Skip to content

BUG: Fix Index.droplevel() stub — use Sequence[Never] to allow only droplevel([]) (GH#1678) #1732

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
loicdiridollou merged 8 commits into from
Apr 25, 2026
+6 −1
Merged

BUG: Fix Index.droplevel() stub — use Sequence[Never] to allow only droplevel([]) (GH#1678) #1732

Show file tree
Hide file tree
Changes from 2 commits
Commits
Show all changes
8 commits
Select commit Hold shift + click to select a range
d693e0b
BUG: Remove Index.droplevel() stub — only MultiIndex.droplevel() is v…
tinezivic Apr 22, 2026
b1cafc2
BUG: Use Sequence[Never] for Index.droplevel() — allows droplevel([])…
tinezivic Apr 24, 2026
45abdd8
CLN: drop explanatory comments per review (cmp0xff)
tinezivic Apr 25, 2026
55f4473
Move idx droplevel lines back to original position
tinezivic Apr 25, 2026
025400d
Restore np.integer subtype in test and pyrefly ignore in multi.pyi
tinezivic Apr 25, 2026
4aa6f43
Update pandas-stubs/core/indexes/base.pyi
tinezivic Apr 25, 2026
3af1913
Update tests/indexes/test_indexes.py
tinezivic Apr 25, 2026
ca916c2
Fix ignore comment for droplevel() — call-arg/reportCallIssue
tinezivic Apr 25, 2026
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
6 changes: 5 additions & 1 deletion pandas-stubs/core/indexes/base.pyi
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -431,7 +431,6 @@ class Index(IndexOpsMixin[S1], ElementOpsMixin[S1]):
@property
def nlevels(self) -> int: ...
def get_level_values(self, level: int | _str) -> Index: ...
def droplevel(self, level: Level | Sequence[Level] = 0) -> Self: ...
@property
def is_monotonic_increasing(self) -> bool: ...
@property
Expand All @@ -452,6 +451,11 @@ class Index(IndexOpsMixin[S1], ElementOpsMixin[S1]):
def dropna(self, how: AnyAll = "any") -> Self: ...
def unique(self, level: Hashable | None = None) -> Self: ...
def drop_duplicates(self, *, keep: DropKeep = "first") -> Self: ...
# droplevel() always raises ValueError on a plain (single-level) Index
# unless you pass an empty sequence — then it's a no-op.
# Use Sequence[Never] so type checkers allow droplevel([]) but reject
# anything else (droplevel(0), droplevel([0]), droplevel("name"), ...).
Comment thread
cmp0xff marked this conversation as resolved.
Outdated
def droplevel(self, level: Sequence[Never] = ...) -> Self: ...
Comment thread
tinezivic marked this conversation as resolved.
Outdated
def duplicated(self, keep: DropKeep = "first") -> np_1darray_bool: ...
def __and__(self, other: Never) -> Never: ...
def __rand__(self, other: Never) -> Never: ...
Expand Down
2 changes: 1 addition & 1 deletion pandas-stubs/core/indexes/multi.pyi
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -140,7 +140,7 @@ class MultiIndex(Index):
@property
def is_monotonic_decreasing(self) -> bool: ...
def dropna(self, how: AnyAll = "any") -> Self: ...
def droplevel(self, level: Level | Sequence[Level] = 0) -> MultiIndex | Index: ... # type: ignore[override] # pyrefly: ignore[bad-override]
def droplevel(self, level: Level | Sequence[Level] = 0) -> MultiIndex | Index: ... # type: ignore[override] # return type is wider than Self
Comment thread
tinezivic marked this conversation as resolved.
Outdated
def get_level_values(self, level: str | int) -> Index: ...
@overload # type: ignore[override]
def unique( # pyrefly: ignore[bad-override]
Expand Down
11 changes: 9 additions & 2 deletions tests/indexes/test_indexes.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1608,15 +1608,22 @@ def test_index_set_names() -> None:


def test_index_droplevel() -> None:
idx = pd.Index([1, 2])
check(assert_type(idx.droplevel([]), "pd.Index[int]"), pd.Index, np.integer)
Comment thread
tinezivic marked this conversation as resolved.
mi = pd.MultiIndex.from_arrays([[1, 2, 3], [4, 5, 6]], names=["elk", "owl"])
check(assert_type(mi.droplevel([]), pd.MultiIndex | pd.Index), pd.MultiIndex)
check(assert_type(mi.droplevel([0]), pd.MultiIndex | pd.Index), pd.Index)
check(assert_type(mi.droplevel((0,)), pd.MultiIndex | pd.Index), pd.Index)
check(assert_type(mi.droplevel(["elk"]), pd.MultiIndex | pd.Index), pd.Index)
check(assert_type(mi.droplevel(("elk",)), pd.MultiIndex | pd.Index), pd.Index)
check(assert_type(mi.droplevel(0), pd.MultiIndex | pd.Index), pd.Index)
# droplevel([]) on a plain Index is a documented no-op — returns the same Index
idx = pd.Index([1, 2])
check(assert_type(idx.droplevel([]), "pd.Index[int]"), pd.Index)
Comment thread
cmp0xff marked this conversation as resolved.
Outdated
if TYPE_CHECKING_INVALID_USAGE:
# Any non-empty argument raises ValueError at runtime, so type checkers
# should reject these — Sequence[Never] catches them all.
idx.droplevel(0) # type: ignore[arg-type] # pyright: ignore[reportArgumentType]
Comment thread
tinezivic marked this conversation as resolved.
idx.droplevel([0]) # type: ignore[arg-type] # pyright: ignore[reportArgumentType]
idx.droplevel("name") # type: ignore[arg-type] # pyright: ignore[reportArgumentType]


def test_index_setitem() -> None:
Expand Down