feat: Support List type in approx_distinct

datafusion/functions-aggregate/src/approx_distinct.rs

@@ -800,6 +800,7 @@ impl AggregateUDFImpl for ApproxDistinct {
             | DataType::Binary
             | DataType::BinaryView
             | DataType::FixedSizeBinary(_)
+            | DataType::List(_)
             | DataType::LargeBinary => Box::new(HLLAccumulator::new()),
             DataType::Null => {
                 Box::new(NoopAccumulator::new(ScalarValue::UInt64(Some(0))))
@@ -871,6 +872,7 @@ fn is_hll_groups_type(data_type: &DataType) -> bool {
             | DataType::BinaryView
             | DataType::FixedSizeBinary(_)
             | DataType::LargeBinary
+            | DataType::List(_)
     )
 }
 

datafusion/sqllogictest/test_files/aggregate.slt

@@ -1969,6 +1969,43 @@ SELECT g, approx_distinct(arrow_cast(arrow_cast(s, 'Binary'), 'FixedSizeBinary(1
 4 1
 
 
+# List
+statement ok
+CREATE TABLE approx_distinct_list_test (g INT, l INT[]) AS VALUES
+  (1, [1, 2]), (1, [1, 2]), (1, [3, 4]),
+  (2, [5, 6]), (2, NULL),
+  (3, NULL), (3, NULL),
+  (4, [7, 8]);
+
+# List non-grouped
+query I
+SELECT approx_distinct(l) FROM approx_distinct_list_test WHERE g = 1;
+----
+2
+
+# List grouped
+# Group 1 -> {[1,2],[3,4]}=2,
+# Group 2 -> {[5,6]}=1 (NULL excluded),
+# Group 3 -> all null=0,
+# Group 4 -> {[7,8]}=1
+query II
+SELECT g, approx_distinct(l) FROM approx_distinct_list_test GROUP BY g ORDER BY g;
+----
+1 2
+2 1
+3 0
+4 1
+
+# The non-group path must agree with the grouped path on
+# the same data, and distinct lists across groups are still counted overall.
+query I
+SELECT approx_distinct(l) FROM approx_distinct_list_test;
+----
+4
+
+statement ok
+DROP TABLE approx_distinct_list_test;
+
 # Integers (Int32): group 1 -> {10,20}=2, group 2 -> {30,40}=2, group 3 -> 0, group 4 -> {50}=1
 query II
 SELECT g, approx_distinct(i) FROM approx_distinct_group_test GROUP BY g ORDER BY g;