klion26 commented on code in PR #10157:
URL: https://github.com/apache/arrow-rs/pull/10157#discussion_r3474142064
##########
parquet-variant-compute/src/type_conversion.rs:
##########
@@ -76,58 +83,201 @@ macro_rules! impl_primitive_from_variant {
}
macro_rules! impl_timestamp_from_variant {
- ($timestamp_type:ty, $variant_method:ident, ntz=$ntz:ident, $cast_fn:expr
$(,)?) => {
+ ($timestamp_type:ty, $shred_fun:expr, $variant_method:expr,
ntz=$ntz:ident, $cast_fn:expr $(,)?) => {
impl TimestampFromVariant<{ $ntz }> for $timestamp_type {
- fn from_variant(variant: &Variant<'_, '_>) -> Option<Self::Native>
{
- variant.$variant_method().and_then($cast_fn)
+ fn from_variant(variant: &Variant<'_, '_>, shred: bool) ->
Option<Self::Native> {
+ let value = match shred {
+ true => ($shred_fun)(variant),
+ false => $variant_method(variant),
+ };
+
+ value.and_then($cast_fn)
}
}
};
}
-impl_primitive_from_variant!(datatypes::Int32Type, as_int32);
-impl_primitive_from_variant!(datatypes::Int16Type, as_int16);
-impl_primitive_from_variant!(datatypes::Int8Type, as_int8);
-impl_primitive_from_variant!(datatypes::Int64Type, as_int64);
-impl_primitive_from_variant!(datatypes::UInt8Type, as_u8);
-impl_primitive_from_variant!(datatypes::UInt16Type, as_u16);
-impl_primitive_from_variant!(datatypes::UInt32Type, as_u32);
-impl_primitive_from_variant!(datatypes::UInt64Type, as_u64);
-impl_primitive_from_variant!(datatypes::Float16Type, as_f16);
-impl_primitive_from_variant!(datatypes::Float32Type, as_f32);
-impl_primitive_from_variant!(datatypes::Float64Type, as_f64);
-impl_primitive_from_variant!(datatypes::Date32Type, as_naive_date, |v| {
- Some(datatypes::Date32Type::from_naive_date(v))
-});
-impl_primitive_from_variant!(datatypes::Date64Type, as_naive_date, |v| {
- Some(datatypes::Date64Type::from_naive_date(v))
-});
-impl_primitive_from_variant!(datatypes::Time32SecondType, as_time_utc, |v| {
- // Return None if there are leftover nanoseconds
- if v.nanosecond() != 0 {
- None
- } else {
- Some(v.num_seconds_from_midnight() as i32)
+fn convert_to_timestamp_nano(value: &Variant) -> Option<DateTime<Utc>> {
+ match *value {
+ Variant::TimestampNanos(d) | Variant::TimestampMicros(d) => Some(d),
+ _ => None,
}
-});
-impl_primitive_from_variant!(datatypes::Time32MillisecondType, as_time_utc,
|v| {
- // Return None if there are leftover microseconds
- if v.nanosecond() % 1_000_000 != 0 {
- None
- } else {
- Some((v.num_seconds_from_midnight() * 1_000) as i32 + (v.nanosecond()
/ 1_000_000) as i32)
+}
+
+fn convert_to_timestamp_ntz_nano(value: &Variant) -> Option<NaiveDateTime> {
+ match *value {
+ Variant::TimestampNtzNanos(d) | Variant::TimestampNtzMicros(d) =>
Some(d),
+ _ => None,
+ }
+}
+
+enum NumericKind {
+ Integer,
+ Float,
+}
+
+trait DecimalCastTarget: NumCast + Default {
+ const KIND: NumericKind;
+}
+
+macro_rules! impl_decimal_cast_target {
+ ($raw_type: ident, $target_kind:expr) => {
+ impl DecimalCastTarget for $raw_type {
+ const KIND: NumericKind = $target_kind;
+ }
+ };
+}
+
+impl_decimal_cast_target!(i8, NumericKind::Integer);
+impl_decimal_cast_target!(i16, NumericKind::Integer);
+impl_decimal_cast_target!(i32, NumericKind::Integer);
+impl_decimal_cast_target!(i64, NumericKind::Integer);
+impl_decimal_cast_target!(u8, NumericKind::Integer);
+impl_decimal_cast_target!(u16, NumericKind::Integer);
+impl_decimal_cast_target!(u32, NumericKind::Integer);
+impl_decimal_cast_target!(u64, NumericKind::Integer);
+impl_decimal_cast_target!(f16, NumericKind::Float);
+impl_decimal_cast_target!(f32, NumericKind::Float);
+impl_decimal_cast_target!(f64, NumericKind::Float);
+
+/// Converts a boolean or numeric variant(integers, floating-point, and
decimals)
+/// to the specified numeric type `T`.
+///
+/// Uses Arrow's casting logic to perform the conversion. Returns `Some(T)` if
+/// the conversion succeeds, `None` if the variant can't be casted to type `T`.
+fn as_num<T>(variant: &Variant) -> Option<T>
+where
+ T: DecimalCastTarget,
+{
+ match *variant {
+ Variant::BooleanFalse => single_bool_to_numeric(false),
+ Variant::BooleanTrue => single_bool_to_numeric(true),
+ Variant::Int8(i) => num_cast(i),
+ Variant::Int16(i) => num_cast(i),
+ Variant::Int32(i) => num_cast(i),
+ Variant::Int64(i) => num_cast(i),
+ Variant::Float(f) => num_cast(f),
+ Variant::Double(d) => num_cast(d),
+ Variant::Decimal4(d) => {
+ cast_decimal_to_num::<Decimal32Type, T, _>(d.integer(), d.scale(),
|x| x as f64)
+ }
+ Variant::Decimal8(d) => {
+ cast_decimal_to_num::<Decimal64Type, T, _>(d.integer(), d.scale(),
|x| x as f64)
+ }
+ Variant::Decimal16(d) => {
+ cast_decimal_to_num::<Decimal128Type, T, _>(d.integer(),
d.scale(), |x| x as f64)
+ }
+ _ => None,
+ }
+}
+
+fn cast_decimal_to_num<D, T, F>(raw: D::Native, scale: u8, as_float: F) ->
Option<T>
+where
+ D: DecimalType,
+ D::Native: NumCast + ArrowNativeTypeOp,
+ T: DecimalCastTarget,
+ F: Fn(D::Native) -> f64,
+{
+ let base: D::Native = NumCast::from(10)?;
+
+ let div = base.pow_checked(<u32 as From<u8>>::from(scale)).ok()?;
+ match T::KIND {
+ NumericKind::Integer => raw
+ .div_checked(div)
+ .ok()
+ .and_then(<T as NumCast>::from::<D::Native>),
+ NumericKind::Float => T::from(single_decimal_to_float_lossy::<D, _>(
+ &as_float,
+ raw,
+ <i32 as From<u8>>::from(scale),
+ )),
+ }
+}
+
+fn cast_naive_date(value: &Variant<'_, '_>) -> Option<NaiveDate> {
+ value.as_naive_date()
+}
+
+fn cast_time_utc(value: &Variant<'_, '_>) -> Option<NaiveTime> {
+ value.as_time_utc()
+}
+
+// helper function for the types that would never be the shred target type.
+fn always_none<T>(_input: &Variant) -> Option<T> {
+ None
+}
+
+impl_primitive_from_variant!(datatypes::Int32Type, Variant::as_int32, as_num);
+impl_primitive_from_variant!(datatypes::Int16Type, Variant::as_int16, as_num);
+impl_primitive_from_variant!(datatypes::Int8Type, Variant::as_int8, as_num);
+impl_primitive_from_variant!(datatypes::Int64Type, Variant::as_int64, as_num);
+impl_primitive_from_variant!(datatypes::UInt8Type, always_none, as_num);
+impl_primitive_from_variant!(datatypes::UInt16Type, always_none, as_num);
+impl_primitive_from_variant!(datatypes::UInt32Type, always_none, as_num);
+impl_primitive_from_variant!(datatypes::UInt64Type, always_none, as_num);
+impl_primitive_from_variant!(datatypes::Float16Type, always_none, as_num);
+impl_primitive_from_variant!(datatypes::Float32Type, Variant::as_f32, as_num);
+impl_primitive_from_variant!(datatypes::Float64Type, Variant::as_f64, as_num);
+impl_primitive_from_variant!(
+ datatypes::Date32Type,
+ Variant::as_naive_date,
+ cast_naive_date,
+ |v| { Some(datatypes::Date32Type::from_naive_date(v)) }
+);
+impl_primitive_from_variant!(
+ datatypes::Date64Type,
+ Variant::as_naive_date,
+ cast_naive_date,
+ |v| { Some(datatypes::Date64Type::from_naive_date(v)) }
+);
+impl_primitive_from_variant!(
+ datatypes::Time32SecondType,
+ always_none, // would never shred to Time32SecondType
+ cast_time_utc,
+ |v| {
+ // Return None if there are leftover nanoseconds
+ if v.nanosecond() != 0 {
+ None
+ } else {
+ Some(v.num_seconds_from_midnight() as i32)
+ }
+ }
+);
+impl_primitive_from_variant!(
Review Comment:
Align the cast logic with arrow-cast for Time related variant will in a
followup pr
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
