def test_numNonsharedFields_separateDim_differentNumberOfSlices(self):
# Doing diffs with a separate_dim, where variable has a different number
# of slices in the two variables. There should be one nonshared field
# for each missing slice. Note that we treat this as non-shared, rather
# than a dim-diff; this feels like the right thing to do, both in terms
# of:
# - By analogy to the fact that we treat each time slice as a separate
# variable for the sake of analysis
# - It seems like this is more likely to do the Right Thing in the case
# of intentional / expected differences in the number of time slices in
# the file
data1 = np.array([[1,2,3,4],[5,6,7,8]])
data2 = np.array([[1,2],[5,6]])
file1 = NetcdfFileFake(
self.FILENAME1,
variables = {'var1': NetcdfVariableFake(data1, ('dim1', 'dim2'))})
file2 = NetcdfFileFake(
self.FILENAME2,
variables = {'var1': NetcdfVariableFake(data2, ('dim1', 'dim2'))})
mydiffs = FileDiffs(file1, file2, separate_dim='dim2')
num_nonshared = mydiffs.num_nonshared_fields()
self.assertEqual(num_nonshared, 2)
# Also make sure that other counts are correct in this case
self.assertEqual(mydiffs.num_vars(), 4)
self.assertEqual(mydiffs.num_vars_differ(), 0)
self.assertEqual(mydiffs.num_dims_differ(), 0)
def test_numNonsharedFields_separateDim_differentNumberOfSlices(self):
# Doing diffs with a separate_dim, where variable has a different number
# of slices in the two variables. There should be one nonshared field
# for each missing slice. Note that we treat this as non-shared, rather
# than a dim-diff; this feels like the right thing to do, both in terms
# of:
# - By analogy to the fact that we treat each time slice as a separate
# variable for the sake of analysis
# - It seems like this is more likely to do the Right Thing in the case
# of intentional / expected differences in the number of time slices in
# the file
data1 = np.array([[1, 2, 3, 4], [5, 6, 7, 8]])
data2 = np.array([[1, 2], [5, 6]])
file1 = NetcdfFileFake(
self.FILENAME1,
variables={'var1': NetcdfVariableFake(data1, ('dim1', 'dim2'))})
file2 = NetcdfFileFake(
self.FILENAME2,
variables={'var1': NetcdfVariableFake(data2, ('dim1', 'dim2'))})
mydiffs = FileDiffs(file1, file2, separate_dim='dim2')
num_nonshared = mydiffs.num_nonshared_fields()
self.assertEqual(num_nonshared, 2)
# Also make sure that other counts are correct in this case
self.assertEqual(mydiffs.num_vars(), 4)
self.assertEqual(mydiffs.num_vars_differ(), 0)
self.assertEqual(mydiffs.num_dims_differ(), 0)
def test_numDimsDiffer_with3Vars(self):
file1 = NetcdfFileFake(
self.FILENAME1,
variables = {'var1': NetcdfVariableFake(np.array([1,2,3])),
'var2': NetcdfVariableFake(np.array([4,5,6])),
'var3': NetcdfVariableFake(np.array([7,8,9]))})
file2 = NetcdfFileFake(
self.FILENAME2,
variables = {'var1': NetcdfVariableFake(np.array([1,2])), # differs
'var2': NetcdfVariableFake(np.array([4,5,6])), # same
'var3': NetcdfVariableFake(np.array([7,8,9,10]))}) # differs
mydiffs = FileDiffs(file1, file2, separate_dim=None)
num_differ = mydiffs.num_dims_differ()
self.assertEqual(num_differ, 2)
def test_numDimsDiffer_with3Vars(self):
file1 = NetcdfFileFake(self.FILENAME1,
variables={
'var1':
NetcdfVariableFake(np.array([1, 2, 3])),
'var2':
NetcdfVariableFake(np.array([4, 5, 6])),
'var3':
NetcdfVariableFake(np.array([7, 8, 9]))
})
file2 = NetcdfFileFake(
self.FILENAME2,
variables={
'var1': NetcdfVariableFake(np.array([1, 2])), # differs
'var2': NetcdfVariableFake(np.array([4, 5, 6])), # same
'var3': NetcdfVariableFake(np.array([7, 8, 9, 10]))
}) # differs
mydiffs = FileDiffs(file1, file2, separate_dim=None)
num_differ = mydiffs.num_dims_differ()
self.assertEqual(num_differ, 2)