0.0, 37.5, 0.0, 0.0]
PointsX: [37.5, 50.0, 62.5, 75.0], coeffsX: [37.5, 37.5, 0.0, 0.0]
PointsX: [75.0, 87.5, 100.0, 112.5], coeffsX: [75.0, 37.5, 0.0, 0.0]
PointsX: [112.5, 125.0, 137.5, 150.0], coeffsX: [112.5, 37.5, 0.0, 0.0]
PointsX: [150.0, 162.5, 175.0, 187.5], coeffsX: [150.0, 37.5, 0.0, 0.0]

This makes the 3rd and 2nd degree coefficients equal to 0. I'll have to improve the math a bit. Stay tuned.

(*) this also true of the bottom sine, however, since the wavelength is different, the range on X is not divisible exactly by 3, so woth the round off errors, the coefficients are small but not null.

---

OK, fixed, see v0.3 (just uploaded to SF)

Thanks for the report and the test-case.