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.

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

feat(rust, python): Decimal arithmetic #9123

Merged
merged 3 commits into from
May 31, 2023
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
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
16 changes: 16 additions & 0 deletions polars/polars-arrow/src/compute/arithmetics/decimal/add.rs
Original file line number Diff line number Diff line change
@@ -0,0 +1,16 @@
use super::*;

pub fn add(
lhs: &PrimitiveArray<i128>,
rhs: &PrimitiveArray<i128>,
) -> PolarsResult<PrimitiveArray<i128>> {
commutative(lhs, rhs, |a, b| a + b)
}

pub fn add_scalar(
lhs: &PrimitiveArray<i128>,
rhs: i128,
rhs_dtype: &DataType,
) -> PolarsResult<PrimitiveArray<i128>> {
commutative_scalar(lhs, rhs, rhs_dtype, |a, b| a + b)
}
Original file line number Diff line number Diff line change
@@ -0,0 +1,89 @@
use arrow::array::PrimitiveArray;
use arrow::datatypes::DataType;
use polars_error::*;

use super::{get_parameters, max_value};
use crate::compute::{binary_mut, unary_mut};

pub fn commutative<F>(
lhs: &PrimitiveArray<i128>,
rhs: &PrimitiveArray<i128>,
op: F,
) -> PolarsResult<PrimitiveArray<i128>>
where
F: Fn(i128, i128) -> i128,
{
let (precision, _) = get_parameters(lhs.data_type(), rhs.data_type()).unwrap();

let max = max_value(precision);
let mut overflow = false;
let op = |a, b| {
let res = op(a, b);
overflow |= res.abs() > max;
res
};
let out = binary_mut(lhs, rhs, lhs.data_type().clone(), op);
polars_ensure!(!overflow, ComputeError: "Decimal overflowed the allowed precision: {precision}");
Ok(out)
}

pub fn commutative_scalar<F>(
lhs: &PrimitiveArray<i128>,
rhs: i128,
rhs_dtype: &DataType,
op: F,
) -> PolarsResult<PrimitiveArray<i128>>
where
F: Fn(i128, i128) -> i128,
{
let (precision, _) = get_parameters(lhs.data_type(), rhs_dtype).unwrap();

let max = max_value(precision);
let mut overflow = false;
let op = |a| {
let res = op(a, rhs);
overflow |= res.abs() > max;
res
};
let out = unary_mut(lhs, op, lhs.data_type().clone());
polars_ensure!(!overflow, ComputeError: "Decimal overflowed the allowed precision: {precision}");

Ok(out)
}

pub fn non_commutative<F>(
lhs: &PrimitiveArray<i128>,
rhs: &PrimitiveArray<i128>,
op: F,
) -> PolarsResult<PrimitiveArray<i128>>
where
F: Fn(i128, i128) -> i128,
{
Ok(binary_mut(lhs, rhs, lhs.data_type().clone(), op))
}

pub fn non_commutative_scalar<F>(
lhs: &PrimitiveArray<i128>,
rhs: i128,
op: F,
) -> PolarsResult<PrimitiveArray<i128>>
where
F: Fn(i128, i128) -> i128,
{
let op = move |a| op(a, rhs);

Ok(unary_mut(lhs, op, lhs.data_type().clone()))
}

pub fn non_commutative_scalar_swapped<F>(
lhs: i128,
rhs: &PrimitiveArray<i128>,
op: F,
) -> PolarsResult<PrimitiveArray<i128>>
where
F: Fn(i128, i128) -> i128,
{
let op = move |a| op(lhs, a);

Ok(unary_mut(rhs, op, rhs.data_type().clone()))
}
43 changes: 43 additions & 0 deletions polars/polars-arrow/src/compute/arithmetics/decimal/div.rs
Original file line number Diff line number Diff line change
@@ -0,0 +1,43 @@
use super::*;

#[inline]
fn decimal_div(a: i128, b: i128, scale: i128) -> i128 {
// The division is done using the numbers without scale.
// The dividend is scaled up to maintain precision after the
// division
Comment on lines +5 to +7
Copy link
Contributor

Choose a reason for hiding this comment

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

The a * scale will overflow very often if dealing with high-precision decimals. Basically what you're doing here implies that using scale greater than MAX_SCALE / 2 is not possible and will lead to overflow errors, I think.

Usually it's done in the 256-bit space (IIRC that's how arrow2 does it too).


// 222.222 --> 222222000
// 123.456 --> 123456
// -------- ---------
// 1.800 <-- 1800
a * scale / b
}

pub fn div(
lhs: &PrimitiveArray<i128>,
rhs: &PrimitiveArray<i128>,
) -> PolarsResult<PrimitiveArray<i128>> {
let (_, scale) = get_parameters(lhs.data_type(), rhs.data_type())?;
let scale = 10i128.pow(scale as u32);
non_commutative(lhs, rhs, |a, b| decimal_div(a, b, scale))
}

pub fn div_scalar(
lhs: &PrimitiveArray<i128>,
rhs: i128,
rhs_dtype: &DataType,
) -> PolarsResult<PrimitiveArray<i128>> {
let (_, scale) = get_parameters(lhs.data_type(), rhs_dtype)?;
let scale = 10i128.pow(scale as u32);
non_commutative_scalar(lhs, rhs, |a, b| decimal_div(a, b, scale))
}

pub fn div_scalar_swapped(
lhs: i128,
lhs_dtype: &DataType,
rhs: &PrimitiveArray<i128>,
) -> PolarsResult<PrimitiveArray<i128>> {
let (_, scale) = get_parameters(lhs_dtype, rhs.data_type())?;
let scale = 10i128.pow(scale as u32);
non_commutative_scalar_swapped(lhs, rhs, |a, b| decimal_div(a, b, scale))
}
40 changes: 40 additions & 0 deletions polars/polars-arrow/src/compute/arithmetics/decimal/mod.rs
Original file line number Diff line number Diff line change
@@ -0,0 +1,40 @@
use arrow::array::PrimitiveArray;
use arrow::datatypes::DataType;
use commutative::{
commutative, commutative_scalar, non_commutative, non_commutative_scalar,
non_commutative_scalar_swapped,
};
use polars_error::{PolarsError, PolarsResult};

mod add;
mod commutative;
mod div;
mod mul;
mod sub;

pub use add::*;
pub use div::*;
pub use mul::*;
pub use sub::*;

/// Maximum value that can exist with a selected precision
#[inline]
fn max_value(precision: usize) -> i128 {
10i128.pow(precision as u32) - 1
}

fn get_parameters(lhs: &DataType, rhs: &DataType) -> PolarsResult<(usize, usize)> {
if let (DataType::Decimal(lhs_p, lhs_s), DataType::Decimal(rhs_p, rhs_s)) =
(lhs.to_logical_type(), rhs.to_logical_type())
{
if lhs_p == rhs_p && lhs_s == rhs_s {
Ok((*lhs_p, *lhs_s))
} else {
Err(PolarsError::InvalidOperation(
"Arrays must have the same precision and scale".into(),
))
}
} else {
unreachable!()
}
}
33 changes: 33 additions & 0 deletions polars/polars-arrow/src/compute/arithmetics/decimal/mul.rs
Original file line number Diff line number Diff line change
@@ -0,0 +1,33 @@
use super::*;

#[inline]
fn decimal_mul(a: i128, b: i128, scale: i128) -> i128 {
// The multiplication is done using the numbers without scale.
// The resulting scale of the value has to be corrected by
// dividing by (10^scale)

// 111.111 --> 111111
// 222.222 --> 222222
// -------- -------
// 24691.308 <-- 24691308642
a * b / scale
Copy link
Contributor

Choose a reason for hiding this comment

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

Same here, a * b will overflow too often. This is usually done with i256.

}

pub fn mul(
lhs: &PrimitiveArray<i128>,
rhs: &PrimitiveArray<i128>,
) -> PolarsResult<PrimitiveArray<i128>> {
let (_, scale) = get_parameters(lhs.data_type(), rhs.data_type())?;
let scale = 10i128.pow(scale as u32);
commutative(lhs, rhs, |a, b| decimal_mul(a, b, scale))
}

pub fn mul_scalar(
lhs: &PrimitiveArray<i128>,
rhs: i128,
rhs_dtype: &DataType,
) -> PolarsResult<PrimitiveArray<i128>> {
let (_, scale) = get_parameters(lhs.data_type(), rhs_dtype)?;
let scale = 10i128.pow(scale as u32);
commutative_scalar(lhs, rhs, rhs_dtype, |a, b| decimal_mul(a, b, scale))
}
19 changes: 19 additions & 0 deletions polars/polars-arrow/src/compute/arithmetics/decimal/sub.rs
Original file line number Diff line number Diff line change
@@ -0,0 +1,19 @@
use super::*;

pub fn sub(
lhs: &PrimitiveArray<i128>,
rhs: &PrimitiveArray<i128>,
) -> PolarsResult<PrimitiveArray<i128>> {
non_commutative(lhs, rhs, |a, b| a - b)
}

pub fn sub_scalar(lhs: &PrimitiveArray<i128>, rhs: i128) -> PolarsResult<PrimitiveArray<i128>> {
non_commutative_scalar(lhs, rhs, |a, b| a - b)
}

pub fn sub_scalar_swapped(
lhs: i128,
rhs: &PrimitiveArray<i128>,
) -> PolarsResult<PrimitiveArray<i128>> {
non_commutative_scalar_swapped(lhs, rhs, |a, b| a - b)
}
2 changes: 2 additions & 0 deletions polars/polars-arrow/src/compute/arithmetics/mod.rs
Original file line number Diff line number Diff line change
@@ -0,0 +1,2 @@
#[cfg(feature = "dtype-decimal")]
pub mod decimal;
1 change: 1 addition & 0 deletions polars/polars-arrow/src/compute/arity.rs
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@

50 changes: 50 additions & 0 deletions polars/polars-arrow/src/compute/mod.rs
Original file line number Diff line number Diff line change
@@ -1,7 +1,57 @@
use arrow::array::PrimitiveArray;
use arrow::datatypes::DataType;
use arrow::types::NativeType;

use crate::utils::combine_validities_and;

pub mod arithmetics;
pub mod arity;
pub mod bitwise;
#[cfg(feature = "compute")]
pub mod cast;
#[cfg(feature = "dtype-decimal")]
pub mod decimal;
pub mod take;
pub mod tile;

#[inline]
pub fn binary_mut<T, D, F>(
lhs: &PrimitiveArray<T>,
rhs: &PrimitiveArray<D>,
data_type: DataType,
mut op: F,
) -> PrimitiveArray<T>
where
T: NativeType,
D: NativeType,
F: FnMut(T, D) -> T,
{
assert_eq!(lhs.len(), rhs.len());
let validity = combine_validities_and(lhs.validity(), rhs.validity());

let values = lhs
.values()
.iter()
.zip(rhs.values().iter())
.map(|(l, r)| op(*l, *r))
.collect::<Vec<_>>()
.into();

PrimitiveArray::<T>::new(data_type, values, validity)
}

#[inline]
pub fn unary_mut<I, F, O>(
array: &PrimitiveArray<I>,
mut op: F,
data_type: DataType,
) -> PrimitiveArray<O>
where
I: NativeType,
O: NativeType,
F: FnMut(I) -> O,
{
let values = array.values().iter().map(|v| op(*v)).collect::<Vec<_>>();

PrimitiveArray::<O>::new(data_type, values.into(), array.validity().cloned())
}
Loading