Merge pull request #6 from adambear82/adambear82-colorByGroup

Create colorByGroup.py

workflow/colorByGroup.py

@@ -0,0 +1,97 @@
+'''
+COLOR BY GROUP - VISUALISE GROUPS AS IN GROUP BY KEY
+'''
+__author__ = 'Adam Bear - adam@ukbear.com'
+__twitter__ = '@adambear82'
+__version__ = '1.0.0'
+
+# IMPORTS / REFERENCES:
+import clr
+clr.AddReference('DSCoreNodes')
+import DSCore
+clr.AddReference('ProtoGeometry')
+from Autodesk.DesignScript.Geometry import *
+clr.AddReference('Display')
+from Display import *
+
+# IN PORTS:
+geometry = IN[0]
+
+# define all the colors of the rainbow in a spectrum list
+c00 = red = DSCore.Color.ByARGB(255, 255, 000, 000)
+c02 = yellow = DSCore.Color.ByARGB(255, 255, 255, 000)
+c04 = green = DSCore.Color.ByARGB(255, 000, 255, 000)
+c06 = cyan = DSCore.Color.ByARGB(255, 000, 255, 255)
+c08 = blue = DSCore.Color.ByARGB(255, 000, 000, 255)
+c10 = magenta = DSCore.Color.ByARGB(255, 255, 000, 255)
+spectrum = c00, c02, c04, c06, c08, c10
+
+# similar to color range node indicies need to be mapped to each color
+# red (0.0), yellow(0.2), green(0.4), cyan(0.6), blue(0.8), magenta(1.0)
+# if we assumed the spectrum values did not change this could simply be
+# sU = sV = 0.0, 0.2, 0.4, 0.6, 0.8, 1.0
+# in designscript we would use the following
+# 0..1..#List.Count(spectrum)
+# for a rigorous approach lets look at how to acheive this in python
+# for all the values in the range of the length of spectrum
+# each value is multipled by the inverse of the length of the spectrum-1
+# if we did not -1 then the factor would be (1/6=0.16) not (1/5=0.20)
+# for more info refer to
+# https://github.com/Amoursol/dynamoPython/blob/master/concepts/numberSequences.py
+sU = [ value * 1.0 / (len(spectrum) - 1) for value in range( len(spectrum) ) ]
+# although the spectrum UV is 2D by stating sU=sV the color range is 1D
+# for more complex requirements a different set of values could be given
+sV = sU
+
+# create the empty list sUV to hold UV coordinates of the spectrum
+sUV = []
+# for every item in lists sU and sV
+# append to sUV the result of UV.ByCordinates
+for u, v in zip(sU, sV) :
+	sUV.append(UV.ByCoordinates(u, v))
+
+# the color range of the spectrum is mapped to indices using UV coords
+colorrange2d = DSCore.ColorRange2D.ByColorsAndParameters(spectrum, sUV)
+
+# as above for spectrum UVs, geometry UVs are listed
+# in design script this would be
+# 0..1..#List.Count(geometry)
+# because the geometry is variable we must calculate this list
+gU = [ value * 1.0 / (len(geometry) - 1) for value in range( len(geometry) ) ]
+# although the geometry UV is 2D by stating gU = gV the mapping is 1D
+# for more complex requirements a different set of values could be given
+gV = gU
+
+# create the empty list gUV to hold UV coordinates of the geometry
+gUV = []
+# the color range of the geometry is mapped to values using UV coords
+for u, v in zip(gU, gV) :
+	gUV.append(UV.ByCoordinates(u, v))
+
+# create an empty list colors to hold values of colors to be displayed
+colors = []
+# for each item in the geometry UV
+for i in gUV :
+	# append the coresponding spectrum UV
+	colors.append(colorrange2d.GetColorAtParameter(i))
+
+# create an empty list colorByGroup to hold geometry displayed in color
+colorByGroup = []
+# for each item 'geom' in the geometry
+for geom, c in zip(geometry, colors) :
+	for g in geom :
+		colorByGroup.append(Display.ByGeometryColor(g, c))
+
+
+# to output the displayed geometry in the same groups as it is inputed
+# clockwork provides a good example of how to chop lists evenly
+# https://github.com/andydandy74/ClockworkForDynamo/blob/master/nodes/1.x/List.Chop+.dyf
+def ListChopEvenly(l, n):
+	# Andreas provides reference from stack overflow with python 2 example
+	# https://stackoverflow.com/questions/312443/how-do-you-split-a-list-into-evenly-sized-chunks
+	return [l[i:i + n] for i in xrange(0, len(l), n)]
+# clockworks chop definition is used to chop the geometry
+choppedGroups = ListChopEvenly(colorByGroup, len(geom))
+
+# the choppedGroups are sent to the OUT port
+OUT = choppedGroups