Skip to content

REGR: non-unique, masked dtype index raising IndexError #57061

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.

Sign up for GitHub

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 8 commits into from
Jan 30, 2024
+44 −32
Merged

REGR: non-unique, masked dtype index raising IndexError #57061

Show file tree
Hide file tree
Changes from 3 commits
Commits
Show all changes
8 commits
Select commit Hold shift + click to select a range
0ec13d3
fix masked indexing regression
lukemanley Jan 25, 2024
94318ac
Merge remote-tracking branch 'upstream/main' into masked-nonunique-in…
lukemanley Jan 25, 2024
fb0d7bb
fix test
lukemanley Jan 25, 2024
868140d
fix test
lukemanley Jan 25, 2024
390ebad
Merge remote-tracking branch 'upstream/main' into masked-nonunique-in…
lukemanley Jan 26, 2024
ec56a51
dedup resizing logic
lukemanley Jan 26, 2024
1bf99d5
add types
lukemanley Jan 26, 2024
3a6b070
Merge remote-tracking branch 'upstream/main' into masked-nonunique-in…
lukemanley Jan 29, 2024
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
1 change: 1 addition & 0 deletions doc/source/whatsnew/v2.2.1.rst
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -14,6 +14,7 @@ including other versions of pandas.
Fixed regressions
~~~~~~~~~~~~~~~~~
- Fixed regression in :func:`merge_ordered` raising ``TypeError`` for ``fill_method="ffill"`` and ``how="left"`` (:issue:`57010`)
- Fixed regression in :meth:`DataFrame.loc` raising ``IndexError`` for non-unique, masked dtype indexes where result has more than 10,000 rows (:issue:`57027`)
- Fixed regression in :meth:`Series.pct_change` raising a ``ValueError`` for an empty :class:`Series` (:issue:`57056`)

.. ---------------------------------------------------------------------------
Expand Down
6 changes: 5 additions & 1 deletion pandas/_libs/index.pyx
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1220,6 +1220,7 @@ cdef class MaskedIndexEngine(IndexEngine):
n_alloc *= 2
if n_alloc > max_alloc:
n_alloc = max_alloc
result = np.resize(result, n_alloc)
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Is this required to resize here? If so can we get rid of the resize that happens after this branch is completed? Might be misreading but it looks duplicative

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I think this is required. The result size isn't known so I believe the resize check needs to be done each time a new value is set. I don't see a way to avoid it with the current algo.

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Ah yea that makes sense. I think its worth making the conditional resize its own function then so we don't have as much duplication

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

good idea - done


result[count] = na_idx
count += 1
Expand All @@ -1236,14 +1237,17 @@ cdef class MaskedIndexEngine(IndexEngine):
n_alloc *= 2
if n_alloc > max_alloc:
n_alloc = max_alloc
result = np.resize(result, n_alloc)

result[count] = j
count += 1
continue

# value not found
if count >= n_alloc:
n_alloc += 10_000
n_alloc *= 2
if n_alloc > max_alloc:
n_alloc = max_alloc
result = np.resize(result, n_alloc)
result[count] = -1
count += 1
Expand Down
12 changes: 12 additions & 0 deletions pandas/tests/indexing/test_loc.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3347,3 +3347,15 @@ def test_getitem_loc_str_periodindex(self):
index = pd.period_range(start="2000", periods=20, freq="B")
series = Series(range(20), index=index)
assert series.loc["2000-01-14"] == 9

def test_loc_nonunique_masked_index(self):
# GH 57027
ids = list(range(11))
index = Index(ids * 1000, dtype="Int64")
df = DataFrame({"val": np.arange(len(index))}, index=index)
result = df.loc[ids]
expected = DataFrame(
{"val": index.argsort(kind="stable")},
index=Index(np.array(ids).repeat(1000), dtype="Int64"),
)
tm.assert_frame_equal(result, expected)