"colour.SpectralDistribution.align" method with non-uniform shapes?

[briend]

With a weird shape like this (below), the "shape" property doesn't seem to be useful or accurate. I'm trying to plug this non-uniform distribution in the Spectral_to_XYZ recovery functions but I cannot align the CMFS and illuminant to this funky SpectralDistribution. Is there something I'm missing or do I need to manually do the alignment somehow? Thank you for any advice!

data = {
467: 0.0,
495: 0.0,
532: 0.0,
570: 0.0,
580: 0.0,
590: 0.0,
630: 0.0
}SpectralDistribution(data).shape
SpectralShape(467.0, 630.0, 10.0)

[KelSolaar]

Hi @briend,

There is no direct way to do that indeed. You could however manually extract the values as follows:

from colour import SpectralDistribution, ILLUMINANTS_SDS

data = {
    467: 0.0,
    495: 0.0,
    532: 0.0,
    570: 0.0,
    580: 0.0,
    590: 0.0,
    630: 0.0
}
sd = SpectralDistribution(data)

illuminant = SpectralDistribution(ILLUMINANTS_SDS['D65'][sd.wavelengths], sd.wavelengths)

print(illuminant)

[[ 467.      115.746 ]
 [ 495.      109.082 ]
 [ 532.      107.0322]
 [ 570.       96.3342]
 [ 580.       95.788 ]
 [ 590.       88.6856]
 [ 630.       83.2886]]

[briend]

Thanks, that worked out well after tweaking the related recovery functions to use the custom shape for CMFS and illuminant

[KelSolaar]

Awesome thanks! Closing this one then :)