feat: Add percentile_cont aggregate function (apache#17988)

## Summary

Adds exact `percentile_cont` aggregate function as the counterpart to
the existing `approx_percentile_cont` function.

## What changes were made?

### New Implementation
- Created `percentile_cont.rs` with full implementation
- `PercentileCont` struct implementing `AggregateUDFImpl`
- `PercentileContAccumulator` for standard aggregation
- `DistinctPercentileContAccumulator` for DISTINCT mode
- `PercentileContGroupsAccumulator` for efficient grouped aggregation
- `calculate_percentile` function with linear interpolation

### Features
- **Exact calculation**: Stores all values in memory for precise results
- **WITHIN GROUP syntax**: Supports `WITHIN GROUP (ORDER BY ...)` 
- **Interpolation**: Uses linear interpolation between values
- **All numeric types**: Works with integers, floats, and decimals
- **Ordered-set aggregate**: Properly marked as
`is_ordered_set_aggregate()`
- **GROUP BY support**: Efficient grouped aggregation via
GroupsAccumulator

### Tests
Added comprehensive tests in `aggregate.slt`:
- Error conditions validation
- Basic percentile calculations (0.0, 0.25, 0.5, 0.75, 1.0)
- Comparison with `median` function
- Ascending and descending order
- GROUP BY aggregation
- NULL handling
- Edge cases (empty sets, single values)
- Float interpolation
- Various numeric data types

## Example Usage

```sql
-- Basic usage with WITHIN GROUP syntax
SELECT percentile_cont(0.75) WITHIN GROUP (ORDER BY column_name) 
FROM table_name;

-- With GROUP BY
SELECT category, percentile_cont(0.95) WITHIN GROUP (ORDER BY value)
FROM sales
GROUP BY category;

-- Compare with median (percentile_cont(0.5) == median)
SELECT percentile_cont(0.5) WITHIN GROUP (ORDER BY price) FROM products;
```

## Performance Considerations

Like `median`, this function stores all values in memory before
computing results. For large datasets or when approximation is
acceptable, use `approx_percentile_cont` instead.

## Related Issues

Closes apache#6714

🤖 Generated with [Claude Code](https://claude.com/claude-code)

---------

Co-authored-by: Claude <[email protected]>

Author: 2 people

datafusion-testing

@@ -20,34 +20,36 @@ use std::fmt::{Debug, Formatter};
 use std::mem::size_of_val;
 use std::sync::Arc;
 
-use arrow::array::{Array, RecordBatch};
+use arrow::array::Array;
 use arrow::compute::{filter, is_not_null};
 use arrow::datatypes::FieldRef;
 use arrow::{
     array::{
         ArrayRef, Float32Array, Float64Array, Int16Array, Int32Array, Int64Array,
         Int8Array, UInt16Array, UInt32Array, UInt64Array, UInt8Array,
     },
-    datatypes::{DataType, Field, Schema},
+    datatypes::{DataType, Field},
 };
 use datafusion_common::{
-    downcast_value, internal_err, not_impl_datafusion_err, not_impl_err, plan_err,
-    Result, ScalarValue,
+    downcast_value, internal_err, not_impl_err, plan_err, DataFusionError, Result,
+    ScalarValue,
 };
 use datafusion_expr::expr::{AggregateFunction, Sort};
 use datafusion_expr::function::{AccumulatorArgs, StateFieldsArgs};
 use datafusion_expr::type_coercion::aggregates::{INTEGERS, NUMERICS};
 use datafusion_expr::utils::format_state_name;
 use datafusion_expr::{
-    Accumulator, AggregateUDFImpl, ColumnarValue, Documentation, Expr, Signature,
-    TypeSignature, Volatility,
+    Accumulator, AggregateUDFImpl, Documentation, Expr, Signature, TypeSignature,
+    Volatility,
 };
 use datafusion_functions_aggregate_common::tdigest::{
     TDigest, TryIntoF64, DEFAULT_MAX_SIZE,
 };
 use datafusion_macros::user_doc;
 use datafusion_physical_expr_common::physical_expr::PhysicalExpr;
 
+use crate::utils::{get_scalar_value, validate_percentile_expr};
+
 create_func!(ApproxPercentileCont, approx_percentile_cont_udaf);
 
 /// Computes the approximate percentile continuous of a set of numbers
@@ -164,7 +166,8 @@ impl ApproxPercentileCont {
         &self,
         args: AccumulatorArgs,
     ) -> Result<ApproxPercentileAccumulator> {
-        let percentile = validate_input_percentile_expr(&args.exprs[1])?;
+        let percentile =
+            validate_percentile_expr(&args.exprs[1], "APPROX_PERCENTILE_CONT")?;
 
         let is_descending = args
             .order_bys
@@ -214,45 +217,15 @@ impl ApproxPercentileCont {
     }
 }
 
-fn get_scalar_value(expr: &Arc<dyn PhysicalExpr>) -> Result<ScalarValue> {
-    let empty_schema = Arc::new(Schema::empty());
-    let batch = RecordBatch::new_empty(Arc::clone(&empty_schema));
-    if let ColumnarValue::Scalar(s) = expr.evaluate(&batch)? {
-        Ok(s)
-    } else {
-        internal_err!("Didn't expect ColumnarValue::Array")
-    }
-}
-
-fn validate_input_percentile_expr(expr: &Arc<dyn PhysicalExpr>) -> Result<f64> {
-    let percentile = match get_scalar_value(expr)
-        .map_err(|_| not_impl_datafusion_err!("Percentile value for 'APPROX_PERCENTILE_CONT' must be a literal, got: {expr}"))? {
-        ScalarValue::Float32(Some(value)) => {
-            value as f64
-        }
-        ScalarValue::Float64(Some(value)) => {
-            value
-        }
-        sv => {
-            return not_impl_err!(
-                "Percentile value for 'APPROX_PERCENTILE_CONT' must be Float32 or Float64 literal (got data type {})",
-                sv.data_type()
-            )
-        }
-    };
-
-    // Ensure the percentile is between 0 and 1.
-    if !(0.0..=1.0).contains(&percentile) {
-        return plan_err!(
-            "Percentile value must be between 0.0 and 1.0 inclusive, {percentile} is invalid"
-        );
-    }
-    Ok(percentile)
-}
-
 fn validate_input_max_size_expr(expr: &Arc<dyn PhysicalExpr>) -> Result<usize> {
-    let max_size = match get_scalar_value(expr)
-        .map_err(|_| not_impl_datafusion_err!("Tdigest max_size value for 'APPROX_PERCENTILE_CONT' must be a literal, got: {expr}"))? {
+    let scalar_value = get_scalar_value(expr).map_err(|_e| {
+        DataFusionError::Plan(
+            "Tdigest max_size value for 'APPROX_PERCENTILE_CONT' must be a literal"
+                .to_string(),
+        )
+    })?;
+
+    let max_size = match scalar_value {
         ScalarValue::UInt8(Some(q)) => q as usize,
         ScalarValue::UInt16(Some(q)) => q as usize,
         ScalarValue::UInt32(Some(q)) => q as usize,
@@ -262,7 +235,7 @@ fn validate_input_max_size_expr(expr: &Arc<dyn PhysicalExpr>) -> Result<usize> {
         ScalarValue::Int16(Some(q)) if q > 0 => q as usize,
         ScalarValue::Int8(Some(q)) if q > 0 => q as usize,
         sv => {
-            return not_impl_err!(
+            return plan_err!(
                 "Tdigest max_size value for 'APPROX_PERCENTILE_CONT' must be UInt > 0 literal (got data type {}).",
                 sv.data_type()
             )

datafusion/functions-aggregate/src/approx_percentile_cont.rs

@@ -81,13 +81,15 @@ pub mod hyperloglog;
 pub mod median;
 pub mod min_max;
 pub mod nth_value;
+pub mod percentile_cont;
 pub mod regr;
 pub mod stddev;
 pub mod string_agg;
 pub mod sum;
 pub mod variance;
 
 pub mod planner;
+mod utils;
 
 use crate::approx_percentile_cont::approx_percentile_cont_udaf;
 use crate::approx_percentile_cont_with_weight::approx_percentile_cont_with_weight_udaf;
@@ -123,6 +125,7 @@ pub mod expr_fn {
     pub use super::min_max::max;
     pub use super::min_max::min;
     pub use super::nth_value::nth_value;
+    pub use super::percentile_cont::percentile_cont;
     pub use super::regr::regr_avgx;
     pub use super::regr::regr_avgy;
     pub use super::regr::regr_count;
@@ -171,6 +174,7 @@ pub fn all_default_aggregate_functions() -> Vec<Arc<AggregateUDF>> {
         approx_distinct::approx_distinct_udaf(),
         approx_percentile_cont_udaf(),
         approx_percentile_cont_with_weight_udaf(),
+        percentile_cont::percentile_cont_udaf(),
         string_agg::string_agg_udaf(),
         bit_and_or_xor::bit_and_udaf(),
         bit_and_or_xor::bit_or_udaf(),