how the density range of image was calculated

[emmaai]

I'm not sure how the logic of this line works.

https://github.com/opendatacube/eo-datasets/blame/d5cf4a44b8ae3bbbe89cf026aa0e244a9badc8a3/eodatasets3/images.py#L1284

Shouldn't it get lowest of low and highest of high of all bands (e.g., rgb) instead?

atm, it caused trouble because I have three bands with different (min, max),

8 45
0 35
40 84

it gives the density range (40, 35), which doesn't make sense in the following function https://github.com/opendatacube/eo-datasets/blame/d5cf4a44b8ae3bbbe89cf026aa0e244a9badc8a3/eodatasets3/images.py#L1292

Is it a bug or something not supposed to be triggered?

[emmaai]

The bug has happened since some version post 0.19.2.