pymath.interpolate_cubic

Interpolates a smooth cubic curve (B-Spline) through a list of points, returned as cubic Bézier segments.

pymath.interpolate_cubic(type, points [, start_tangent, end_tangent])

Parameter	Type	Description
type	string	"open" or "closed" (as in pytha.create_polyline).
points	list of points	Array of points to interpolate. Each point can be either an array of length d for d-dimensional space or a number in 1D.
start_tangent	point (optional)	For an "open" curve only, the tangent direction at the first point — the same shape as a point
end_tangent	point (optional)	For an "open" curve only, the tangent direction at the last point.

Return value

Type

Description

curve

A flat list of control points describing the cubic Bézier segments, or nil if the curve could not be built (see Notes). Each control point has the same shape as an input point (a d-vector, or a plain number when points was given as a flat list of numbers). Successive segments share their joint control point: points 1..4 are the first segment, 4..7 the second, and so on. An "open" curve returns 3 * (n - 1) + 1 control points, a "closed" curve 3 * n (its last segment closes back to the first point).

Notes:

• The curve passes through every input point: control point 3 * k + 1 equals input point k + 1.
• Each group of four consecutive control points is a cubic Bézier segment that can be passed straight to pymath.eval_bezier.
• For "closed", give the distinct vertices — do not repeat the first point at the end.
• start_tangent / end_tangent are only allowed for an "open" curve; passing one with "closed" raises an error. Without them the curve leaves its endpoints with a zero derivative. Only the direction is used — the tangent's length is ignored.
• nil is returned when two consecutive points coincide.

Example:

local pts = {{0, 0}, {1, 2}, {3, 1}, {4, 3}}
local cp = pymath.interpolate_cubic("open", pts)
-- #cp == 10                       -- 3 * (4 - 1) + 1 control points
-- cp[1] == {0, 0}, cp[4] == {1, 2}, cp[7] == {3, 1}, cp[10] == {4, 3}

-- sample the first Bézier segment at its midpoint
local mid = pymath.eval_bezier(0.5, cp[1], cp[2], cp[3], cp[4])

-- prescribe the end tangent directions (open curve): leave horizontally, arrive vertically
local cp2 = pymath.interpolate_cubic("open", pts, {1, 0}, {0, 1})

-- 1-D points as a flat list of numbers -> a flat list of numbers back
local y = pymath.interpolate_cubic("open", {0, 1, 3, 4})
-- #y == 10, y[1] == 0, y[4] == 1, y[7] == 3, y[10] == 4  (all plain numbers)

Version Support:

Minimum PYTHA Version: V27

See also:

pymath, pymath.eval_bezier, pymath.eval_bezier_deriv