Skip to content
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

fix: Cast empty lists and lists of only nulls without an inner dtype #15525

Closed
wants to merge 5 commits into from
Closed

fix: Cast empty lists and lists of only nulls without an inner dtype #15525

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
f7044d9
fix: Cast empty lists in Series of lists without an inner dtype
petrosbar Apr 3, 2024
ec8a227
mypy
petrosbar Apr 7, 2024
589df5e
move branch outside from_iter
petrosbar Apr 11, 2024
7b68752
else if
petrosbar Apr 12, 2024
01064c3
ease branch prediction
petrosbar Apr 14, 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
6 changes: 6 additions & 0 deletions crates/polars-core/src/chunked_array/upstream_traits.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -236,6 +236,12 @@ impl FromIterator<Option<Series>> for ListChunked {
builder.append_series(first_s).unwrap();

for opt_s in it {
if opt_s.is_some() {
debug_assert_eq!(
opt_s.clone().unwrap().dtype(),
first_s.dtype()
);
}
builder.append_opt_series(opt_s.as_ref()).unwrap();
}
builder.finish()
Expand Down
12 changes: 11 additions & 1 deletion crates/polars-core/src/series/any_value.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -493,9 +493,19 @@ fn any_values_to_list(
let mut valid = true;
#[allow(unused_mut)]
let mut out: ListChunked = if inner_type == &DataType::Null {
let mut first_value_dtype = DataType::Null;
let mut set_first = false;
avs.iter()
.map(|av| match av {
AnyValue::List(b) => Some(b.clone()),
AnyValue::List(b) => {
if !set_first && first_value_dtype != *b.dtype() {
first_value_dtype = b.dtype().clone();
set_first = true;
} else if b.is_empty() || b.null_count() == b.len() {
return b.cast(&first_value_dtype).ok();
}
Some(b.clone())
},
AnyValue::Null => None,
_ => {
valid = false;
Expand Down
15 changes: 15 additions & 0 deletions py-polars/tests/unit/datatypes/test_list.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -831,3 +831,18 @@ def test_take_list_15719() -> None:
)

assert_frame_equal(df, expected)


@pytest.mark.parametrize(
("values", "dtype", "full_dtype"),
[
([["a"], [], [None], None], pl.List, pl.List(pl.String)),
([[1], [], [None], None], pl.List, pl.List(pl.Int64)),
([[], None, ["a"]], pl.List, pl.List(pl.String)),
],
)
def test_list_anyvalue_empty_list(
values: list[Any], dtype: PolarsDataType, full_dtype: PolarsDataType
) -> None:
s = pl.Series("a", values, dtype=dtype)
assert_series_equal(s, pl.Series("a", values, dtype=full_dtype))
Loading