From f45d1679364024091fabda93812814c3f830382f Mon Sep 17 00:00:00 2001
From: RIchard Baah <richiebbaah@Gmail.com>
Date: Thu, 27 Aug 2026 11:15:49 -0400
Subject: [PATCH 1/2] feat: support dictionary in approx_distinct

---
 .../src/approx_distinct.rs                    | 39 +++++++++++++++++--
 1 file changed, 36 insertions(+), 3 deletions(-)

diff --git a/datafusion/functions-aggregate/src/approx_distinct.rs b/datafusion/functions-aggregate/src/approx_distinct.rs
index 1746edd8239f2..03e2dbdf923f3 100644
--- a/datafusion/functions-aggregate/src/approx_distinct.rs
+++ b/datafusion/functions-aggregate/src/approx_distinct.rs
@@ -841,6 +841,7 @@ impl AggregateUDFImpl for ApproxDistinct {
             | DataType::Struct(_)
             | DataType::Union(_, _)
             | DataType::LargeBinary => Box::new(HLLAccumulator::new()),
+            DataType::Dictionary(_, _) => Box::new(HLLAccumulator::new()),
             DataType::Null => {
                 Box::new(NoopAccumulator::new(ScalarValue::UInt64(Some(0))))
             }
@@ -919,6 +920,7 @@ fn is_hll_groups_type(data_type: &DataType) -> bool {
             | DataType::Map(_, _)
             | DataType::Struct(_)
             | DataType::Union(_, _)
+            | DataType::Dictionary(_, _)
     )
 }
 
@@ -932,10 +934,10 @@ mod tests {
         use super::*;
         use arrow::array::{
             AsArray, Decimal32Array, Decimal64Array, Decimal128Array, Decimal256Array,
-            Int64Array, IntervalDayTimeArray, IntervalMonthDayNanoArray,
-            IntervalYearMonthArray, StringViewArray,
+            DictionaryArray, Int32Array, Int64Array, IntervalDayTimeArray,
+            IntervalMonthDayNanoArray, IntervalYearMonthArray, StringArray, StringViewArray,
         };
-        use arrow::datatypes::{IntervalDayTime, IntervalMonthDayNano, i256};
+        use arrow::datatypes::{Int32Type, IntervalDayTime, IntervalMonthDayNano, i256};
         use std::sync::Arc;
         // A string longer than the 12-byte inline limit
         const LONG: &str = "this string is definitely longer than twelve bytes";
@@ -1248,6 +1250,37 @@ mod tests {
             assert_eq!(counts.value(0), 3);
         }
 
+        #[test]
+        fn dictionary_int32_utf8_per_row_and_groups_acc() {
+            // {"a", "b", "b", "c", "c", "c"} — 3 distinct logical values
+            let keys = Int32Array::from(vec![0, 1, 1, 2, 2, 2]);
+            let values = StringArray::from(vec!["a", "b", "c"]);
+            let array: ArrayRef = Arc::new(
+                DictionaryArray::<Int32Type>::try_new(keys, Arc::new(values)).unwrap(),
+            );
+
+            assert!(is_hll_groups_type(array.data_type()));
+
+            let mut per_row = HLLAccumulator::new();
+            per_row.update_batch(&[Arc::clone(&array)]).unwrap();
+            assert_eq!(distinct_count(&mut per_row), 3);
+
+            let group_indices = vec![0usize; array.len()];
+            let mut groups = HllGroupsAccumulator::new();
+            groups
+                .update_batch(&[Arc::clone(&array)], &group_indices, None, 1)
+                .unwrap();
+            let result = groups.evaluate(EmitTo::All).unwrap();
+            assert_eq!(
+                result
+                    .as_any()
+                    .downcast_ref::<UInt64Array>()
+                    .unwrap()
+                    .value(0),
+                3
+            );
+        }
+
         /// Regression: a short (≤ 12-byte) Utf8View string must hash identically
         /// regardless of which batch it appears in — all-inline or mixed.
         #[test]

From 6335771ae88f9bf32c25385b03a1d8c4229672de Mon Sep 17 00:00:00 2001
From: RIchard Baah <richiebbaah@Gmail.com>
Date: Thu, 27 Aug 2026 11:20:18 -0400
Subject: [PATCH 2/2] trim

---
 datafusion/functions-aggregate/src/approx_distinct.rs | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/datafusion/functions-aggregate/src/approx_distinct.rs b/datafusion/functions-aggregate/src/approx_distinct.rs
index 03e2dbdf923f3..318a9422b48cc 100644
--- a/datafusion/functions-aggregate/src/approx_distinct.rs
+++ b/datafusion/functions-aggregate/src/approx_distinct.rs
@@ -935,7 +935,8 @@ mod tests {
         use arrow::array::{
             AsArray, Decimal32Array, Decimal64Array, Decimal128Array, Decimal256Array,
             DictionaryArray, Int32Array, Int64Array, IntervalDayTimeArray,
-            IntervalMonthDayNanoArray, IntervalYearMonthArray, StringArray, StringViewArray,
+            IntervalMonthDayNanoArray, IntervalYearMonthArray, StringArray,
+            StringViewArray,
         };
         use arrow::datatypes::{Int32Type, IntervalDayTime, IntervalMonthDayNano, i256};
         use std::sync::Arc;
@@ -1252,7 +1253,7 @@ mod tests {
 
         #[test]
         fn dictionary_int32_utf8_per_row_and_groups_acc() {
-            // {"a", "b", "b", "c", "c", "c"} — 3 distinct logical values
+            // 3 distinct logical values
             let keys = Int32Array::from(vec![0, 1, 1, 2, 2, 2]);
             let values = StringArray::from(vec!["a", "b", "c"]);
             let array: ArrayRef = Arc::new(