Skip to content

BUG: RangeIndex.difference missed case #44107

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 2 commits into from
Oct 20, 2021
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
21 changes: 16 additions & 5 deletions pandas/core/indexes/range.py
Original file line number Diff line number Diff line change
Expand Up @@ -705,13 +705,21 @@ def _difference(self, other, sort=None):
else:
return super()._difference(other, sort=sort)

elif len(overlap) == 2 and overlap[0] == first[0] and overlap[-1] == first[-1]:
# e.g. range(-8, 20, 7) and range(13, -9, -3)
return self[1:-1]

if overlap.step == first.step:
if overlap[0] == first.start:
# The difference is everything after the intersection
new_rng = range(overlap[-1] + first.step, first.stop, first.step)
elif overlap[-1] == first[-1]:
# The difference is everything before the intersection
new_rng = range(first.start, overlap[0], first.step)
elif overlap._range == first[1:-1]:
# e.g. range(4) and range(1, 3)
step = len(first) - 1
new_rng = first[::step]
else:
# The difference is not range-like
# e.g. range(1, 10, 1) and range(3, 7, 1)
Expand All @@ -725,16 +733,19 @@ def _difference(self, other, sort=None):
if overlap.step == first.step * 2:
if overlap[0] == first[0] and overlap[-1] in (first[-1], first[-2]):
# e.g. range(1, 10, 1) and range(1, 10, 2)
return self[1::2]
new_rng = first[1::2]

elif overlap[0] == first[1] and overlap[-1] in (first[-1], first[-2]):
# e.g. range(1, 10, 1) and range(2, 10, 2)
return self[::2]
new_rng = first[::2]

# We can get here with e.g. range(20) and range(0, 10, 2)
else:
# We can get here with e.g. range(20) and range(0, 10, 2)
return super()._difference(other, sort=sort)

# e.g. range(10) and range(0, 10, 3)
return super()._difference(other, sort=sort)
else:
# e.g. range(10) and range(0, 10, 3)
return super()._difference(other, sort=sort)

new_index = type(self)._simple_new(new_rng, name=res_name)
if first is not self._range:
Expand Down
18 changes: 18 additions & 0 deletions pandas/tests/indexes/ranges/test_setops.py
Original file line number Diff line number Diff line change
Expand Up @@ -377,6 +377,24 @@ def test_difference_mismatched_step(self):
result = obj[::-1].difference(obj[1::2], sort=False)
tm.assert_index_equal(result, expected[::-1], exact=True)

def test_difference_interior_overlap_endpoints_preserved(self):
left = RangeIndex(range(4))
right = RangeIndex(range(1, 3))

result = left.difference(right)
expected = RangeIndex(0, 4, 3)
assert expected.tolist() == [0, 3]
tm.assert_index_equal(result, expected, exact=True)

def test_difference_endpoints_overlap_interior_preserved(self):
left = RangeIndex(-8, 20, 7)
right = RangeIndex(13, -9, -3)

result = left.difference(right)
expected = RangeIndex(-1, 13, 7)
assert expected.tolist() == [-1, 6]
tm.assert_index_equal(result, expected, exact=True)

def test_difference_interior_non_preserving(self):
# case with intersection of length 1 but RangeIndex is not preserved
idx = Index(range(10))
Expand Down