I noted this issue recently when working with several datasets which have scales=(0.0001,) but offsets=(0.0,) resulting in the default rescale=True not applying the scale value.
To facilitate future testing of rio_reader functionality, I'd also propose incorporating some file based integration tests similar to those used in rioxarray which might cover these types of edge cases.
I noted this issue recently when working with several datasets which have
scales=(0.0001,)butoffsets=(0.0,)resulting in the defaultrescale=Truenot applying the scale value.To facilitate future testing of
rio_readerfunctionality, I'd also propose incorporating some file based integration tests similar to those used inrioxarraywhich might cover these types of edge cases.