From f13fc87137de38b6cb5870607c728787e0f22646 Mon Sep 17 00:00:00 2001
From: Andy Grove <agrove@apache.org>
Date: Tue, 26 May 2026 09:27:02 -0600
Subject: [PATCH] chore(audit): audit any_value and expand tests

---
 .../spark_expressions_support.md              |   3 +
 .../expressions/aggregate/any_value.sql       | 104 ++++++++++++++++++
 2 files changed, 107 insertions(+)
 create mode 100644 spark/src/test/resources/sql-tests/expressions/aggregate/any_value.sql

diff --git a/docs/source/contributor-guide/spark_expressions_support.md b/docs/source/contributor-guide/spark_expressions_support.md
index 65f941210a..962afc3d34 100644
--- a/docs/source/contributor-guide/spark_expressions_support.md
+++ b/docs/source/contributor-guide/spark_expressions_support.md
@@ -33,6 +33,9 @@
 
 - [x] any
 - [x] any_value
+  - Spark 3.4.3 (2026-05-26)
+  - Spark 3.5.8 (2026-05-26)
+  - Spark 4.0.1 (2026-05-26)
 - [ ] approx_count_distinct
 - [ ] approx_percentile
 - [ ] approx_top_k
diff --git a/spark/src/test/resources/sql-tests/expressions/aggregate/any_value.sql b/spark/src/test/resources/sql-tests/expressions/aggregate/any_value.sql
new file mode 100644
index 0000000000..421d93b08f
--- /dev/null
+++ b/spark/src/test/resources/sql-tests/expressions/aggregate/any_value.sql
@@ -0,0 +1,104 @@
+-- Licensed to the Apache Software Foundation (ASF) under one
+-- or more contributor license agreements.  See the NOTICE file
+-- distributed with this work for additional information
+-- regarding copyright ownership.  The ASF licenses this file
+-- to you under the Apache License, Version 2.0 (the
+-- "License"); you may not use this file except in compliance
+-- with the License.  You may obtain a copy of the License at
+--
+--   http://www.apache.org/licenses/LICENSE-2.0
+--
+-- Unless required by applicable law or agreed to in writing,
+-- software distributed under the License is distributed on an
+-- "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
+-- KIND, either express or implied.  See the License for the
+-- specific language governing permissions and limitations
+-- under the License.
+
+-- any_value is a RuntimeReplaceableAggregate that Spark's optimizer rewrites
+-- to first(expr, ignoreNulls) before physical planning, so these tests exercise
+-- the same Comet path as first_last.sql but pin the any_value name explicitly.
+
+-- ============================================================
+-- Setup: shared tables
+-- ============================================================
+
+statement
+CREATE TABLE test_any_value(i int, grp string) USING parquet
+
+statement
+INSERT INTO test_any_value VALUES (1, 'a'), (2, 'a'), (3, 'a'), (NULL, 'b'), (4, 'b')
+
+statement
+CREATE TABLE test_any_value_ignore_nulls(id int, val int, grp string) USING parquet
+
+statement
+INSERT INTO test_any_value_ignore_nulls VALUES
+  (1, NULL, 'a'),
+  (2, 10,   'a'),
+  (3, 20,   'a'),
+  (4, NULL, 'b'),
+  (5, 30,   'b'),
+  (6, NULL, 'b')
+
+statement
+CREATE TABLE test_any_value_all_nulls(val int, grp string) USING parquet
+
+statement
+INSERT INTO test_any_value_all_nulls VALUES (NULL, 'a'), (NULL, 'a'), (NULL, 'b'), (1, 'b')
+
+-- ============================================================
+-- any_value: basic (default behavior includes nulls)
+-- ============================================================
+
+query
+SELECT any_value(i) FROM test_any_value
+
+query
+SELECT grp, any_value(i) FROM test_any_value GROUP BY grp ORDER BY grp
+
+-- ============================================================
+-- any_value with isIgnoreNull literal
+-- ============================================================
+
+-- any_value(expr, true) ignores nulls.
+query
+SELECT any_value(val, true) FROM test_any_value_ignore_nulls
+
+query
+SELECT grp, any_value(val, true) FROM test_any_value_ignore_nulls GROUP BY grp ORDER BY grp
+
+-- any_value(expr, false) respects nulls (default).
+query
+SELECT grp, any_value(val, false) FROM test_any_value_ignore_nulls GROUP BY grp ORDER BY grp
+
+-- ============================================================
+-- any_value: all-null group with ignoreNulls
+-- ============================================================
+
+-- Group 'a' has all nulls so result is NULL even when ignoring nulls.
+query
+SELECT grp, any_value(val, true) FROM test_any_value_all_nulls GROUP BY grp ORDER BY grp
+
+-- ============================================================
+-- any_value alongside other aggregates
+-- ============================================================
+
+query
+SELECT grp,
+  any_value(val, true),
+  count(val),
+  sum(val)
+FROM test_any_value_ignore_nulls GROUP BY grp ORDER BY grp
+
+-- ============================================================
+-- any_value matches first when called identically
+-- ============================================================
+
+-- Same input expression and same ignoreNulls flag should yield identical results
+-- because Spark rewrites any_value(x, b) to first(x, b).
+query
+SELECT
+  any_value(val, true) = first(val, true) AS ignore_nulls_match,
+  any_value(val, false) = first(val, false) AS respect_nulls_match
+FROM test_any_value_ignore_nulls