Many Tests Fail due to Small Tolerances on Almost Equal Comparisons

[swainn]

I've marked many of the tests with xfail (expected fail) b/c most of them use almost equal comparisons on files with very small tolerances. As a result, the tests fail on many systems due to the natural variance from system to system.

When marked as xfail, these tests will still run, but won't cause the test suite to fail when they fail.

[swainn]

Oh and at least two of the xfail tests are failing b/c they are using deprecated xarray calls.

[snowman2]

If I remember correctly, my logic here was to get the tests passing with the strictest tolerance possible. With all of the changes to GDAL and PROJ since I touched them, I would recommend regenerating the test files and manually inspecting them to see if they work.

[snowman2]

Also, the tolerance is arbitrary. Maybe 2-3 decimal places would be better.

[swainn]

@snowman2 I agree, but I’m not familiar enough with the code to regenerate the test files and know if they are valid. I recommend either you do it or someone at ERDC that has an interest in those parts of the library. The core ORM functionality is working and test are passing, which is all I need at this point.