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

Jump to bottom

feat(rust): Implement strict/nonstrict conversion for primitive AnyValues #14186

Merged
merged 2 commits into from
Mar 2, 2024
Merged

feat(rust): Implement strict/nonstrict conversion for primitive AnyValues #14186

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
8319ba9
Add tests
stinodego Feb 18, 2024
d8c6051
Allow mixed integers
stinodego Feb 19, 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
110 changes: 90 additions & 20 deletions 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 @@ -3,12 +3,6 @@ use std::fmt::Write;
use crate::prelude::*;
use crate::utils::get_supertype;

fn any_values_to_primitive<T: PolarsNumericType>(avs: &[AnyValue]) -> ChunkedArray<T> {
avs.iter()
.map(|av| av.extract::<T::Native>())
.collect_trusted()
}

#[cfg(feature = "dtype-decimal")]
fn any_values_to_decimal(
avs: &[AnyValue],
Expand Down Expand Up @@ -225,36 +219,36 @@ impl Series {
) -> PolarsResult<Series> {
let mut s = match dtype {
#[cfg(feature = "dtype-i8")]
DataType::Int8 => any_values_to_primitive::<Int8Type>(av).into_series(),
DataType::Int8 => any_values_to_integer::<Int8Type>(av, strict)?.into_series(),
#[cfg(feature = "dtype-i16")]
DataType::Int16 => any_values_to_primitive::<Int16Type>(av).into_series(),
DataType::Int32 => any_values_to_primitive::<Int32Type>(av).into_series(),
DataType::Int64 => any_values_to_primitive::<Int64Type>(av).into_series(),
DataType::Int16 => any_values_to_integer::<Int16Type>(av, strict)?.into_series(),
DataType::Int32 => any_values_to_integer::<Int32Type>(av, strict)?.into_series(),
DataType::Int64 => any_values_to_integer::<Int64Type>(av, strict)?.into_series(),
#[cfg(feature = "dtype-u8")]
DataType::UInt8 => any_values_to_primitive::<UInt8Type>(av).into_series(),
DataType::UInt8 => any_values_to_integer::<UInt8Type>(av, strict)?.into_series(),
#[cfg(feature = "dtype-u16")]
DataType::UInt16 => any_values_to_primitive::<UInt16Type>(av).into_series(),
DataType::UInt32 => any_values_to_primitive::<UInt32Type>(av).into_series(),
DataType::UInt64 => any_values_to_primitive::<UInt64Type>(av).into_series(),
DataType::Float32 => any_values_to_primitive::<Float32Type>(av).into_series(),
DataType::Float64 => any_values_to_primitive::<Float64Type>(av).into_series(),
DataType::UInt16 => any_values_to_integer::<UInt16Type>(av, strict)?.into_series(),
DataType::UInt32 => any_values_to_integer::<UInt32Type>(av, strict)?.into_series(),
DataType::UInt64 => any_values_to_integer::<UInt64Type>(av, strict)?.into_series(),
DataType::Float32 => any_values_to_f32(av, strict)?.into_series(),
DataType::Float64 => any_values_to_f64(av, strict)?.into_series(),
DataType::String => any_values_to_string(av, strict)?.into_series(),
DataType::Binary => any_values_to_binary(av, strict)?.into_series(),
DataType::Boolean => any_values_to_bool(av, strict)?.into_series(),
#[cfg(feature = "dtype-date")]
DataType::Date => any_values_to_primitive::<Int32Type>(av)
DataType::Date => any_values_to_primitive_nonstrict::<Int32Type>(av)
.into_date()
.into_series(),
#[cfg(feature = "dtype-datetime")]
DataType::Datetime(tu, tz) => any_values_to_primitive::<Int64Type>(av)
DataType::Datetime(tu, tz) => any_values_to_primitive_nonstrict::<Int64Type>(av)
.into_datetime(*tu, (*tz).clone())
.into_series(),
#[cfg(feature = "dtype-time")]
DataType::Time => any_values_to_primitive::<Int64Type>(av)
DataType::Time => any_values_to_primitive_nonstrict::<Int64Type>(av)
.into_time()
.into_series(),
#[cfg(feature = "dtype-duration")]
DataType::Duration(tu) => any_values_to_primitive::<Int64Type>(av)
DataType::Duration(tu) => any_values_to_primitive_nonstrict::<Int64Type>(av)
.into_duration(*tu)
.into_series(),
#[cfg(feature = "dtype-decimal")]
Expand Down Expand Up @@ -460,6 +454,82 @@ impl Series {
}
}

fn any_values_to_primitive_nonstrict<T: PolarsNumericType>(values: &[AnyValue]) -> ChunkedArray<T> {
values
.iter()
.map(|av| av.extract::<T::Native>())
.collect_trusted()
}

fn any_values_to_integer<T: PolarsIntegerType>(
values: &[AnyValue],
strict: bool,
) -> PolarsResult<ChunkedArray<T>> {
fn any_values_to_integer_strict<T: PolarsIntegerType>(
values: &[AnyValue],
) -> PolarsResult<ChunkedArray<T>> {
let mut builder = PrimitiveChunkedBuilder::<T>::new("", values.len());
for av in values {
match av {
av if av.is_integer() => {
let opt_val = av.extract::<T::Native>();
let val = match opt_val {
Some(v) => v,
None => return Err(invalid_value_error(&T::get_dtype(), av)),
};
builder.append_value(val)
},
AnyValue::Null => builder.append_null(),
av => return Err(invalid_value_error(&T::get_dtype(), av)),
}
}
Ok(builder.finish())
}
if strict {
any_values_to_integer_strict::<T>(values)
} else {
Ok(any_values_to_primitive_nonstrict::<T>(values))
}
}

fn any_values_to_f32(values: &[AnyValue], strict: bool) -> PolarsResult<Float32Chunked> {
fn any_values_to_f32_strict(values: &[AnyValue]) -> PolarsResult<Float32Chunked> {
let mut builder = PrimitiveChunkedBuilder::<Float32Type>::new("", values.len());
for av in values {
match av {
AnyValue::Float32(i) => builder.append_value(*i),
AnyValue::Null => builder.append_null(),
av => return Err(invalid_value_error(&DataType::Float32, av)),
}
}
Ok(builder.finish())
}
if strict {
any_values_to_f32_strict(values)
} else {
Ok(any_values_to_primitive_nonstrict::<Float32Type>(values))
}
}
fn any_values_to_f64(values: &[AnyValue], strict: bool) -> PolarsResult<Float64Chunked> {
fn any_values_to_f64_strict(values: &[AnyValue]) -> PolarsResult<Float64Chunked> {
let mut builder = PrimitiveChunkedBuilder::<Float64Type>::new("", values.len());
for av in values {
match av {
AnyValue::Float64(i) => builder.append_value(*i),
AnyValue::Float32(i) => builder.append_value(*i as f64),
AnyValue::Null => builder.append_null(),
av => return Err(invalid_value_error(&DataType::Float64, av)),
}
}
Ok(builder.finish())
}
if strict {
any_values_to_f64_strict(values)
} else {
Ok(any_values_to_primitive_nonstrict::<Float64Type>(values))
}
}

fn any_values_to_bool(values: &[AnyValue], strict: bool) -> PolarsResult<BooleanChunked> {
if strict {
any_values_to_bool_strict(values)
Expand Down
20 changes: 17 additions & 3 deletions py-polars/tests/unit/constructors/test_any_value_fallbacks.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -15,6 +15,8 @@
@pytest.mark.parametrize(
("dtype", "values"),
[
(pl.Int64, [-1, 0, 100_000, None]),
(pl.Float64, [-1.5, 0.0, 10.0, None]),
(pl.Boolean, [True, False, None]),
(pl.Binary, [b"123", b"xyz", None]),
(pl.String, ["123", "xyz", None]),
Expand All @@ -32,6 +34,8 @@ def test_fallback_with_dtype_strict(
@pytest.mark.parametrize(
("dtype", "values"),
[
(pl.Int64, [1.0, 2.0]),
(pl.Float64, [1, 2]),
(pl.Boolean, [0, 1]),
(pl.Binary, ["123", "xyz"]),
(pl.String, [b"123", b"xyz"]),
Expand All @@ -41,16 +45,26 @@ def test_fallback_with_dtype_strict_failure(
dtype: pl.PolarsDataType, values: list[Any]
) -> None:
with pytest.raises(pl.SchemaError, match="unexpected value"):
PySeries.new_from_any_values_and_dtype("", values, pl.Boolean, strict=True)
PySeries.new_from_any_values_and_dtype("", values, dtype, strict=True)


@pytest.mark.parametrize(
("dtype", "values", "expected"),
[
(
pl.Int64,
[False, True, 0, -1, 0.0, 2.5, date(1970, 1, 2), "5", "xyz"],
[0, 1, 0, -1, 0, 2, 1, 5, None],
),
(
pl.Float64,
[False, True, 0, -1, 0.0, 2.5, date(1970, 1, 2), "5", "xyz"],
[0.0, 1.0, 0.0, -1.0, 0.0, 2.5, 1.0, 5.0, None],
),
(
pl.Boolean,
[False, True, 0, 1, 0.0, 2.5, date(1970, 1, 1)],
[False, True, False, True, False, True, None],
[False, True, 0, -1, 0.0, 2.5, date(1970, 1, 1), "true"],
[False, True, False, True, False, True, None, None],
),
(
pl.Binary,
Expand Down
4 changes: 2 additions & 2 deletions py-polars/tests/unit/testing/test_assert_frame_equal.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -250,8 +250,8 @@ def test_compare_frame_equal_nested_nans() -> None:
{
"id": 2,
"struct": [
{"x": "text", "y": [nan, 1], "z": ["!"]},
{"x": "text", "y": [nan, 1], "z": ["?"]},
{"x": "text", "y": [nan, 1.0], "z": ["!"]},
{"x": "text", "y": [nan, 1.0], "z": ["?"]},
],
},
]
Expand Down
Loading