Line handling of nans

[EricThomson]

When data contains nans rendering doesn't work for line objects. Images work (i.e., the nans just show up as zeros).

xs = np.linspace(0, 30, 500)
ys = np.sin(xs)
ys[30:50] = np.nan
data1 = np.dstack([xs, ys])[0]

plot_l = Plot()
plot_l.line(data=data1, size=2, cmap="jet")
plot_l.show()

Compare to following which shows up fine (line renders except the nans which are just blank -- obviously could be handled differently this is just one option):

import matplotlib.pyplot as plt
plt.plot(xs, ys)

In practice this comes up frequently as the caiman coordinates have nans in them fairly often (right now I just filter out those nan-containing rows).

[kushalkolar]

Yes I see, I encountered this issue with the HistogramGraphic previously and fixed it in #28

We could just implement it in the base Graphic when data are converted to float32, do you want to do your first PR? :D https://github.com/kushalkolar/fastplotlib/blob/master/fastplotlib/graphics/_base.py#L19

[EricThomson]

Seems like a good one for start with for sure :) Might take me a little while to get to it (weekend likely).

[kushalkolar]

thinking about this more, for lines we don't want to show it as zeros though becuase this will actually move the position of the line vertex, maybe we can set the alpha value for the color at those points to zero

[EricThomson]

Yes the zero strategy won’t work for lines. Either need to interpolate or make the relevant bits disappear. For the latter, could split into multiple lines or the alpha trick sound good. Having to track multiple separate line objects would probably be a pain!

[kushalkolar]

splitting into multiple lines would be a major pain, definitely try the alpha method!

[kushalkolar]

I made a PR in pygfx which should fix this, it was a camera issue because autoscaling isn't possible when the bbox is nan:

pygfx/pygfx#540

[kushalkolar]

fixed in pygfx/pygfx#540