migrate approx_percentile_cont, approx_distinct, and approx_median to UDAF

Ref: approx_distinct apache/datafusion#10851
Ref: approx_median apache/datafusion#10840
Ref: approx_percentile_cont and _with_weight apache/datafusion#10917

Author: Michael-J-Ward

python/datafusion/functions.py

@@ -1233,9 +1233,9 @@ def flatten(array: Expr) -> Expr:
 
 
 # aggregate functions
-def approx_distinct(arg: Expr) -> Expr:
+def approx_distinct(expression: Expr) -> Expr:
     """Returns the approximate number of distinct values."""
-    return Expr(f.approx_distinct(arg.expr, distinct=True))
+    return Expr(f.approx_distinct(expression.expr))
 
 
 def approx_median(arg: Expr, distinct: bool = False) -> Expr:
@@ -1244,20 +1244,22 @@ def approx_median(arg: Expr, distinct: bool = False) -> Expr:
 
 
 def approx_percentile_cont(
-    expr: Expr,
+    expression: Expr,
     percentile: Expr,
-    num_centroids: int | None = None,
+    # num_centroids: int | None = None,
     distinct: bool = False,
 ) -> Expr:
     """Returns the value that is approximately at a given percentile of ``expr``."""
+    # TODO: enable num_centroids
+    num_centroids = None
     if num_centroids is None:
         return Expr(
             f.approx_percentile_cont(expr.expr, percentile.expr, distinct=distinct)
         )
 
     return Expr(
         f.approx_percentile_cont(
-            expr.expr, percentile.expr, num_centroids, distinct=distinct
+            expr.expr, percentile.expr, distinct=distinct
         )
     )
 

src/functions.rs

@@ -37,6 +37,57 @@ use datafusion_expr::{
     lit, Expr, WindowFunctionDefinition,
 };
 
+#[pyfunction]
+pub fn approx_distinct(expression: PyExpr) -> PyExpr {
+    functions_aggregate::expr_fn::approx_distinct::approx_distinct(expression.expr).into()
+}
+
+#[pyfunction]
+pub fn approx_median(expression: PyExpr, distinct: bool) -> PyResult<PyExpr> {
+    // TODO: better builder pattern
+    let expr = functions_aggregate::expr_fn::approx_median(expression.expr);
+    if distinct {
+        Ok(expr.distinct().build()?.into())
+    } else {
+        Ok(expr.into())
+    }
+}
+
+#[pyfunction]
+pub fn approx_percentile_cont(
+    expression: PyExpr,
+    percentile: PyExpr,
+    distinct: bool,
+) -> PyResult<PyExpr> {
+    // TODO: better builder pattern
+    let expr =
+        functions_aggregate::expr_fn::approx_percentile_cont(expression.expr, percentile.expr);
+    if distinct {
+        Ok(expr.distinct().build()?.into())
+    } else {
+        Ok(expr.into())
+    }
+}
+
+#[pyfunction]
+pub fn approx_percentile_cont_with_weight(
+    expression: PyExpr,
+    weight: PyExpr,
+    percentile: PyExpr,
+    distinct: bool,
+) -> PyResult<PyExpr> {
+    let expr = functions_aggregate::expr_fn::approx_percentile_cont_with_weight(
+        expression.expr,
+        weight.expr,
+        percentile.expr,
+    );
+    if distinct {
+        Ok(expr.distinct().build()?.into())
+    } else {
+        Ok(expr.into())
+    }
+}
+
 #[pyfunction]
 pub fn sum(args: PyExpr) -> PyExpr {
     functions_aggregate::expr_fn::sum(args.expr).into()
@@ -727,13 +778,6 @@ array_fn!(list_resize, array_resize, array size value);
 array_fn!(flatten, array);
 array_fn!(range, start stop step);
 
-aggregate_function!(approx_distinct, ApproxDistinct);
-aggregate_function!(approx_median, ApproxMedian);
-aggregate_function!(approx_percentile_cont, ApproxPercentileCont);
-aggregate_function!(
-    approx_percentile_cont_with_weight,
-    ApproxPercentileContWithWeight
-);
 aggregate_function!(array_agg, ArrayAgg);
 aggregate_function!(avg, Avg);
 aggregate_function!(corr, Correlation);