This repository has been archived on 2023-03-25. You can view files and clone it, but cannot push or open issues or pull requests.
mightyscape-1.1-deprecated/extensions/fablabchemnitz/fablabchemnitz_exponential_distort.py

108 lines
4.7 KiB
Python
Raw Normal View History

2020-07-30 01:16:18 +02:00
#!/usr/bin/env python3
from __future__ import print_function
import sys
import math
import inkex
from inkex.paths import CubicSuperPath
class TransformExponential(inkex.Effect):
def __init__(self):
inkex.Effect.__init__(self)
#self.arg_parser.add_argument('-a', '--axis', default='x', help='distortion axis. Valid values are "x", "y", or "xy". Default is "x"')
self.arg_parser.add_argument('-x', '--exponent', type=float, default=1.3, help='distortion factor. 1=no distortion, default 1.3')
self.arg_parser.add_argument('-p', '--padding_perc', type=float, default=0, help='pad at origin. Padding 100% runs the exponential curve through [0.5 .. 1.0] -- default 0% runs through [0.0 .. 1.0]')
def x_exp(self, bbox, x):
""" reference implementation ignoring padding. unused. """
xmin = bbox[0] # maps to 0
xmax = bbox[1] # maps to 1
w = xmax-xmin # maps to 1
# convert world to math coordinates
xm = (x-xmin)/w
# apply function with properties f(1.0) == 1.0 and f(0.0) == 0.0
xm = xm**self.options.exponent # oh, parabola or logarithm?
# convert back from math to world coordinates.
return x*w + xmin
def x_exp_p(self, bbox, x):
""" parabola mapping with padding
CAUTION: the properties f(1.0) == 1.0 and f(0.0) == 0.0
do not really hold, as our x does not run the full range [0.0 .. 1.0]
FIXME: if you expect some c**xm here, instead of xm**c, think about c==1 ...
"""
xmin = bbox[0] # maps to 0 when padding=0,
xmax = bbox[1] # maps to 1
xzero = xmin - (xmax-xmin)*self.options.padding_perc*0.01 # maps to 0, after applying padding
w = xmax - xzero
w = w * (1+self.options.padding_perc*0.01)
# convert world to math coordinates
xm = (x-xzero)/w
# apply function with properties f(1.0) == 1.0 and f(0.0) == 0.0
xm = xm**self.options.exponent # oh, parabola or logarithm?
return xm
def x_exp_p_inplace(self, bbox, xm):
""" back from mat to world coordinates, retaining xmin and xmax
Algorithm: (pre)compute a linear mapping function by explicitly
running x_exp_p for the two points xmin and xmax.
Then use the resulting linear function to map back any xm into world coordinates x.
An obvious speedup by factor 3 is waiting for you here.
"""
xmin = bbox[0]
xmax = bbox[1]
## assert that xmin maps to xmin and xmax maps to xmax, whatever x_exp_p() does to us.
f_xmin = self.x_exp_p(bbox, xmin)
f_xmax = self.x_exp_p(bbox, xmax)
f_x = self.x_exp_p(bbox, xm)
x = (f_x - f_xmin) * (xmax-xmin) / (f_xmax-f_xmin) + xmin
return x
def computeBBox(self, pts):
""" 'improved' version of simplepath.computeBBox, this one includes b-spline handles."""
xmin = None
xmax = None
ymin = None
ymax = None
for p in pts:
for pp in p:
for ppp in pp:
if xmin is None: xmin = ppp[0]
if xmax is None: xmax = ppp[0]
if ymin is None: ymin = ppp[1]
if ymax is None: ymax = ppp[1]
if xmin > ppp[0]: xmin = ppp[0]
if xmax < ppp[0]: xmax = ppp[0]
if ymin > ppp[1]: ymin = ppp[1]
if ymax < ppp[1]: ymax = ppp[1]
return (xmin, xmax, ymin, ymax)
def effect(self):
if len(self.svg.selected) == 0:
2020-08-20 13:12:34 +02:00
inkex.errormsg("Please select an object to perform the " +
"exponential-distort transformation on.")
2020-07-30 01:16:18 +02:00
return
for id, node in self.svg.selected.items():
type = node.get("{http://sodipodi.sourceforge.net/DTD/sodipodi-0.dtd}type", "path")
if node.tag != '{http://www.w3.org/2000/svg}path' or type != 'path':
inkex.errormsg(node.tag + " is not a path. Type="+type+". Please use 'Path->Object to Path' first.")
else:
pts = CubicSuperPath(node.get('d'))
bbox = self.computeBBox(pts)
## bbox (60.0, 160.0, 77.0, 197.0)
## pts [[[[60.0, 77.0], [60.0, 77.0], [60.0, 77.0]], [[60.0, 197.0], [60.0, 197.0], [60.0, 197.0]], [[70.0, 197.0], ...
for p in pts:
for pp in p:
for ppp in pp:
ppp[0] = self.x_exp_p_inplace(bbox, ppp[0])
node.set('d', str(pts))
if __name__ == '__main__':
TransformExponential().run()