The finished product, a tribute to Bridget Riley's Rustle 6 (2015).

import math

canvasX = 600
canvasY = 600
nFrames = 120
side = canvasX/14
height = (side * sqrt(3))/2
center = (side/sqrt(3))/2
nX = 14
nY = 15
offsetX = side
offsetY = height

from datetime import *
time = datetime.now()

def interpolate(pt1, pt2, ratio):
    assert ratio <= 1
    return (pt1[0] + (ratio * (pt2[0] - pt1[0])), pt1[1] + (ratio * (pt2[1] - pt1[1])))
    
def triangle(s, h, c, inout):
    assert inout >= 0 and inout <= 1
    start = (-s/2, h/2)
    pivot = (0, h/2 - (c * 3))
    target = (s/2, h/2)
    control = interpolate((s/2, -c/2), (0, h/2 - c), inout)
    radius = s
    # optimal distance to control points is (4/3)*tan(pi/(2n); thank you https://stackoverflow.com/users/16673/suma for answering this question https://stackoverflow.com/questions/1734745/how-to-create-circle-with-b%C3%A9zier-curves
    control_ratio = ((4/3 * tan(pi/12)) * radius) / (2 * c)
    controlP = interpolate(pivot, control, control_ratio)
    controlT = interpolate(target, control, control_ratio)
    triangle = BezierPath()
    triangle.moveTo(start)
    triangle.lineTo(pivot)
    triangle.curveTo(controlP, controlT, target)
    triangle.closePath()
    drawPath(triangle)

def frame(x, y, m):
    f = BezierPath()
    g = BezierPath()
    f.rect(0, 0, x, y)
    g.rect(m, m, x - (2 * m), y - (2 * m))
    f = f.difference(g)
    f.closePath()
    drawPath(f)

# original Riley plan for Rustle 6    
plan = [
    [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
    [0, 0, 0, 1, 0, 0, 0, -1, 0, 0, 0, 1, 0, 0],
    [-1, 0, -1, 0, 0, 0, 0, 0, 1, 0, -1, 0, 0],
    [1, 0, 1, -1, 0, 0, -1, 0, 0, 0, 1, 0, 1, 0],
    [0, -1, 0, -1, 0, 0, 0, 1, 0, -1, 0, -1, 0],
    [0, 1, -1, 1, 0, 1, 0, 0, 0, 0, -1, 1, 0, 0],
    [-1, 0, -1, 0, 0, 0, 0, 0, 0, 0, -1, 0, -1],
    [1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0],
    [0, -1, 0, 0, 0, 0, 0, 0, 0, -1, 0, -1, 0],
    [0, -1, 0, 1, 0, 1, 0, 1, 0, 0, 0, 1, -1, 0],
    [-1, 0, -1, 0, 0, 0, 0, 0, 0, 0, -1, -1, -1],
    [-1, 0, -1, 0, 1, 0, 1, 0, 0, 0, 1, -1, -1, 0],
    [0, -1, 0, 0, 0, 0, 0, 0, 0, -1, 0, -1, -1, 0],
    [0, 0, 0, -1, 0, 0, 0, 0, 0, 0, 0, -1, 0, 0],
    [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
    ]

for n in range(nFrames):
    newPage(canvasX, canvasY)
    frameDuration(1/30)
    ex = sin(radians(n * (360/nFrames)))
    fill(1)
    rect(0, 0, canvasX, canvasY)
    with savedState():
        translate(canvasX/2, canvasY/2)
        fill(0)
        translate(-(nX - 1) * offsetX/2, (nY - 1) * offsetY/2)
        for i in range(nY):
            with savedState():
                translate(0, i * -offsetY)
                if i % 2 == 0:
                    for j in range(nX - 1):
                        with savedState():
                            translate(offsetX/2 + (j * offsetX), 0)
                            triangle(side, height, center, .5 - (ex * (plan[i][j] * .5)))
                else:
                    for j in range(nX):
                        with savedState():
                            translate(j * offsetX, 0)
                            triangle(side, height, center, .5 - (ex * (plan[i][j] * .5)))
    fill(1)
    frame(canvasX, canvasY, side/2)

saveImage('~/Desktop/riley_tribute_test_01_%s.gif' %time)

The reference image:

I mean to do so at some point—probably on GitHub, probably this summer when I can take a little time to clean up code and organize the sketches.

And here's yet another—I forget the name of the illusion just now.

canvas = 500 #750
number = 15
shape = canvas/15
small = shape/2
offset = shape
nFrames = 90 #240
r = 5

def semi(pos, rot):
    s = BezierPath()
    if pos == 'top':
        s.arc((0, 0), shape/2, 0, 180, False)
    elif pos == 'bottom':
        s.arc((0, 0), shape/2, 0, 180, True)
    s.closePath()
    s.rotate(rot)
    drawPath(s)
    
for n in range(nFrames):
    newPage(canvas, canvas)
    frameDuration(1/30)
    fill(.85)
    rect(0, 0, canvas, canvas)
    translate(canvas/2, canvas/2)
    save()
    translate(-(number - 1) * offset/2, (number - 1) * offset/2)
    fill(.2)
    for i in range(number):
        save()
        translate(0, i * -offset)
        if (i % 2) == 0:
            v = 1
        elif (i % 2) == 1:
            v = -1
        for j in range(number):
            save()
            translate(j * offset, 0)
            if (j % 2) == 0:
                semi('top', r * (sin((2 * pi) * (n/nFrames))) * v)
            elif (j % 2) == 1:
                semi('bottom', r * (sin((2 * pi) * (n/nFrames))) * v)
            restore()
        restore()
    restore()
    translate(0, (number - 1) * offset/2)
    for i in range(number):
        save()
        translate(0, i * -offset)
        if (i % 2) == 0:
            fill(1)
            oval((cos((2 * pi) * (n/nFrames)) * canvas/2) - small/2, 0, small, small)
        elif (i % 2) == 1:
            fill(1)
            oval((cos((2 * pi) * (n/nFrames)) * -canvas/2) - small/2, 0, small, small)
        restore()
saveImage('~/Desktop/op_semi_lines.gif')

With both sketches, a larger canvas and more frames (240+) makes for a better effect.

Here's code for another, the 'intertwining illusion'.

canvas = 500
nFrames = 60 #240
size = canvas/36
num1 = 60
num2 = 44
num3 = 30
num4 = 14
r = 15

def mod(rot):
    m = BezierPath()
    m.rect(-size/2, -size/2, size, size)
    m.closePath()
    m.rotate(rot)
    drawPath(m)
    
for n in range(nFrames):
    newPage(canvas, canvas)
    frameDuration(1/30)
    fill(.5)
    rect(0, 0, canvas, canvas)
    strokeWidth(3)
    translate(canvas/2, canvas/2)
    save()
    for i in range(num1):
        if i % 2 == 0:
            stroke(1)
        else:
            stroke(.15)
        save()
        translate(.8 * canvas/2, 0)
        mod(r * (sin(2 * pi * n/nFrames)))
        restore()
        rotate(360/num1)
    restore()
    save()
    for j in range(num2):
        if j % 2 == 0:
            stroke(1)
        else:
            stroke(.15)
        save()
        translate(.6 * canvas/2, 0)
        mod(-r * (sin(2 * pi * n/nFrames)))
        restore()
        rotate(360/num2)    
    restore()
    save()
    for k in range(num3):
        if k % 2 == 0:
            stroke(1)
        else:
            stroke(.15)
        save()
        translate(.4 * canvas/2, 0)
        mod(r * (sin(2 * pi * n/nFrames)))
        restore()
        rotate(360/num3)    
    restore()
    save()
    for l in range(num4):
        if l % 2 == 0:
            stroke(1)
        else:
            stroke(.15)
        save()
        translate(.2 * canvas/2, 0)
        mod(-r * (sin(2 * pi * n/nFrames)))
        restore()
        rotate(360/num4)    
    restore()

saveImage('~/Desktop/opart_intertwine.gif')

The more frames, the more subtle the shift.

Good news about the pen functionality—glad I saw that, it'll come in handy for a couple things I'm working on …

@michelangelo Ned, please feel free to email me at maurice@mauricemeilleur.net. This is a bit of a specialty of mine.

Is it possible to add units and resolution as options for .svg export in saveImage()? I have a very niche reason for asking: I want to use exported .svg files with a plotter, and it might make the files easier to use if the units/resolution (effectively, scale) information in the .svg file.

Specifically, I want to plot .svg outputs with an AxiDraw, and I'm having trouble convincing Inkscape that the files should be treated as 72dpi rather than stubbornly insisting (1) that they are sized in pixels, not DTP, and (2) that it should treat them as if they're 96ppi.

Why does applying a Gaussian blur in different places in the code move the image it applies to? Example code, output, revised code putting the blur call before the 1st image, output.

size(550, 300)
# initiate a new image object
im = ImageObject()
# draw in the image
# the 'with' statement will create a custom context
# only drawing in the image object
with im:
   # set a size for the image
   size(200, 200)
   # draw something
   fill(1, 0, 0)
   rect(0, 0, width(), height())
   fill(1)
   fontSize(30)
   text("Hello World", (10, 10))
# draw in the image in the main context
image(im, (10, 50))
# apply some filters
im.gaussianBlur()
# get the offset (with a blur this will be negative)
x, y = im.offset()
# draw in the image in the main context
image(im, (300+x, 50+y))

yields:

size(550, 300)
# initiate a new image object
im = ImageObject()
# draw in the image
# the 'with' statement will create a custom context
# only drawing in the image object
with im:
   # set a size for the image
   size(200, 200)
   # draw something
   fill(1, 0, 0)
   rect(0, 0, width(), height())
   fill(1)
   fontSize(30)
   text("Hello World", (10, 10))
# apply some filters
im.gaussianBlur()
# draw in the image in the main context
image(im, (10, 50))
# get the offset (with a blur this will be negative)
x, y = im.offset()
# draw in the image in the main context
image(im, (300+x, 50+y))

yields:

@frederik https://www.drawbot.com/_/downloads/en/stable/pdf/, and my bad: I was using an old version of the documentation (3.121). I see the 3.126 version has a working URL. (The line break is breaking the link in the pdf, though.)

Ran into this same problem—@gferreira's updated URL works, but someone should update the DB pdf documentation for image(), etc. to include it and @frederik's detail about a direct link to image data.

That's what I suspected. It seemed like the kind of question students would ask and I wanted to make sure the answer I gave was the right one. Thanks, Frederik.

Just a matter of convenience, if I'm only using one path?

Wim, thanks—as soon as I get a chance I'll see if this works for me. A potential problem is that I'll have polygons (not just triangles but squares, pentagons, etc) rotating, which means that any given shape in the larger figure will be interacting with multiple other shapes in changing ways.

Funny, not for me—can't find a package for NodeBox, and errors for nodebox-opengl and nodebox-color:

These are similar to the error I got and reported on PageBot's GitHub repo trying to install PageBot (or re-install it, since I thought we'd solved the problems I was having last October and I'd installed it successfully):

Collecting pagebot
  Using cached https://files.pythonhosted.org/packages/ce/b5/85ecaa46445effb02ea7e127aae95136a813bd44565d7a6f48e14989d64e/pagebot-0.9.3-py3-none-any.whl
Collecting booleanOperations
  Downloading https://files.pythonhosted.org/packages/fc/c6/c4cae54f482465a33c5f011d95ec64293dce9e012dac7873147c2dc85396/booleanOperations-0.9.0-py3-none-any.whl
Collecting tornado
  Using cached https://files.pythonhosted.org/packages/30/78/2d2823598496127b21423baffaa186b668f73cd91887fcef78b6eade136b/tornado-6.0.3.tar.gz
    ERROR: Command errored out with exit status 1:
     command: /Applications/DrawBot.app/Contents/MacOS/python -c 'import sys, setuptools, tokenize; sys.argv[0] = '"'"'/private/tmp/pip-install-anyjj7fg/tornado/setup.py'"'"'; __file__='"'"'/private/tmp/pip-install-anyjj7fg/tornado/setup.py'"'"';f=getattr(tokenize, '"'"'open'"'"', open)(__file__);code=f.read().replace('"'"'\r\n'"'"', '"'"'\n'"'"');f.close();exec(compile(code, __file__, '"'"'exec'"'"'))' --no-user-cfg egg_info --egg-base /private/tmp/pip-install-anyjj7fg/tornado/pip-egg-info
         cwd: /private/tmp/pip-install-anyjj7fg/tornado/
    Complete output (6 lines):
    Traceback (most recent call last):
      File "<string>", line 1, in <module>
      File "setuptools/__init__.pyc", line 20, in <module>
      File "setuptools/dist.pyc", line 30, in <module>
      File "setuptools/extern/__init__.pyc", line 61, in load_module
    ImportError: The 'packaging' package is required; normally this is bundled with this package so if you get this warning, consult the packager of your distribution.
    ----------------------------------------
ERROR: Command errored out with exit status 1: python setup.py egg_info Check the logs for full command output.

In case it's relevant, here is the result of the search: