CLN: remove need for assert_matching in pandas/tests/indexes/test_multi.py

[jreback]

xref #19058

I believe this can be subsumed with simple result.equals(expected) index comparisions.
no need to have a non-standard comparison method.

[jreback]

cc @Mofef

[databasedav]

Index.equals uses:

pandas/pandas/core/dtypes/missing.py

Lines 258 to 264 in bbfbe48

	def array_equivalent(left, right, strict_nan=False):
	"""
	True if two arrays, left and right, have equal non-NaN elements, and NaNs
	in corresponding locations. False otherwise. It is assumed that left and
	right are NumPy arrays of the same dtype. The behavior of this function
	(particularly with respect to NaNs) is not defined if the dtypes are
	different.

which assumes that the dtypes of each array is the same whereas some of the assert_matching calls in test_multi.py specify the checking of dtypes. Should I add an argument for checking dtypes in Index.equals?

[jreback]

oh sorry, I meant to say tm.assert_index_equals which does full comparisons.

[databasedav]

Ok so all of the assert_matching's are comparing .level's and .label's

pandas/pandas/tests/indexes/test_multi.py

Lines 180 to 198 in f6c0f8a

	def test_set_levels(self):
	# side note - you probably wouldn't want to use levels and labels
	# directly like this - but it is possible.
	levels = self.index.levels
	new_levels = [[lev + 'a' for lev in level] for level in levels]
	def assert_matching(actual, expected, check_dtype=False):
	# avoid specifying internal representation
	# as much as possible
	assert len(actual) == len(expected)
	for act, exp in zip(actual, expected):
	act = np.asarray(act)
	exp = np.asarray(exp)
	tm.assert_numpy_array_equal(act, exp, check_dtype=check_dtype)
	# level changing [w/o mutation]
	ind2 = self.index.set_levels(new_levels)
	assert_matching(ind2.levels, new_levels)
	assert_matching(self.index.levels, levels)

In the first test, a slightly modified levels new_levels is being set on a new index ind2 and then assert_matching is checking if ind2's levels have been set correctly without mutating the original. So does using tm.assert_index_equal make sense in this case since it would require manually creating a new index to compare levels when all we want to compare are the levels themselves?

[jreback]

I prefer to use the stock routines rather than write a custom routine.