One of my goals is to add regression tests to make sure that each step in the new pipeline matches the output from the old. While this will be useful to make sure that things are working, by itself this will not be useful for tracking down how the new output differs from the original.
To that end, it would be helpful to have helper functionality that will produce useful output to show how the files differ.
Reading the CDF file into an xarray object would hopefully let us make use of functionality already available across the scientific pythoniverse. If I'm remembering right, we could likely do this with cdflib. I'll have to look into which tools already exist.
One of my goals is to add regression tests to make sure that each step in the new pipeline matches the output from the old. While this will be useful to make sure that things are working, by itself this will not be useful for tracking down how the new output differs from the original.
To that end, it would be helpful to have helper functionality that will produce useful output to show how the files differ.
Reading the CDF file into an xarray object would hopefully let us make use of functionality already available across the scientific pythoniverse. If I'm remembering right, we could likely do this with cdflib. I'll have to look into which tools already exist.